Search

Image analysis of power equipment has important practical significance for power-line inspection and maintenance. This paper proposes an image recognition method for power equipment based on multitask sparse representation. In the feature extraction stage, based on the two-dimensional (2D) random projection algorithm, multiple projection matrices are constructed to obtain the multilevel features of the image. In the classification process, considering that the image acquisition process will inevitably be affected by factors such as light conditions and noise interference, the proposed method uses the multitask compressive sensing algorithm (MtCS) to jointly represent multiple feature vectors to improve the accuracy and robustness of reconstruction. In the experiment, the images of three types of typical power equipment of insulators, transformers, and circuit breakers are classified. The correct recognition rate of the proposed method reaches 94.32%. In addition, the proposed method can maintain strong robustness under the conditions of noise interference and partial occlusion, which further verifies its effectiveness.

The scraper conveyor is the key coal transportation equipment on the coal mining face. The conveyor often encounters the influence of abnormal working conditions, such as stuck chains, broken chains, and impact loads, which have a great effect on the smooth operation of the scraper chain and seriously affect the safety of coal production. If the chain breaks during the test, the broken link may fly out and hurt the tester. The test risk of the scraper conveyor under abnormal working conditions is relatively large; hence, the test research is impossible to carry out. The combination of experiment and simulation is used to carry out the research. The dynamic characteristic test bed of the chain drive system is created to collect the speed of the sprocket and the acceleration of the scraper. The two groups of data are obtained by simulation, and the accuracy of the simulation is verified by comparing the test data with the simulation data. The dynamic characteristics of the chain drive system under abnormal working conditions are analyzed by simulation. Results show that the longitudinal velocity of the link dramatically changes, the contact force increases by 5.4 times, and the link vibrates under impact loads. When the chain is stuck, the speed of the running direction suddenly changes, and the chainring vibrates. The contact force between the chainrings increases by 7.2 times compared with the smooth operation. When the chain is broken, the acceleration in the running direction of the link dramatically changes. This study provides a basis and reference for the further optimization design of the chain drive system and detection of the abnormal working conditions.

An obstacle perception system for intelligent vehicle is proposed. The proposed system combines the stereo version technique and the deep learning network model, and is applied to obstacle perception tasks in complex environment. In this paper, we provide a complete system design project, which includes the hardware parameters, software framework, algorithm principle, and optimization method. In addition, special experiments are designed to demonstrate that the performance of the proposed system meets the requirements of actual application. The experiment results show that the proposed system is valid to both standard obstacles and non-standard obstacles, and suitable for different weather and lighting conditions in complex environment. It announces that the proposed system is flexible and robust to the intelligent vehicle.

Traditional unsupervised seismic facies analysis techniques need to assume that seismic data obey mixed Gaussian distribution. However, field seismic data may not meet this condition, thereby leading to wrong classification in the application of this technology. This paper introduces a spectral clustering technique for unsupervised seismic facies analysis. This algorithm is based on on the idea of a graph to cluster the data. Its kem is that seismic data are regarded as points in space, points can be connected with the edge and construct to graphs. When the graphs are divided, the weights of the edges between the different subgraphs are as low as possible, whereas the weights of the inner edges of the subgraph should be as high as possible. That has high computational complexity and entails large memory consumption for spectral clustering algorithm. To solve the problem this paper introduces the idea of sparse representation into spectral clustering. Through the selection of a small number of local sparse representation points, the spectral clustering matrix of all sample points is approximately represented to reduce the cost of spectral clustering operation. Verification of physical model and field data shows that the proposed approach can obtain more accurate seismic facies classification results without considering the data meet any hypothesis. The computing efficiency of this new method is better than that of the conventional spectral clustering method, thereby meeting the application needs of field seismic data.

The optical electronic nose (e-nose) is drawing more and more attention from the academia and industry in Internet of Things applications. Since the charge-coupled Device (CCD)-based portable spectrometer is commonly used in gas sensing system of e-nose, the baseline drift interference is easy to be introduced, which reduces the accuracy and sensitivity of the system and makes the sensing performance worse. For this issue, first, the baseline drift data are analyzed and the characteristics of the baseline drift interference brought by the portable spectrometer are found in this article. Then, according to the characteristics of the drift, a baseline drift suppression technique for portable spectrometer is proposed. Specifically, it mainly includes four steps: data preprocessing, dimension reduction using principal component analysis, sources separation using independent component analysis, and reconstruction after interference suppression. Finally, experimental results show that the proposed technique suppressed the interference effectively and even better compared with classical existing methods.

