Search

Pre-trained language models perform well in various natural language processing tasks. However, their large number of parameters poses significant challenges for edge devices with limited resources, greatly limiting their application in practical deployment. This paper introduces a simple and efficient method called Autocorrelation Matrix Knowledge Distillation (AMKD), aimed at improving the performance of smaller BERT models for specific tasks and making them more applicable in practical deployment scenarios. The AMKD method effectively captures the relationships between features using the autocorrelation matrix, enabling the student model to learn not only the performance of individual features from the teacher model but also the correlations among these features. Additionally, it addresses the issue of dimensional mismatch between the hidden states of the student and teacher models. Even in cases where the dimensions are smaller, AMKD retains the essential features from the teacher model, thereby minimizing information loss. Experimental results demonstrate that BERTTINY-AMKD outperforms traditional distillation methods and baseline models, achieving an average score of 83.6% on GLUE tasks. This represents a 4.1% improvement over BERTTINY-KD and exceeds the performance of BERT4-PKD and DistilBERT4 by 2.6% and 3.9%, respectively. Moreover, despite having only 13.3% of the parameters of BERTBASE, the BERTTINY-AMKD model retains over 96.3% of the performance of the teacher model, BERTBASE.

The electric network frequency (ENF), often referred to as the industrial heartbeat, plays a crucial role in the power system. In recent years, it has found applications in multimedia evidence identification for court proceedings and audio–visual temporal source identification. This paper introduces an ENF region classification model named UniTS-SinSpec within the UniTS framework. The model integrates the sinusoidal activation function and spectral attention mechanism while also redesigning the model framework. Training is conducted using a public dataset on the open science framework (OSF) platform, with final experimental results demonstrating that, after parameter optimization, the UniTS-SinSpec model achieves an average validation accuracy of 97.47%, surpassing current state-of-the-art and baseline models. Accurate classification can significantly aid in ENF temporal source identification. Future research will focus on expanding dataset coverage and diversity to verify the model’s generality and robustness across different regions, time spans, and data sources. Additionally, it aims to explore the extensive application potential of ENF region classification in preventing crimes such as telecommunications fraud, terrorism, and child pornography.

Well placement optimization is one of the most important means to control the decline of oilfields and improve the recovery rate in the development process of deep and heterogeneous reservoirs, such as deep buried carbonate oil reservoirs. However, the mapping relationship from deployed well positions to actual profits is non-linear and multi-modal. At the same time, the injection and production relationship of new wells also affects the contribution of well positions to final profits. Currently, common algorithms include gradient-based and heuristic non-gradient algorithms, which have advantages, but face problems of high computational complexity, slow optimization speed, and difficulty in convergence. We propose an evolutionary algorithm for well placement optimization in carbonate reservoirs. This algorithm improves well placement optimization and computational speed by constraining the sampling process to effective sampling spaces, integrating prior knowledge to enhance sampling efficiency, strengthening local optima exploration, and utilizing parallel computing. Additionally, it refines the optimized variable content based on actual control factors, enhancing the algorithm’s robustness in practical applications. A case study from a carbonate reservoir in northwestern China demonstrated that this algorithm not only improved the performance by 50% compared to the classic DE algorithm but also achieved 15% higher optimization effectiveness than the current state-of-the-art algorithms.

In response to the challenge of existing wireless communication anti-jamming methods in effectively handling unknown jamming, this paper proposes a multi-parameter control anti-jamming algorithm for wireless communication systems based on the Linear–Quadratic Regulator (LQR). First, the proposed algorithm models the wireless communication system as a linear switched system based on the modulation and coding scheme. Subsequently, a feedback controller design method based on the LQR is introduced. By utilizing the multiple Lyapunov function method combined with linear matrix inequalities, sufficient criteria for the asymptotic stability of the system under unknown jamming conditions are provided. Finally, theoretical analysis and simulation results indicate that the proposed algorithm can rapidly adjust modulation and coding schemes as well as transmission power in complex jamming environments, thereby maintaining bit error rate (BER) stability and enhancing the reliability of the communication system.

Accurate wind power prediction helps to fully utilize wind energy and improve the stability of the power grid. However, existing studies mostly analyze key wind power-related features equally without distinguishing the importance of different features. In addition, single models have limitations in fully extracting input feature information and capturing the time-dependent relationships of feature sequences, posing significant challenges to wind power prediction. To solve these problems, this paper presents a wind power forecasting approach that combines feature weighting and a combination model. Firstly, we use the attention mechanism to learn the weights of different input features, highlighting the more important features. Secondly, a Multi-Convolutional Neural Network (MCNN) with different convolutional kernels is employed to extract feature information comprehensively. Next, the extracted feature information is input into a Stacked BiLSTM (SBiLSTM) network to capture the temporal dependencies of the feature sequence. Finally, the prediction results are obtained. This article conducted four comparative experiments using measured data from wind farms. The experimental results demonstrate that the model has significant advantages; compared to the CNN-BiLSTM model, the mean absolute error, mean squared error, and root mean squared error of multi-step prediction at different prediction time resolutions are reduced by 35.59%, 59.84%, and 36.77% on average, respectively, and the coefficient of determination is increased by 1.35% on average.

The inaccurate positioning of a plat valve is a crucial factor affecting the efficiency and safety of drilling operations. Because of the complicated structure of vertical drilling, it is difficult to obtain the actual position of a plat valve during the working process. This work effectively addresses the challenge of accurately measuring the deflection angle of a plat valve in mechanical automatic vertical drilling. Initially, a comprehensive mathematical model was proposed for determining the angle range of the plat valve, which was founded on the thrust generated by vertical drilling. Furthermore, this work proposed a precise measurement methodology for acquiring the crucial parameters of the MVDT. The actual deviation angle range for the N-MVDT and O-MVDT vertical drilling was calculated using data obtained from measurements of the tools. At the same time, the influence of TID on the performance of the MVDT was determined. Then, a simulation was carried out to compare the calculation results. From this, it was determined that the deflection angle range of the O-MVDT at a higher rotational speed was [86°, 58°], and the deflection angle range of the N-MVDT was [−14°, 14°]. By comparing the two methods, the error in the angle range was derived. The error of the N-MVDT was less than 16.67%, and the error of O-MVDT was less than 9.43%, proving that the measurement method and the mathematical model can accurately estimate the deflection angle range of a plat valve for mechanical automatic vertical drilling. The experimental and mathematical model results consistently showed that the deflection angle φ of the N-MVDT was significantly smaller than that of the O-MVDT, so the TID significantly improved the stability of the MVDT.

