This PDF file contains the front matter associated with SPIE Proceedings Volume 12161, including the Title Page, Copyright information and Table of Contents

In this paper, a point cloud extraction method based on short-range vehicle millimeter wave radar is proposed. This method is based on Cluster CLEAN algorithm. Compared with the traditional peak extraction algorithm, it can obtain more abundant point cloud data, which is conducive to the subsequent target classification. In this paper, the Cluster CLEAN algorithm is improved, and a improved incremental Density-Based Spatial Clustering of Applications with Noise (DBSCAN) algorithm is proposed to improve the clustering process of the Cluster CLEAN algorithm. It is not necessary to repeatedly Cluster the past scattering point data in each iteration, and the computational efficiency of the algorithm can be improved by more than 98%. In addition, in this paper, a modified order statistics based multidimensional clustering (OSMC) algorithm is proposed to directly Cluster the clustering results of Cluster CLEAN algorithm in angle dimension, without the need to perform range-velocity-angle clustering after angle dimension estimation, which accelerates the efficiency of the algorithm and reduces the redundancy of the algorithm. It is verified that compared with the traditional peak extraction algorithm, the signal processing algorithm in this paper can increase the number of scattering points from the same target by 2-7 times while maintaining real-time performance.

Large-scale complex information system software is large in scale and high in complexity. Conventional testing methods will consume huge manpower and material resources. However, there are some similarities among system software defect generation, failure mode triggering, and characterization phenomena. In this paper, a general method and process of software defect prediction is proposed by using machine learning method, which can shorten the test cycle and save a lot of test resources when applied to practical projects.

The progress of science and technology leads to quick replacement of products and causes a large number of abandoned products. As one of the important steps of product recycling, disassembly can maximize the utilization of resources. Considering that most of the existing disassembly lines are operated by humans, it is of great significance to consider the change of human body posture in the disassembly process. This work divides the disassembly posture into the standing posture operation and the sitting posture operation. Then, on the premise of satisfying various constraints, a mathematical model that maximizes the profit and minimizes the number of postural changes is established. A new multi-objective discrete harmony search algorithm based on Pareto is proposed. Compared with the original harmony search algorithm and multi-objective evolutionary algorithm based on decomposition, the results show that the proposed algorithm has better performance in terms of efficiency and quality.

With the widespread application of UAV (Unmanned Aerial Vehicle), the management and control of the UAV’s black flight has been an urgent task for city managers. Navigation spoofing technology is an effective method for the management. Therefore, this paper summarized the navigation spoofing methods of the UAV and the UAV group. Firstly, the background of navigation spoofing is introduced. Secondly, the basic principles and classification of navigation spoofing is introduced. Then, according to the order from UAV to UAV group, the current research status of navigation spoofing is introduced. Finally, in view of the current deficiencies in the research on navigation spoofing, we can conclude that UAV group navigation spoofing is the future research direction

Rapid-exploration Random Tree (RRT) is an efficient algorithm to search non-convex and high-dimensional spaces via randomly constructing spatial filling trees. This algorithm has been widely used in autonomous robot path planning. However, the basic RRT algorithm has some shortcomings. In order to improve the defects of low search efficiency and poor path quality of the RRT algorithm, this paper proposes an A* based RRT path planning algorithm with the advantages of completeness and optimality of the A* algorithm and fast extensibility of the RRT algorithm. During the procedure of random node sampling of the RRT algorithm, A* path is used to formulate the sampling strategy. Meanwhile, the constraint of the path turning angle is added to the nearest neighboring search of the RRT algorithm, which can enhance the rationality of the search tree node selection and improve the obtained path quality. Simulation experiments have been performed to verify the effectiveness of the proposed method for unmanned aerial vehicle path planning.

Traditional beamforming algorithms are hard to be applied to distributed arrays which have special array structures. An improved virtual beamforming algorithm based on non-uniform sparse optimization of multiple transform regions is proposed in this paper. Through this algorithm, adaptive beamforming for distributed arrays could be obtained while the main lobe position is not shifted and the nulls are deepened.

Automatic seizure detection system can greatly reduce the burden of manual diagnosis on epilepsy. In this paper, an epileptic seizure detection method is proposed based on transfer learning of VGGNet-16 and gated recurrent unit network. Evaluated on CHB-MIT EEG dataset, the proposed detection method achieved an average sensitivity of 90.12%, an average specificity of 96.32% and an average accuracy rate of 96.31%. A comparative experiment based on the transfer learning of Resnet50 further demonstrates the good performance of VGGNet-16 in seizure detection

Aiming at the problem that the abnormal values of the measurement noise appear when GPS signals become weak or disappear in the urban environment, which reduces the positioning accuracy of the INS/GPS tight coupled navigation system, an improved adaptive filtering algorithm based on the adjustable Student-t distribution is proposed. This method uses Student-t distribution to model the measurement noise, the Mahalanobis distance of innovation vector to adjust filter’s adaptation, and the variational Bayesian method to better track changes in measurement noises. Experimental results show that the method can achieve a more robust estimation result and a better positioning effect in an urban environment.

