Search

The automation degree of sealing technology in coal mine is insufficient, and the automatic directional drilling rig can improve the automation degree of sealing. A colloidal slip with double friction force for automatic directional drilling machine is designed to solve the problem that the clamping device of automatic directional drilling machine is bad. Two kinds of colloidal structures are proposed to realize the difference of friction between active drill pipe jointing and unjointing, and to ensure the smooth jointing of drill pipe. Using the nonlinear finite element method, Mooney-Rivlin strain energy function was used to describe the mechanical properties of colloid, and the simulation of colloid clamping hole sealer was carried out. The working friction of colloid of different materials was obtained, and the rationality of the two structural colloids was verified. The influence of the compression amount of adhesive strip on the surface of the colloid on the stress of the hole sealer was studied. The effect of the number of adhesive strips on the colloidal surface on the axial friction force was investigated. The results show that natural rubber and polyurethane rubber are reasonable as colloidal materials, the compression of rubber strips is 0.9 mm, and the number of rubber strips is 3.

Objective CTCS2+ATO (Chinese Train Control System Level 2 + Automatic Train Operation) system can meet the requirements for cross-line operation between city (suburban) railway and national mainline and intercity railways. However, it cannot satisfy the demand for the 3-minute headway public transport operation in city (suburban) railway. By optimizing functionality, refining resource management, and simplifying operational processes, the goal of a turn-back interval of less than 3 min within the CTCS framework system is proposed, so as enabling the CTCS2+ATO system to meet the demands of city (suburban) railways in both public-transit operation and interoperability with national railway lines. Method Building on the post-station turn-back technology plan under existing CTCS2+ATO system, this approach breaks away from the traditional route-based interval protection method inside station by implementing resource-based interval protection to shorten the time required for entry and exit from the turn-back siding. Additionally, an automatic procedure is introduced where the train, upon reaching the turn-back siding, automatically switches ends and activates into operational status, significantly reducing the end-changing time. The implementation details of this specific technical plan are expounded, including its application sequence design and applicable scenarios. Result & Conclusion The proposed turn-back method is suitable for intercity and city (suburban) railways designed according to the standards outlined in the Code for Design of Suburban Railway and Code for Design of Intercity Railway. By adopting this method, it is possible to safely meet the operational demand for a 3-minute headway in city (suburban) railway public- transit operations while also ensuring interoperability between this line and the national trains. Train following intervals are reduced, turn-back efficiency is enhanced, and trains are prevented from inadvertently entering non-turnback tracks during unmanned operations.

Information acquisition is the prerequisite for command decision-making. In the era of big data filled with massive information, the useful information is submerged in useless information. How to efficiently obtain the useful information needed by command personnel is the key to improve operational command effectiveness. The author uses the information automatic recommendation algorithm, on the basis of the information requirements proposed by the commander, extracts the useful information and automatically recommends to commander based on the text similarity. It can effectively improve the timeliness of information acquisition and command decision-making, and thus improve the effectiveness of combat command.

Objective The detection of foreign objects in platform door gap is critical to metro operational safety. Therefore, it is essential to develop a new anti-clamping detection system for metro platform door, enhancing the safety and efficiency of future FAO (fully automatic operation) systems. Method Based on the video and LiDAR algorithm fusion technology, a dual-criterion AI detection strategy that combines video image recognition with LiDAR point cloud data is proposed. PointNet algorithm framework is innovatively adopted for the detection of foreign objects in metro platform door gap, implementing a camera video assisted LiDAR working mode. In the event of foreign object detection in door gap, the system triggers an alarm-video synergistic operation and initiates video capture of the incident site immediately. The use of multi-dimensional deep learning techniques reduces the probability of false alarms. Result & Conclusion In system design, a cross-stacking layered sensor installation method is proposed, enabling the redundant detection function of foreign objects in platform door gaps. The cross-verification mechanism significantly enhances the redundancy and reliability of the detection device, and using 2D sensors to achieve 3D detection effects. The developed system provides safety interlocking signals to metro signaling system, sends alarm information to the integrated monitoring system, pushing wristband alerts to on-site operation personnel. This system ensures more accurate and reliable detection of foreign objects in platform door gaps, offering safety support for FAO.

Objective There are no train crew on the train that runs on the fully automatic operation (FAO) line of urban rail transit, leading to very big safety risks in the evacuation scenario. Therefore, it is necessary to study the automatic protection scheme for section evacuation on FAO lines. Method Based on the application of automatic protection in section evacuation on FAO lines （hereinafter referred to as the automatic protection）, the necessity of the automatic protection is analyzed. Three schemes for the automatic protection are proposed, and comparison analysis is conducted on the schemes. Scheme 3 is recommended. Further research is done on the automatic protection measures for section evacuation in Scheme 3. Result & Conclusion Scheme 3 additionally includes the flexible handling measures for signal system and OCC (operation control center) dispatchers, making the protection handling measures in the section evacuation more flexible and much safer.

【Objective】 Crops are vulnerable to drought and waterlogging at their seedling stages. Experience-based irrigation and drainage is not effective to alleviate water stresses that seedlings could face. This paper presents an automatic drainage and irrigation system we developed for seedlings of direct-seeded rice. 【Method】 The system includes a micro-sprinkler irrigation system and an automatic drainage device. They are controlled by a LoRa wireless communication network; the irrigation - drainage scheduling was calculated by the fuzzy control algorithm, based on the dynamics of soil moisture and growth of the seedlings measured in the field. The applicability of the system in the fields was evaluated using performance indicators and drainage device performance indicators. 【Result】 ① In the terms of performance, the delay in the response of the system was 0.398 s, the decision-making time was 1.348 s, and the errors of the information monitoring was ±1.48%. ② The automatic drainage device took 4-6 s to close or open, and the standby electric current was 21.7 mA. ③ Experimental results showed that the systems can control soil moisture at 85% of the field capacity 15 days after the seed-drilling; the survival rate of the seedlings was 83.9%. 【Conclusion】 The automatic irrigation and drainage system for seedlings of direct-drilled rice is energy-efficient. It can maintain soil moisture in the proximity of the soil surface around the designed value, alleviating water stress and improving the survival of the seedlings of the directly drilled rice.

The construction of automatic meteorological observation stations in China has been continuously improved.Currently， more than 60， 000 automatic meteorological observation stations have been built， providing abundant information of surface meteorological variables for weather and climate research.However， the practical application of ground automatic station data has always been constrained by high uncertainty in the quality of observation data.Strict quality control is a prerequisite for the effective application of automatic station data， but the high spatiotemporal resolution characteristics of automatic station observations bring more difficulties to quality control researches.How to accurately distinguish local small-scale weather information and local variation caused by erroneous data in high-resolution automatic station data has always been a difficult point in the research of quality control methods for spatiotemporal resolution automatic station data.On the basis of analyzing the spatial correlation scale and error characteristics of surface temperature， this study established a quality control method for temperatures from surface automatic station based on EOF （Empirical Orthogonal Function） analysis method， which only relies on observation data.The study conducted quality control experiments using surface automatic station temperature observations from January to May 2022， and compared the differences in surface temperature between the automatic station observation data and the Chinese reanalysis data CRA40 （CMA's global atmospheric Re-Analysis） before and after quality control.The results indicate that the established autonomous quality control method for observation data can effectively identify erroneous observation data， relying solely on the observation data itself， effectively avoiding the impact of background errors on quality control effectiveness.The quality control sub regions determined on the basis of correlation scale analysis further enhance the quality control method's ability to identify small-scale temperature changes in observation data， effectively preserving the reject of temperature extremum data corresponding to extreme events in small areas， the number of quality control modes determined by actual data characteristics can well separate the principal and residual terms of the observed data， significantly improving the accuracy of erroneous extreme value recognition.Further introducing sliding detection methods and overlap rejection standards can also retain as much valuable observation data as possible in areas with steep terrain.The quality control results of 1 month data show that the new quality control method can obviously and stably improve the spatial correlation coefficient between the surface temperature of automatic station data and the corresponding variable of CRA40 （CMA's global atmospheric Reanalysis） reanalysis data， and the average deviation is also reduced.Although the average data rejection rate is only about 8%， the spatial correlation coefficient can reach a maximum increase of about 0.02， which fully proves that proposed quality control method can effectively eliminate erroneous data and improve the spatial continuity of automatic station data.