To address the difficulty in obtaining a large number of labeled jamming signals in complex electromagnetic environments, this paper proposes a small-sample communication jamming signal recognition method based on WDCGAN-SA (Wasserstein Deep Convolution Generative Adversarial Network–Self Attention) and C-ResNet (Convolution Block Attention Module–Residual Network). Firstly, leveraging the DCGAN architecture, we integrate the Wasserstein distance measurement and gradient penalty mechanism to design the jamming signal generation model WDCGAN for data augmentation. Secondly, we introduce a self-attention mechanism to make the generation model focus on global correlation features in time–frequency maps while optimizing training strategies to enhance the quality of generated samples. Finally, real samples are mixed with generated samples and fed into the classification network, incorporating cross-channel and spatial information in the classification network to improve jamming signal recognition rates. The simulation results demonstrate that under small-sample conditions with a Jamming-to-Noise Ratio (JNR) ranging from −10 dB to 10 dB, the proposed algorithm significantly outperforms GAN, WGAN and DCGAN comparative algorithms in recognizing six types of communication jamming signals.

Before the construction of a bridge begins, workers arrange the necessary parts and then cut and process them. The quality of the cutting layout directly affects the material utilization rate and the efficiency of the subsequent processes. During bridge construction, an intelligent part layout can improve work efficiency, save time, and reduce the labor intensity and production costs for the company. In this study, we studied a layout optimization algorithm, focusing on rectangular parts in the material preparation process. A mathematical model for the rectangular layout problem was constructed, and a hybrid genetic whale optimization algorithm is proposed that is a combination of the whale optimization algorithm and the genetic algorithm. Based on the “one size fits all” layout strategy, the materials are divided into strips, which are further divided into stacks, serving as the positioning strategy to determine the positional relationships of the parts. Test cases and actual engineering data were used to compare the layouts generated using different algorithms. The results show that the genetic whale algorithm proposed in this paper results in a high utilization rate and is highly effective.

In order to effectively improve the drilling speed in deep hard rock and save drilling costs, this study explores the transformation mechanism and critical velocity range of ultra-high-speed diamond drilling technology on rock breaking effect, using marble as an example. The study establishes an ultra-high-speed single diamond fragmentation model using the finite element method (FEM) and solves for the unknown critical velocity of marble in this drilling technique. Additionally, small diameter bit drilling experiments were conducted on our self-developed ultra-high-speed diamond drilling test bench. Based on existing simulation results and experimental studies, we discuss the critical velocity problem and compare and analyze the change in the rock-breaking mechanism and mechanical specific energy (MSE) under conventional drilling versus ultra-high-speed rotary drilling conditions. Our results indicate that changes in rock breaking mechanisms under ultra-high-speed diamond drilling conditions are limited to a specific speed range and do not persist with increasing speeds. Furthermore, experimental verification confirms that ultra-high-speed diamond drilling can effectively reduce MSE and increase the rate of penetration (ROP) by altering the rock-breaking mode. It is hoped that these findings will provide valuable insights for applying this technology to various hard rocks.

Semantic segmentation of 3D point clouds in drivable areas is very important for unmanned vehicles. Due to the imbalance between the size of various outdoor scene objects and the sample size, the object boundaries are not clear, and small sample features cannot be extracted. As a result, the semantic segmentation accuracy of 3D point clouds in outdoor environment is not high. To solve these problems, we propose a local dual-enhancement network (LDE-Net) for semantic segmentation of 3D point clouds in outdoor environments for unmanned vehicles. The network is composed of local-global feature extraction modules, and a local feature aggregation classifier. The local-global feature extraction module captures both local and global features, which can improve the accuracy and robustness of semantic segmentation. The local feature aggregation classifier considers the feature information of neighboring points to ensure clarity of object boundaries and the high overall accuracy of semantic segmentation. Experimental results show that provides clearer boundaries between various objects, and has higher identification accuracy for small sample objects. The LDE-Net has good performance for semantic segmentation of 3D point clouds in outdoor environments.

The increase of popularity of reactivity-controlled compression ignition (RCCI) is attributed to its capability of achieving ultra-low nitrogen oxides (NOx) and soot emissions with high brake thermal efficiency (BTE). The complex and nonlinear nature of the RCCI combustion makes it challenging for model-based control design. In this work, a model-based control system is developed to control the combustion phasing and the indicated mean effective pressure (IMEP) of RCCI combustion through the adjustments of total fuel energy and blend ratio (BR) in fuel injection. A physics-based nonlinear control-oriented model (COM) is developed to predict the main combustion performance indicators of an RCCI marine engine. The model is validated against a detailed thermo-kinetic multizone model. A novel linear parameter-varying (LPV) model coupled with a model predictive controller (MPC) is utilized to control the aforementioned parameters of the developed COM. The developed system is able to control combustion phasing and IMEP with a tracking error that is within a 5% error margin for nominal and transient engine operating conditions. The developed control system promotes the adoption of RCCI combustion in commercial marine engines.

Proteins undergo a series of conformational changes when affected by the applied electric field, which changes their functions and properties. The conformational changes in proteins in various electric fields are different due to their internal structures. This study simulates the molecular dynamics of proteins in different amounts and directions of electric fields with gromacs software. According to the root mean square deviation, hydrogen bond, dipole moment, and solvent accessible surface area, it is proved that the conformation change in proteins is more drastic under the simultaneous action of multiple electric fields under various directions, and different fragments unfold with divergent electric fields combined, which is of great importance to control protein function, improve biochemical research and production efficiency in the food and drug safety field.

Industrial system operations usually have dynamic characteristics. If these characteristics are ignored, the performance of fault detection degrades. Herein, the fault-detection algorithm of dynamic global–local preserving projection (DGLPP) is employed to solve the problem mentioned. First, time-delay data are added to the sample to form an augmentation matrix and characterize the system dynamics. Second, the dimensionality of the augmented matrix is reduced using global–local preserving projection. The dimensionality-reduction method can preserve the data’s global and local structures. Then, a DGLPP model is built using the dimensionality-reduced data. Moreover, Hotelling’s T2 and squared prediction error (SPE) statistics are used for fault detection. Finally, this method is used to detect the fault in the Tennessee Eastman (TE) process. The experimental results show that the DGLPP method has an enhanced fault detection rate. Moreover, the fault-detection effects of the DGLPP method are better than those of the principal component analysis (PCA), local preserving projection (LPP), and global–local preserving projection (GLPP) methods.