Distributed networks in an insecure environment may have malicious attack nodes, which will tamper with their observation and communication data and affect the accuracy of parameter estimation. This paper proposes an improved Diffusion Least Mean Square (DLMS) algorithm with an adaptive punishment mechanism under malicious attacks. The algorithm uses median filtering to process the received neighbor node estimates, effectively filtering malicious nodes that have a greater impact. The location of malicious nodes is detected by designing an adaptive threshold. The combination weight of malicious nodes will be correspondingly reduced due to the existence of the punishment factors, thereby effectively weakening the impact of malicious nodes on network attacks. The simulation results show that the algorithm is robust under malicious attacks, can effectively resist data tampering attacks, and improve the overall estimation performance of the system.

During the Eleventh Five Year Plan period, China has included the "ruins search and rescue robot" project in the National 863 key project, which is jointly developed by the State Key Laboratory of robotics of Shenyang Institute of automation, Chinese Academy of Sciences and China Earthquake Emergency Search and rescue center, and has successfully developed the "waste ruins deformable search and rescue robot, robotic life detector and rotor UAV" Three robots. These three robots have been rated by the State Seismological Bureau as one of the 10 most effective scientific and technological achievements since the eleventh five year plan. In this paper, a multi-functional mine search robot scheme based on inertial navigation is proposed. The main controller uses ten axis gyroscope and GPS to establish a three-dimensional geospatial coordinate system and record the search trajectory. The collapse terrain data is obtained through 3D visual camera and radar, integrated into the three-dimensional geospatial coordinates to obtain a three-dimensional terrain space model, and infrared sensors search for the victims, Mark it into the model and transmit the data to the host computer through industrial wireless module. The slave controller can collect the voltage and current data in the voltage and current transformer, measure the data of cable current in each area under the mine, and judge whether the power supply in the local area is normal, which can be used as one of the parameters to judge whether it collapses. Search and rescue personnel can formulate effective rescue plans according to the directions and partial characteristics of the victims displayed by the host computer, so as to improve the efficiency of rescue operations

The bursting rate of bubbles is an indicator of bubble stability. To accurately recognize burst bubbles, the Kinect sensor is first introduced to the froth flotation industrial site, then the bubble depth information is collected accurately. Due to the bubble burst is close related with depth change, depth feature is extracted by our proposed method based on depth difference. The proposed method filters potential burst bubbles and registers single bubble region based on color data, and then recognizes bubble bursts in terms of depth data. the experimental results show that our proposed method can effectively identify bubble bursts, the precision and recall reach 0.8547 and 0.8963 respectively.

Disassembly line plays a crucial role in the recycling of end-of-life products, which can effectively reduce the pressure of resource shortage. Considering the development of intelligent plant, this paper studies the human-robot collaborative disassembly line balancing problem with the optimization objectives of maximizing total profit and minimizing energy consumption. The disassembly process is specified with the AND/OR graph model. In addition, a Pareto improved multiobjective shuffled frog leading algorithm is proposed, which introduces an elitist strategy to improve the searching ability. Finally, the proposed model and algorithm are applied to instances of human-robot collaborative disassembly lines. Through different comparison experiments with the nondominated sorting genetic algorithm II and harmony search, the superiority of the proposed algorithm in performance and quality is verified.

Time-sensitive network technology is a key part of deterministic networks and one of the key technologies for 6G networks. The development of time-sensitive network technology and the security issues and security techniques it faces deserve a summary of our in-depth research. In this paper, we understand and summarize the main technical development, application scenarios and requirements of time-sensitive networks, mainly from two aspects of data flow shaping and scheduling strategies and their related security issues, and investigate the related security technologies in TSN and deterministic networks. It aims to promote the in-depth research in this field and contribute to the development of timesensitive network technology.

The auto-commissioning system of engineering equipment requires real-time identification of the working mode with the whole product. Determine whether there is any abnormality in the commissioning process according to the parameter standards of each working mode, to evaluate the compliance rate of the commissioning. In this article, the working mode of engineering equipment is determined by Toeplitz Inverse Covariance-based Clustering (TICC) and the time series data are segmented and clustered. Each mode in the TICC is defined by a Markov Random Field (MRF), which characterizes the interdependence between different factors in the mode. On this basis, highlight the characteristics of the model and the importance of each factor. By using the TICC method to commissioning the concrete pump truck, the results of automatic identification show that the method has high accuracy in the recognition of the working mode of engineering equipment. The successful application of the online commissioning and monitoring system of the whole machine based on this technology provides a new idea for the development of the industry.

With the development of information technology, robot technology has made great progress in various fields. These new technologies enable robots to be used in industry, agriculture, education and other aspects. In this paper, we propose a drum robot that can automatically complete music transcription in real-time, which is based on AIoT and fog computing technology. Specifically, this drum robot system consists of a cloud node for data storage, edge nodes for real-time computing, and data-oriented execution application nodes. In order to analyze drumming music and realize drum transcription, we further propose a light-weight convolutional neural network model to classify drums, which can be more effectively deployed in terminal devices for fast edge calculations. The experimental results show that the proposed system can achieve more competitive performance and enjoy a variety of smart applications and services.

In the traditional methods for wisdom of crowd, the measure and improvement of crowd consensus is very important. In the process of crowd consensus measure and improvement, it is necessary to modify the decision-maker information that does not meet the consensus threshold. And in the traditional methods for wisdom of crowd, decision-makers do not exchange information and communication. In order to avoid the above problems, this paper proposes the method for wisdom of crowd based on the evolution of decision makers in social network. The method is mainly divided into three parts: Firstly, the decision-makers compare the alternatives in pairs according to their own knowledge and experience to get the evaluation results. Secondly, the bisecting K-means clustering algorithm is used to group the alternatives according to the evaluation results of the decision makers, and the French-Harary-DeGroot model is used to simulate the opinions of the decision makers in the crowd. After negotiation, the consensus is reached. Finally, the opinions of each group are gathered and the best alternative is selected. At the end of the paper, the experiment is used to prove the effectiveness of the proposed method.