Aiming at the problem of disconnection after road classification of remote sensing image, this paper proposes an optimization method for broken road connection considering spatial connectivity. The method extracts the road skeleton based on the binarized image after road extraction, and uses the eight neighborhood detection algorithm to find the road breakpoints after road extraction of high-resolution remote sensing image, and removes the isolated points of the road edge according to mathematical morphology filtering. Secondly, use K-means clustering algorithm to search for road breakpoints, and eliminate invalid breakpoints; then, fit the breakpoints of each category through polynomial curves, and record the mathematics of each fitted curve expression; Finally, the coordinate sequences between each kind of breakpoint is calculated according to each fitted polynomial, and the corresponding pixel is filled with the width of the road to realize automatic detection and connection. In this paper, the images after road extraction based on the U-Net network is used to test the method. The results show that the proposed method can better connect the roads formed by road or building shadows. Especially, the single broken road , has a high integrity of the road shape after repairing. The method proposed in this paper has certain reference significance for the classification and repair of linear objects such as roads, power grids and tracks.

The traditional teaching mode of music teaching is analyzed in this paper. The virtual reality music teaching system is designed and developed in view of the singleness of teaching method and learning situation and the limitation of learning resources. The music teaching system combines music teaching with Moodle teaching platform based on network environment, and takes constructivism theory as the main theoretical basis. In the design of the system, according to the needs of specific courses, the function module plug-in of Moodle platform is developed. The system can meet the needs of specific courses and improve the application of teaching data in actual teaching activities. Teaching activities under the system environment have changed the roles of teachers and students. Experimental data show that learners are more free to choose learning time and space, and their learning methods are more flexible

Addressing the phenomenon of data sparsity in hostile working conditions, which leads to performance degradation in traditional machine learning based fault diagnosis methods, a novel Wasserstein distance based Asymmetric Adversarial Domain Adaptation (WAADA) is proposed for unsupervised domain adaptation in bearing fault diagnosis. A GAN-based loss and asymmetric mapping are integrated to alleviate the difficulty of the training process in adversarial transfer learning, especially when the domain shift is serious. Moreover, simplified lightweight architecture is introduced to enhance the generalization and representation capability and reduce the computational cost. The experiment results show that our method not only achieves outstanding performance with sufficient data but also outperforms these prominent adversarial methods with limited data (both source and target domain), which provides a promising approach to the real industrial bearing fault diagnosis.

In a distributed OFDMA network, wireless channel is divided into multiple time-frequency resource block for transmission. Because there is no information interaction between nodes in a distributed network, and independent resource allocation of each nodes will result in collisions. Reasonable allocation resources can effectively reduce the collision ratio and improve system throughput. For the distributed network, we propose a time-frequency resource allocation method based on deep reinforcement learning (DRL), with definition of the state space, action space, and rewards. Each node uses this algorithm to realize time-frequency resource allocation. The nodes in the network use this algorithm without information interaction between the other nodes. Different nodes have a preference for the occupation of resource blocks, which reduces the probability of collisions in resource allocation. When the traffic intensity fluctuates, we introduce the concept of virtual traffic, which improves the continuity of the node’s selection of resource blocks. This continuity reduces the impact of traffic fluctuations on the node’s decisions. Compared with the random access slotted-aloha protocol, simulation results show the network throughput has been significantly improved. When the traffic intensity remains constant or fluctuates, the throughput can approximate that of the centralized network.

The single-phase H6 PV grid-connected inverter can effectively improve the efficiency and safety of the PV grid-connected system with the characteristics of high efficiency and suppressing leakage current. In order to reduce the current distortion due to the control delay, a predictive deadbeat control method is proposed based on the analysis of the operation of H6 topology and the principle of deadbeat control. The power feed-forward control speeds up the response of output current to the input power changes. The system model is established in the Matlab/Simulink and a prototype is constructed. Simulation and experimental results verify the validity of the proposed control method.

Based on SURF feature detection operator, image feature points are extracted, matched and mosaic Wavelet transform is used to fuse the registered images and eliminate the edges between stitched images. Overall, the splicing effect is well.

