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

As the internet of things is more and more widely used in traffic, in order to avoid accidents caused by the driver's lack of judgment during the lane changing process, a lane-change control strategy is proposed based on the internet of of vehicles. First, the speed difference and distance between vehicles are obtained through the vehicle-vehicle communication, and the lane change warning level is established. Then, according to the relationship between the pedal opening and the torque coefficient in the drive control strategy of a pure electric vehicle, three different pedal modes are analyzed. According to the vehicle dynamics model, the relationship between torque and acceleration is obtained. Finally, the lane-changing scene was constructed in PreScan and verified by PreScan/Simulink co-simulation. All simulations are carried out under the internet of vehicles environment. The simulation results show that this strategy can meet the driver's lane changing needs when the speed difference and the distance between vehicles are different.

The development of economy drives the progress of society, and people's travel needs are also rising, which inevitably leads to the increasing traffic pressure. Although urban rail transit can greatly facilitate our travel, it is still impossible for residents to go door-to-door, so if we want to give full play to the advantages of rail transit, it is inevitable to connect with other modes of transportation. Especially the transfer with buses, the transfer facilities become extremely important. In order to effectively enhance the core competitiveness and the comprehensive transportation efficiency of urban public transportation system, the radiation capacity of rail transit can be met through the perfect bus coverage, and more passengers can be served as much as possible.Bring a wonderful ride experience to passengers. Through the study of urban rail transit and bus transfer facilities, the factors affecting the coordination level of transfer facilities are analyzed, and the two modes of transportation are perfectly connected and complementary in advantages, so as to improve transfer efficiency and better serve passengers.

The mixed platoon composed of human driving vehicles (HVs) and automated vehicles (AVs) will be a long-term traffic phenomenon. Compared with the traditional vehicle platoon composed of the same type of vehicles, the uncontrollability and the limitation of information acquisition of HVs in the mixed platoon make it difficult to apply the existing control algorithm of AVs to the mixed platoon. Therefore, considering the particularity of the mixed platoon, we break down the mixed platoon into different small platoons, and then present three information flow topologies (IFTs) for the AVs in small platoon, and based on these, we propose a control strategy for AVs. The simulation experiment results show that the IFTs structure we propose can adapt to different communication conditions, and the presented control strategy can also ensure the stability of the mixed platoon.

The problems of unintuitive expression of bridge inspection disease information and difficult query of historical disease records are widespread. To solve the above problems, this paper proposes an automatic three-dimensional (3D) visualization expression method of bridge inspection disease information based on BIM (Building Information Modeling) technology. The construction technology of bridge BIM model based on completed drawings is studied. The storage structure of bridge disease information database is designed and the disease information is stored in a time-sharing classification manner. Studying the 3D expression method of common diseases and the conversion relationship model of bridge global coordinates and component local coordinates, the automatic matching and visual expression of disease information and 3D model of bridge are realized. The research results can provide intuitive and visual information query means for bridge maintenance and operation, and help to improve the efficiency of information utilization and rapid formulation of maintenance schemes.

Transportation construction is an important part of infrastructure construction. Before the construction of major transportation construction projects, it is necessary to achieve the implementation of the scheme after the demonstration and selection of multiple schemes. Through the traffic simulation evaluation technology, multiple scenarios and multiple schemes can be evaluated iteratively without changing the existing traffic infrastructure. Then make reasonable suggestions according to the evaluation results, feedback the suggestions to the designers and formulate corresponding measures. Thus improving traffic conditions and saving costs.

Generally, the first factor in the primary consideration of the transportation of materials is to minimize and make the total cost of transport most economical. The model established in this study is applicable to emergency situations such as the spread of infectious diseases. The model considers how to help each province or region to transport the supplies that meet the demand to the destination in the shorted time, while ensuring the transportation cost as low as possible. Therefore, this paper establishes a model to simulate the situation, and proves the rationality and efficiency of the model.

In the problem of traffic flow prediction at traffic intersections, this paper tries to predict traffic flow based on the random characteristics of cloud model. Establish the cloud model of traffic flow through the mapping relationship between qualitative and quantitative. To establish a new prediction mechanism between adjacent intersections to predict future traffic trends. Provide dynamic optimization support for intersection traffic signal timing scheme to evacuate traffic congestion as soon as possible.

This paper is intended to provide valuable references and experiences for the road traffic management department and the sector which formulate the traffic regulations and standards. Based on analysis of the relevant research domestic and overseas, this paper uses the quantitative argumentation mainly and the qualitative analysis as a supplement. It puts forward some concepts about information quantity transferred from highway traffic engineering facilities, mutual information and information entropy. According to the basic elements of traffic engineering facilities, it also builds a basic amount of information computational model which serves for the highway traffic engineering facilities from the angle of the eight types of information (Chinese character information, English character information, Arabic numerals information, Geometric shapes information ,Color information ,Guidance information , Graph or characteristic symbol information and linear facilities information ,et al .Minority characters is not considered temporarily here.). Besides, to analyze the theoretical information content which can be obtained by drivers in the unit road space, this paper proposes a concept called transmission information density of highway traffic engineering facilities and does a case analysis for the component model. The research in this paper has a certain theoretical significance and practical value for improving the service capacity, operational efficiency and safety level of the highway traffic system.

Aiming at the integrated application of Building Information Modeling (BIM) and other Information and Communication Technologies (ICTs) in the Architecture Engineering and Construction (ACE) industry, this paper presents an implementation case in airport infrastructure construction in China using BIM integration with Internet of Things (IoT). Firstly this paper reviews the BIM and other ICTs used in the ACE industry, then presents the application of BIM and IoT in airport infrastructure construction, lastly analyzes the implementation effect. The results from this case show that using BIM integration with IoT in airport infrastructure construction can effectively improve project delivery performance including cost, safety and so on.

This paper aims to propose a multi-objective optimal charging scheduling strategy for large-scale electric vehicles (EVs), considering traffic flow, power grid and charging stations all together. First, based on the characteristics of these three systems, the mathematical model that characterizes the performance of the traffic flow, power grid and charging stations is designed separately. The multi-objective optimization function and constraints are established, taking the speed of traffic flow, charging load of the power grid, and the quantity of EVs in charging stations as the optimization targets. Then, a simulation platform is built, and a practical case is studied within the third ring of Beijing. 24 h simulation test for 242,880 EVs is performed, and the effectiveness of the proposed strategy is verified by comparative analysis.

In the Internet of Vehicles project, multiple sensors are used to detect the trajectory of the vehicle. Due to the detection error of the sensors, the detection results of each group of sensors will not be completely consistent. At the same time, since the local clock of each sensor has a time difference, the detection time of different sensors cannot be strictly consistent. This leads to the problem of temporal and spatial inconsistency of trajectory data. In order to solve the problem of temporal and spatial inconsistency in the data, it is necessary to perform post-fusion processing on the trajectory data. This paper proposes an algorithm based on sliding clustering to post-fusion of multiple data source trajectories. The experimental results prove that the fusion trajectory data with lower error can be obtained in real time using this algorithm.