Aimed at increasing the customer satisfaction of yard warehousing, a model of crane scheduling that is used to minimize the total level of disappointment of all customers is established, and resolved by an improved dragonfly algorithm (IDA) proposed, which overcomes the premature convergence and slow convergent speed of standard dragonfly algorithm by introducing Gaussian neighborhood radius and multiple mating operator. They can balance the global exploration and the local exploitation. Compared with other popular algorithms, simulation results demonstrate that IDA achieves high search efficiency and accuracy.

This paper studies the large-scale management, large-concurrency, and high-reliability of the Internet of Things platform for heterogeneous terminals in the power grid. It sorts out the management framework of different intelligent terminals, intelligent gateways and other heterogeneous terminal equipment in professional fields such as power transmission, power transformation, and distribution, and relies on flexible expansion. Strategies and technologies to meet the management needs of the Internet of Things platform for large-scale terminals; study the container management technology of smart terminals and smart gateways based on the Internet of Things platform, and propose the Internet of Things for the information collection and equipment management requirements of smart terminals and smart gateways by containers. The management technology of micro-applications between the platform and smart terminals and smart gateway containers; research the OTA operation and maintenance service capabilities such as terminal file distribution, terminal file encryption, and terminal update plan release of the Internet of Things platform, and realize the remote upgrade of smart terminal and smart gateway device software.

With the progress of the times and the development of science and technology, Internet of things technology has been gradually applied in many business fields and urban construction in China. The development and maturity of Internet of things technology based on big data has further promoted the development of urban construction in China and made people's life more convenient. With the gradual improvement of people's living standards, food safety has become the focus of people's concern. The research on the application of Internet of things technology can promote the application of Internet of things technology in work and life to be more scientific and extensive. This paper introduces the current situation of food traceability and the application of Internet of things technology in food safety management under the background of big data, and makes relevant analysis and discussion, in order to improve food safety problems in the future and provide a reference for relevant researchers. In addition, this paper also discusses the application of Internet of things technology in logistics management, geological exploration, medical informatization, environmental security and smart city construction, and looks forward to the conclusion, hoping to provide a reference and ideas for the application and development of Internet of things technology

Autonomous driving technology is an important direction in vehicle engineering research, and acceptance, as a reflection of public attitudes toward vehicles, is the basis for promoting the marketing of the technology. The results of recent acceptance surveys have gradually decreased, and acceptance may be related to recent traffic accidents. To explore the influencing factors of declining acceptance, the authors introduced the safety risk factors of autonomous driving and proposed an improved model based on the combination of TPB and TAM for acceptance interpretation and prediction. By analyzing the model and data results statistically, it is verified that safety risk has an impact on acceptance and a more significant impact on actual behavior. It is concluded that people's misunderstanding of unreasonable factors or irrational factors about the safety of autonomous driving technology needs to be eliminated in order to improve public acceptance and promote the accelerated implementation of autonomous driving technology.

There are more than 100000 certificates need to issue in the metrology technical institutions every year. The traditional signature mode on certificate is not only time-consuming and laborious, but also difficult to confirm the seal type according to the authorization ability. In view of this problem, based on the existed measurement management platform, an electronic certificate signature system based on the contract lock electronic signature platform is proposed. The necessity and application scenarios of integrated electronic signature function, and the main realization process are studied and analyzed. The development and test are carried out under the environment of .NET-based development environment with COM interface. The functions of automatic adding electronic signature and verifying authenticity are realized. It is of great significance for the metrology technical institutions to improve service ability and optimize business environment.

The action recognition method based on Wi-Fi is one of the research focuses in recent years. Existing methods have problems with complex noise processing and weak classification ability. To solve the limitation of existing human action recognition methods, this paper proposes a human action recognition solution that is a method based on Wi-Fi signal and the residual network (Wi-Res). First, the method captures human motion information (running, walking, sitting, standing, jumping) using CSI (Channel State Information) signals and constructs motion templates based on time series. Second, this work uses the obtained motion templates as basic input units. Third, the Wi-Res extracts feature through convolutional neural networks and adds residual connections to solve problems, such as gradient disappearance. The method can describe motion on a longer time scale. The human action recognition method based on the residual network receives an accuracy rate of about 97% for human action recognition when tested in indoors.

Bronze inscription is one of the earliest well-established writing systems dating back to Shang dynasty in China. The Recognition of Bronze character recognition plays an important role in the identification and interpretation of Bronze inscription which traditionally is a tough and challenging task. To deal with class imbalance of training data in bronze inscription recognition, we propose a method based on few-shot learning. The recognition process consists of three stages. In the first stage, a model is pretrained in a large-scale character dataset with a novel negative margin loss. In the second stage, the pretrained weights of the backbone network is transferred to the target dataset. In the final stage, the distribution of few-shot classes is calibrated and a new classifier is re-trained accordingly. Through qualitative and quantitative experimental analyses, the proposed method exceeds the state-of-the-art on our Bronze Character dataset.