Robotic grasping has extensive applications in fields such as logistics sorting, automated assembly, and medical surgery. Grasping detection is an important step in robotic grasping. Recently, with the decrease in their costs, depth cameras have been gradually applied for grasping detection, which has promoted the application of pose estimation-based methods for robotic grasping. However, most publicly available RGB-D image-based pose estimation datasets rely on equipment such as expensive 3D laser scanners to obtain 3D models of target objects. Meanwhile, the annotation process relies heavily on manual operation, which is time-consuming, labor-intensive, and unfavorable for the creation of large-scale datasets. To address these issues, this study implements a dataset automatic acquisition and annotation system aimed at developing RGB-D image-based pose estimation methods for robotic grasping. The proposed system deploys easily and does not require an expensive 3D laser scanner. RGB-D image sequences are obtained only by an off-the-shelf depth camera, and the system can automatically acquire the reconstructed 3D model of the target object, annotated pose information, and 2D image segmentation masks. During the process of developing the automatic annotation algorithm for the proposed system, a novel minimum spanning tree-based normal propagation method is proposed to guarantee that consistent normal directions can be acquired so that deformations or tearing on the reconstructed 3D surface caused by inconsistent normal directions can be avoided. During the experiments, the proposed system created a pose estimation dataset containing 84 objects with 8400 RGB-D images. 3D models, image segmentation masks, and 6D poses were annotated by the system in every RGB-D image for each object. To evaluate the accuracy of the annotated segmentation masks, the annotated segmentation masks and the corresponding manually labeled results were compared. Furthermore, the accuracy of the annotation results was also assessed from the performance of an instance segmentation network trained by the annotated image masks. To evaluate the accuracy of the annotated poses, a point cloud registration mission was launched to align the model point cloud and the scene point cloud using the annotated pose parameters. Furthermore, a category-level pose estimation network was trained using the annotated pose parameters, and its performance can directly reflect the accuracy of the annotation results. The experimental results show that the overlapped area between the annotated mask and the manually labeled mask is greater than 98%. Additionally, a 100% alignment rate can be achieved, meaning that the model point cloud can be aligned to any scene point cloud through the corresponding annotated pose parameters. These results demonstrate that the designed and implemented system in this paper can be used to sufficiently create a high-quality dataset for developing real pose estimation-related solutions. A solid data foundation can be provided on the basis of the proposed system for future research and application of deep learning models aimed at robotic grasping detection.

Achieving the best wideband stealth performance under the limited weight constraint is the goal pursued in stealth aircraft radar absorbing coating（RAC） application schemes design. To meet the need for the integrated application of various kinds of RACs on stealth aircraft， an automatic optimization method for the application programme of RAC was established. Firstly， the electromagnetic fields of the model under different frequencies， azimuths and polarizations were simulated based on the finite difference time domain method. The total electromagnetic field was obtained by normalization and weighted calculation of the electromagnetic fields.Then， the total electromagnetic field was divided according to the specific direction and quantity. Each partition was loaded with the corresponding RAC. The reflection coefficient of the RAC was calculated by the transfer matrix method to superimpose the effect of the RAC on the electromagnetic field. The near-far field transformation was performed on the total electromagnetic field after loading with RAC. In this way， fast computation of radar cross section （RCS） was implemented. Based on the improved simplex method and Karush-Kuhn-Tucker condition， an automatic optimization method is established. The objective function was reducing RCS of the target or reducing weight of the absorbing coating. Finally， taking traveling wave plate as an example， the application plan of three thicknesses of RACs was optimized. The goal of optimization was to reduce total weight of the coating. Compared with the 1.5 mm coating plan， the optimization program can reduce the weight by 50%， with the RCS of traveling wave plate increasing by on more than 1 dB. The optimization effect is remarkable.

叶片过滤机是精炼车间脱色工段中非常关键的设备，设备前端后续的工艺阀门很多，有很多操作步骤和时间节点控制阀门的开启和关闭，操作复杂烦琐。为简化人工手动操作的烦琐性及为更多单机系统控制提供技术参考，以立式叶片过滤机为例，采用PLC和触摸屏的方式构建自动控制系统，控制叶片过滤机的单机运行和切换控制。自动控制系统的CPU采用西门子S7-1214C DC/DC/Rly，HMI采用西门子KTP1200触摸屏，并采用西门子TIA PORTAL V18软件进行PLC硬件组态及编程和上位机KTP1200的监控和控制界面的制作，自动控制系统的设计包括两台过滤机的单独控制和切换控制，可根据不同的产量设定切换的时间参数。此系统运行稳定可靠，是独立的控制系统，可以单独应用于生产，还具有以太网通信功能，可以实现与其他控制系统的数据交换。The vane filter is a very critical equipment in the decolorization section of the refining workshop, there are many process valves at the front end and follow-up of the equipment, there are many operation steps and time nodes to control the opening and closing of the valve, and the operation is complex and cumbersome. In order to simplify the tediousness of manual operation and provide technical reference for the control of more stand-alone systems, taking the vertical vane filter as an example, PLC and touch screen were used to build an automatic control system to control the stand-alone operation and switching control of the vane filter. The CPU of automatic control system adopted Siemens 1200 series S7-1214C DC/DC/Rly, HMI adopted Siemens lean series touch screen KTP1200, and Siemens TIA PORTAL V18 software was used for hardware configuration and programming, and the production of monitoring and control interface of host computer KTP1200. The design of the automatic control system includes separate control and switching control of the two sets of filters, and the time parameter of switching can be set according to different outputs. This system operates stably and reliably, and is an independent control system, which can be applied to production alone, also the system has Ethernet communication function, therefore can realize data exchange with other control systems.

【Objective】In the production test process of 50 Gbit/s Small Form Pluggable (SFP) 56 optical module, the number of instruments and equipment to be matched for the test is large. Therefore, the test instruments are used in isolation, and the test process is complex. These issues make the optical module test longer, and the test efficiency is relatively low. It brings a lot of inconvenience to the test personnel, and can not guarantee the consistency of the test products. The purpose of this paper is to enhance debugging and testing efficiency while alleviating testers' workload.【Methods】Python language and its robust third-party libraries, including the VISA and pyserial libraries, are utilized to develop an automatic testing system for SFP56 optical modules compatible with the Windows operating system.【Results】The system can debug and test the optical module automatically. The test instruments can be used in parallel, reducing the waste of time and resources. According to the test requirements, the optical module can also be tested step by step, which greatly improves the efficiency of the debugging and testing of the optical module in the production line. It can also reduce the requirements of the tester. The verification shows that compared with the traditional optical module test, the test time of each optical module is saved by about 70 s, which greatly improves the test efficiency of the optical module. In the optical module test, the result deviation caused by human operation is avoided, and the product consistency can be guaranteed.【Conclusion】In comparison to Labview-based automatic optical module testing systems, which are currently prevalent in the market, the system designed in this study is more scalable and convenient for upgrades and maintenance. Moreover, it can be expanded to support optical module testing at speeds of 100, 200, 400, and 800 Gbit/s, thus laying a technical foundation for future high-speed optical module testing.

To comprehensively meet the test requirements for the common mode rejection ratio (CMRR) across different ECG-particular standards, this paper presents the design of an ECG CMRR automatic test system. The hardware component primarily consists of a test signal generation module, an automatic control network (which includes a resistance-capacitance network control module and a polarization voltage control module), and a noise level switching module. The software portion enables automatic control and user interaction. Experimental results indicate that the system is stable, reliable, and highly automated, capable of satisfying the test requirements of various ECG-particular standards, thus demonstrating a promising application prospect.

ObjectiveManual disinfection in large-scale sheep farm is laborious, time-consuming, and often results in incomplete coverage and inadequate disinfection. With the rapid development of the application of artificial intelligence and automation technology, the automatic navigation and spraying robot for livestock and poultry breeding, has become a research hotspot. To maintain shed hygiene and ensure sheep health, an automatic navigation and spraying robot was proposed for sheep sheds.MethodsThe automatic navigation and spraying robot was designed with a focus on three aspects: hardware, semantic segmentation model, and control algorithm. In terms of hardware, it consisted of a tracked chassis, cameras, and a collapsible spraying device. For the semantic segmentation model, enhancements were made to the lightweight semantic segmentation model ENet, including the addition of residual structures to prevent network degradation and the incorporation of a squeeze-and-excitation network (SENet) attention mechanism in the initialization module. This helped to capture global features when feature map resolution was high, addressing precision issues. The original 6-layer ENet network was reduced to 5 layers to balance the encoder and decoder. Drawing inspiration from dilated spatial pyramid pooling, a context convolution module (CCM) was introduced to improve scene understanding. A criss-cross attention (CCA) mechanism was adapted to acquire context global features of different scales without cascading, reducing information loss. This led to the development of a double attention enet (DAENet) semantic segmentation model was proposed to achieve real-time and accurate segmentation of sheep shed surfaces. Regarding control algorithms, a method was devised to address the robot's difficulty in controlling its direction at junctions. Lane recognition and lane center point identification algorithms were proposed to identify and mark navigation points during the robot's movement outside the sheep shed by simulating real roads. Two cameras were employed, and a camera switching algorithm was developed to enable seamless switching between them while also controlling the spraying device. Additionally, a novel offset and velocity calculation algorithm was proposed to control the speeds of the robot's left and right tracks, enabling control over the robot's movement, stopping, and turning.Results and DiscussionsThe DAENet model achieved a mean intersection over union (mIoU) of 0.945 3 in image segmentation tasks, meeting the required segmentation accuracy. During testing of the camera switching algorithm, it was observed that the time taken for the complete transition from camera to spraying device action does not exceed 15 seconds when road conditions changed. Testing of the center point and offset calculation algorithm revealed that, when processing multiple frames of video streams, the algorithm averages 0.04 to 0.055 per frame, achieving frame rates of 20 to 24 frames per second, meeting real-time operational requirements. In field experiments conducted in sheep farm, the robot successfully completed automatic navigation and spraying tasks in two sheds without colliding with roadside troughs. The deviation from the road and lane centerlines did not exceed 0.3 meters. Operating at a travel speed of 0.2 m/s, the liquid in the medicine tank was adequate to complete the spraying tasks for two sheds. Additionally, the time taken for the complete transition from camera to spraying device action did not exceed 15 when road conditions changed. The robot maintained an average frame rate of 22.4 frames per second during operation, meeting the experimental requirements for accurate and real-time information processing. Observation indicated that the spraying coverage rate of the robot exceeds 90%, meeting the experimental coverage requirements.ConclusionsThe proposed automatic navigation and spraying robot, based on the DAENet semantic segmentation model and center point recognition algorithm, combined with hardware design and control algorithms, achieves comprehensive disinfection within sheep sheds while ensuring safety and real-time operation.