With the development of intelligent transport, big data has become an important means to analyze traffic problems. Aiming at the problem of commuting demands in suburban areas without rail transit coverage, this paper proposes a data fusion method based on the data of online carpooling and the AFC data of subway, analyzes the travel characteristics of online carpooling and subway multimodal transit, and estimates the fuel saving benefits. The conclusion proves that the multimodal transit is a suitable way to satisfy commuting demands and can save 78.1% energy consumption on average compared with driving alone. This research will provide support for transport intelligent operation management.

In the two lane scenario, by considering the influence of driver’s optimal estimation for flux difference information on macroscopic traffic flow, a new lattice model is proposed. The dynamic performance of stability for the developed model is analyzed. Also, computer simulation method used to prove the linear and nonlinear analyses. The research results indicate that driver’s optimal estimation for flux difference information can efficiently suppress the emergence of traffic congestion and stabilize the operation of vehicles in two lanes.

The paper is based on the HCM's left-turn particular road, and the special channel passage capability model. The head timetable and operational characteristics of the left turn and the head vehicle are analyzed. Taking the main road of Jinan City, the traffic flow of Ten Road has been studied the influencing factors of the proportion of different traffic flow ratios of left-turn vehicles and lanes, and the linear regression method determines the difference between the head, and establishes the relevant calculation model. Combined with the turning head adjustment factor, study the influence factor of the power out of the vehicle and the vehicle traffic flow, and established the traffic capacity calculation model of the intersection of left and turned to the turning lane.

In order to explore the causes of road traffic accidents and analyze the influencing factors of the severity of the accident, this paper is based on the traffic accident record data of Dallas in 2019. Firstly, it uses Python to preprocess and render the data, and initially analyze the relationship between the independent variables and the severity of the traffic accident. Based on the preprocessed 12,392 pieces of accident data, it is divided into general accidents and major accidents according to the severity; Secondly, the adaptive K-DBSCAN density clustering algorithm is introduced to cluster the effective density according to the severity, to eliminate the noise point data, and reflect the hot spots of traffic accidents in Dallas state; Finally, the parameters of the established XGBoost classification model are traversed, and the parameters with the highest accuracy are selected, the characteristics that have an important impact on the accident severity prediction results in the feature set such as road and environment are sought. The feature importance ranking is obtained, and practical suggestions and measures are put forward from the perspective of model prediction.

Questionnaires are the traditional solution to obtain samples of urban rail passengers' route choice, but large-scale questionnaires are required to obtain a sufficient number of valid samples to calibrate a discrete choice model. The rail travel trajectory data of cell phone users provide a large amount of real travel data, which provides a new method to obtain rail route choice samples. Based on cell phone users' rail travel trajectory data, we propose a route choice model of rail passengers. First, this paper analyzes the issues and effects of the midway point loss of the track data of cell phone users, and explains the necessity of developing a discrete choice model. Second, this paper proposes a Logit model, and then describes the route travel time calculation method and the route choice dataset generation method, the latter of which fully considers the disadvantages and advantages of cell phone data. The model is calibrated and the result is evaluated based on the trajectory data of cell phone users in Shenzhen. The results show that the model proposed in this paper has high accuracy and can be utilized in actual rail passenger flow assignment.

True 3D (three-dimension) display technology is the key technology of Digital Railway Location System (DRLS) based on virtual environment. Based on the principle of stereo perspective projection image generation, the true 3D display of virtual geographic environment of DRLS is realized on the microcomputer and projection platform. The realization method of stereo display based on OpenGL and hardware configuration scheme of DRLS are given. A realistic case for the stereo alignment location of a railway in western China verified that the alignment location in the virtual geographical environment based on stereo display stimulates the creativity of railway alignment designers. Meanwhile, the efficiency, quality and interest of railway location are improved.

In order to deal with the emergency of railway dangerous goods transportation, provide decision support for decision makers to design robust emergency facilities under uncertain environment, consider uncertain demand and multi-stage standby coverage of emergency facilities, establish a robust optimization model of multi-stage reserve coverage under uncertain railway dangerous goods transport demand, compare and analyze the experimental results under different safety parameters. A simulation example is given to verify the feasibility of the model and algorithm. The results show that the model can effectively solve the problem of constructing the location and layout network of emergency facilities for railway dangerous goods transportation under uncertain demand, and can ensure that the location decision has good robustness. Decision makers can determine the optimal location scheme according to their conservative degree, risks caused by uncertain factors and reality.

The accessibility of the port collection and distribution system is an important factor affecting the scope of the port hinterland. Based on the BPR function, this paper proposes economic, time and environmental cost factors applicable to multiple transportation modes, and constructs an accessibility measurement model based on the generalized cost impedance function. Finally, we take the Douala port as an example to verify the model, the results shown that the accessibility of each logistics node is reduced by 73.66% on average after the environmental cost factor is added, and different nodes show different degrees of decline, meanwhile the advantages of waterway and railway transportation in environmental cost factors and logistics cost factors are more prominent. In summary, the accessibility measurement model based on the generalized cost impedance can more comprehensively reflect the accessibility characteristics of the port collection and distribution system.

The road network is an important carrier for the realization of spatial transfer and flow of passenger flow, cargo flow and information flow, and plays an important supporting role in social and economic development. Accessibility is an important indicator for evaluating regional traffic conditions. It is of great significance for the system to comprehensively analyze the accessibility of a certain region. Therefore, this paper uses GF-6 remote sensing image, DEM and road network data in the research area, and uses the time cost distance model to analyze the road accessibility in the research area. The research shows that the regional accessibility of Pingtou Village in Fengyi Town of Mao County is the best, followed by the area along National Highway 213, and the accessibility of the rest areas is poor. The overall spatial pattern of accessibility is significantly different. Terrain and economic factors have significant impacts on road accessibility. The research results can provide protection and management services for National Highway 213 in Mao County, and also provide data support for disaster prevention and relief work.

Taking the Zhongshanmen Street-Southern Polytechnic Intersection as the research object, based on microscopic simulation technology and traffic conflict technology, the operation efficiency and safety status of the intersection are studied. A VISSIM micro-simulation model for conflict simulation is established, and SSAM software is used to analyze the simulation conflict of the simulation model to verify the rationality of the optimization scheme. The simulation results show that the road service level of the intersection has increased from C to B, and the TTC has also increased from 0.29 to 0.58.