Following the development of research on machine learning in general, feature selection, as a process utilized in machine learning, has also gained some attention. As a common process in machine learning, it would be normal to be enhanced on, just to increase the overall efficiency of the processing. However, to achieve that, one must gain a deep understanding of the topic itself first. In this paper, we will offer a general and comprehensive review and analysis for feature selection. Stating from three basic methods of feature selections: Wrapper, Filter, and embedded, we introduced how those methods work and analyze them as well. Then, we started introducing the steps necessary to perform feature selection: Generation Procedure, Evaluation Function, Stopping Criterion, and Validation Procedure. Finally, we would review and analyze a few recent developments and important concepts related to feature selection: MIC Formulation, Distance Correlation, Model-Based Ranking, Recursive Feature Elimination, and Laplacian score. Those are the methods that represent the recent development of feature selection. This paper is a useful review and reference for people who just started entering the field of feature selection and machine learning in general. It may help one understand some important concept from the ground in feature selection or machine learning in general.

Convolutional neural network is widely used in the field of image denoising, and FFDNet model has excellent performance in the field of image denoising. The denoising of remote sensing image is also one of the most basic preprocessing methods of remote sensing image. In this paper, FFDNet model is applied to remote sensing image denoising. Select a remote sensing image data set (UC combined land use data set), replace the natural noise with additive Gaussian white noise (AWGN), process it with different methods, compare it with DnCNN and VDNet, and analyze the comparison results. FFDNet has better performance.

This paper proposes an improved electrical betweenness method to identify vulnerable lines in the power grid. This method is based on the line power flow obtained after power flow calculations, which overcomes the limitation of the previous methods assuming that electrical power flow flows along the shortest path, and simplifies the steps of each unit power injection. Taking into account the directionality of the electrical power flow of the power grid, the proposed percentage of power loss flow is used as an indicator to evaluate the stability of the power grid, and the dynamic impact on the power grid after the removal of generators and load nodes are cut off. Simulation analysis on IEEE 39 and 118-bus system shows that the total active power flow of the grid will decrease rapidly if the high improved electrical betweenness is deliberately attacked, which verifies the feasibility and effectiveness of the method to identify vulnerable lines.

This paper created Structure-Behavior Coalescence (SBC) State Analysis Ontology (SAO) for model-based systems engineering. Since the state analysis capability has been utilized in SBC state analysis ontology, the mapping from the ontology to the systems model can be effectively achieved.

Aiming at the problems of error and poor recognition effect in the data collected by sensors in current motion posture recognition, this paper proposes a self-attention network for motion posture recognition based on data fusion to solve these problems. This method uses the attitude angle information output by the gyroscope to correct the attitude angle obtained by the acceleration sensor using kalman filtering, which effectively improves the accuracy of the attitude angle; at the same time, the attitude angle and acceleration sensor data are used to construct an attention convolutional neural long short term memory artificial neural network (CNN-LSTM) of the attention mechanism to recognize the motion state. The experimental results show that the use of data fusion method can correspond to the accuracy of physical signs, and compared with the traditional network, the accuracy of the network frame recognition proposed in this paper is improved

In practical radar detection applications, due to the limitation of the beam width of the pattern, limited field of view (FOV) lacks the overall perception ability of the area of interest (AOI). Especially, when unknown and time-varying targets appear in AOI, it can easily lead to missing even wrong tracking of key objects. In view of the above problems, the radar network is adopted to fuse the observation data of limited multi-view to obtain the global field of view information, and then realize the trajectories estimation of multi-object in the fusion center. Based on FInite Set STatistics (FISST) framework, mapping the newborn and death process of multiple targets within FOVs as multi- Bernoulli process, the posteriori density of multi objects is propagated recursively followed Bayesian criterion in time. The simulation results of multi-object trajectories estimation with four kinds of multi-Bernoulli (MB) filters are given under three scenarios, which illustrates that the number of interest objects and the accuracy of trajectories estimation are improved, along with the increase of the number of local observation fields of view. Furthermore, the tracking performance of labeled multi-Bernoulli (LMB) filter is superior to that of unlabeled filter.

In this paper, we designed a dynamic route planning scheme for scenic spots. Specifically, in this scheme, the initial value of each scenic spot is determined in the scenic spot. Then, the map containing the relative distance data between the scenic spots in the scenic spot is processed to obtain the path information between the scenic spots. The tourist’s time requirements are obtained for the scenic area visit in order to get the default optimal route. Furthermore, the default optimal path is obtained to meet the expectations of tourists. Further, based on the current time point, the precise value of the remaining scenic spots in the tourist's selection requirements, and the path information between the remaining scenic spots, the optimal value solving operation is performed to obtain the optimal path for the current time point. Based on the known data, combined with the multi-definition algorithm to solve the optimal path, the best tour route matching the current location can be obtained in real time according to the actual tour mode of the tourists, thereby avoiding time waste and improving the tour experience.

The identification of minerals in rocks from thin section images is a basic task of geoscience. Compared with traditional manual interpretation, machine recognition is widely used in mineral classification for its advantages of speed and objectivity. It is an important scientific issue to choose which mineral feature to use for automatic classification. Based on this, the texture features of thin mineral images were specially studied in frequency domain. Firstly, the primary texture classification variables were obtained by simulating the radial statistical analysis of images and mineral samples; then, the separability was verified by variance analysis, and the variables were combined based on the factor analysis method; lastly, classification verification of mineral samples was carried out by discriminant analysis. The experimental results show that the low frequency information accounts for about 95% of the energy in the sample spectrum, and the classification efficiency is significantly higher than the test threshold. The total classification accuracy of Texture Contour Factor (TCF) and Texture Detail Factor (TDF) is 89.6%, which is obtained by factor analysis. The results show that the frequency features in the thin section mineral images can effectively reflect the changes of mineral texture and have a good effect on the automatic classification of mineral images.