It is an effective hole protection measure to ensure the effect of gas drainage in the mining process of the working face by using the screen pipe to complete the borehole after the directional long hole drilling in the coal seam roof. In order to study the feasibility of using the large-diameter drill pipe re-entered into the drill hole and the hydraulic transportation of the grouping screen pipe automatic docking in the drill pipe to complete the full-hole section screen pipe protection, the hydraulic transportation and automatic docking simulation tests of the grouping screen pipe in the drill pipe were carried out using the designed hydraulic transportation screen pipe testing device and the grouping screen pipe docking device, and the hydraulic transportation differential throttling pressure and transportation force of the grouping screen pipe under different pump volumes were obtained, and compared with the theoretical calculation results. At the same time, the docking reliability of the grouping screen pipe docking device and the automatic docking success rate of the grouping screen pipe are analyzed and tested. The test results show that the measured values of differential throttling pressure and hydraulic transportation force are close to the theoretical values and show an exponential growth trend under different pump volumes. When the pump volume is 390 L/min, the measured differential throttling pressure and hydraulic transportation force of the grouping screen pipe with ϕ82 mm throttling booster pipe in the ϕ102/ϕ83 mm large-diameter drill pipe reach 1.29 MPa and 6 860 N respectively, and the single docking success rate of the grouping screen pipe docking device and the automatic docking success rate of the grouping screen pipe are not less than 94.18% and 82.23% respectively.

Considering the lack of a general multi-carrier signal dataset in urban multipath channels, and the challenge of recognizing the modulation types of distorted signals at low signal-to-noise ratio (SNR), a differentiable architecture search-based (DARTS) automatic modulation recognition algorithm for multi-carrier signals was proposed. Firstly, the received signal datasets of commonly used multi-carrier signals, i.e., orthogonal frequency division multiplexing, filter bank multi-carrier, and orthogonal time frequency space, were generated over typical urban multipath channels. The time-frequency images, which were insensitive to modulation parameters, were selected as feature vectors to train the neural network. Secondly, DARTS was employed to automatically search the optimal network architecture. Finally, a joint attention mechanism was introduced into the feature learning process. This mechanism spatially transforming distorted signal features to mitigate the impact of multipath interference, while also calculating and sorting the information weights for each channel of the feature maps to improve the recognition performance of the relevant feature map channels. Simulation results demonstrate that the proposed algorithm improves accuracy in urban multipath channels, especially at low SNR, while simultaneously providing better robustness to modulation parameter variations and small-sample scenarios.

The traditional filter weighing method has a cumbersome operation process, a long detection cycle, and a low degree of automation. Although it has high accuracy, it cannot meet the needs of real-time detection of dust concentration. An automatic dust mass concentration detection device was designed, and the temperature and humidity compensation model was established to replace the steps of filter membrane drying after sampling in the manual weighing method, so as to further reduce the dust concentration detection time and the volume of the dust concentration detection device. The prototype device was built and the experiment was conducted. The test results showed that the standard deviation of the CCZ-20A dust sampler commonly used in coal mines and the converted concentration of dust detected by this device was within 5%, and the fitting correlation of the experimental data under the univariate linear regression fitting analysis was good. In order to further improve the detection accuracy of the device, an error compensation method based on the combination of Fourier series, linear fitting and periodic fitting was studied, and the concentration calculation process of the device was established and brought into the original data, and the detection concentration error of the device was concentrated from the original (−7.20%, −1.26%) to (−3.64%, 3.65%). After the introduction of the device concentration calculation process, several comparative experiments were carried out, and the experimental results showed that the detection errors of the device concentration were within this range, which verified the reliability of the device concentration calculation process. The device shortens the time required for dust concentration detection, and at the same time controls the detection error within a reasonable range, which provides a reference for the research of filter membrane weighing method in the online monitoring of dust concentration.

The traditional net-laying operation is highly dependent on manual labor, which not only has high labor intensity and high safety risks, but also has low efficiency, which has become an important factor restricting the balance of mining and excavation. Therefore, the core objective of this study is to design and implement a standardized mesh-laying strategy for underground coal mine robots, which improves the efficiency of mesh-laying operation, reduces the labor load of workers, enhances the safety of operation through the introduction of automation and intelligent technology, and then promotes the efficient, green and sustainable development of coal mine production. In order to realize the above research objectives, this study adopts a combination of simulation and experimental methods, based on the special environment and operating characteristics of underground, and designs a remote-controlled manipulator structure that assists the net-laying operation of digging-anchor integrated machine. The structure adopts a seven-axis articulated robotic arm with a top gripper in order to realize the multi-angle and high-precision laying of anchor nets. In addition, advanced computer simulation technology is used to analyze the dynamic performance, motion trajectory planning and force characteristics of the robotic arm, and the structural design and control algorithm of the robotic arm are optimized iteratively to ensure its stability and reliability in the complex underground environment. At the same time, a set of adaptive adjustment strategy is developed for the motion control of the robotic arm, so that the robotic arm can automatically adjust the motion parameters according to the actual working conditions (tunnel shape, mesh material), and realize accurate and efficient net-laying operation. The results show that the automated net-laying equipment significantly improves the efficiency of net-laying operation, compared with the traditional manual net-laying method, the efficiency is increased by more than 20%. Through automated operation, it realizes the goal of reducing the number of workers by more than 33%, which effectively alleviates the problem of tense human resources in underground coal mines. The labor intensity of workers has been greatly reduced, reducing the high degree of climbing work, reducing the risk of work-related injuries, and the labor intensity has been reduced by more than 80%. At the same time, the automated net-laying process reduces the possibility of human error and significantly improves operational safety. In the future, with the continuous iteration and improvement of the technology, this technology is expected to play a more important role in the intelligent construction of coal mines, and promote the development of coal mine production in the direction of safer, greener and more efficient.

According to the engineering geological characteristics and production technology level of Qianyingzi Coal Mine, the stability control technology of surrounding rock of the cutting roof and pressure relief roadway is proposed. For the advanced section, lagging section and stable section of the mining roadway, technical measures of advanced reinforcement support, side wall protection and lagging temporary support were proposed. Through the real-time monitoring system, the stress of anchor cable and surrounding rock deformation of goaf roadway are monitored, the deformation characteristics of roadway and stope surrounding rock are studied, and the law of mine pressure development in self-formed roadway without coal pillar is revealed by the comparison and analysis with the existing goaf roadway. The results show that the deformation rate of surrounding rock reaches the maximum at the position of 170 m away from the working face, and the deformation of surrounding rock is the most intense at this stage, and then tends to be gentle, and the deformation of surrounding rock reaches the stable stage when the working face lags 220 m. The application effect of the pre-cracking blasting cutting roof for gob-side entry retaining in the W3233 working face of Qianyingzi Coal Mine has reached the expected goal.

At present, the locomotive automatic driving device has been applied to multiple lines and has achieved certain phased results. Due to factors such as complex application scenarios, strong non-linearity and discreteness of control objects, some problems and challenges have arisen during the operation assessment, including impulsive and surging phenomena during train interval operation, non-compliance with station benchmark parking accuracy requirements, failure to start on steep slopes, and difficulty in evaluating air braking performance. These problems seriously affect the smoothness of operation and driving safety. The article studies the automatic driving control technology for key scenarios. Through in-depth analysis of operating procedures and historical operating data, optimization of train dynamics model parameters, establishment of accurate basic resistance models and air braking models, design of expert knowledge base for smooth operation strategy, intelligent parking control strategy, air braking terminal predictive control method, model and data-driven safety evaluation method, the problems encountered in on-site operation are ultimately solved. By the proposed strategies, it achieves the goal of 100% fully automatic control for the main line operation of 20 000 ton combined trains, providing a technical foundation for the normalized application of locomotive automatic driving.