Methane is considered as a clean energy that is widely used in places with high environmental requirements. The increasing demand for methane exploration in polar and deep sea extreme environments has a positive role in carbon neutrality policies. As a result, there will be a gradual increase in exploration activities for deep sea methane resources. Methane sensors require high reliability but are prone to faults, so fault diagnosis and forecasting of gas sensors are of vital practical significance. In this work, a Gaussian-linearized transformer model with a tranquilized time-series decomposition method is proposed for fault diagnosis and forecasting tasks. Since the traditional transformer model requires more computational expense with time complexity of O (N2) and is not applicable to continuous-sequence prediction tasks, two blocks of the transformer are improved. First, a Gaussian-linearized attention block is modified for fault-diagnosis tasks so that its time complexity can be changed to O (N), which can reduce computational resources. Second, a model with proposed attention for fault forecasting replaces the traditional embedding block with a decomposed block, which can input the continuous sequence data to the model completely and preserve the continuity of the methane data. Results show that the Gaussian-linearized transformer improves the accuracy of fault diagnosis to 99% and forecasting with low computational cost, which is superior to that of traditional methods. Moreover, the least mean-square-error loss of fault forecasting is 0.04, which is lower compared with the traditional time series prediction models and other deep learning models, highlighting the great potential of the proposed transformer for fault diagnosis and fault forecasting of gas sensor arrays.

As underground engineering extends into the western and deeper regions of China, more and more Luohe Formation sandstone layers will be encountered, which have weak cementation and high water content. It is a significant challenge to apply the open TBM, and the support system is crucial in determining whether TBM can excavate quickly and safely. Therefore, in order to optimize the support scheme, this paper analyzes the pore structure and porosity through CT scanning, the results indicate that the volume percentage of pores ≥34 μm is 2.3% and 1.5%, respectively, the large pore apertures are predominant, the surrounding rock has strong permeability, and there is a high risk of rock burst and roof collapse accidents, hence requiring reinforced support measures. On this basis, numerical simulations were conducted to evaluate the support effectiveness. The results show that replacing the “bolt + mesh” with a “bolt + cable + mesh + steel belt”, and replacing the top three bolts with 7.3 m anchor cables, can better control the deformation and provide sufficient thrust force for the TBM, ensuring excavation speed. After the implementation of this scheme at the Kekegai coal mine in Shaanxi, China, the TBM excavation speed increased by 70%, from the previous 10 m/day to 17 m/d, significantly reducing the project duration and construction costs.

Traditional maize seedling detection mainly relies on manual observation and experience, which is time-consuming and prone to errors. With the rapid development of deep learning and object-detection technology, we propose a lightweight model LW-YOLOv7 to address the above issues. The new model can be deployed on mobile devices with limited memory and real-time detection of maize seedlings in the field. LW-YOLOv7 is based on YOLOv7 but incorporates GhostNet as the backbone network to reduce parameters. The Convolutional Block Attention Module (CBAM) enhances the network’s attention to the target region. In the head of the model, the Path Aggregation Network (PANet) is replaced with a Bi-Directional Feature Pyramid Network (BiFPN) to improve semantic and location information. The SIoU loss function is used during training to enhance bounding box regression speed and detection accuracy. Experimental results reveal that LW-YOLOv7 outperforms YOLOv7 in terms of accuracy and parameter reduction. Compared to other object-detection models like Faster RCNN, YOLOv3, YOLOv4, and YOLOv5l, LW-YOLOv7 demonstrates increased accuracy, reduced parameters, and improved detection speed. The results indicate that LW-YOLOv7 is suitable for real-time object detection of maize seedlings in field environments and provides a practical solution for efficiently counting the number of seedling maize plants.

Tight sandstone reservoirs, as typical unconventional reservoirs, have attracted more and more attention worldwide. Compared to conventional reservoirs, the complex microscopic pore structures and micro-fractures of the tight reservoirs are developed, and fluid flow in the tight sandstone reservoir is very complicated. Therefore, studying the influence of fractures on the rock topological structure and the flow characteristics between micro-fractures and the matrix is critical to the development of tight sandstone reservoirs. In light of the intricate nature of micro-fractures, digital rock analysis technology is utilized to address this issue instead of traditional experimental methods. Firstly, the selected tight rock is scanned by X-ray computed tomography (XCT), and then the natural tight digital rock is reconstructed by filtering and binary segmentation. Secondly, micro-fracture with different shapes is added to the digital rock. Then, a pore network model extracted by the maximum ball method is adopted to conduct the flow simulation. The study finds that the distribution of pore throat radius changes with increasing fracture aperture and length and that the permeability and capillary pressure curves are influenced by fracture aperture and length in the direction of connectivity. The impact of micro-fracture dip angle on pore throat structure is significant with changes in permeability and fluid flow in different directions. Overall, the study emphasizes the importance of fracture aperture and length in determining the properties and behavior of tight sandstone reservoirs and highlights the usefulness of digital rock analysis technology in providing insights into complex microstructures.

Convolutional neural networks (CNNs) have been developed as an effective strategy for hyperspectral image (HSI) classification. However, the lack of feature extraction by CNN networks is due to the network failing to effectively extract global features and poor capability in distinguishing between different feature categories that are similar. In order to solve these problems, this paper proposes a novel approach to hyperspectral image classification using a multidimensional spectral transformer with channel-wise correlation. The proposed method consists of two key components: an input mask and a channel correlation block. The input mask is used to extract relevant spectral information from hyperspectral images and discard irrelevant information, reducing the dimensionality of the input data and improving classification accuracy. The channel correlation block captures the correlations between different spectral channels and is integrated into the transformer network to improve the model’s discrimination power. The experimental results demonstrate that the proposed method achieves great performance with several benchmark hyperspectral image datasets. The input mask and channel correlation block effectively improve classification accuracy and reduce computational complexity.

Shaped cooling holes have received considerable attention in recent years due to their potential to improve heat transfer while minimizing pressure drop. In this study, the effects of lobe-shaped cooling holes on film cooling performance and turbulent flow characteristics are investigated using large-eddy simulation (LES). The LES approach was validated by comparing the calculated film cooling effectiveness with experimental data. The results show that well-designed lobe-shaped cooling holes can improve the cooling efficiency by up to 37% and reduce the total pressure drop. Additionally, the cooling jets of lobe-shaped cooling holes have a smaller angle to the wall surface, indicating better wall attachment. However, poorly designed lobe-shaped cooling holes can reduce the cooling efficiency. Overall, this study suggests that lobe-shaped cooling holes are a promising technology for improving gas turbine engine performance.