The redundant information contained in feature can be reduced and the accuracy of data analysis is improved via extracting the features from the data set. The existing methods to extract the feature ignoring the information contained in the data vector of feature. In this paper, the similarity between data features is firstly calculated via multiple methods to form the similarity vector of feature. Then the adaptive weighted clustering ensemble is proposed to cluster the similarity vector of feature to partitioning the feature subspaces. Secondly, utilizing the characteristics of the data vector of feature, the weights of the features in the subspace are calculated, and then the effective features are extracted using the linear weighted method. In the experiments, the result shows that the proposed method can significantly improve the accuracy of the data analysis.

When the traditional ORB algorithm is applied in the field of image mosaic, it is susceptible to the interference of light factors and cannot extract high-quality features, which leads to the inaccuracy of the perspective transformation matrix and the problem of poor mosaic effect. In response to this problem, this paper proposes an improved Unsuperpoint feature extraction network to apply to image stitching. By designing the backbone network of the unsupervised model, the points of interest and the descriptor loss function are optimized to improve the efficiency of the network and the accuracy of feature point extraction. Compared with the traditional ORB algorithm, the image stitching algorithm in this paper reduces the image matching time by half and increases the matching accuracy by 7%.The images stitched by this method avoid the phenomenon of cracks and black lines at the joints, and the image transitions are natural and high in definition.

Helmet wearing is a major concern for the safety and protection of people on the construction site. Statistic data demonstrate that injuries and accidents occur mainly due to not following prescribed procedures, i.e., not wearing helmet. Camera-based surveillance system can conduct online monitoring task to detect such abnormalities through captured images with image processing system analysis. Although deep learning-based method can achieve higher image identification performance, it requires extensive hardware support of the computational resources. Therefore, it is imperative to design a lightweight network with lower hardware requirement to address such problem. In this paper, a GhostNet, YOLOv5 and a lightweight network are combined to design a model to analyze the image for online monitoring with faster processing speed. The performance of the proposed model is compared with those of the mainstream lightweight models. Experimental results have demonstrated that the proposed model has higher detection accuracy and flexible adaptability.

There is an urgent need for efficient magnetic tile image classification methods in the complex industrial background. Broad learning system (BLS) is a network structure that does not require depth. In order to further reduce the number of parameters that need to be calculated and improve the training efficiency of the network, the data set is enhanced and then the principal component analysis (PCA) is performed. Compared with broad learning, the method we proposed reduces the dimensionality of the data set and shortens the training and testing time from 2.93 to 0.63s while the classification accuracy is not much different.

The light microscope image of glioblastoma cell line A172 has the characteristics of low contrast and high density. An improved algorithm based on Mask R-CNN is proposed. The residual neural network of this algorithm is introduced into deformable convolution to enhance the segmentation ability of multi-cell shapes, and at the same time, based on the feature pyramid network, the high-level semantic structure information is transferred to the bottom layer to form dense connections to adapt to the detection of dense feature images. This instance segmentation algorithm has been verified in the human glioblastoma A172 cell line in the LIVECell-2021 data set. Comparative experiments show that our method performs better in COCO evaluation indicators and visual segmentation effects. Among them, in terms of detection performance, AP increased by 0.319%, AP50 increased by 0.107%, and AP50 increased by 0.136% in segmentation performance.

The accuracy and reliability of the geodetic GPS receiver is directly related to the quality of the project. The article refers to the JJF 1118-2004 Global Positioning System (GPS) receiver (geometric and navigation type) calibration specification to discuss and study the influence of the GPS phase center Consistency and factors of measurement error. Through experiments and theoretical derivation, a method for accurately calculating the consistency of the phase center is obtained.

The ionosphere is the main error affecting the positioning of the global navigation satellite system. When the ionosphere is strongly disturbed, the ionospheric electron density gradient and its irregularity will seriously affect the overall operation performance of the positioning system and various radio communication systems. As one of the most severe surface weather changes, it is particularly important to study the ionospheric disturbance characteristics during typhoon. Based on the ionospheric Total Electron Content (TEC) data provided by China Earthquake Administration, this paper explores the ionospheric disturbances in Guangdong and Guangxi during the super typhoon Mangkhut 2018. The moving quartile method was used to detect and analyze the ionospheric TEC during the typhoon. The results show that the super typhoon Mangkhut produced abnormal disturbances on the ionospheric TEC in Guangdong and Guangxi from generation to extinction, and produced about 2 TECu negative abnormal disturbances on the day of typhoon generation, and about 2 TECu positive abnormal disturbances on the second day of typhoon upgrading to super typhoon (doy255). Moreover, on the day of the typhoon’s demise, there was a relatively obvious negative anomaly disturbance to the ionospheric TEC near it.

With the development and application of new technology and new system, the important field of ship radar has been greatly developed. Taking the current situation of domestic ship radar as the research background, this paper briefly describes the development status of ship radar, discusses the integration path between radar and related equipment through the analysis of the composition of relevant ship instruments and equipment, introduces the current situation of data fusion between ship radar and ARPA, AIS and ECDIS systems, and explores the possibility of ship equipment integration from two aspects of hardware and software, The necessity of integrated development of marine instruments is pointed out.