In order to solve the problems of spout hole, collapse hole, plugging hole and buried drilling in gas extraction drilling operation in broken soft coal seam, a type of ZDY4500LFK automatic drilling machine suitable for broken soft coal seam is developed. This paper focuses on the difficulties in the automatic construction of broken soft coal seam, introduces the mechanical structure and control principle of the drill, and studies and analyzes the technical principles of adaptive drilling, automatic positioning and correction of drill pipe, cooperative drill pipe loading and unloading path planning in broken soft coal seam. The field test verifies the advantages of the drill such as high porosity, high per capita efficiency, high safety and high rod carrying capacity, which provides technical support for the safe and efficient production of soft crushed coal seam.

To process the high-density and high-frequency mass data generated by ground automatic weather stations, a real-time stream processing system based on Storm is designed and implemented in the Meteorological Big Data Cloud Platform (Tianqing). It leverages the advantages of large-scale parallel computing to enhance processing speed. For BUFR messages, a Storm topology is designed to process the standardized BUFR format data transmitted by RabbitMQ directly on the service, reducing the intermediate steps from transmission to processing of observations. In the spout design, the manual confirmation mode of RabbitMQ messages is adopted to ensure that each message is effectively processed. In the decoding process, bolt is anchored to the spout using message identification (ID) to ensure reliable processing of each message. Format and time checks are performed during data decoding to filter out abnormal data. A batch timing monitoring strategy is applied to address the issue of data ingestion loss caused by port occupancy during extensive monitoring data transmission. A startup strategy with a configurable number of spout and bolt is designed for quick optimization and adjustment based on system resources. During cluster deployment, some resources are reserved to enable automatic task migration without disrupting business operations in case of node corruption within the cluster. System design involves automatically reconnecting message queues and databases to enhance system stability and enable self-healing capabilities. Application results show that the service efficiency of 2442 national stations has decreased from 175 s with CIMISS to 78 s with Tianqing. Additionally, the service efficiency of hourly data from over 60000 regional stations has decreased from 5 min with CIMISS to 2 min with Tianqing. After switching the data source of the ART (analysis of real time) system to Tianqing, the number of stations that can be retrieved simultaneously is doubled compared to CIMISS. It can effectively improve the quality of ART live products while keeping other conditions unchanged. By implementing specialized stream processing, it can effectively handle various business scenarios where data access process of the provincial Tianqing ground automatic weather stations differ from that of other provinces. It enables the provincial Tianqing to quickly process nationwide data from ground automatic weather stations. In December 2021, Storm-based stream processing is implemented in the national and provincial meteorological information departments alongside Tianqing. It has been running smoothly over two years, delivering reliable ground automatic weather station data to users, including MICAPS4, SWAN2.0, ART systems and others.

The conventional reservoir automatic history matching method requires multiple computationally time-consuming reservoir numerical simulations. Deep learning agent models can perform alternative reservoir numerical simulation calculations with approximate accuracy and greater computational efficiency. In the reservoir automatic history matching method based on the deep learning agent model， the reservoir uncertainty parameters adjusted by the reservoir automatic history matching method are usually used as the input parameters of the deep learning agent model. Existing deep learning agent models are often single input-output neural network model architectures. They do not consider that reservoir automatic history matching methods require the adjustment of multiple reservoir uncertainty parameters. Multiple deep learning agent models need to be trained to predict water saturation field distribution and pressure field distribution in reservoirs. For solving this problem， a reservoir automatic history matching method based on a dual input-output convolutional neural network agent model was proposed to simultaneously predict the water saturation field distribution and pressure field distribution in a reservoir by using a dual input-output convolutional neural network， with the reservoir permeability field distribution and phase permeability parameters as input. The production was calculated with the help of the Peaceman equation. It was coupled to the ensemble smoother with multiple data assimilation （ES-MDA） methods to invert the reservoir permeability field distribution and phase permeability parameters to achieve a more efficient reservoir automatic history matching solution. The results of the study show that the prediction accuracy of the reservoir water saturation field distribution and pressure field distribution is above 93% at the specified time step based on the dual input-output convolutional neural network agent model. Compared with the traditional reservoir automatic history matching method， the proposed reservoir automatic history matching method based on a dual input-output convolutional neural network agent model avoids the time-consuming computation of multiple calls to the reservoir numerical simulator and improves the efficiency of the matching.

In response to the problems of current electromagnetic compatibility testing methods for elevator control cabinets, including low intelligence, cumbersome operation, and lack of elevator system status monitoring, this paper proposes an electromagnetic compatibility testing method for elevator control cabinets and designs a set of electromagnetic compatibility testing and control system for elevator control cabinets. The author constructed an "upper computer+ lower computer" control system architecture, developed an embedded controller based on ARM, designed application software for upper and lower computers of the automatic testing process and control system, implemented such functions as sensor signal acquisition and CAN bus communication for elevator control cabinets, and realized the real-time monitoring of electromagnetic compatibility testing of elevator control cabinets and elevator system status. Then, the author used this system to carry out the conducted immunity, voltage dip, voltage interruption, electrical fast transient pulse group and surge tests on elevator control cabinets, and conducted simulation tests on elevator control cabinets for sensor signal abnormalities, power supply failures, and communication anomalies. The results show that the electromagnetic compatibility testing and control system of the elevator control cabinet operates stably, with accurate data collection and simple operation, accurately and efficiently identifies various types of faults, meets the needs of elevator manufacturers and testing institutions for electromagnetic compatibility testing, and compared with previous testing methods, reduces testing manpower input by about 67%.

Sampling and recognizing the wind pressure performance curve from the wind performance curve image, and then fitting the wind pressure performance function, is a key technology for solving the mine ventilation network. Currently, manual methods are commonly used to recognize wind pressure performance curves, which have low efficiency and poor accuracy. This study proposes an automatic recognition method for the wind pressure performance curve of ventilator based on image processing technology. The method uses bilateral filtering, image sharpening, and binarization techniques to preprocess the original ventilator wind pressure performance curve image, in order to improve image quality. The method extracts grid lines and coordinate text from the performance curve image of the ventilator based on corrosion algorithm and contour detection algorithm. The method uses logical operation, median filtering, contour detection, and K3M algorithm to extract the wind pressure performance curve. The pixel coordinates of the wind pressure performance curve are recognized using a row by row pixel recognition method. The method uses template matching algorithm to recognize coordinate numbers, and then complete the conversion from pixel coordinates to physical coordinates, achieving wind pressure performance curve recognition. The automatic recognition method for the wind pressure performance curve of the ventilator is integrated into the ventilation network calculation software. The recognition experiment is conducted on the wind pressure performance curve of the ventilator. The results show that the sampling speed of the method for a wind pressure performance curve is 24 Samples/s. The recognized wind pressure performance curve has a high overlap with the original curve. The maximum error between the wind pressure fitting value and the original value is only 0.88%. Compared to manual recognition methods, the method greatly improves the efficiency and accuracy of the ventilation network calculation .

Objective: To solve the problems of low level of automation and high labor intensity of workers in the fermented grains discharging process of solid-state fermentation in the ground-pot. Methods: An automatic fermented grains scooping strategy for the compound robot for fermented grains scooping was proposed. Based on the pre-processing of the ground-pot image and the reconstruction of the point cloud inside the inner edge, the point cloud segmentation algorithm was used to extract the point cloud of fermented grains. The scooping plane was constructed. The safe working distance for scooping was solved. Preset scooping points were set. Afterwards, the scooping amount of the preset scooping point was evaluated by the pixel surface element voxel algorithm, to select the optimal scooping point. Results: The method of planning the scooping point based on the safe working distance is reasonable, which ensures that the bucket does not collide with the ground-pot wall while scooping fermented grains from the wall at different depths of the scooping plane. The strategy can effectively reduce the number of robot scooping, thereby improving the efficiency of fermented grains scooping. Conclusion: The strategy can effectively guide the compound robot to realize the automatic and intelligent fermented grains discharging.