In response to China’s “transportation power”, “carbon peak by 2030”, “carbon neutral by 2060” and other policies, the urban rail transit system is gradually becoming digital, intelligent, and green development with the support of technologies such as “Internet +” and “shared travel”. Residents' needs for green travel, intelligent services, and information integration have also appeared. Transportation and travel pay more attention to the construction of a humancentric integrated travel chain system. "Mobility as a Service" (MaaS) has improved urban traffic problems to a certain extent. It has also met diversified travel needs and improved the quality of travel services. Based on the current research status and background at home and abroad, this paper clarifies the connotation of MaaS concept and system framework, and focuses on summarizing the key technologies of the MaaS-oriented urban rail transit smart service platform: passenger multi-mode identity authentication and aggregate payment technology and equipment, reversible Using enterprise-level urban rail transit service middle-station technology, blockchain-based multi-agent trusted transportation travel management, and one-stop accurate information services based on carbon credit incentives and behavioral profiling, the above content can solve the data fragmentation in the transportation field. Difficult to associate; parties do not trust each other; value mining is simple; business disconnection issues. Finally, it puts forward suggestions on building a MaaS smart service platform at the government, enterprise, and individual levels.

Aiming at the problem of real-time safe trajectory planning of automated vehicles in a complex mixed traffic environment, this paper proposes a trajectory planning method based on risk prediction. Firstly, the motion of surrounding vehicles is predicted by the fusion of Constant Turn Rate and Acceleration model (CTRA) and Unscented Kalman Filter (UKF). Combining the time and space relationship between vehicles to assess driving risk. Then the initial and multiple final states of the trajectory are determined by driving behaviors to generate a set of candidate trajectories. After that, a multi-index cost function is constructed, to obtain an optimal trajectory without collision. Finally, the simulation results show that this proposed method meets the needs of vehicle safety trajectory planning in the highway scenario

Automatic safe lane changing is the key to the realization of unmanned vehicles. To accurately identify the lane changing state of driving vehicles to ensure driving safety, this paper establishes a vehicle automatic lane changing behavior recognition model based on the multi-class support vector machine. This paper selects vehicle trajectory data from the NGSIM data set for classification processing and uses genetic algorithm optimized particle swarm optimization (GA-PSO) to optimize and calibrate the penalty parameter C and the kernel parameter g in the multi-class support vector machine model. Using sample data to train and test lane-changing behavior recognition models and the research shows that the model can well recognize the behavior of the vehicle during the automatic lane changing process and provide support for the study of the vehicle lane changing phase.

Based on the analysis of relevant information on motor vehicle traffic accidents and non-motorized vehicles and pedestrian traffic accidents in Shenzhen, a Logistic model was established to analyze the influencing factors of deaths in motor vehicle traffic accidents and non-motorized and pedestrian traffic accidents. The model was tested to quantitatively analyze the impact of various factors on the deaths in the accident, and the factors that have different effects on motor vehicle accidents and non-motorized vehicles and pedestrian accidents were compared and analyzed. The results in this paper have certain reference significance for the relevant departments to put forward road traffic safety management measures.

In rail traffic engineering, the pantograph strip is the device for high-speed trains to obtain power energy, which is a part of the electric contact system. The system consists of catenary wire whose main material is copper, and pantograph strip whose main material is carbon. The arc often occurs on the carbon strip. As the speed of high-speed trains increases, the impact and vibration of the electric contact system intensify, which increases more arc. The electric erosion of arc is an important factor in the damage of pantograph carbon strip, so it is urgent to do the study of its damage characteristics. In this paper, pure electric carbon material is used, instead of pantograph carbon strip, and pure copper material is used, instead of catenary wire in the arc erosion experiment. The damage process of pure carbon material was analyzed during arc burning, by adjusting the arc burning time length between catenary wire and pantograph strip, that is between pure copper bar and pure carbon strip. The influence of arc erosion is discussed, which is on the appearances, electrical and mechanical properties of pure carbon material. With the increase of arc burning time, the surface erosion degree of the pure carbon material increases, and the electrical and mechanical properties of the carbon material weaken. This research will be helpful for finding effective measures to decrease the arc erosion of pure carbon material and prolong its service life.

The passenger drop-off area is an important part of the comprehensive transportation hub, as in the travel rush hour, the on-ramp traffic flow cannot enter the drop-off area smoothly. To relieve traffic congestion and improve efficiency of service during peak hours, this article analyzed the actual vehicle operating characteristics based on the vehicle trajectory extraction of video survey data, and the traffic situation during peak hours is simulated through the car-following model, and then, the dynamic control method and algorithm of the on-ramp in passenger drop-off area are presented. Finally, the actual case is simulated by using this algorithm, the results show that the dynamic control method can improve the actual traffic capacity effectively.

Urban traffic is a complex system composed of multiple roads and intersections. The unimpeded road traffic directly affects the development of urban economy. The intersection is the pivot of urban road traffic and plays an important role in the management of the road traffic. Traffic at an intersection often is influenced by other intersections and is a result of a joint impact of them, there must be the relationship between an intersection and several other intersections. This relationship is difficult to express with determined mathematical formula due to the stochasticity of resident travel. However, the copula function can connect the distribution function of traffic flow at an intersection with the joint distribution function of traffic flow at several upstream intersections to establish a mathematical model and predict the traffic flow at this intersection. This traffic flow can be regarded as a traffic control parameter for the intersection, and can effectively improve the coordinated control in the region.This method has achieved good results with experimental validation.

The transportation industry is a concentration of energy consumption and carbon emissions, and the development of lowcarbon transportation becomes a new industrial form for developing a low-carbon society. This paper constructs a lowcarbon transportation evaluation index system from two aspects of transportation basic energy efficiency and development operation with low-carbon transportation theory as the background. On this basis, the more used comprehensive evaluation methods are listed, and through comparison, the applicable principles, advantages and disadvantages of various evaluation methods are analyzed to lay a theoretical foundation for the subsequent evaluation of the effectiveness of low-carbon transportation in cities.

In the urban intelligent transportation system, the traffic department can better grasp the current situation of traffic flow through the collection and analysis of road traffic information. In the description of the overall traffic situation, most scholars still rely on the observation data in a certain period of time to give subjective evaluation, and can only roughly describe the overall change of traffic data in a certain period of time on a road. The traffic flow model can be used to simulate the traffic congestion situation more accurately, and combined with the developed evaluation index, any instantaneous traffic congestion situation can be evaluated at multiple points, providing a scientific and effective theoretical basis for the traffic diversion and traffic control of smart cities.

Big data of urban public transportation contains rich spatial and temporal information, which is the data basis for passenger travel characteristics analysis and evaluation of urban transportation service capacity. In this paper, we take the big data of Beijing bus swipe card and taxi track as the research object, store and calculate these two types of big data based on Hadoop distributed system, build the calculation model of passenger flow extraction, extract the hot ride areas and establish the visualization system based on WebGIS for the visual expression of data analysis results.