Urban traffic congestion causes huge economic losses and many social problems, such as, environmental pollution, energy consumption and longer commuting time. Accurate simulation and perception of traffic congestion scenes can help to analyze and solve the problems of traffic congestion. This paper presents a data-driven traffic congestion simulation method, which, by fitting the function relationship between traffic performance index (TPI) and the flow value of simulation input, simulates stable traffic scenes of the specified TPI. Three different virtual reality ways are applied to perceive the results of the simulation. The digital sand table is used to achieve the ”God’s perspective” perception of overall congestion scenes; the virtual driving system realizes drivers’ ”first-perspective” perception of traffic congestion scenes; a Myo-based virtual flight system is presented to roam and perceive congestion scenes from a ”free perspective”. The simulation and perception of traffic congestion scenes, serving as tool support, make contributions to the popularization education of traffic congestion, design evaluation of signal timing and driving behavior analysis. Experiments on various traffic scenes are conducted to verify the validity of the proposed method.

According to the requirements of tractor automatic steering hydraulic system, a proportional integrated control valve was designed. Based on the non-linearity of the hydraulic system, the state equation of the integrated proportional control valve that works in the open-center hydraulic steering systems was established. With the establishment of the MATLAB/Simulink models, the static and dynamic performance of the integrated proportional control valve were simulated. Furthermore, the indoor test platform was built on the basis of the proposed test scheme, on which the properties of steady flow, load flow and pressure loss of the integrated proportional control valve were analyzed. The simulation and experimental results show that the valve flow can be better to follow the drive voltage; when the spool opening unchanged, load step change, the simulation and test results of the flow is about 10L/min. The differential pressure relief valve pressure compensation performance is better. Finally, the actual test of the proportional integrated control valve is carried out. The experimental results show that the average deviation of the target tracking is less than 2.192° and the maximum deviation is less than 2.44°. The valve proposed in this study performed well referring to the tracing and quick response characteristics, which meets the requirements of tractor navigation control.

Aiming at the problem of mass data processing in intrusion detection model, a kind of intrusion detection model algorithm based on minimum rule self-organizing map (SOM) elliptical fuzzy pa...

In order to estimate the probability distribution of process level fault in Smart Substation, a tree based probability distribution model of process level fault in Smart Substation is proposed. Taking 14 years' process level fault data of Smart Substation in a regional power grid as sample data, this paper compares and analyzes various probability models. The probability distribution of process level fault scale of Smart Substation in this area is calculated by using the Boley Tanner branching process model in tree structure. The experimental results show that the model can estimate the probability distribution of process level fault size of Smart Substation well. Under the same confidence requirement, the sample data needed by the method based on the Boley Tanner model to estimate the fault probability distribution is one order of magnitude less than that directly based on the actual fault data.

With the gradual extension of artificial intelligence applications in terminal equipment, intelligent terminals based on edge computing are also widely used in power systems. The new embedded terminal device with local decision function can process the collected raw data locally and only report the key information, which can reduces the requirement of network bandwidth, so makes the wireless transmission meet the needs of more business scenarios. In order to collect a large number of terminal data in time, traditional substations need high network bandwidth. The data can only be transmitted over wire. The data in new transformer substation with intelligent terminal can be aggregated by wireless transmission. But the data security of wireless networks such as IEEE802.11ah, LoRa , NBIoT and WiFi cannot be guaranteed. At the same time, with the needs of social development, data security has risen to the height of national security. In order to realize the "security and controllability" of information systems, China Cryptographic Administration has published a series of encryption and decryption algorithms such as SM2, SM3, SM4. In this paper, based on the IEEE802.11ah standard, we study the application of SM2, SM3 and SM4 in LAN of transformer substation, and mainly focus on the identity authentication, key distribution and data encryption and decryption. In view of the three different stages, the system scheme designs are proposed respectively, and then we analyze their security and the additional system overhead in detail. The research of the SM2, SM3 and SM4 algorithms in LAN of transformer substation will further promote the popularization and application of intelligent terminal in substaion, and this will also ehance the level of intelligence of substation.