With the rapid development of intelligent vehicle safety verification, scenario-based testing methods have received increasing attention. As the space of driving scenarios is vast, the challenge in scenario-based testing is the generation and selection of high-value testing scenarios to reduce the development and validation time. This paper proposes a method for generating challenging test scenarios. Our method quantifies the challenges in these scenarios by estimating the risks based on ISO 26262. We formulate the problem as a Markov decision process and quantify the challenges in the current state using the three risk factors provided in ISO 26262: exposure, severity, and controllability. We then employ reinforcement learning algorithms to identify the challenges and use the state–action value matrix to select motions for a background vehicle to generate critical scenarios. The effectiveness of the approach is validated by testing the generated challenge scenarios using a simulation model. The results show that our method can ensure both accuracy and coverage, and the larger the state space is, the more accident-prone the generated scenarios are. Our proposed method is general and easily adaptable to other cases.

With the development of additive manufacturing (AM), the Ti-6Al-4V alloy manufactured by laser powder bed fusion (LPBF) is becoming more widely studied. Fatigue fracture is the main failure mode of such components. During LPBF processing, porosity defects are unavoidable, which hinders the exploration of the relationship between fatigue performance and microstructure. In this study, a laser remelting method was used to reduce porosity defects inside the Ti-6Al-4V alloy. Three annealing treatments (AT) and three solution-aging treatments (SAT) were used to study the effect of the two-phase zone (α + β) microstructure on fatigue life and fatigue crack growth behavior. Fatigue life and crack growth rate (CGR) curves were obtained, and fatigue fracture surface and crack growth fracture surface were analyzed. The results show that microstructure influences fatigue life but has little effect on CGR. Compared with the as-built specimen, the fatigue life of the AT and SAT specimens increased significantly at 850℃ by 101 and 63.7 times, respectively. The thickness of the α lath and the location of crack nucleation together affect the fatigue life. In the stable growth stage, the layered microstructure of α colonies is the most resistant to crack growth.

Recommending loan products to applicants would benefit many financial businesses and individuals. Nevertheless, many loan products suffer from the cold start problem; i.e., there are no available historical data for training the recommendation model. Considering the delayed feedback and the complex semantic properties of loans, methods for general cold start recommendation cannot be directly used. Moreover, existing loan recommendation methods ignore the default risk, which should be evaluated along with the approval rate. To solve these challenges, we propose CSRLoan for cold start loan recommendation. CSRLoan employs pretraining techniques to learn the embeddings of statements, which captures the intrinsic semantic information of different loans. For recommendation, we design a dual neural matrix factorization (NMF) model, which can not only capture the semantic information of both loan products and applicants but also generate the recommendation results and default risk at the same time. Extensive experiments have been conducted on real-world datasets to evaluate the effectiveness and efficiency of the proposed CSRLoan.

In Chinese megacity centers, substandard housing is being removed. The inhabitants are relocated to peripheral large-scale residential areas, and we call them “relocatees”. After the relocation, their housing quality is improved but travel quality may worsen. Little attention has been paid to the change in travel equity over time for shopping-related trips of relocatees and non-relocatees in these areas in rapidly developing countries, especially for the change before and after the major built environment development, such as a shopping plaza’s opening. To fill this research gap, we conducted two surveys before and after Wanda Shopping Plaza’s opening. We used statistical analysis, estimated discrete choice models focusing on mode choice, and calculated consumer surplus as a measurement of travel equity. Results show a relocatee has a lower travel quality for shopping-related trips. Travel equity between relocatees and non-relocatees for shopping-related trips becomes worse after Wanda’s opening. Differences in consumer surplus between relocatees and non-relocatees provide a reference for the travel subsidy policy.

Owing to its vibration reduction and indirect signal amplification characteristics, a nonlinear energy sink (NES) is used for crack identification of structures. The equation of a cracked simply supported beam with an NES is derived and the influences of the crack properties on the dynamics of the system are investigated. A pixel method is applied to estimate phase diagram areas and crack parameters are identified based on the geometric features of the phase diagram. It is shown that the phase diagram of the NES varies significantly with crack parameters. This is because of the higher amplitudes generated by the absorbing NES with respect to the primary structure. In addition, the NES can reduce the amplitude of the cracked beam without hindering crack identification. Moreover, through introducing the NES, the system can produce a strongly modulated response, which can be utilized for convenient crack detection. Two sets of crack identification indices are defined for the system, generating a single-period response and a strongly modulated response. Contour maps of identification indices based on the area of phase diagrams are obtained by varying crack location and depth. The crack parameters are identified through the intersection of contour lines.

The shortage of fossil fuels has led to increasing attention on new energy technologies, such as wind power and photovoltaic energy technologies, and the volatility of new energy has become the biggest obstacle for its participation in the process of power-system restoration. This paper presents an optimal control method of the parallel closing angle of the loop network in the process of power system recovery considering the uncertainty of wind-power output. Firstly, based on solving the uncertainty of wind-power output, a probability-prediction model of wind-power output based on the quantile regression of long-term and short-term memory networks (LSTMs) is established. Based on predicting the future wind-power output interval, the probability density function of the output at each time point in the future is obtained by a kernel density estimation. Secondly, by adjusting the output of conventional units and restoring the feeder load, a multiobjective optimal-control model is established to minimize the output change of conventional units and restore the most important feeder load. Based on considering the output probability of new energy, the optimal control of the phase angle difference at both ends of the line to be paralleled is realized, and the multiobjective optimization algorithm is used to solve the established model. Finally, the effectiveness of the proposed method and model is verified by IEEE 39-bus system simulation.

A tunable and compact mid-infrared optical parametric oscillator (OPO) based on BaGa4Se7 (BGSe) crystal with a repetition rate up to 250 Hz was demonstrated. A high energy and more approachable side-pumped Q-switched Nd:YAG laser was employed as the pump for the BGSe OPO. Due to the pump double-pass single-resonant oscillator (DP-SRO) configuration, the maximum average power of 250 mW and the maximum pulse energy of 1.28 mJ at 4.06 μm was achieved with the repetition rate of 250 Hz and 100 Hz, respectively. The tunable mid-infrared output from 3.42–4.73 μm was obtained. The influence of repetition rate and cavity length was studied and the thermal effect was analyzed.