The emergence of "Smart Earth" is accompanied by the continuous emergence and development of complex new technologies, but at the same time, the emergence and accumulation of a large amount of information has led to the explosive growth of information. In this fast-developing society, all kinds of information of library users also urgently need new technology to handle and satisfy, but traditional services have gradually been replaced. This paper studies the information system of the city study image based on cloud computing, understands the relevant theories of the information system of the city study image based on the literature, and then designs the information system of the city study image based on cloud computing. The designed system is tested, and the test results show that the average response time of the system designed in this article is 1.8s in retrieval, and the average response time in borrowing is 3.9s.

Neural networks have the problems of excessive number of parameters and computation due to data noise and redundant filters, which limits the application of neural networks. A feature map pruning method using channel attention (FPC) is proposed to address the problem of excessive accuracy loss of existing pruning methods. Experiments show that the feature graph pruning method using channel attention has better pruning effect with the same accuracy loss.

With the continuous expansion of cloud data center scale, the flow of network traffic generated by virtual machine scheduling also increased, resulting in a sharp increase of energy consumption in cloud data center. Under certain network topology, different virtual machine scheduling strategies would produce different network flow. Therefore, how to optimize the scheduling strategy and reduce network flow became the key factors to reduce the energy consumption of cloud datacenter. In this paper, based on the cloud computing simulation platform CloudSim, the network module was extended, adding the network communication module of virtual machine migrating. In order to ensure the stability of the network data center, a virtual machine scheduling algorithm NPABFD for network traffic and energy consumption aware was proposed. The experimental results showed that the NPABFD algorithm could effectively reduce the transfer traffic of migrating VMs in high-level switches, reduce the global NDC's global energy consumption and the migrating number of VMs.

In recent years, with the improvement of the country's economic level and the improvement of people's consciousness of national culture, the macro-control of the country's cultural soft power, people's recognition of Chinese national culture continues to improve.Cultural and museum-exploration programs, with National Treasure as the main one, have been highly praised by the general public.The program not only innovates the form of today's TV programs, but also enables more people to know about traditional culture.Cultural relics are the material carrier of traditional culture, which carries profound connotation and contains rich cultural deposits.This paper takes the innovation and development of cultural and exhibition exploration programs as the main line, based on the TV program "National Treasure", carries on the program overview and case analysis.In general, from the narrative object, analyzed narrative skills, etc, from the museum to the stage of modernization transformation of the booth design with background music and so on aspects of case scenario analysis, convey analysis, etc., for the show way of cultural communication and convey the results brought by the second comprehensive program development status and existing problems, mainly from the program, open Angle of view, In terms of communication mode and creative team, the author deeply explored and analyzed the inspiration of the program to cultural and museological exploration TV programs, making cultural and museological TV programs more active.

Although the traditional Convolutional neural network is applied to image segmentation successfully, it has some limitations. That's the context information of the long-range on the image is not well captured. With the success of the introduction of self-attentional mechanisms in the field of natural language processing (NLP), people have tried to introduce the attention mechanism in the field of computer vision. It turns out that self-attention can really solve this long-range dependency problem. This paper is a summary on the application of self-attention to image segmentation in the past two years. And we think about whether the self-attention module in this field can replace convolution operation in the future. The answer to this review is yes, so it is recommended that the focus of future research be on the self-attention module

In today’s era when the original data is huge, useful information can only be extracted from the data through calculations and corresponding processing. Nowadays, the computing power of wireless access points, laptops, and mobile phones is comparable to that of computers more than a decade ago. However, most of the time, these computing resources are often idle, which greatly causes a waste of resources. In the Internet of Vehicles scenario, this problem has always been the focus of attention by scholars. The V2V (Vehicle to Vehicle) technology has solved this problem to a certain extent. This article considers the link failure time LET (Link Expiration Time) and the limit of the task allowable delay. Most of the equipment on the road uses external power supply and user equipment is powered by energy. If the transmission of the task cannot be completed within the effective time of the link, it will cause more energy consumption. This article aims at link failure time and task allowable delay Under the limitation of V2V multi-hop task, the total energy loss in the unloading process of the V2V multi-hop task is minimized.

With the acceleration of the digital construction of the power grid, many important data is collected in data centers. If a fault is not found in time, it may cause serious information security incidents. In terms of the above problems, a digital twin modeling of data center computer room (DC computer room) based on long short-term memory (LSTM) network is proposed in this paper to monitor and early warn the failures of important equipment in computer rooms. The model adopts a five-layer architecture of the equipment layer, data interaction layer, model construction layer, simulation analysis layer, and application layer. Meanwhile, the evaluation characteristics of the equipment in the data center room are constructed, and the time sequence parameters of the equipment are predicted in real time based on the long-term and short-term memory network, and the equipment that may fail is warned in advance to assist the maintenance personnel in equipment maintenance.

The cybersecurity testbed is of great importance to cybersecurity practitioners and is a necessary platform for conducting cybersecurity. To better demonstrate the offensive and defensive postures in network security tests, all data in the tests must be quickly collected, standardized, securely stored, and allowed to be quickly retrieved. Based on Elastic Stack products, virtualization technology tools, and commercial link collection systems, this paper designs an efficient data collection method based on the network security test cloud platform to provide data support for real-time situational monitoring by the guide in the test. After testing and verification, the method can achieve efficient collection of host behavior logs, virtual network data, and real network link data within the network security test cloud platform.

Deploying distributed simulations on the cloud can obtain many benefits, including lower cost, higher efficiency, easier access and so on. However, due to the virtualization technology adopted by cloud computing, improper assignments of tightly coupled simulation tasks may degrade the overall user experience. At the same time, considering the frequent changes of the status of tasks and hosts, the scheduler should be able to give a good enough solution in time. In this paper, we mainly focus on the efficient scheduling of simulation tasks in the cloud, which has been recognized as an NP-hard combinational optimization problem. Besides mapping such a task-host matching problem as a min-cost max-flow problem, we also design an incremental flow-based task scheduler to deal with the dynamic changes of tasks and hosts. Simulation experiments on Alibaba cluster trace show that our design is adequate to this scenario.

At present, MCU technology has broad prospects and application market. The upgrading of technology and products is faster and faster, and the difficulty and complexity are also increasing. High performance MCU will become a new trend. Through the research on the current commonly used high-performance MCU chip verification methods, this paper designs and implements a chip verification system based on ARM Development Studio (ADS) platform for high-performance MCU with cortex-A7 core and its ads based debugging method. The verification system and debugging method for highperformance MCU chip in the sample verification stage after the completion of the tape, the rapid construction of software and hardware verification environment, through interactive, software and hardware collaborative way to verify the functions of each module of MCU chip real-time and reliable. Efficient verification method can improve MCU development, shorten verification time, improve verification efficiency, find chip design defects in time, and shorten chip development cycle.