The temporal and spatial variation characteristics of precipitation in the basin are obtained by analyzing the precipitation data from 1960 to 2016 in the North Canal. The results show that: (1) The annual precipitation in the North Canal Basin has a weak downward trend as a whole, and there are alternating rising and falling characteristics between years. Uneven distribution during the year, precipitation is mainly concentrated in June to August, the average precipitation during this period accounted for 72% of the year;(2)The spatial distribution of precipitation in the basin is generally the most precipitation in the middle reaches, followed by the upper mountainous areas and the lower plains; In the seasons, the precipitation in the upper reaches of the summer is higher than that in the lower plains, and the opposite is true in winter.(3)The precipitation variation of the Northern Canal Basin is obvious, and the mutation point appeared in 1996 in the mid-1990s. This time node is consistent with the large regional pattern change of the Haihe River Basin; The periodic analysis shows that there are three periods of "bump-dry" change in the North Canal watershed, and the six oscillations on the 8-17a time scale are the most stable and significant, and the first main period is 13a.

In this paper, a method for increasing the isolation of transceiver antenna in CW radar is presented. By adding a virtual element in the proper position between the transmitting and receiving antennas, the isolation of the same frequency between the transmitting and receiving antennas is simply but effectively improved. It controls the virtual element secondary radiation from the transmitting antenna to reach the receiving antenna's amplitude and phase, making it close to the same frequency electromagnetic wave from the transmitting antenna to the receiving antenna directly, and cancels each other out in phase. In the case of this paper, the isolation of transmitting and receiving antennas is improved by 9dB compared with that without virtual elements.

Pressurized tree-type water distribution network (WDN) is widely used in rural water supply projects. Optimization of this network has direct practical significance to reduce the capital cost. This paper developed a discrete nonlinear model to obtain the minimum equivalent annual cost (EAC) of pressurized tree-type WDN. The pump head and pipe diameter were taken into account as the double decision variables, while the pipe head loss and flow velocity were the constraint conditions. The model was solved by using the improved decomposition&ndash, dynamic programming aggregation (DDPA) algorithm and applied to a real case. The optimization results showed that the annual investment, depreciation and maintenance cost (W1) were reduced by 22.5%, however, the pumps&rsquo, operational cost (p) increased by 17.9% compared to the actual layout. Overall, the optimal EAC was reduced by 15.2% with the optimized pump head and optimal diameter distribution of the network. This method demonstrated an intrinsic trade-off between investment and operational cost, and provided an efficient decision support tool for least-cost design of pressurized tree-type WDN.

It is difficult to realize the effective separation of signal and noise in the vibration signals, owing to the traditional signal filtering method. This paper puts forward an algorithm which was based on Hasudorff distance and nonlocal means (Hasudorff-Non Local Means, HDNLM). This algorithm makes use of Hasudorff distance more efficient than Euclidean distance distance in preserving signal details. It is aim to improve the weight model of traditional NLM of Euclidean Distance. Experimental and simulation show that: The improved NLM is superior than the simple morphology, traditional NLM, and db8 wave transform in the noise denoising of vibration signal aspect. The newly method is combined with the power spectral density, and it is well applied in the fault feature extraction of on-load tap changers.

In order to effectively monitor the operation of the on-load tape-changer (OLTC) online, an optimized empirical mode decomposition algorithm based on improved mask signal is proposed to analyze the vibration signals on the surface of the switched on load changers. An improved mask signal is added to the collected original signal, which can effectively eliminate the modal aliasing in the process of Empirical Mode Decomposition. Then the maximum power characteristic matrix is obtained according to the decomposed intrinsic mode function (IMF), further monitoring condition of the on-load tap switch. The results show that the maximum power characteristic matrix of OLTC in different operating states has obvious difference. When different typical faults occur in OLTC, the interval maximum power characteristic matrix index can effectively discriminate the degree of vibration signal difference.

In this paper, a practical strategy for adjusting PID parameters of hydro-governors in hydro-dominant power systems is presented, in order to suppress the ultra-low frequency oscillation (ULFO) in the range of 0.05 Hz. Firstly, practical indexes are proposed to select sensitive generators of ULFO. Then, Practical PID parameters optimization methods are introduced, considering damping characteristic of governor. Afterwards, general process of adjusting parameters for hydro-governors is presented, which could be referenced by engineering and technical personnel. Finally, the proposed strategies are applied to Southwest China Power Grid (SCPG) which is going to be asynchronous interconnection with Central China Power Grid. Simulation and field test results show effectiveness and practicability of the proposed measures.