An algorithm for automatically planning trajectories designed for painting large objects is proposed in this paper to eliminate the difficulty of painting large objects and ensure their surface quality. The algorithm was divided into three phases, comprising the target point acquisition phase, the trajectory planning phase, and the UR5 robot inverse solution acquisition phase. In the target point acquisition phase, the standard triangle language (STL) file, algorithm of principal component analyses (PCA), and k-dimensional tree (k-d tree) were employed to obtain the point cloud model of the car roof to be painted. Simultaneously, the point cloud data were compressed as per the requirements of the painting process. In the trajectory planning phase, combined with the maximum operating space of the UR5 robot, the painting trajectory of the target points was converted into multiple traveling salesman problem (TSP) models, and each TSP model was created with a genetic algorithm (GA). In the last phase, in conformity with the singularities of the UR5 robot’s motion space, the painting trajectory was divided into a recommended area trajectory and a non-recommended area trajectory and created by the analytical method and sequential quadratic programming (SQP). Finally, the proposed algorithm for painting large objects was deployed in a simulation experiment. Simulation results showed that the accuracy of the algorithm could meet the requirements of painting technology, and it has promising engineering practicability.

Fruit quality has long been well known as an important prerequisite for evaluating pepper resources. In the study, 75 cultivars belonging to four Capsicum species were investigated on the bases of fruit morphology and concentrations of fruit secondary metabolites. The results showed that C. annuum had a greater variation than C. chinense and C. frutescens in terms of fruit weight, size and contents of carotenoids, anthocyanidins, and capsaicinoids. Moreover, there were significant differences in the concentrations of total phenol, total soluble sugar, total soluble solids, titratable acid, and total soluble protein of all samples. Capsaicin and dihydrocapsaicin were the most important capsaicinoids components in pepper fruits, of which C. chinense both had the highest levels, whereas some accessions of C. annuum were not detected. Eighteen fatty acids were detected in pepper fruits, and the dominant member was linoleic acid (a polyunsaturated fatty acid) therein. We integrated a set of methods for metabolites, and the results indicated that there was a positive correlation between concentrations of capsaicinoids and fatty acids. Four accessions had the highest contents of quality-related compounds, two of which belonged to C. chinense (S23 and S24) and possessed high levels of capsaicin. However, S67 had a high level of β-carotene and S68 contained higher contents of total fatty acid and ascorbic acid, and both of them belonged to C. annuum. This work could provide a valid experimental basis for the potential application value of the four accessions mentioned above.

Meta-learning aims to teach the machine how to learn. Embedding model-based meta-learning performs well in solving the few-shot problem. The methods use an embedding model, usually a convolutional neural network, to extract features from samples and use a classifier to measure the features extracted from a particular stage of the embedding model. However, the feature of the embedding model at the low stage contains richer visual information, while the feature at the high stage contains richer semantic information. Existing methods fail to consider the impact of the information carried by the features at different stages on the performance of the classifier. Therefore, we propose a meta-learning method based on adaptive feature fusion and weight optimization. The main innovations of the method are as follows: firstly, a feature fusion strategy is used to fuse the feature of each stage of the embedding model based on certain weights, effectively utilizing the information carried by different stage features. Secondly, the particle swarm optimization algorithm was used to optimize the weight of feature fusion, and determine each stage feature’s weight in the process of feature fusion. Compared to current mainstream baseline methods on multiple few-shot image recognition benchmarks, the method performs better.

The multi-UAV system has stronger robustness and better stability in combat. Therefore, the collaborative penetration of UAVs has been extensively studied in recent years. Compared with general static combat scenes, the dynamic tracking and interception of equipment penetration are more difficult to achieve. To realize the coordinated penetration of the dynamic-tracking interceptor by the multi-UAV system, the intelligent UAV model is established by using the deep deterministic policy-gradient algorithm, and the reward function is constructed using the cooperative parameters of multiple UAVs to guide the UAV to proceed with collaborative penetration. The simulation experiment proved that the UAV finally evaded the dynamic-tracking interceptor, and multiple UAVs reached the target at the same time, realizing the time coordination of the multi-UAV system.

An innovative hybrid energy system consisting of a waste-to-energy unit and a coal-fired power unit is designed to enhance the energy recovery of waste and decrease the investment costs of waste-to-energy unit. In this integrated design, partial cold reheat steam of the coal-fired unit is heated by the waste-to-energy boiler’s superheater. The heat required for partial preheated air of waste-to-energy unit and its feedwater are supplied by the feedwater of CFPU. In addition, an additional evaporator is deployed in the waste-to-energy boiler, of which the outlet stream is utilized to provide the heat source for the urea hydrolysis unit of coal-fired power plant. The stand-alone and proposed designs are analyzed and compared through thermodynamic and economic methods. Results indicate that the net total energy efficiency increases from 41.84% to 42.12%, and the net total exergy efficiency rises from 41.19% to 41.46% after system integration. Moreover, the energy efficiency and exergy efficiency of waste-to-energy system are enhanced by 10.48% and 9.92%, respectively. The dynamic payback period of new waste-to-energy system is cut down from 11.39 years to 5.48 years, and an additional net present value of $14.42 million is got than before.

MEMS-based LIDAR has the advantages of fast-scanning, simple structure, small volume, and light weight, which make it a hot topic for 3D imaging. In order to overcome the limited scanning angle of the current MEMS mirrors, a large angle MEMS-based optical scanning system was designed using telephoto structure. In this design an f-θ lens group is used for flat field scanning and a symmetrical cemented lens group is used for angle expansion. The principle of this system is discussed in detail and an optical design is implemented by ZEMAX software. The designed system realized a scanning angle as large as 57° and is in excellent linear relation with the tilt angle of the MEMS mirror. At the target distance of 100 m, the maximum RMS radius of the light spot is only 7 cm. This optical scanning system shows a significant application potentiality for 3D imaging LIDAR.

In recent years, much progress has been made on the development of metal mirrors based on additive manufacturing (AM). The sandwich mirror is well known for its excellent mechanical properties and challenging machining. Now, AM can be used to fabricate this complex structure and reduce the processing time and cost. In addition, with the aid of some new design methods for additive manufacturing, such as lattice, topology optimization (TO), and Voronoi, the freedom of mirror structure design is enormously improved. The common materials of mirrors include ceramics (SiC), glasses (glass ceramics, fused silica), and metals (aluminum, beryllium). Among them, the AM technology of metals is the most mature and widely used. Researchers have recently extensively developed the new-generation metal mirror to improve performance and lightweight rate. This review focuses on the following topics: (1) AM technologies and powder materials for metal mirrors, (2) recent advances in optomechanical design methods for AM metal mirrors, (3) challenges faced by AM metal mirrors in fabricating, and (4) future trends in AM metal mirrors.