With the development of computer technology, smart cities can not only provide more information services for city management, but also provide citizens with more personalized and intelligent services. This paper analyzes the probability of citizens’ demand for various cultures by means of questionnaire surveys, screens out the main culture dissemination needs, and draws on the construction methods of data warehouses to study and design a high-cohesion, low-coupling three-tier culture dissemination system framework based on a smart city platform. For the bottom data layer, through three data acquisition methods of outsourcing data, built-in data and crawling data, it realizes the construction of cultural databases and the collection and intelligent analysis of cultural demand data. And finally in the presentation layer, it realizes the output of standard cultural data and personalized cultural data.

VANET is a special mobile ad hoc network, which is an indispensable part of ITS. One of its main functions is to provide security messages for the vehicle nodes. The high mobility of the nodes leads to frequent changes in the topology of the network, so it is necessary to design efficient and reliable MAC for VANETs. This paper improves on MoMAC by proposing an adaptive MAC protocol based on node density, called A-MoMAC. The protocol is no longer fixed for the time slot division. Before reserving time slots, the node will first adjust the time slot division according to the current traffic conditions in each lane. After the adjustment of frame partition, the node will determine a group of reservable time slots and then attempt to reserve a random time slot. Simulation results show that A-MoMAC is well suited to scenarios where nodes are non-uniform, improving channel utilization and reducing access collision.

Through the scheduling of demand-side response resources such as electric water heater load, the adverse impact of distributed power output fluctuation on the operation of power system can be reduced and the consumption of renewable energy can be promoted. Firstly, based on the operation framework of load aggregator, the cost and benefit of load aggregator are analyzed, and the demand response compensation strategy based on user comfort is proposed among the components of load aggregator cost. In this study, a flexibility index is established to measure the ability of the system to deal with the output fluctuation of distributed power generation. Finally, an optimal load scheduling model of electric water heater is proposed. The objective function of the model is that the total income of load aggregator and system flexibility are maximum, and the user comfort temperature, wind and photovoltaic output fluctuation and aggregator income constraints are taken into account. The simulation results of numerical examples show that the proposed model has good economic benefits and improves the stabilization effect of distributed wind power fluctuations.

Recently, the application of Front Collision Warning (FCW) on traffic safety has become an active research subject. In this study, electroencephalography (EEG) was used to measure drivers’ brain signals in vehicle-pedestrian conflicting events under FCW by conducting a driving simulation experiment to measure driving performance. Dividing the whole collisionavoidance process into three driving stages, we found that EEG signals in different driving stages have different effects on collision-avoidance performance, and proper releasing time of FCW could play an important role in reducing the risk of vehicle-pedestrian accidents by intervening in EEG signals during perception stage. With critical EEG features related to collision identified by logistic regression, EEG signals are validated to be reliable indicators of vehicle-pedestrian accidents.

Roadside Light Detection and Ranging (LiDAR) can provide over the horizon perception information for connected vehicles (CV). However, its performance may be affected by the weather, especially in rainy and snowfall weather. To improve all-weather working ability, a combined denoising algorithm is proposed in this paper after analyzing the shortcomings of the existing filters. The filter is composed of crop box filter, ray ground filter, voxel filter, and statistical outlier filter. By combining multiple point clouds filters, the snowfall points are removed and the effectiveness of general filters in complex scenes is verified. The experiment shows that it not only can retain the traffic objects’ features, but also realize denoising on real point clouds data of snowfall weather.

Baidu Migration Data is a massive location service data developed by Baidu, and it is also a product of the smart city industry in the era of intelligent networking. Based on the big data of Baidu's migration during the National Day in 2021, the inter-city travel spatiotemporal OD matrix and the overall inter-city travel network of the Bohai Rim city group are constructed. Through the spatiotemporal feature extraction method of singular value decomposition, three main National Day intercity travel patterns are identified from the constructed spatiotemporal OD matrix, and the pattern analysis is carried out from the perspective of time and space. From the perspective of complex network theory, explore the spatial structure characteristics of the National Day intercity travel network through the judgment of the characteristics of the small world, the exploration of urban agglomerations, and the analysis of urban centrality. Research shows: 1) During the National Day, intercity travel consists of three modes: National Day overall travel, staggering travel in some areas, and post-holiday return travel. It has observable time-varying characteristics of the three stages as departure, journey and return. 2) Intercity travel has the spatial flow law of short-distance travel from large cities to neighboring prefecture-level cities, and divergence from populated cities and back from tourist cities. 3)The structure of the intercity travel network is stable with obvious small-world characteristics, and a "core-periphery" radial structure has been formed with the main administrative regions taken as the core. 4)Three economic zones of Beijing-Tianjin-Hebei, south central peninsula of Liaoning and Shandong peninsula in the Bohai Rim City Group are extremely closely connected. 5)Cities with high administrative levels, developed economy, and densely populated cities in the intercity travel network are more central and have huge travel demands.

With the vigorous promotion of urban energy internet and smart grid, the risk of various malicious cyber attacks on existing cyber physics power systems has increased significantly after they are transformed into integrated energy cyber physics systems（IEGS）. In order to ensure the safe and reliable operation of urban integrated electricity-gas system, this paper proposes a loss assessment model for integrated electricity-gas system under cyber-physical coordinated attack. Firstly, the branch fault scenario is randomly generated, and then the real operation state of the branch is concealed by injecting false data. Secondly, the DC power flow model and probability model are used to simulate the cascading failure of the power system, and the loss caused by the coordinated attack on the urban integrated electricity-gas system is evaluated according to the physical operation characteristics of IEGS, the IEGS vulnerability branch is evaluated. Finally, the correctness and effectiveness of the model are verified by the integrated electricity-gas system composed of IEEE 30 nodes and Belgium 20 nodes.

The problem of noise in obtaining CACC vehicles cruises data through technologies of automatic control and wireless communication, e.g., V2X communication. Based on the Kalman Filter theory of the speed and azimuth angle between vehicles is proposed, this method analyzes the characteristics of the CACC vehicles driving in a curve. This model only considers the relationship between the previous moment and the current moment, and it applies the iterative method to deal with Kalman filtering, simplifying the calculation process. This method analyzes the characteristics of the vehicles driving in a curve, uses the Kalman filter to process the noise of the acquired data, and analyzes the influence of the presence or absence of data noise on the CACC vehicles motion law. The simulation of CarSim and Simulink shows that the proposed data processing method can improve the accuracy of the data, e.g., MSE increased by 81.03% and RMSE increased by 56.36%. The model make the CACC platoon more in line with the current situation, helping improve the efficiency of vehicles in the curved area and improve the level of traffic safety.

Smart city is a new trend of urban development. The difference with traditional city construction lies in its construction mode, operation mode and service effect. The government plays an extremely important role in the construction of smart cities, and only the integration of government and social data can form a strong synergy of urban governance. This paper focuses on the effectiveness of government governance in the development of smart cities by analyzing the problems of government governance in the process of smart city construction and proposing countermeasures to improve the effectiveness of government governance with the background of smart cities and the theory of good governance as the theoretical support.