ACP100 NPP designed by CNNC (China National Nuclear Corporation) is a 125MWe, multi-purpose small modular reactor based on pressurized water reactor technology; it adopts the integrated reactor technology. Different application scenarios bring up different design requirements: some require high compactness, but others care more about a longer cycle length, and some may require a fully mature and conservative design; thus, multiple design choices need to be proposed. Also, the same and most important thing cared by all users is that, the design needs to be validated to satisfy the current nuclear safety standards, and lower cost would be always preferred. Core nuclear design is a key part of the whole NPP design. Basically, nuclear design target of ACP100 is to achieve a reasonable good balance during longer cycle length, larger discharge exposure for fuel assemblies, and maximally using the mature technologies, and of course, with sufficient reactivity control ability for safety assurance. Aiming at satisfying all these different needs maximally, a strategy of supplying multiple nuclear design choices is proposed for ACP100: choice 1. Boron-free plan, this is a compact design with no need for chemistry and volume system, no need for daily boron adjustment and relative waste storage; choice 2. Boron and rod co-controlled plan, this is similar with large commercial PWRs, with a lower power peak factor and suitable for broad location sites. Both choices load 57 units of the same type fuel assemblies CF3S (with height reduced from CF3 fuel assemblies) per cycle, and both adopt partial reload and shuffle fuel management strategy to achieve larger discharge exposure. Gd is loaded in the fuel rods in both choices to help control reactivity. Choice 1 loads much more control rod clusters than choice 2, and of course, reactivity adjustment and compensation during operation is totally different between them. Using suitable and reliable software to simulate the core, through large amount of optimization, both choices achieve a 24-month fuel cycle length; the average discharge exposure of fuel assemblies reach about 40000MWd/tU, which is competitive among SMRs, especially for boron-free ones; and sufficient reactivity control ability and safety margin is validated to fully meet the reactor safety requirements.

The advent of simultaneous wireless information and power transfer (SWIPT) offers a promising approach to providing cost-effective and perpetual power supplies for energy-constrained mobile devices in heterogeneous cellular networks (HCNs). As energy efficiency (EE) has been envisioned as a key performance metric in 5G wireless networks, we consider a multiple-input single-output (MISO) femtocell cochannel overlaid with a Macrocell to exploit the advantages of SWIPT while promoting the EE. The femto base station sends information to information decoding (ID) femto users (FUs) and transfers energy to energy harvesting (EH) FUs simultaneously, and also suppresses its interference to Macro users. We maximize the information transmission efficiency (ITE) of ID FUs and energy harvesting efficiency (EHE) of EH FUs, respectively, with the QoS of all users, and investigate their relationship. We formulate these problems as fractional programming, which are nontrivial to solve due to the nonconvexity of ITE and EHE. To tackle these problems, we devise two beamformers namely zero-forcing (ZF) and mixed beamforming (MBF), and then propose an efficient algorithm to obtain the optimal power under both beamformers. Simulation results demonstrate that MBF provides better ITE and EHE than ZF, and there exists a tradeoff between ITE and EHE in general.

The reservoir operation method is used for the storage routing, due to the existence of forecast information errors and other uncertain factors, it may bring risks to the reservoir and its upstream and downstream flood prevention safety. Therefore, the risk analysis is to design and implement the reservoir flood prevention and dispatch the key to the way. This paper introduces a single factor risk analysis method for reservoir operation. Taking Qingshan Reservoir as an example, the risk of reservoir operation using forecast is analysed. The results displayed that taking into account the flood forecast error, when the reservoir adopts the forecast operation method, the risk rate of the reservoir dam body safety.

SF6 gas is widely used in various power equipment. In this paper, a baseline correction algorithm based on spectral detection method based on optimal wavelet basis is proposed. The algorithm selects the optimal wavelet base, removes the strong spectrum information by threshold method, eliminates the influence of continuum fitting, and performs the least squares fitting on the residual signal to obtain the continuum, and removes the continuum from the original spectrum to obtain the required Line spectrum. The accuracy of the quantitative analysis of the line spectrum intensity after processing is greatly improved, and the baseline is effectively estimated and corrected, and the fitting accuracy is higher than the traditional iterative wavelet baseline correction algorithm.