BackgroundThe prevalence of depressive disorder in adolescents is on the rise. There have been studies on the pairwise relations between dysfunctional attitude， negative automatic thoughts， positive coping style and depressive symptoms in the past. However， the impact of the intrinsic relations among dysfunctional attitude， negative automatic thoughts and positive coping style on depressive symptoms is still unclear.ObjectiveTo explore the influence of dysfunctional attitude on adolescent depressive symptoms and examine the action path of negative automatic thoughts and positive coping style on it， in order to provide references for intervention for adolescent patients with depressisve disorder.MethodsThis study involved 162 adolescent patients with depressive disorder， who met the diagnostic criteria for depressive episodes in the International Classification of Diseases， 10th edition （ICD-10） and received treatment in Taiyuan Psychiatric Hospital from October 1， 2022 to October 31， 2023. These patients were evaluated using Self-rating Depression Scale （SDS）， Dysfunction Attitude Scale （DAS）， Positive Coping Style Subscale in Simplified Coping Style Questionnaire（SCSQ） and Automatic Thoughts Questionnaire （ATQ）. Pearson correlation analysis was adopted to examine the correlation among scores of scales above. Model 6 in Process 3.4.1 was adopted to test the acting path of negative automatic thoughts and positive coping style between dysfunctional attitude and adolescent depression symptoms.ResultsA total of 148 adolescent patients with depressive disorder completed an effective questionnaire survey， with a response rate of 91.36%. The direct effect value of dysfunctional attitude on depressive symptoms was 0.423 and the effect size was 63.32%. Negative automatic thoughts and positive coping style affected as acting path between dysfunctional attitude and depressive symptoms， with effect values of 0.156 （accounting for 23.35% of the total effect） and 0.045 （accounting for 6.74% of the total effect） respectively. Meanwhile， negative automatic thoughts and positive coping style affected as a chain reaction pathway between dysfunctional attitude and depressive symptoms， with an effect value of 0.044， accounting for 6.59% of the total effect.ConclusionDysfunctional attitude can not only directly affect the depressive symptoms of adolescent patients with depressive disorder， but also affect the depressive symptoms of adolescent patients with depressive disorder through the independent path or chain path of negative automatic thoughts and positive coping style.

Objective The existing target speed planning algorithm for the automatic driving of urban rail transit train cannot handle in real time the temporary change of the line speed limit in case of emergency due to the large computational load and long computation time. With regard to the problem, a real-time target speed planning algorithm based on global optimization is proposed to generate the speed planning curve in real time. Method Based on the conditions of train current position, current speed and speed limit of the forward line, the speed planning curve is firstly generated in the shortest time through point by point calculation. Then the traction and braking levels are being adjusted to keep the train running at uniform acceleration or deceleration, optimizing the comfort index of the train running. Next, the cruising speed in the maximum speed limit section is adjusted to reduce unnecessary traction and braking time, optimizing the energy consumption index of the train running. Finally, the train speed planning curve is output. Result & Conclusion The simulation results show that the real-time speed planning curve generated by the proposed algorithm satisfies the basic constraints of safety, punctuality and accurate parking. Compared with the traditional algorithm, it improves the comfort level and reduces the energy consumption during train operation. Meanwhile, the proposed algorithm can effectively handle the temporary change of the line speed limit in emergency and optimize several operation indicators.

The Shenmu-Shuozhou Railway of Shenhua Baoshen Railway Group Co., Ltd. operates 10,000t heavy-haul trains using centralized traction on intricate railway conditions. The application of the automatic train operation (ATO) mode has accentuated train impulse issues arising from frequent traction adjustments during train operation, underscoring the need for enhanced stability of ATO. This paper presents an optimization control algorithm based on a dynamic evaluation model of train states, tailored to the specific conditions and operational requirements of the Shenmu-Shuozhou Railway. This algorithm features a multi-particle model and a train state evaluation model for heavy-haul trains, evaluating the development trend of train states while incorporating constraints related to the loading/unloading slope in levels. The proposed algorithm was applied in an experiment on the Shenmu-Shuozhou Railway. During the departing stage, the optimization algorithm effectively improved operational stability on steep grades, and increased the departing speed by about 11.1%. During the cruising stage, train impulses was effectively suppressed pre and post neutral-zone crossings, and the train running speed increased by about 0.9%. Moreover, the trend of simulated coupling force changes aligned closely with actual traction loading, and the maximum locomotive-wagon coupling force was reduced by about 26.3%. During the parking stage, the maximum amplitude of train acceleration decreased by 71.4%, and the average impulse count decreased by 60%. The experimental results demonstrate the effectiveness of the proposed algorithm in improving the ATO stability for 10,000t heavy-haul trains operating under centralized traction.

The sodium-cooled fast reactor uses liquid metal sodium as the coolant, and the entire external and internal refueling operations are carried out in a fully enclosed environment, and the operation process of the components is not visible. To ensure the correctness of the sodium-cooled fast reactor components, the design converted the component number into a two-dimensional DataMatrix (DM) code and marked it on the head position of the component through direct identification technology (DPM). An industrial advanced and reliable code reading device (reader) was installed on the new component transfer equipment (new component loader). When the new component was put into the reactor, the code was automatically recognized by the reader, and the component code was converted into a component number and transmitted to the fuel operation monitoring system through communication. The monitoring system compared it with the component numbers in the refueling plan to determine whether the transported new components were correct, improving the reliability and safety of component operations. To determine the size and depth of component codes, 16 component codes consisting of 4 sizes and 4 depths were provided as the test sample library based on the engineering site conditions. In order to determine the optimal component code from the experimental sample library, combined with the engineering site conditions, four main factors affecting the determination of component code were analyzed and summarized through enumeration method: installation angle of code reading equipment, offset angle, component type, and component movement speed. Since these four factors had no interaction with each other, a single factor experimental plan was determined that each experiment only changed one influencing factor, and the rest were fixed as fixed values, which were constant factors. There are a total of 7 tests: control rod assembly code reading test, fuel rod assembly code reading test, reader installation angle code reading test, control rod coding offset angle test, fuel rod coding offset angle test, control rod coding motion test, and fuel rod coding motion test. The judgment basis for each experiment is that the average reading time and maximum reading time of the code reader are less than 500 ms, and the success rate of reading is 100%. Through the first two experiments, the experimental sample library was preliminarily optimized, and then the optimized sample library was used for testing the installation angle of the code reader to screen out the optimal installation angle. Then, the sample library was further streamlined and optimized through encoding offset angle experiments, and finally, the optimal component encoding was determined through encoding motion experiments.

In response to the problems of long debugging cycle, high debugging cost, high safety risk, and difficult quality evaluation in the current automatic forming and cutting control of roadways, a digital twin driven virtual debugging method for roadway automatic forming and cutting is proposed. Firstly, by using real-time appearance mapping (RTAP-MAP) technology to reconstruct the three-dimensional environment of the roadway, a control system model of the roadheader is constructed to form a virtual debugging environment. The virtual sensor technology is used to achieve precise mapping from physical space to virtual environment state. A performance evaluation method for roadway automatic forming cutting has been established to address the problem of difficulty in quantitatively evaluating the quality of section forming. Based on the recording of the section forming cutting control process in the data transmission center, the evaluation indicators of section forming precision, cutting efficiency and the number of oil cylinder switches, hard rock cutting adjustment, and over excavation and under excavation are mainly calculated. This provides precise feedback signals for the iterative optimization of deep learning algorithms. An automatic cutting control strategy that integrates reinforcement learning is proposed to improve the adaptability and precision of automated operations. To verify the effectiveness and accuracy of the virtual debugging method, an automatic control experimental platform for roadheader is built. The virtual debugging system is applied to the automatic control program for forming cutting of excavation roadway. The virtual simulation results show the following points. ① The maximum positioning errors of the X, Y, and Z axes of the debugged software at the control key points are 74.8, 72.93, 123.67 mm, respectively. It indicates that the positioning precision of the virtual debugging method meets the performance requirements. ② The trajectory of the virtual prototype is basically consistent with that of the physical prototype, indicating that this debugging method has achieved mapping to the physical space. The application result shows the following points. ① The reinforcement learning controller adapts to complex environments in virtual excavation testing, effectively converts virtual sensor inputs into precise control instructions, and verifies the feasibility of simulation reality transfer training. By processing real-time feedback on excavation precision and avoiding over excavation and under excavation, the controller learns and optimizes the strategy. ② The improved section forming cutting control performance has been improved. According to the control quantity timestamp records in the database, it takes 126 seconds, which is 8 seconds less than before the improvement. ③ After improvement, the maximum error in tracking the end trajectory of the cutting section is 6.0 mm, which is 0.3 mm lower than before. This avoids the under excavation caused by the shaking of the cutting trajectory and makes the trajectory and section smoother.

Aiming at the situation of unstable control and high error during recoil compensation control of gunship borne automatic weapon, the recoil compensation control algorithm of gunship borne automatic weapon is proposed. According to the specific firing conditions of airborne automatic weapons, combined with the internal and external factors that affect the recoil of weapons, the force analysis is carried out. Meanwhile, the attitude parameters of gunship, including coordinate parameters and attitude Angle parameters, are calculated. On this basis, the control model of gunship is established, corresponding controller is designed, nonlinear interference observer is introduced, and adaptive control law is designed. The recoil compensation control of airborne automatic weapon is realized. The experimental results show that the proposed recoil compensation control algorithm has small control delay, can respond to compensation control changes in time, and has small control error. The overall fault tolerance of the control algorithm has been improved.