In this paper, a shallow–deep feature fusion (SDFF) method is developed for pedestrian detection. Firstly, we propose a shallow feature-based method under the ACF framework of pedestrian detection. More precisely, improved Haar-like templates with Local FDA learning are used to filter the channel maps of ACF such that these Haar-like features are able to improve the discriminative power and therefore enhance the detection performance. The proposed shallow feature is also referred to as weighted subset-haar-like feature. It is efficient in pedestrian detection with a high recall rate and precise localization. Secondly, the proposed shallow feature-based detection method operates as a region proposal. A classifier equipped with ResNet is then used to refine the region proposals to judge whether each region contains a pedestrian or not. The extensive experiments evaluated on INRIA, Caltech, and TUD-Brussel datasets show that SDFF is an effective and efficient method for pedestrian detection.

Flow characteristics and pollutant dispersion characteristics of a mechanical upper vent of a tunnel under various discharge patterns and ambient wind were studied by using the computational fluid dynamics (CFD) numerical method and an experimental model. On this basis, variations in the environmental impact radius rcriti changing with the ambient wind velocity U for pollutants from vertical and horizontal outlets were also analyzed with the set discharge capacity. According to our findings, the pollutant emission of the upper vent under a vertical discharge pattern can be considered as free buoyant jet diffusion under the impact of the momentum of ambient wind, while the emission under the horizontal discharge pattern can be regarded as semi-confined buoyant jet diffusion under the impact of the momentum of ambient wind. Moreover, the rcriti of the vertical outlet increased with the increase of U when U ≤ 3 m/s and was minimally affected by U for U > 3 m/s. Additionally, the rcriti of the horizontal outlet decreased with the increase of U. When the discharge capacity was constant, the rcriti of the upper vent under both vertical and horizontal discharge patterns declined with the increase of the discharge velocity V0. Note that the rcriti of the vertical outlet was smaller than that of the horizontal one for the range of parameters considered. These results can provide a theoretical reference for optimizing the design of a vent on the top of an urban tunnel.

The current status of the research of Ballistocardiography (BCG) and Seismocardiogram (SCG) in the field of medical treatment, health care and nursing was analyzed systematically, and the important direction in the research was explored, to provide reference for the relevant researches. This study, based on two large databases, CNKI and PubMed, used the bibliometric analysis method to review the existing documents in the past 20 years, and made analyses on the literature of BCG and SCG for their annual changes, main countries/regions, types of research, frequently-used subject words, and important research subjects. The results show that the developed countries have taken a leading position in the researches in this field, and have made breakthroughs in some subjects, but their research results have been mainly gained in the area of research and development of the technologies, and very few have been actually industrialized into commodities. This means that in the future the researchers should focus on the transformation of BCG and SCG technologies into commercialized products, and set up quantitative health assessment models, so as to become the daily tools for people to monitor their health status and manage their own health, and as the main approaches of improving the quality of life and preventing diseases for individuals.

At present, millimeter wave radar imaging technology has become a recognized human security solution in the field. The millimeter wave radar imaging system can be used to detect a concealed object; multiple-input multiple-output radar antennas and synthetic aperture radar techniques are used to obtain the raw data. The analytical Fourier transform algorithm is used for image reconstruction. When imaging a target at 90 mm from radar, which belongs to the near field imaging scene, the image resolution can reach 1.90 mm in X-direction and 1.73 mm in Y-direction. Since the error caused by the distance between radar and target will lead to noise, the original reconstruction image is processed by gamma transform, which eliminates image noise, then the image is enhanced by linearly stretched transform to improve visual recognition, which lays a good foundation for supervised learning. In order to flexibly deploy the machine learning algorithm in various application scenarios, ShuffleNetV2, MobileNetV3 and GhostNet representative of lightweight convolutional neural networks with redefined convolution, branch structure and optimized network layer structure are used to distinguish multi-category SAR images. Through the fusion of squeeze-and-excitation and the selective kernel attention mechanism, more precise features are extracted for classification, the proposed GhostNet_SEResNet56 can realize the best classification accuracy of SAR images within limited resources, which prediction accuracy is 98.18% and the number of parameters is 0.45 M.

The multi-blade centrifugal fan is commonly used in modern building ventilation and air-conditioning system. However, it does not readily satisfy the increasing demand for energy saving, high efficiency or noise reduction. Its performance is inherently limited by the geometrical structure of single circular arc blades. Q35-type multi-blade centrifugal fan studied as an example by combining the disturbance CST function to parameterize the blades. The optimization parameter change range is confirmed, and test samples are extracted before establishing an RBF proxy model. The NSGA-II algorithm is incorporated, and multi-objective optimization is performed with flow rate and total pressure efficiency as optimization goals. The results show that the fan performance is effectively improved. At the design working point, the air volume of the multi-blade centrifugal fan increases by 1.4 m3/min; at the same time, the total pressure efficiency increases by 3.1%, and the noise is reduced by 1.12 dB, applying the proposed design. The obtained higher fan efficiency can effectively improve performance of the whole ventilation and air-conditioning system. This novel optimization method also has relatively few parameters, which makes it potentially valuable for designing multi-wing centrifugal and other types of fans, providing a new idea for energy saving and emission reduction design of fan.

In recent years, many groups and institutions have been committed to the research of integrated quantum photonic circuit technologies, of which the key components are waveguide coupled single photon sources. In this study, we propose an on-chip waveguide-coupled single photon source that is easily implemented as the waveguide is directly made from the quantum dot membrane. In order to scatter light out of the on-chip waveguide plane into the detection apparatus, grating output couplers are made at both ends of the waveguide. The photon statistics of the on-chip photon source were investigated by second-order correlation function g(2)(τ) measurements using a Hanbury Brown and Twiss interferometer. From the spectra and cross-correlation experiments by collecting emission at the point of quantum dot and out coupler, the emitting of single photons from the same quantum dot and propagating via the waveguide to the out couplers was confirmed. These results show that we have achieved an on-chip single photon source that is easily implemented and easily integrated into quantum photonic circuits.