With the popularity of electric vehicles, the charging load increases gradually. A large number of disorderly connected electric vehicles will have a great impact on the power grid. Electric vehicle is a kind of schedulable energy storage resource with the dual characteristics of power supply and load, which has the ability of active regulation. In view of the above problems, using multi-agent consensus algorithm, taking the incremental cost of generator sets and the incremental benefit of aggregate electric vehicle as consensus variables, a strategy for aggregate electric vehicles to participate in power system economic dispatch is designed, and the economic dispatch problem is solved by distributed optimization. An example simulation is carried out on the IEEE 39 bus system, and the aggregate electric vehicles are connected to the power grid step-by-step to verify the effectiveness of the proposed strategy.

Design and implement a high-performance absolute-value detector through the CMOS process, and realize the function of taking the absolute value of the input data and then comparing it, which can be applied in a variety of circuits, such as A/D conversion. The circuit structure is simple, which adopts the structure of the mirror adder and the multiplexer. This paper also looks for the critical path and uses logical effort theory to find the minimum delay. The next work is to adjust the gate size and determine the optimal size of the device and the optimal power supply voltage. Finally, this paper attempts to achieve the minimum energy consumption.

An optical frequency-domain reflectometry (OFDR) based on non-tunable laser is presented in this article. Instead, a common narrow-linewidth laser modulated by an external phase modulator is used for source. By swiftly changing the modulation frequency thus sweeping the frequency of modulated light, the linear chirp probe light needed in OFDR can be made. Comparing to the common tunable laser, this external modulation scheme has a better repeatability. The corresponding experimental system is constructed to verify the system performance, and the detection distance is longer than 4km. Firstly, the end and junction reflection of the optical fiber at 4.1km was detected. Secondly, 30kHz vibration signal is detected at this position. Finally, the stress near this position is detected.

Material transfer and material status monitoring have always been an important part of operation and maintenance in the workshop. Too much reliance on manual operation will greatly reduce work efficiency and cause frequent errors. This article is based on Residual Neural Network (ResNet) transfer learning (TL) for model training. The status of material points in the workshop, namely empty frame, no frame, and full-frame, has been well recognized by using a small amount of surveillance video stream data for image analysis, which realizes the reverse optimization of the model. The accuracy of material point status recognition is as high as 99.7%. Based on the material point status recognition technology, the manufacturing operation management system interface can be called through the HTTP protocol to issue tasks, and the intelligent logistics system can be combined to realize the automatic circulation of materials in the workshop and improve production efficiency.

This paper proposed a novel dynamic event-triggered integral sliding mode control (ISMC) method for the consensus tracking problem of multi-agent systems (MASs) with disturbances. A mixed event-triggered scheme with two different dynamic trigger parts is addressed with the ISMC controller to avoid data consumption and reduce the trigger times. Two types of integral sliding mode switching functions are developed to analyze the reachability of sliding mode surface and design the dynamic event-triggered integral sliding mode controller. Finally, the validity of obtained theoretical results are verified by simulation.

Microchips are widely used in a series of electronic products, such as integrated circuits, consumer electronics, industrial power supplies, household appliances and many other fields. To achieve the function of detecting and conversing electrical signals, it is of significance to design an absolute value detector. Compared with dynamic digital circuits, traditional analog circuits have the poor signal capability, along with large transmission delay and huge power consumption. Therefore, the content of this research is to modify a complementary metal-oxide semiconductor (CMOS) absolute value detector circuit by transforming it to a domino circuit and gate sizing. To verify the feasibility of the circuit, this research compares the performance of the modified circuit with the traditional circuit in terms of delay and energy, with methods mainly focusing on simulation and theoretical calculation.

This article studies the performance of different CMOS comparators. Comparators play an important role in signal processing and other electronic applications. Its speed, energy-efficient and accuracy are largely depending on its noise, kickback and sensitivity. A dynamic bias comparator adds a tail capacitor to store the energy. However, it brings relatively high time delay. A Triple-Tail fully dynamic comparator separates the comparing option into three stages to minimize the resolving time. An improved Triple-latch feed-forward (TLFF) fully dynamic comparator adds extra feed-forward and backward paths to reduce the delay. An innovation Edge-Pursuit Comparator (EPC) adopts a ring structure, allowing two input signals to chase each other and achieves extremely low energy consumption. The dynamic Floating Inverter Amplifier (FIA) comparator combines strongARM latch with the preamplifier stage and achieves high speed and low power consumption. In this article, the parameters and performance of these comparators have been analyzed and compared. Further conclusions are made.

With the continuous increase of signal transmission rate, the impact of soldering transition section between surface mount technology (SMT) radio frequency (RF) connector and printed circuit board (PCB) on the entire transmission path becomes more and more serious. In this paper, the transmission characteristics of soldering transition section are analyzed by means of scattering matrix, time domain reflectometry (TDR) and other tools in signal integrity analysis, and an equivalent circuit model based on stub lines is established. At the same time, the methods to improve impedance continuity and adjust cutoff frequency are proposed, such as reducing the radius of the pad, cutting out a circle on layer below the pad, biasing the center of circle cutout, increasing the impedance of the pad and the connecting segment of the transmission line, so as to optimize the transmission characteristics of 40-50GHz window. Full-wave simulation results show that these optimization methods can effectively reduce the reflection coefficient and improve the transmission coefficient in the optimization window. The experimental results show that the optimized scheme controls the S11 below -10dB from DC to 50GHz, which is much better than that of -5dB before optimization, and proves the effectiveness of the optimization.