Objective In order to ensure that FAO (fully automatic operation system) can effectively guarantee the safe evacuation of passengers in case of emergency and quickly resume operations after evacuation completion, it is proposed to study the evacuation solution for FAO system based on central remote supervision and command. Method Based on the line conditions (such as evacuation platform, trackbed, etc.) and vehicle conditions (providing side door and end door two evacuation modes) that affect the safety of passenger evacuation, and targeting the safety risks that may be encountered during the passenger evacuation process (such as passengers falling, train collision, transition switch injury, third rail electric shock, etc.), the evacuation activation, evacuation recovery, as well as the FAO multi-system linkage mechanism and protection method for coupling train evacuation are designed in detail from two different application scenarios of platform evacuation and interval evacuation. Result & Conclusion Line and vehicle conditions required for the realization of FAO system evacuation function are proposed. During the evacuation process, the central dispatcher can remotely monitor and direct the evacuation of passengers through CCTV (video surveillance system), PA (broadcast system), emergency intercom and other devices. At the same time, through the ATS (automatic train supervision) system, the central dispatcher can realize the linkage control of multiple systems, such as CC (car controller), ZC (zone controller), and CI (computer interlocking), in order to ensure the safety protection and evacuation recovery of passengers when evacuating through the side/end doors of trains with different marshalling on platforms or intervals. This solution significantly improves the safety level and efficiency of evacuation.

The analysis points out that the current hydraulic pressure technology in coal mines faces challenges such as the inability to quickly and accurately adjust the output pressure and flow rate of fracturing pumps, the need to improve the remote safety monitoring effect, and the level of automation. An underground hydraulic fracturing automatic control system in coal mine has been designed. Based on the hydraulic fracturing technology and system composition in coal mines, it is clear that the key technologies of the control system include rapid and precise adjustment of the output flow and pressure of high-pressure and high flow fracturing pumps, remote high reliability and safety, high-speed real-time monitoring, one click start stop, and graphical analysis. Based on the KXH12 intrinsic safety controller as the core, combined with frequency converters, combination switches, monitoring hosts, electric valves and other equipment, as well as a dual line redundant communication scheme with fiber optic and CAN bus, the control system hardware has been developed. The fracturing pump controller, water tank controller, and central controller software have been developed. It achieves functions such as fast variable flow water injection, stable pressure maintenance, remote high-speed real-time monitoring and alarm of the fracturing system. Industrial tests are conducted on the hydraulic fracturing automatic control system at the coal mine site, and the results show that the pressure control precision of the system is 0.1 MPa, and the flow control precision is 0.1 m3/h. During 14 consecutive fracturing processes, good pressure retention effects are achieved after each coal layer crack. The operation is simple with high safety, meeting the requirements of hydraulic fracturing technology in coal mines underground.

[Objective] Setting up underpass or overpass in front or inside the urban rail transit depot for personnel's access to the automatic zone requires large project investment, being prone to affect personnel evacuation and cause safety hazards due to the long walking distance and extremely inconvenient operation. Therefore, it is necessary to make further research on protection technology that does not use underpasses or overpasses in the depot automatic zone. [Method] Due to the problems such as hazards of fire safety, inconvenient access, complicated management procedure, large project investment and construction difficulty under traditional protection strategy for the depot automatic zone in FAO (fully automatic operation) vehicle base, the active protection technology is put forward. The depot automatic zone is arranged according to the concept of active protection technology. The active protection operation process and the system architecture of this technology are introduced in detail. [Result & Conclusion] The active protection technology for the depot automatic zone in the FAO vehicle base is as follows: the electric retractable door in the middle channel is adjusted from normally closed state to normally open state. The underpass or overpass is canceled and the personnel and maintenance vehicles can enter the automatic zones through the middle channel. After the active protection technology is used, the personnel can pass the middle channel normally when no vehicle entering or exiting in the depot. Using active protection technology reduces the walking distance of both the personnel and maintenance vehicles, eliminates fire safety hazards, improves operation management efficiency and reduces project investment and construction difficulty.

According to different objectives, shielding calculation problems can be divided into regional problem, source-detector problem, and global problem. To solve the deep penetration problem of the three kinds of problems, the MCShield research team systematically proposed a series of variance reduction methods. According to these works, this article constructed a variance reduction technique system based on the automatic importance sampling (AIS) method, and conducted some verification work. For the source-detector problem, the NUREG/CR-6115 PWR pressure vessel calculation benchmark was used to verify the small detector automatic importance sampling (SDAIS) method. The result shows that the SDAIS method is more efficient than the AIS method for the source-detector problem. In addition, an AIS-based Monte Carlo coupled variance reduction (AIS-CADIS) method was also proposed which introduces the AIS method into Monte Carlo adjoint calculations and was verified against benchmark problems. For global problems, the grid-based AIS method was proposed and verified using a simplified reactor shielding calculation problem. The results show that compared with the AIS method and analog Monte Carlo methods, the grid-based AIS method can improve calculation efficiency.

High-speed automatic train operation (ATO) systems inherently exhibit strong nonlinearity and uncertainty. In view of the characteristics of nonlinearity and time-variability of high-speed train model parameters, this paper proposes a feed forward adaptive-generalized model predictive control (FA-GPC) method for dynamic optimization control of the ATO system along with a constrained multi-object predictive controller. Based on a multi-particle train model, an initial analysis explores the impacts of additional resistance changes on train operation. Subsequently, a multi-object performance indicator function containing control input constraints is constructed, combined with key indicators during the operation of high-speed trains, such as speed tracking accuracy, stopping accuracy, and riding comfort. Furthermore, a feed forward generalized prediction speed tracking control algorithm is designed based on the multi-object function, aiming to solve controller overshoot due to additional resistance changes and enhance control convergence rates. Taking into account various factors including influences of external environments and passenger movement during train operation, the resistance changes greatly, making it difficult to establish an accurate mathematical model, a constrained variable forgetting factor-recursive least square method is incorporated to identify the controlled auto-regressive integrated moving average model (CARIMA) of the train control system under different operational conditions. This approach aims to improve the robustness of the control system. Simulation results show that, compared with traditional GPC in the absence of the feed forward functionality and PID controller, the proposed feed forward generalized controller demonstrates good cruise control speed tracking accuracy within a ±0.5 km/h range under different line conditions and strong robustness thanks to the adaptive modification to the feed forward generalized predictive control algorithm for better performance in strong disturbance conditions.

Aiming at the problem that the data-driven automatic correlation methods which are difficult to adapt to the automatic correlation of oil-bearing strata with large changes in lateral sedimentary facies and strata thickness, an intelligent automatic correlation method of oil-bearing strata based on pattern constraints is formed. We propose to introduce knowledge-driven in automatic correlation of oil-bearing strata, constraining the correlation process by stratigraphic sedimentary patterns and improving the similarity measuring machine and conditional constraint dynamic time warping algorithm to automate the correlation of marker layers and the interfaces of each strata. The application in Shishen 100 block in the Shinan Oilfield of the Bohai Bay Basin shows that the coincidence rate of the marker layers identified by this method is over 95.00%, and the average coincidence rate of identified oil-bearing strata reaches 90.02% compared to artificial correlation results, which is about 17 percentage points higher than that of the existing automatic correlation methods. The accuracy of the automatic correlation of oil-bearing strata has been effectively improved.

The traditional coal seam cutting path planning predicts the height of the drum through geometric control, planning calculation, and other methods. But there are problems with large data errors in planning and prediction and inability to adapt to changes in geological conditions. In order to solve the above problems, a shearer automatic height adjustment technology based on a cutting roof and floor height prediction model is proposed. Firstly, the factors affecting the height of the cutting roof and floor are analyzed. It is pointed out that the main factors affecting the height of the cutting roof and floor include the fluctuation data of the coal seam, historical cutting data, elevation data of the scraper conveyor, and manual operation records. The above four types of data are fused and processed to establish a cutting roof and floor height prediction algorithm model based on long short term memory (LSTM) model and gray Markov model. The height of the cutting roof and floor is predicted through an algorithmic model. Secondly, based on the height data of the cutting roof and floor, combined with the position and posture and spatial coordinates of the shearer, a geometric model for calculating the height of the drum is established. At the same time, correction is made according to factors such as the sliding amount of the scraper conveyor and whether the addition and subtraction process is carried out. Finally, the height sequence of the roof and floor is converted into a drum height sequence. The cutting roof and floor height is converted into the target height of the shearer drum, which is executed by the shearer to the target height, achieving automatic adjustment of the drum height. The industrial test results show the following points. ① Under the control of automatic height adjustment technology, 90% of the predicted height deviation values of the roof and floor drums are within 10 cm of the actual height. The predicted height of the drums is significantly consistent with the actual height. ② Compared with traditional manual control methods, the number of manual intervention height adjustment times for cutting coal in the middle has decreased from 49 to 21. It indicates that the height prediction model for cutting the roof and floor and the geometric model for calculating the height of the drum are accurate and reasonable, and the automatic height adjustment technology for the shearer drum is feasible.