At present, there are many problems in the operation and maintenance of equipment and facilities in the urban rail transit industry, such as untimely online monitoring, independent monitoring systems, low efficiency of traditional operation and maintenance and low level of intelligent operation and maintenance. Based on the analysis of the status of operation and maintenance application, mode and technology of the industry, this paper puts forward the overall technical framework of intelligent operation and maintenance of urban rail transit based on cloud platform. The paper also elaborates key technologies such as Internet of Things data monitoring, failure diagnosis, intelligent operation and maintenance, micro-services and cloud-side collaboration. And it has reference significance for improving the level of operation and maintenance management and security ability.

It briefly introduces the research status of domestic and foreign traffic signal transition, and summarizes the characteristics of these algorithms. Based on the judgment that the time difference between the previous and last two cycles does not exceed a certain limit, and the adjustment time is the key parameter, the smooth transition and rapid transition are combined and unified, and a traffic signal smoothing and Fast transition algorithm. The results show that the algorithm can quickly coordinate the transition before and after the transition within a few minutes by adjusting the cycle difference constraint range. When the cycle difference restriction range is larger, the adjustment speed is faster, the required time is shorter, the smoothing effect is worse, and the transition quality is lower; when the cycle difference restriction range is smaller, the adjustment speed is slower and the required time is longer , The better the smoothing effect, the higher the transition quality. Since this algorithm takes into account the influence of different timing schemes before and after the transition on the transition timing, when the timing scheme changes before and after the transition is obvious, the fast coordinated transition effect performed by this algorithm is more obvious.

Traffic infrastructure road property survey is the basic work of the intelligent management of traffic facilities, and traditional survey techniques have obvious shortcomings. Based on the investigation of a provincial expressway, this paper introduces the composition of the vehicle-borne mobile mapping system and the operation process and technical scheme of the highway infrastructure survey, and analyzes the acquisition results. It verifies the reliability and practicability of the vehicle-borne mobile measurement system in highway infrastructure surveys, and provides an efficient and accurate technical means for highway infrastructure surveys.

The frequent occurrence of natural disasters and man-made attacks greatly increases the failure rate of urban integrated energy system in the face of such uncertain events, which seriously threatens the security and reliability of urban energy supply. Aiming at the resilience of urban electricity-gas integrated energy system under extreme events, a three-stage robust optimization model is proposed. The model aims to minimize the load shedding of integrated energy system under extreme events. The line fault fuzzy set is constructed through the component fault probability interval under extreme events, and the defense-attack-defense three-layer framework is established, then C&CG algorithm decomposes it into outer main problem and inner sub problem for solution. Finally, the effectiveness of the proposed method is verified by the improved IEEE-30 bus power system and 7-bus natural gas system.

Intercity transportation in urban agglomerations has the characteristics of large travel scale, diversified travel modes, diversified travel purposes, unbalanced temporal and spatial distribution and so on. Based on the mobile signaling data, this paper calculates the intercity travel volume and external traffic travel volume of the Beijing-Tianjin-Hebei urban agglomeration through data cleaning, processing, fusion and mining. Using the spatial autocorrelation analysis method, the spatial effect of intercity highway traffic travel is analyzed. The results indicate that the intercity highway travel volume of adjacent cities shows a significant positive correlation, and intercity highway travel is positively correlated with population, economic development level, urbanization development level and residents' income level.

Nowadays, scientific and reasonable traffic volume prediction plays an important role especially in the traffic infrastructure planning. In the recent research, establishing a robust mathematical model for traffic volume prediction becomes a challenging problem. In our research, Hidden Markov Model (HMM) is constructed based on the numeral characteristics of monthly traffic volume for each freeway in Jiangsu Province. By analyzing the Markov property of the monthly flat peak traffic volume and the nonlinear effect of the monthly peak traffic volume, we further predict the future monthly traffic volume. Compared with the traditional models, our proposed model has significant advantages in some evaluation indicator, such as MRE, MAE, RMSE. Further more, The construction of this model only depends on the numerical characteristics of historical traffic volume data, which has the advantages of convenience as well as broad application prospects.

An adaptive traffic signal control model for isolated intersections is proposed based on reinforcement learning. The signal control problem is formulated as a Markov Decision Process because its probabilistic features match well with the random nature of traffic system. To improve the design of reward, which is usually an ad-hoc combination of several traffic measures, this paper develops a reinforcement learning algorithm that draws connections between the performance evaluation with the stage pressure-based P0 control policy in the reward design process. Numerical results show that the reinforcement learning control method has lower total delay and higher throughput compared with the fixed-time control. Moreover, the P0 policy performs better in improving throughput under heavy traffic conditions.

In this paper we describe a hardware-in-the-loop platform (HIL) for evaluating the performance of On-Board-Unit (OBU) in C-V2X scenario. We use CARLA as the simulation environment to obtain high fidelity scenario data. In this platform we design the basic evaluation scenarios based on current standards and then design random scenario parameters for obtaining a large number of test scenarios with different parameters to test the robustness of the algorithm. The key scenario parameters are stored while the scenario is running and are used to analyze and evaluate the performance of the algorithm at the end of the scenario and to produce an evaluation report, a use case is provided in the text to demonstrate the overall usefulness of the platform.

With the development of BIM(Building Information Modeling) technology, CIM(City Information Modeling) is accompanied by the development of Internet of Things technology, and the development of CIM has become an academic hotspot. In recent years, due to the limitations of BIM and CAD, it is impossible to transform the spatial and temporal relationship information well. Therefore, the emergence of CIM represents the progress of the construction industry, especially the integration of BIM and CIM makes it complementary, which can maximize the construction of smart cities.

This paper proposes a rapid construction technology of human machine interface (HMI) based on components in order to solve the problems such as inconsistent interface style, poor code reusability and low development efficiency in the development of traditional information system HMI. Based on analyzing a large number of HMIs of information systems, the technology extracts interface elements to form a reusable general component library. Adopting the user-defined control management model based on Qt platform, the general components of information system HMI are integrated. The topic management technology, attribute visual editing technology and component communication technology based on signal/slot mechanism, which support one-key switching of interface style, are put forward. Therefore, the rapid construction of information system HMI is realized. This paper selects an interface in a practical project and compares the effect of the interface constructed by the same developer with and without this technology. The results show that the rapid construction technology of HMI can effectively solve the above problems. At present, this technology has been widely used in command-and-control system, combat system, state monitoring system and other complex information systems

The Electroencephalography (EEG) based brain-computer interfaces is a convenient way to use brain waves to investigate different emotions and some mental disorders. With the development of research and design, it is even possible to use brain waves to control devices (such as robot arms) to improve the life of the disabled. On this basis, based on the summary of previous research results, this paper focuses on the analog front-end of wearable brain computer interface (especially electrode and amplifier) and its related algorithms. The algorithms are developed on the basis of some classical machine learning algorithms, which are more suitable for EEG signals like common spatial paternal and long short-term memory network. This article may provide some convenience and inspiration for future development of EEG-based brain-machine interfaces.