For the battery-powered implantable stimulators, the energy optimization of neural stimulation is critical for extending its lifetime. However, due to the nonlinearity and complexity of the relationship between the neural electrical stimulation and the neural response, optimizing the stimulation parameters to achieve desired neural response with minimal energy consumption is a challenge. The circuit-probability theory is a novel method to investigate the quantitative relationship between electrical stimulations and neural responses. Based on the circuit-probability theory, this paper proposed a method to investigate the energy optimization problem at different neural response strength. The results show that the pulse width of the optimal current waveform with minimal energy consumption is not a constant, but a value that increases with the neural stimulation strength. In the future, our study can be applied for the optimization of current pulses of implantable neural stimulators to prolong the lifetime and reduce the recharge intervals of batteries, thereby reducing the volume of implanted pulse generators and the costs and risks of battery-replacement surgeries.

The inherent resolution of traditional PWM (pulse width modulation) RF (radio frequency) modulation is not high to achieve 256QAM modulation in the 5G OFDM (orthogonal frequency division multiplexing) 700MHz UHF (ultra-high frequency) bands system. This paper proposes a method that can significantly improve the resolution of PWM modulation, which is to introduce a dither scheme in the PWM hardware circuit. Under the same pulse width resolution frequency, the modulation of the 5G system can be increased from QPSK to 256QAM with our proposal. This paper completes the system design of the PWM modulation. With the logic circuit level simulation, the correctness of the dither scheme proposed in this paper is verified, and the RMS EVM (root mean square error vector magnitude) is reduced to about 1%, which satisfies 3GPP requirements. The dither scheme in this paper enables PWM digital RF modulation to be applied in the 5G UHF bands

The location of electric vehicle charging stations is a hot issue in the development of electric vehicles (EVs). How to determine the locations and scales of EV charging stations are the main research problems. For this purpose, an improved Non-dominated Sorting Genetic Algorithm II(NSGA2) based on double-layer coding is proposed, which is named INSGA2. The INSGA2 adopts a multi-objective optimization for the locations of EV charging stations. Taking the comprehensive cost of charging stations and users as the two objective functions, and the service scope and scale of charging stations as the constraints, a multi-objective optimization model is established. The results reveal that it has better effectiveness and practicable in solving the locations and scales of the EV charging stations.

The existing indoor positioning methods of firefighters are generally based on GPS single positioning data, but in complex indoor environment, the signal stability is reduced and the positioning accuracy is not high. In order to improve the stability of positioning signal and optimize the accuracy of positioning data in indoor environment, a method of indoor positioning for firefighters based on GPS and sensor network was proposed. According to the differences of GPS positioning data and sensor network, the uncertain information transition analysis model, and then according to the continuity of the two data model is calculated, and the GPS and the sensor network data exchange noise reduction calculation, GPS and indoor positioning signal sensing network latency difference correction, GPS and indoor positioning signal fusion calculation of sensor networks. The simulation results show that the indoor positioning error of the proposed method is only 1.23%, which is more suitable for the indoor application scenarios of firefighters.

In this paper, a new sliding mode control strategy is proposed to be applied to the direct instantaneous torque control (DITC) of SRM. Based on the terminal sliding mode control, the load torque estimation value is added to the control law to improve the anti-interference ability of the system, and the controller parameters are adjusted online through the fuzzy control strategy to sup-press the torque ripple caused by SRM in the dynamic process, Finally, the convergence characteristics of the system are analyzed by Lyapunov stability theory. In order to verify the effectiveness of the system, simulation experiments are carried out in MATLAB/Simulink. The results show that the controller mentioned in this paper has good dynamic characteristics. While ensuring the response speed, it can well suppress the torque ripple when the motor starts and the given parameters change, sup-press the rising trend of current, and optimize the operation state of SRM.

In view of the problem that traditional collaborative robots mostly work according to the prescribed path through teaching or offline programming cannot obtain the feedback information of the robot in real time, resulting in low intelligence. A multi-decision pressure compensation robot massage control system based on stereo vision imaging is proposed. It could reduce the execution time and positioning error of the robot when conducting human body positioning massage, and perform force compensation based on the massage pressure feedback in real time to improve the level of human-computer interaction. Firstly, hand-eye calibration is carried out through binocular vision sensor and collaborative robot, and the robot positioning control is realized by using high-precision real-time point cloud generated by structured light compensation. The massage pressure is fed back through the pressure sensor on the massage end which the massage force is trimmed to ensure the safety and comfort during the massage process. In the visual working range, the experimental data shows that the depth positioning error is within 0.277mm and the accuracy of the robot massage in the appropriate pressure range can reach 97.35%, which effectively solves the problems in robot massage and increases the standard of human-computer interaction and the immersive experience which is made innovations in the direction of stereo imaging technology and medical treatment.

Comparators contribute a significant role to analog to digital converters (ADC). Considering the comparator design, several performances should be deal with. Several optimizing methods for comparator design in recent years are reviewed in this work. Based on the traditional amplifiers, FIA (Floating Inverter Amplifier) technology is used to improve the preamplifier, thus achieving low power consumption. Triple-Latch Feedforward Dynamic Comparator by adding a feedforward path to reduce delay and a Rail-to-Rail Dynamic Voltage Comparator combines NAND logic circuit and NOR logic circuit to enable the comparator to operate in full scale of input voltage also proposed especially for low-input applications.