Under coupled compression–shear loading, the failure and instability behavior of inclined pillars is different from that of horizontal pillars. To enhance the reliability and accuracy of pillar strength design, the influence of different inclination angles and loading rates on mechanical property and the failure behavior of inclined pillar should be studied. In this paper, the combined compression and shear test (C-CAST) system was developed, and mechanical properties and macro failure behavior of coal samples under different inclination angles and loading rates were studied, and acoustic emission (AE) technology was used to determine the internal cracking mode of the sample. The results show that with the increase of inclination angle, the peak shear stress of coal sample increases gradually, while the peak axial stress and elastic modulus slightly increase first and then decrease, and reach the maximum value at an inclination angle of 5°. Within the inclination angle range of 0°–15°, with the increase of loading rate, the peak axial stress and elastic modulus of coal samples first increase and then decrease, while the loading rate corresponding to peak axial stress and elastic modulus decreases. Within the inclination angle range of 20°–25°, the peak axial stress and elastic modulus of the sample gradually decrease with the increase of loading rate. The failure mode of coal samples changes from tension-splitting failure (0°–5°), tension–shear composite failure (10°) to single shear failure (15°–25°). Meanwhile, the loading rate has little effect on the failure mode of coal samples, but has a significant effect on the failure degree. When the loading rate is 1.0 and 10 mm/min and the inclination angle ranges from 0°–5°, the proportion of tensile crack is significantly greater than that of the shear crack, and tensile failure is the main failure mode; when the inclination angle ranges from 10°–25°, the proportion of shear crack is more than 50% and increases gradually with the increase of inclination angle, and shear failure is the main failure mode. This law is consistent with the macroscopic failure mode of the sample.

The Kaiser effect of rock has been extensively studied due to its wide application in in-situ stress measurement and rock damage quantification. The uniaxial cyclic loading and unloading (UCLU) test is commonly employed to examine the rock Kaiser effect. However, how the two critical parameters, including prescribed stress in the first loading cycle (σA) and loading strain rate (lsr), affect the appearance of the Kaiser effect lacks thorough understanding. We systematically performed UCLU tests on 75 sandstone specimens under 25 combinations of σA and lsr. σA spans from 0.5σc (σc is the uniaxial compressive strength) to 0.9σc, and lsr ranges from 10−5 s−1 to 10−3 s−1, respectively. The acoustic emission characteristics of all the rock specimens are continuously monitored over the entire tests. We find that the Kaiser effect is unanimously observed in the stable crack growth stage, corresponding to the stress levels of 0.5σc to 0.7σc because under a lower stress, the Kaiser effect is easily covered by the acoustic emissions generated by microcrack friction. The loading strain rate also heavily affects the occurrence of the Kaiser effect. When lsr does not exceed 10−4 s−1, the Felicity ratio (FR) rises quickly as lsr ascends, whereas FR increases less pronouncedly once lsr exceeds 10−4 s−1. Therefore, a relatively high loading strain rate, i.e., lsr higher than 10−4 s−1, is suggested to facilitate the appearance of the Kaiser effect.

The ability of the constitutive model to simulate the ratcheting behavior of metastable austenitic stainless steel S30408 is significant to ensure the safety of the liquefied natural gas (LNG) semi-trailer tanks in the lightweight process of the inner containers. This is because the lightweight inner vessels often encounter cyclic stresses due to the road inertia loads together with high mean stresses due to internal pressures. In this study, we performed cryogenic uniaxial tension experiments and a series of ratcheting experiments to investigate the cyclic plasticity behavior of the metastable austenitic stainless steel S30408. Based on the Ohno-Wang II model, we proposed a new cyclic plasticity constitutive model with martensitic transformation, which relates the content of deformation-induced martensite with isotropic hardening and kinematic hardening. The ratcheting behaviors of S30408 were first simulated by the proposed model with the incremental loading method using MATLAB. The results showed that the model could reasonably predict the ratcheting behavior of S30408, and the evolution law of martensite content could well predict the content of deformation-induced martensite. Under the assumption of the von Mises yield criterion and normal plasticity flow rule, we developed a numerical algorithm of plastic strain with the proposed model to implement the finite element calculation of the model. Internal iteration in the numerical algorithm was implemented with the Euler backward method, which calculated the trial strain for each equilibrium iteration using the consistent tangent matrix. With a user subroutine, the proposed model was programmed into ANSYS for a user - executable version. By simulating the uniaxial ratcheting of a S30408 round bar, we found that the calculated results were in good agreement with the experimental results, which promises further applications in the design of structures, such as LNG semi-trailer tanks.

With its superior performance, the unmanned combat air vehicle (UCAV) will gradually become an important combat force in the future beyond-visual-range (BVR) air combat. For the problem of UCAV using the BVR air-to-air missile (AAM) to intercept the highly maneuvering aerial target, an autonomous attack guidance method with high aiming precision is proposed. In BVR air combat, the best launching conditions can be formed through the attack guidance and aiming of fighters, which can give full play to the combat effectiveness of BVR AAMs to the greatest extent. The mode of manned fighters aiming by manual control of pilots is inefficient and obviously not suitable for the autonomous UCAV. Existing attack guidance control methods have some defects such as low precision, poor timeliness, and too much reliance on manual experience when intercepting highly maneuvering targets. To address this problem, aiming error angle is calculated based on the motion model of UCAV and the aiming model of BVR attack fire control in this study, then target motion prediction information is introduced based on the designed model predictive control (MPC) framework, and the adaptive fuzzy guidance controller is designed to generate control variable. To reduce the predicted aiming error angle, the algorithm iteratively optimizes and updates the actual guidance control variable online. The simulation results show that the proposed method is very effective for solving the autonomous attack guidance problem, which has the characteristics of adaptivity, high timeliness, and high aiming precision.

With the development of unmanned combat air vehicles (UCAVs) and artificial intelligence (AI), within visual range (WVR) air combat confrontations utilizing intelligent UCAVs are expected to be widely used in future air combats. As controlling highly dynamic and uncertain WVR air combats from the ground stations of the UCAV is not feasible, it is necessary to develop an algorithm that can generate highly intelligent air combat strategies in order to enable UCAV to independently complete air combat missions. In this paper, a 1-vs.-1 WVR air combat strategy generation algorithm is proposed using the multi-agent deep deterministic policy gradient (MADDPG). A 1-vs.-1 WVR air combat is modeled as a two-player zero-sum Markov game (ZSMG). A method for predicting the position of the target is introduced into the model in order to enable the UCAV to predict the target’s actions and position. Moreover, to ensure that the UCAV is not limited by the constraints of the basic fighter maneuver (BFM) library, the action space is considered to be a continuous one. At the same time, a potential-based reward shaping method is proposed in order to improve the efficiency of the air combat strategy generation algorithm. Finally, the efficiency of the air combat strategy generation algorithm and the intelligence level of the resulting strategy is verified through simulation experiments. The results show that an air combat strategy using target position prediction is superior to the one that does not use target position prediction.