With increasing attention to cardiovascular diseases in recent years, various portable electrocardiograph (ECG) monitoring devices have been designed and manufactured. As a mature design, the three operational amplifiers (3 op-amps) structure has high input impedance and a high common-mode rejection ratio; therefore, it is widely used in the circuit design of portable ECG equipment. In the actual circuit design, in addition to the 3 op-amp structure, some extra improvements are always needed to meet the measurement requirements of ECG signals. This paper first presents the technical specs and common structures of instrument amplifiers. It then expands the topic to portable ECG monitors, providing not only technical requirements of the portable ECG equipment but also a summary of the existing improvements, including drivenright-leg circuit, notch circuit, self-calibration circuit, electrode design and 5 op-amp model, which are helpful to the designers of portable ECG detection devices.

In view of the low recognition accuracy of traditional weather recognition methods and the serious imbalance in the number of weather images in various categories in the weather image data set, a weather image classification algorithm based on generative adversarial network and transfer learning is proposed to solve the above problems. The proposed method mainly includes two parts: data equalization based on generative adversarial network and image classification based on transfer learning. This paper uses generative adversarial network to amplify the data of a few categories of weather images, so as to obtain a relatively balanced weather image data set.The method of transfer learning is used to fine-tune the model to realize the classification of weather images. The experimental results show that the method proposed in this paper is better than the traditional method, effectively solving the problem of low model classification accuracy caused by the imbalance of training samples, and realizing the recognition and classification of four types of weather images: sunny, foggy, rainy, and snowy.

The Electrocardiogram (ECG) is a simple test used to check heart rate and electrical activity. Deep learning has been widely used in disease classification, disease prediction, and complex disease decision-making. Since 1980, the collection of public ECG data has been carried out continuously, which plays a vital role in the analysis of some specific conditions, such as arrhythmia, cardiac infarction, and cardiac ischemia. Therefore, at the same time, private institutions or organizations have also begun to build large ECG databases (5 times as much as public databases) to promote the absorption of deep learning models. With people’s continuous optimization of deep learning algorithms, its accuracy, stability, and efficiency have greatly increased, and now deep learning has better versatility that can be competent in new clinical scenarios. This paper focuses on the latest technology of ECG analysis, especially the latest technologies of unbalanced data and data classification. Meanwhile, the limitations of these technologies and the areas for future improvements are also discussed. This article summarizes the deep learning techniques currently used for ECG.

Combine the networking process and text emotion analysis technology to build an early warning system of public opinion, which provides a basis for judging the emotional tendency of police Weibo and points out the direction for the future emotional analysis and research. Based on the emotion analysis method of emotion dictionary, this paper constructs the analysis model of emotion polarity, takes the related topics of police work as an example to verify the availability of the model, and finally carries on the correlation analysis to these information of Weibo and the emotion polarity. The emotional polarity analysis model of this paper has availability, there is a significant positive correlation between Weibo comments and retweets, and when the retweets are low, there is a significant negative correlation between comments and emotional polarity. And when the number of likes on Weibo is greater than the number of comments, Weibo content itself has a positive emotional tendency.

In order to reduce the complicated daily inspection tasks of power system staffs and improve the inspection efficiency and accuracy, an indoor orbital inspection robot applied to the power industry is developed. The inspection robot includes robot body, track and background monitoring system. The walking system, lifting system and camera platform system of the robot are introduced, and the hardware control circuit and software system are developed. Based on the simulation environment Gazebo of the Robot Operating System(ROS), the robot model is created by two method. The first method is to create robot by using Gazebo model editor, and the second method is to create robot by software SolidWorks then model is exported to URDF file. The simulation results show, the robot model established by Gazebo model editor is more convenient for simulation optimization analysis.

Gesture recognition is the latest human-computer interaction (HCI) technology, which allows users to naturally control electronic devices through the movement of fingers and palms without operating redundant devices. Radar gesture recognition technology offers significant advantages in terms of privacy and security, device reliability and design flexibility. In this paper, a model GestureNet suitable for radar gesture recognition is designed by using the smooth pseudo Wigner Ville processing of millimeter wave radar gesture echo and the knowledge of hybrid zero convolution neural network in deep learning. The results show that the recognition accuracy of the validation set of GestureNet reached 97.35% and the recognition accuracy of the test set reached 91.75%, indicating that the model has good generalisation ability, thus providing a strong guarantee for radar gesture recognition.

The increasing computing power emerged in the network drives the further explosion of data. However, the in-network computing resources are not well utilized efficiently by cloud or edge computing to process the big data. Dispersed computing, as a promising complementary paradigm, can gather all the in-network dispersed computing resources to build a near real-time, location-aware computing paradigm. Task scheduling for dispersed computing faces the problems of resource heterogeneity and dynamics. Traditional scheduling algorithms cannot be well adapted to the dispersed computing environment due to the lack of learning. In this paper, we model the task scheduling process as a Markov Decision Process (MDP) and propose a Q-Learning-based task scheduling algorithm for dispersed computing. The simulation results display the feasibility and effectiveness of the algorithm, which can effectively reduce makespan and latency compared with the baseline algorithm and can effectively sense and utilize the dispersed resources.

With the progress of telecommunication technology, remote acquisition and data analysis have become possible. Electrocardiogram (ECG) is the measure of the electrochromic activity of the heart and is an important human body data. Therefore, portable continuous ECG acquisition is an important part of family health care, which can be realized by wearable devices such as smartwatches. This paper aims to review the development of wearable ECG devices from aspects of hardware and software design. In the hardware part, this paper focuses on the design of electrodes that directly receive signals from the human body and the integrated circuit inside the wearable devices. The extraction, process and transmission of the obtained ECG signal are discussed in the software part. This article can provide a reference for those who are interested in understanding wearable ECG technologies or designing new wearable devices.

At present, garbage classification mainly depends on manual work, with large classification workload, low efficiency and low user participation rate. Under the background of the gradual popularization of smart home in people's life, aiming at the unsatisfactory implementation of waste classification, a kind of traceable intelligent garbage classification system based on voice interaction was developed in this work. The voice interactive flexible garbage sorting bin in the system can realize automatic movement of the partition to adjust the regional capacity, and automatically recognize the garbage type by voice, so that users can start classification at home. In addition, the encrypted QR code was adopted to track the user's information so as to supervise the garbage classification and solve the first half of the problem of garbage classification.