In order to detect the grease content of greasy semi-solid in gearbox, a method of changing the value of the plate capacitance due to the change of grease content is adopted. This paper makes a theoretical analysis of the flat capacitance sensor, carries out COMSOL simulation experiment on the detection of grease content in the gearbox by the flat capacitance sensor, studies the influence of the structural parameters of the flat capacitance sensor, the grease in the gearbox and the environment in the gearbox on the detection of flat capacitance. As a result, a flat capacitive sensor and detection circuit for detecting the grease content in the gearbox are designed. Combined with theory, simulation and practice, the algorithm of the detection system is optimized. Through the experimental verification of the grease height in the gearbox, the experimental data in the range of 0mm ~ 50mm are analyzed. The results show that the detection method works stably, has linear output, and can accurately measure the content of grease in the gearbox.

Supervisory Control and Data Acquisition (SCADA) system is the core component of the cryogenic propellant loading process control system, used to collect field data in real time and monitor and control scattered field devices. Aiming at the lacking authentication capabilities and other security measures of SCADA system, we proposed a lightweight security authentication scheme mainly using Cryptography for SCADA system. Based on symmetrical encryption technology and message authentication code (HMAC) technology, the proposed scheme could achieve bidirectional identity authentication and message confidentiality between the control server and the controller which could guarantee the availability, confidentiality, integrity of the transmitted data between different devices. Our scheme could also resist security threats such as replay attacks.

Disassembly systems play an important role in the remanufacturing processes. It has been widely used to systematically disassemble valuable and reusable parts and raw materials from wasted or end-of-life products through a series of operations. In this work, a disassembly line balancing model is established based on an AND/OR graph. It takes precedence relation, cycle time restriction, failure risk, and time uncertainty into consideration and aims to maximize the dismantling profit and minimize the energy consumption. Then, a multi-objective discrete bat optimizer based on Pareto rules is designed. A precedence preserving crossover operator, a single point mutation operator, and a 2-optimization operator are used in the search stage. To speed up the convergence, this paper proposes an elite strategy to maintain the non-dominate solutions in the external files and verify the effectiveness of the algorithm in solving the disassembly line balancing problem by comparing it with the current popular multi-objective optimization algorithm.

In this paper, the amount of induced charge on the electrode of the gas-solid two-phase flow with different flow patterns is studied. The induction mechanism model of rod electrode is established based on Poisson equation and Gauss law,and the flow pattern is divided into slices. To reveal the relationship between the flow pattern and the charge, the simulation is provided. The amount of electrostatic induced charge of diffrrent flow pattern in rod is obtained. The influence of different flow patterns in the pipeline on the amount of induced charge is analyzed, and the correctness of the results is verified by design simulation.

With rapid development of artificial intelligence, cognitive technology and big data technology, intelligent sensing system has attracted extensive attention of researchers at home and abroad. This paper studies the related development status of intelligent sensing technology, puts forward the basic concept, system framework and technical characteristics of intelligent sensing system, and analyzes the related enabling key technologies.

The special requirements of computer information processing system for safety and availability determine the importance of terminal sensor quality management in information acquisition system in the process of engineering construction. Industrial process instruments and equipment have the characteristics of large quantity, many types, low price, most suppliers are agents, simple process, short manufacturing cycle and few nonconformities. How to ensure the quality of purchased instruments has become the primary task of instrument procurement in the whole process. Therefore, purchasing personnel must strictly abide by quality management standards and use effective quality management methods to ensure the high quality, efficiency and safety of purchased items. This paper adopts the quality control method of Total Quality Management to explore the quality management of process instrument procurement and take a nuclear power plant for an example.

The safe and stable operation of the Internet of Things (IOT) system is strongly related to the economy and daily life. IOT equipment damage accidents have caused major economic losses, which impacted on people's lives, and endanger industrial and social safety. However, accidents caused by various accidental factors cannot be completely avoided, effective monitoring and fault diagnosis of the IOT system, in particular in power system are very important. From the perspective of large-scale heterogeneous terminals and heterogeneous terminals across domains, this paper conducts fault diagnosis on the operation of the IOT based on big data and artificial intelligence methods. Firstly, based on the datadriven method, the information collected by different sensors is processed through the deep neural network for dimensionality reduction and feature extraction. Secondly, the extracted multi-source information features are feature fused. Finally, the Softmax function is used to diagnose the faults of the fused features.

A recursive principal component analysis (RPCA) method was presented in this paper for reliable fault detection of nuclear power systems equipment. Existing fault detection methods for nuclear power equipment are still stay in the theoretical research, as well as experimental analysis. Due to the special working environment of nuclear power equipment, a planned repairs and maintenance for each equipment is a normal operation. However, with the growth in installed capacity of nuclear power units, they suffer from several drawbacks. Offline detection of nuclear power system equipment can never truly reveal the actual situation, leading to maloperation or a waste use of equipment. To tackle such problems, this paper introduces RPCA methods for fault detection. A simple control chart is established for intuitive visualization of the false working condition. A recursive PCA scheme is proposed as a reliable extension of the PCA method to reduce the false alarms for time-varying process. The proposed RPCA approach are verified by detecting abnormal working status occurring in a simulated nuclear power system.