This paper presents a measurement method for high-precision cutting edge radius of single point diamond tools using an atomic force microscope (AFM) and a reverse cutting edge artifact based on the edge reversal method. Reverse cutting edge artifact is fabricated by indenting a diamond tool into a soft metal workpiece with the bisector of the included angle between the tool’s rake face and clearance face perpendicular to the workpiece surface on a newly designed nanoindentation system. An AFM is applied to measure the topographies of the actual and the reverse diamond tool cutting edges. With the proposed edge reversal method, a cutting edge radius can be accurately evaluated based on two AFM topographies, from which the convolution effect of the AFM tip can be reduced. The accuracy of the measurement of cutting edge radius is significantly influenced by the geometric accuracy of reverse cutting edge artifact in the proposed measurement method. In the nanoindentation system, the system operation is optimized for achieving high-precision control of the indentation depth of reverse cutting edFigurege artifact. The influence of elastic recovery and the AFM cantilever tip radius on the accuracy of cutting edge radius measurement are investigated. Diamond tools with different nose radii are also measured. The reliability and capability of the proposed measurement method for cutting edge radius and the designed nanoindentation system are demonstrated through a series of experiments.

In this study, a one-step rapid heating novel process was used to prepare high-purity anhydrous scandium chloride molten salt with low-purity scandium oxide. High-purity anhydrous ScCl3 molten salt was used as the Sc-bearing raw material for preparing the Sc-bearing master alloy. Inert gas was used to enhance the purity of anhydrous scandium chloride and reduce the hydrolysis rate of scandium. The results show that high-purity scandium chloride (purity, 99.69%) with the scandium content of 29.61%, was obtained, and the hydrolysis rate of scandium was 1.19% under the conditions used: removing ammonium chloride; residual crystal water temperature of 400 °C; m(Sc2O3):m(NH4Cl) = 1:2.5; holding-time of 90 min; heating-rate of 12 °C/min; and argon flow of 7.5 L/min. XRD, SEM, and EPMA analyses further verified that anhydrous scandium chloride crystallization condition was relatively good and the purity of high-purity anhydrous scandium chloride approached the theory purity of anhydrous scandium chloride.

The intrusion detection system (IDS) which is used widely in enterprises, has produced a large number of logs named alerts, from which the intrusion patterns can be mined. These patterns can be used to construct the intrusion scenarios or discover the final objectives of the malicious actors, and even assist the forensic works of network crimes. In this paper, a novel algorithm for the intrusion pattern mining is proposed which aimsto solve the difficult problems of the intrusion action sequence such as the loss of important intrusion actions, the disorder of the action sequence and the random noise actions. These common problems often occur in the real production environment which cause serious performance decrease in the analyzing system. The proposed algorithm is based on the online analysis of the intrusion action sequences extracted from IDS alerts, through calculating the influences of a particular action on the subsequent actions, the real intrusion patterns are discovered. The experimental results show that the method is effective in discovering pattern from the complex intrusion action sequences.

With the development of intrusion detection, a number of the intelligence algorithms (e.g., artificial neural networks) are introduced to enhance the performance of the intrusion detection systems. However, many intelligence algorithms should be trained before being used, and retrained regularly, which is not applicable for continuous online learning and analyzing. In this paper, a new online intrusion scenario discovery framework is proposed and the intelligence algorithm HTM (Hierarchical Temporal Memory) is employed to improve the performance of the online learning ability of the system. The proposed framework can discover and model intrusion scenarios, and the constructed model keeps evolving with the variance of the data. Additionally, a series of data preprocessing methods are introduced to enhance its adaptability to the noisy and twisted data. The experimental results show that the framework is effective in intrusion scenario discovery, and the discovered scenario is more concise and accurate than our previous work.

Tobacco (Nicotiana tabacum L.) is a crop that is able to accumulate metals. In this study, humic acid was selected as a Cd, Cu, Pb, and Zn passivator, and added to calcareous field soil in amounts of 6.4, 10.3, and 14.8 kg·ha−1. Its impact on the soil fractions of the metals in the soil was extracted by the Community Bureau of Reference (BCR) sequential extraction method, and their accumulation of the metals in tobacco leaves was investigated. Application of 14.8 kg·ha−1 humic acid decreased the DTPA-extracted concentrations of Pb, Cd, Zn, and Cu by 39%, 37%, 29%, and 18%, respectively, as compared with untreated soil. The fractions of Cd, Pb, Cu, and Zn in soil were extracted by the BCR sequential extraction method, and the relationship between the difference metal fractions in the soil and the metal contents in the plant materials were analyzed. The exchangeable fractions of Cd, Pb, Cu, and Zn and the reducible fractions of Pb and Cu are the main bioavailable fractions. Additionally, the reducible fractions of Cd and Zn, the oxidizable fractions of Pb, Cu, and Zn, and all residual fractions of metals were nonbioavailable fractions in the soil. The soils were treated with humic acid (HA) to shift bioavailable metals to stable phases that were less bioavailable. The available Cd, Pb, Cu, and Zn were strongly retained in the soil after the application of humic acid, which decreased the uptake in tobacco in the upper, middle, and lower leaves.

Due to the light scattering effect, it is difficult to directly achieve optical focusing and imaging in turbid media, such as milk and biological tissue. The turbidity suppression of a scattering medium and control of light through the scattering medium are important for imaging on biological tissue or biophotonics. Optical phase conjugation is a novel technology on turbidity suppression by directly creating phase conjugation light waves to form time-reversed light. In this work, we report a digital optical phase conjugation system based on off-axis holography. Compared with traditional digital optical phase conjugation methods, the off-axis holography acquires the conjugation phase using only one interference image, obviously saving photo acquisition time. Furthermore, we tested the optical phase conjugate reduction performance of this system and also achieved optical focusing through the diffuser. We also proved that the reversing of random scattering in turbid media is achievable by phase conjugation.