With the high proportion of renewable energy access to the distribution network, the volatility and randomness of renewable energy output will have a serious impact on the distribution network, especially at the peak of renewable power output, the reverse trend phenomenon is obvious. The active distribution network displays the “power” characteristics externally, which makes the voltage problem and transmission congestion problem in the distribution network very serious. This paper establishes a distributed energy storage optimization configuration model that takes into account the voltage quality, and considers the energy storage operation strategy optimization in the energy storage site selection and volume model, which makes the energy storage configuration more reasonable, improves the voltage quality and enhances the renewable energy consumption capacity, improve the economics of system operation.

Wi-Fi HaLow is a wireless networking protocol: IEEE 802.11ah, which was published on May 10th 2017. The technology's low power consumption competes with LoRa, ZigBee, Bluetooth. It uses sub 1 GHz (S1G) license-exempt bands, allowing the creation of large groups of stations or sensors, to provide extended range Wi-Fi networks and support the concept of Internet of Things (IoT). This paper gives an overview of the Wi-Fi HaLow technology, which is able to solve the weakness of Wi-Fi in penetrability and transmission distance up to 1km at the default transmission power of 200mW.

On-load tap-changer (OLTC) is easy to fail because of the frequency switching operation. The vibration signals produced by the operation process show obvious nonlinear characteristics, but most of the existing methods of extracting OLTC running feature are based on the linear theory, and there are some limitations. In order to realize the mechanical fault diagnosis of OLTC under the nonlinear vibration signal, a kind of OLTC mechanical fault diagnosis method based on the multiscale permutation entropy (MPE) nonlinear analysis method is proposed. Firstly, the vibration signals are collected in the different mechanical faults of the OLTC in the laboratory. Secondly, the Cao's method is used to determine the MPE parameters and the MPE characteristic of the mechanical failure state of the OLTC is extracted. Finally, the K-means++ algorithm is used to identify the mechanical state of the OLTC and compares it with the clustering effect of the traditional K-means algorithm. The results show that Mechanical fault diagnosis method for OLTC based on MPE and K-means++ algorithm can effectively extract the mechanical fault characteristics of OLTC, and can correctly identify the OLTC mechanical fault, and its diagnosis effect is better than the traditional K-means algorithm, and the stability is high.

In order to achieve simultaneous saturation attack on a same target of multi-missile cooperatively, and designs remaining impact time estimation method based on the velocity prediction information under time-varying velocity and the missile impact-time-control method are designed. Classical PNG law is adpoted in the longitudinal plane, at the same time, the missile adjusts the remaining time through maneuvering in lateral plane. In addition, three-dimensional model of coupled longitudinal and vertical planes is adopted in this paper. Simulation results verify that the method proposed in this paper can accurately estimate the remaining impact time based on priori knowledge of future velocity profile, and then effectively control the impact time.

Ground assault system construction is a necessary guarantee for the rapid formation of combat effectiveness. Based on the analysis of ground assault system requirements, this paper proposes an assessment method based on finite synthesis. This method effectively limits the "dimension disaster" problem caused by the aggregation of multi-level and multi-category indicators by limiting the scope and degree of comprehensive integration. In the form of capability maps, the calculation results of the equipment system are showed which are convenient for judging an equipment and decision support.

This paper proposes a framework for the calibration of a multi-camera and multi-LiDAR system. It utilizes the Apriltags to build a calibration environment to solve the poses and extrinsics of cameras and then deploy ICP-like algorithm to solve the extrinsics of LiDARs. Compared to previous extrinsic calibration methods, this proposed framework naturally applies to systems with different numbers and configurations of cameras and LiDARs. The calibration procedure produces extrinsics not only accurately and robustly, but efficiently without inconvenient human manipulation. The framework provides general, efficient and standardizable calibration solution for autonomous driving platforms with cameras and LiDARs.

In response to the application and requirements of all things seamless interconnection, International Institute of Electrical and Electronics Engineers (IEEE) has put forward a new generation of standard 802.11 ah, working in Sub-1GHz frequencies, achieving low power consumption and larger wireless local area network coverage. It improves the performance of Wi-Fi which easily affected by the building block, transmission distance and coverage. The new features of 802.11ah makes it more suitable for use in the smart grid. Intelligent substation adopts advanced, reliable, integrated and environment-friendly intelligent equipment. It supports grid real-time controlling, smart adjustment, online analysis and decision-making, collaboration and interaction. All the basic operations, including the information collection, measurement, control, protection, metering and monitoring, are taken automatically. In this paper, the feasibility of the 802.11ah wireless communication applying in intelligent substation is discussed. Besides, the actual performance of 802.11ah module prototype in 110kV substation has been tested. The results are analyzed, further studies and improvements are suggested.