旨在为仓储物流行业及油脂加工厂中仓库设计提供借鉴，以库容量10万箱中小包装食用油（5 L小包装、10 L中包装成品油各5万箱）仓库为研究案例，系统分析了两种仓库（平面仓库和自动化立体库）的主要特点，从占地面积、生产成本、建设投资、运营费用等方面对两种仓库的技术方案进行了比较，并对两种仓库进行了经济分析。结果表明：自动化立体库相较于平面仓库能减少66.4%的占地面积，每年能够节约运营成本近50万元；自动化立体库的增量投资收益率为19.01%，增量投资回收期为5.26年，费用现值相比平面仓库低114万元。针对多品种、多拣选、年吞吐量大、产品可追溯性高、衔接工厂MES/ERP系统且持续运营在6年以上的仓库，使用自动化立体库能够显著降低项目总持有成本，降低工人劳动强度，提高发货效率，改进客户体验。In order to provide reference for the warehouse logistics industry and the warehouse design in oil processing plant, warehouse with a storage capacity of 100 000 boxes (each 50 000 boxes of 5 L small packaged oil and 10 L medium packaged oil) was used as a research case, the main features of two types of warehouse (automatic three-dimensional warehouse and flat warehouse) were systematically analyzed, the technical schemes of the two warehouses were compared through warehouse floor space, production costs, construction investment, operation costs and other aspects, their economic analysis was also conducted. The results showed that compared with flat warehouse, automatic three-dimensional warehouse could reduce 66.4% of the floor space, and save operation costs nearly 500 000 yuan per year. The incremental investment return rate of automatic three-dimensional warehouse was 19.01%, the incremental investment payback period was 5.26 years, and the cost present value was 1.14 million yuan lower than flat warehouse. For multi-species, multi-picking, large annual throughput, high product traceability, connecting factory MES/ERP system, and continuous operation in more than 6 years of warehouses, the use of automatic three-dimensional warehouse can significantly reduce the total cost of ownership of the project and the intensity of labor, improve delivery efficiency and customer experience.

[Objective] Current daily metro track inspections, including manual detection and track inspection vehicle method, suffer from issues such as low efficiency and high costs, failing to meet the increasing demands for rail transit operation safety. Therefore, it is necessary to develop an automatic identification and early-warning system for on-board rail corrugation based on image processing technology. [Method] The overall solution for automatic rail corrugation identification is elucidated, covering the denoising and uneven illumination correction of rail surface images, the positioning and segmentation of rail surface images, the localization of rail corrugation interval, and the estimation of rail corrugation cycles. The automatic rail corrugation identification and assessment method based on image processing is elaborated. The construction of the defect automatic identification and early-warning system is described from both software and hardware perspective based on system composition and logic architecture, and the demonstration application situation of the system in Beijing Subway is introduced. [Result & Conclusion] The system enables real-time detection, localization, assessment and early-warning of rail corrugation defects. Demonstrative application test results indicate that the system achieves an identification accuracy of over 97% for rail corrugation defects. In comparison to conventional manual inspection and dedicated track inspection vehicles, the system exhibits a significant advantage in detection efficiency.

Absolute observation and relative record are two methods in geomagnetic station observation. The actual values of geomagnetic field elements can be obtained after the relative record data is corrected by baseline data. Fluxgate theodolite is one of the basic tools required to obtain baseline values and plays an important role in geomagnetic observations. At present, fluxgate theodolite used in stations requires manual operation, and human factors affect the observation data quality. Because absolute geomagnetic observation cannot be realized in remote areas or areas with harsh environments, some blank areas in the coverage of geomagnetic absolute observation exist. The second phase of the Chinese Meridian Project will carry out automatic geomagnetic absolute observation of magnetic declination and inclination for the first time in China's mainland, which will promote the innovation and development of geomagnetic absolute observation technology. The automatic fluxgate theodolite is a self-developed equipment from China for automatic observation of magnetic declination and inclination. It adopts non-magnetic materials and piezoelectric motors, effectively overcomes and eliminates the systematic errors by optimizing the design, improves the machining process, and compensates for the multi-parameter error algorithm, so as to ensure measurement accuracy. The equipment has been applied to long-term practical observation in stations such as Shexian Station in Hebei Province, Helong Station in Jilin Province, Qianling Station in Shanxi Province and Baijiatuan Station in Beijing. Compared with the observation results of artificial fluxgate theodolite in these stations, the results show that the main performance of the automatic fluxgate theodolite is the same as that of the artificial fluxgate theodolite. The equipment has also passed a precursor equipment standard test from the China Earthquake Administration, and its functions and performance meet the requirements of absolute geomagnetic observation for geomagnetic stations.

Background With the aging population, the incidence of pressure injury (PI) is gradually increasing. This not only severely impacts the quality of life for patients but also increases healthcare expenditures. However, the early detection and accurate staging of PI heavily depend on specialized training. Objective To construct and validate an artificial intelligence model for the automatic detection and staging of PI aimed at enhancing the real-time nature, accuracy, and objectivity of PI diagnostics. Methods A total of 693 PI images from the electronic management system of pressure ulcers at Changshu No.1 People's Hospital were selected from January 2021 to February 2024, the images were randomly divided into a training set (551 images) and a test set (142 images), and categorized into six stages according to National Pressure Ulcer Advisory Panel (NPUAP) guidelines: StageⅠ (154 images), StageⅡ (188 images), StageⅢ (160 images), StageⅣ (82 images), deep tissue injury (57 images), and unstageable (52 images). A deep learning object detection model for PI was established using five different versions of the YOLOv8 [nano (n), small (s), medium (m), large (l) and extra large (x) ] neural network and transfer learning. The model evaluation metrics included accuracy, sensitivity, specificity, false positive rate, and detection speed. Finally, the model was deployed to a mobile application via the Ultralytics Hub platform, facilitating the application of the AI model in clinical practice. Results During the evaluation of a test set containing 142 PI images, the YOLOv8l version demonstrated high accuracy (0.827) and fast inference speed (68.49 fps), achieving the best balance between precision and speed among the YOLO versions. Specifically, it achieved an overall accuracy of 93.18% across all categories, a sensitivity of 76.52%, a specificity of 96.29%, and a false positive rate of 3.72%. Among the six stages of PI, the model achieved the highest accuracy for StageⅠat 95.97%. The accuracies for StageⅡ, StageⅢ, StageⅣ, deep tissue injury, and unstageable were 91.28%, 91.28%, 91.95%, 95.30%, and 93.29%, respectively. In terms of processing speed, YOLOv8l took a total of 2.07 seconds to process 142 images, averaging 68.49 PI images per second. Conclusion The AI model based on the YOLOv8l network can quickly and accurately detect and stage PI. Deploying this model to a mobile app allows for portable use in clinical practice, demonstrating significant potential for clinical application.

Objective To develop a deep learning network based on attention mechanism to identify the number of the vertical mandibular third molar(MTM) roots(single or double) on panoramic radiographs in an automatic way. Methods The sample consisted of 1 045 patients with 1 642 MTMs on paired panoramic radiographs and Cone-beam computed tomography(CBCT) and were randomly grouped into the training(80%), the validation(10%), and the test(10%). The evaluation of CBCT was defined as the ground truth. A deep learning network based on attention mechanism, which was named as RN-MTMnet, was trained to judge if the MTM on panoramic radiographs had one or two roots. Diagnostic performance was evaluated by accuracy, sensitivity, specificity, and positive predict value(PPV), and the receiver operating characteristic(ROC)curve with the area under the ROC curve(AUC). Its diagnostic performance was compared with dentists’ diagnosis, Faster-RCNN, CenterNet, and SSD using evaluation metrics. Results On CBCT images, single-rooted MTM was observed on 336(20.46%) sides, while two-rooted MTM was 1 306(79.54%). The RN-MTMnet achieved an accuracy of 0.888, a sensitivity of 0.885, a specificity of 0.903, a PPV of 0.976, and the AUC value of 0.90. Conclusion RN-MTMnet is developed as a novel, robust and accurate method for detecting the numberof MTM roots on panoramic radiographs.