With the rapid development of wireless sensor networks, the precision of time synchronization between nodes is becoming increasingly attractive. However, low cost, low power, low technology and other requirements restrict the use of synchronization algorithms in wireless sensor networks. TPSN time synchronization algorithm is a two-way synchronization algorithm, which is widely used in wireless sensor network. But due to the media access mechanism and the channel stability of the wireless sensor network, the standard TPSN algorithm can not meet the requirement of high precision of time synchronization. In this paper, the standard TPSN algorithm have been improved based on doze mechanism. Under the control of the central node, the nodes are doze and wake up sporadically to help with getting the transmission delay accurately. Thus, the time synchronization precision between each node is improved and the overall network power consumption is reduced. The experimental results show that the improved TPSN algorithm can effectively improve the accuracy of time synchronization.

According to the circuit structure and working principle of CMOS APSs, the single event effect and its producing mechanism are discussed in detail. Through a set of detailed experimental data and chart analysis, introduces a set of effective single event test methods for CMOS APSs. The experimental data and analysis results are not only helpful to improve the design of RHBD of CMOS APSs but also effectively carry out some research on ray monitoring and detection.

Virtual reality provides a new means for sensing and displaying Building Information Modeling (BIM) from two-dimension to three-dimension. This work uses virtual reality technology to connect BIM models with human perception based on the modeling and integration of BIM models. A variety of virtual reality methods are provided to perceive the highway BIM, such as holographic imaging, panoramic publishing, 3D printing, virtual driving, virtual flight, virtual cycling, virtual walking, augmented reality construction model delivery, etc. Virtual reality realizes people-oriented evaluation and optimization for highway project design, and improves the scientific nature of highway planning and design.

Accurate and reliable segmentation of liver tissue and liver tumor is essential for the follow-up of hepatic diagnosis. In this paper, we present a method for liver segmentation and a method for liver tumor segmentation. The two methods are grounded on a novel unified level set method (LSM), which incorporates both region information and edge information to evolve the contour. This level set framework is more resistant to edge leakage than the single-information driven LSMs for liver segmentation and surpasses many other models for liver tumor segmentation. Specifically, for liver segmentation, a hybrid image preprocessing scheme is used first to convert an input CT image into a binary image. Then with manual setting of a few seed points on the obtained binary image, the following region-growing is performed to extract a rough liver region with no leakage. The unified LSM is proposed at last to refine the segmentation result. For liver tumor segmentation, a local intensity clustering based LSM coupled with hidden Markov random field and expectation-maximization (HMRF-EM) algorithm is applied to construct an enhanced edge indicator for the unified LSM. With this development, expected segmentation results can be obtained via the unified LSM, even for complex tumors. The two methods were evaluated with various datasets containing a local hospital dataset, the public datasets SLIVER07, 3Dircadb, and MIDAS via five measures. The proposed liver segmentation method outperformed other previous semiautomatic methods on the SLIVER07 dataset and required less interaction. The proposed liver tumor segmentation method was also competitive with other state-of-the-art methods in both accuracy and efficiency on the 3Dircadb database. Our methods are evaluated to be accurate and efficient, which allows their adoptions in clinical practice.

In Autonomous Driving (AD), detection and tracking of obstacles on the roads is a critical task. Deep-learning based methods using annotated LiDAR data have been the most widely adopted approach for this. Unfortunately, annotating 3D point cloud is a very challenging, time- and money-consuming task. In this paper, we propose a novel LiDAR simulator that augments real point cloud with synthetic obstacles (e.g., cars, pedestrians, and other movable objects). Unlike previous simulators that entirely rely on CG models and game engines, our augmented simulator bypasses the requirement to create high-fidelity background CAD models. Instead, we can simply deploy a vehicle with a LiDAR scanner to sweep the street of interests to obtain the background point cloud, based on which annotated point cloud can be automatically generated. This unique "scan-and-simulate" capability makes our approach scalable and practical, ready for large-scale industrial applications. In this paper, we describe our simulator in detail, in particular the placement of obstacles that is critical for performance enhancement. We show that detectors with our simulated LiDAR point cloud alone can perform comparably (within two percentage points) with these trained with real data. Mixing real and simulated data can achieve over 95% accuracy., Comment: 10 pages