Background and purpose: Radiotherapy is an important means to treat pelvic tumors. It is important to improve the accuracy of radiotherapy and enhance the effectiveness. This study aimed to compare and analyze the influence of positioning errors during the radiotherapy of patients using surface guided radiotherapy (SGRT) automatic positioning versus traditional surface marking positioning. Methods: We performed a retrospective analysis of 50 patients with pelvic tumors who received radiotherapy and had simple positioning fixation methods at the Fudan University Shanghai Cancer Center from April 2023 to April 2024. A total of 25 patients were treated using SGRT (observation group), and 25 patients were treated using traditional surface marking (control group). After positioning according to the two positioning techniques, cone beam computed tomography (CBCT) images were collected. These data were compared with the initial reference computed tomography (CT) images within 6 dimensions. The correction error values of the six dimensions were recorded. We calculated the averages, standard deviations, and the number of cases and probabilities in different interval distributions. SPSS 27.0 was used to perform statistical analysis of the data from the two groups of patients. We also analyzed the correlation of patients in the observation group within the 6 dimensions. Results: The non-parametric test showed significant differences in only the LNG and PITCH directions (P

Objectives To simulate the complex nonlinear interactions between the free surface and instantaneous wetted hull surface during ship navigation, an automatic mesh generation method is proposed.MethodsThe method takes a whole-hull triangular mesh file as the input and automatically generates meshes for the instantaneously transforming free surface and wetted hull surface. It is able to handle arbitrary hullforms with complex bow and stern geometries while exhibiting strong robustness. Simulations of waves generated by a KCS hull at various speeds are then conducted using the proposed method. ResultsThe predicted ship-side wave profiles, wave patterns around the hull, and wave-making resistance show good agreement with the experimental results. Moreover, at higher speeds, the simulated wave surface at the KCS stern consistently stays below the transom, exhibiting a dry transom characteristic. ConclusionsThe results demonstrate that the proposed method can accurately model nonlinear ship waves and has the potential to simulate the flow fields around transom-stern ships.

推特上不乏使用者貼出描述心情的各式推文，分析這些推文，可協助瞭解個體的心理狀態，對促進大眾心理健康的研究將有所助益。本研究擬對推文中關於心理健康素養方面的簡短文本，進行自動分類。使用包括傳統機器學習以及BERT、SetFit、GPT-3、 GPT-4 等人工智慧的技術，將其自動分類到五個面向中的11個題項，每個題項都有五個相關強度分數。期望在有限的人工標記的訓練資料下，機器預測的成效要到0.8以上，達到機器有效協助心理健康研究的目的。研究結果顯示使用SetFit 進行自動分類，多數題項都能達到MacroF1約0.8的標準，只有兩個題項成效在0.65左 右。本研究的貢獻之一，在呈現並比較多種自然語言處理派典，在這些困難任務中，其文字理解與分析上的成效。 Users on Twitter often post various tweets describing their moods. Analyzing these tweets can aid in understanding an individual’s psychological state, which will be beneficial to research aimed at promoting public mental health. This study intends to perform automatic classification on tweets related to mental health literacy. Techniques including traditional machine learning as well as AI technologies like BERT, SetFit, GPT-3, and GPT-4 are used to automatically classify them into 11 items across five dimensions, with each item having five related intensity scores. The goal is to achieve a machine prediction effectiveness of over 0.8 with limited human-annotated training data, to ensure the machine can effectively assist in mental health research. The results show that using SetFit, most items can achieve a Macro F1 score of about 0.8, with only two items scoring around 0.65. T he contribution of this study lies in presenting and comparing the effectiveness of various natural language processing paradigms in text comprehension and analysis on these difficult tasks.

For the purpose of the peaceful, safe and reliable application of nuclear technology, the monitoring of radioactive gas is one of the most important items for on-site inspection of CTBTO and safety monitoring of nuclear installations, which could rapidly and correctly discriminate nuclear activity of violation treaty of CTBT (Comprehensive Nuclear-Test-Ban Treaty) and identify the leakage of nuclear installations. Since radio-xenon isotopes (mainly 131m,133,133m,135Xe) have distinguished features including high fission yields, leaking easily and spreading worldwide, notable decay properties and low ambient concentrations, which are seen as important monitoring objects. Based on the proposed principle combined with these techniques including dynamic adsorption at -110 ℃ ultra low temperature, impurity removal with hollow fiber membranes, and gamma spectrum monitoring with low background, a rapid and highly sensitive on-site automatic monitoring system for radioxenon was developed, which could sample, separate and measure radioxenon on-site with a rapid and efficient way. During the development process, the technological flow diagram for whole system was established, and the critical hardware components including impurity removal column, adsorption column, lead shield chamber, and sample container were designed. And then these elements were assembled as four modules about sampling, preprocessing, refrigeration and measurement units combined with other components such as three hollow fiber membranes, four flowmeters, two pressure sensors and a HPGe detector and so on. Meanwhile, two control softwares including upper computer and off-line control versions were designed. And the methods of efficiency calibration, recovery test, calculation of activity concentration and MDC (minimum detectable activity) of xenon, and uncertainty evaluation were studied. Then these parameters were tested including detection efficiency, recovery of stable xenon, MDCs of 133Xe and 135Xe, and uncertainties of measuring results. The experimental results show that the recovery of stable xenon is 79.9%, the MDC of 133Xe is less than 0.26 Bq/m3 within a 120 min cycle including sampling, enrichment and measurement, when the activity concentration of 222Rn and sampling volume are 20 Bq/m3 and 2 m3, respectively. Compared with technical specifications of other systems, the elapsed time of a cycle of the system developed in this paper decreases larger than 10 times of SAUNA-OSI and XESPM-Ⅲ systems, while the MDC of the systems is smaller than other rapid systems. Because this developed system has following features about high sensitivity, shorter monitoring period, automation during total process, it will provide a dominant and straightforward contribution to the on-site inspection (OSI) of CTBT, safety monitoring of the gaseous effluents and emergency monitoring for nuclear facilities, and will be a powerful supplement for current monitoring technique and systems.

This study summarizes the application of automatic recognition technologies for patient-ventilator asynchrony (PVA) during mechanical ventilation. In the early stages, the method of setting rules and thresholds relied on manual interpretation of ventilator parameters and waveforms. While these methods were intuitive and easy to operate, they were relatively sensitive in threshold setting and rule selection and could not adapt well to minor changes in patient status. Subsequently, machine learning and deep learning technologies began to emerge and develop. These technologies automatically extract and learn data characteristics through algorithms, making PVA detection more robust and universal. Among them, logistic regression, support vector machines, random forest, hidden Markov models, convolutional autoencoders, long short-term memory networks, one-dimensional convolutional neural networks, etc., have all been successfully used for PVA recognition. Despite the significant advancements in feature extraction through deep learning methods, their demand for labelled data is high, potentially consuming significant medical resources. Therefore, the combination of reinforcement learning and self-supervised learning may be a viable solution. In addition, most algorithm validations are based on a single dataset, so the need for cross-dataset validation in the future will be an important and challenging direction for development.

In order to achieve high-precision pixel-level detection of multiple surface diseases in metro shield tunnels, a semantic segmentation model SU-ResNet++ based on deep learning is proposed. First, the encoder SE-ResNet50 based on residual unit and attention mechanism is designed and pre-trained, using as the backbone network of U-Net++ to design a new neural network model. Then, through original data collection, data preprocessing, and manual annotation, a shield tunnel surface multiple diseases dataset with 4 500 pictures is constructed. Finally, the proposed method is trained, verified, and tested on a dataset, and applied to practical engineering detection, achieving high-precision pixel-level diseases semantic segmentation. The experimental results indicate that the proposed SU-ResNet++ algorithm is applicable to the detection of shield tunnel disease data, and can automatically and accurately identify the disease category and form. Compared with the traditional semantic segmentation models, its disease identification precision is significantly improved, which meets the practical engineering requirements.

Addressing the issue of inter-class similarity is a challenging task in the research of autonomous driving scene classification, which primarily focuses on learning the distinctive features of targets in real-world complex traffic scenarios with high similarity, and constructing the overall correlation between features for scene classification. To this end, a multi-scale adaptive feature selection network for autonomous driving scene classification is proposed. Initially, a dual multi-scale feature extraction module is utilized for preliminary processing to extract inter-class similar features at different scales. Subsequently, a feature differentiation screening module is designed to complete the screening of scene-similar features, enabling the network to focus more on the typical and easily distinguishable features of different scene categories. Then, the feature screening results and multi-scale feature maps are transferred to the feature fusion classification module for scene classification, and the correlation between scene features is captured. Finally, an adaptive learning algorithm dynamically adjusts the training parameters through the output results, accelerating the network's convergence speed and improving accuracy. The proposed method is compared with existing network methods on three datasets: BDD100k, BDD100k+ and self-made dataset. Compared with the Top2 networks, it leads in accuracy by 3.29%, 5.59% and 12.65% (relatively), respectively. Experimental results demonstrate the effectiveness of the proposed method and its strong generalization capability. The scene classification method presented in this paper aims to learn the typical and easily distinguishable features and their correlations under different complex scene categories, reducing the impact of inter-class similarity among multiple targets, thereby making the scene classification results in real-world traffic scenario datasets more accurate.