Intelligent connected vehicle (ICV) is an important application area of artificial intelligence and has received more and more attention. In the current autonomous development of intelligent driving, it is a key and hot issue to improve the driving safety and autonomous decision-making ability of ICV under dynamic traffic flow. The existing driving safety theory mainly considers the vehicle's attributes, which cannot reflect the influence of traffic elements, and driver's behavior on driving safety. Based on the field theory, a driving risk field model considering traffic elements, vehicle state and driver behavior under dynamic traffic flow is established, and the influencing factors of each parameter in the driving risk field model are analyzed. The results of the study can provide an important theoretical basis for ICV autonomous driving decision-making.

In order to solve the problems of urban expressway congestion and pollution in China, combined with the defects of the existing expressway control system, this paper innovatively introduces the interaction between ramp and variable speed limit, based on this, the traditional METANET model is improved, a more accurate macro traffic flow prediction model is established. The objective function is traffic benefit and environmental benefit, and the hierarchical sequence method based on particle swarm optimization is used for the optimization calculation. On that basis, the intelligent collaborative control system of ramp and variable speed limit based on MPC is established. It can be seen from the experimental simulation results, the system can predict and control the traffic flow, it is very real time and intelligent. Compared with the traditional control, it can control the traffic flow better, in addition to this, it is generally applicable to a variety of situations, at the same time, it also provides a richer theoretical basis for the follow-up research and intelligent control of China's expressway.

At present, the three-ratio method and artificial intelligence algorithm are widely used in transformer fault diagnosis. However, the three-ratio method based on DGA has problems such as lack of coding and insufficient anti-interference ability of various artificial intelligence methods, resulting in a high rate of misdiagnosis. This paper proposes a transformer fault diagnosis model based on FI (Feature Image)-CNN. It takes the percentage of dissolved gas in the oil as an input parameter, constructs a feature vector, and extracts the two most widely distributed data as two-dimensional coordinates based on the characteristics of the fault. Axis, and then uses the RGB principle to construct a three-layer neuron structure with the remaining gas data to form a “RGB dynamic map” of the training sample; build a convolutional neural network model, combine the faulty DGA data, and use the DNN full-link network for fitting, and finally Realize the fault diagnosis of the power transformer. Finally, it is compared with the diagnostic performance of various optimization algorithms. The results show that the FI-CNN method can obtain high-dimensional curves that divide the fault categories according to the "RGB dynamic map" of the training samples and the corresponding spatial locations, and has higher diagnostic accuracy.

Based on the analysis of the current situation of pure electric bus enterprises and battery industry in China, and combining the characteristics of pure electric bus and the theory of battery capacity and power, this paper expounds the necessity of battery box selection for pure electric bus and the purpose of battery box selection for pure electric bus enterprises. The idea of matching optimization design is put forward to liberate pure electric drive, so that the whole vehicle space and the number of battery boxes can be matched and optimized to the maximum. Taking pure electric bus as the carrier, this paper puts forward the problems and solutions of pure electric drive in battery box selection of pure electric bus, and explains the importance of combining the whole vehicle space with the size and number of battery boxes. In this paper, the measures and performance of pure electric drive in pure electric bus enterprises are elaborated in detail.

In the present work, power generation roads in the field of intelligent transportation and intelligent road were considered, and a kind of vibration energy collector based on piezoelectric was designed. On this basis, its temperature sensitivity was studied. Specifically, according to the characteristics of road load, a kind of energy collector structure based on an elastic block of disc spring was proposed. In addition, considering the influence of ambient temperature on energy collection efficiency, the physical and mathematical models of the heat transfer process were constructed for the designed collector structure. By importing different piezoelectric material parameters under different temperatures into the model and applying vehicle axle-load, the correlation between ambient temperature and the output voltage of the energy collector was successfully developed. This work provides essential guiding significance for designing/optimizing the structure of road energy collectors and selecting appropriate piezoelectric materials， which are suitable for road load and ambient temperature.

Traditional resident travel survey methods, such as paper questionnaire, telephone interview and mail inquiry, have disadvantage of low data accuracy, difficult organization and limited sampling size. GPS-based travel survey method is playing an increasingly important role in modern transportation planning. This paper proposes an innovative method for detecting individual trip mode recognition by using mobile phone GPS positioning data. First, a smartphone GPS sensorbased application is developed for multi-mode travel trajectory data collection. Data characteristics of different trip modes are deeply analyzed and the characterization indexes of different modes are put forward. Second, a support vector machine (SVM) algorithm is proposed for trip mode detection. SVM can map low-dimensional data to high-dimensional space for segmentation, is especially suitable for traffic mode recognition. Results show that the average mode detection accuracy reaches 92% for walk, bike, bus and car. This paper can provide solid data support for urban traffic planning.

With the constraints of limited computing and communication resources in V2X networks, V2V communication for task offloading needs to adapt to the dynamic V2X environment while satisfying the requirements of low delay and high reliability. Aiming at balancing the trade-off between delay and power consumption in V2V-enabled task offloading, this paper proposes a V2V-enabled multi vehicle task offloading algorithm based on deep reinforcement learning Deep Deterministic Policy Gradient (DDPG). Firstly, we construct a V2V-enabled multi vehicle task offloading system model, which leverages the surrounding vehicles with computing resources as the relay node. In this model, multi hop communication is used to offload tasks to multiple vehicles for coordinated computation, and we assume the transmission power of tasks is a tunable continuous variable. Then, DDPG algorithm is proposed to deal with the continuous high-dimensional action space in task offloading. The proposed task offloading algorithm based on DDPG is simulated and compared with other deep reinforcement learning algorithms. Experimental results show that the algorithm can effectively balance the delay and power performance while improving the success rate of task offloading.

With the development of automobile industry, automobile residents of automobile and technology. The proportion of simple operating automatic transmission cars is also getting much more, called the first choice for more families. However, there is still a certain gap between reliability, service life and fuel consumption compared with the manual transmission. At the same time, the power response and driving pleasure reflected by manual transmission cars in driving are also irreplaceable by automatic transmission cars. However, manual transmission semi-linkage control in the process of shift is often difficult to master. In this paper, for the control of semi-linkage, we put forward a manual transmission car clutch semi-linkage automatic control mechanism to solve the frustration in the process of shift, make driving more smooth, reduce fuel consumption, reduce gear wear and tear, and extend the service life of the transmission.

The paper considers vehicles’ multi-hop connectivity as inter-grid connectivity and draws some important conclusions by theoretical modeling. Based on these conclusions, a new RSU scheme called Pro_Dep algorithm is proposed which tries to find maximum high connectivity grids where a RSU can cover more coverage area at a given minimum connectivity probability, so as to decrease the number of RSU installment. The simulations, which come from traces of real spatial temporal GPS positions, study the relationship between numbers of RSU, number of grids covered, number of nodes covered and minimum required connectivity probability, ratio of side length of grid to vehicles’ communication range. The experiment results prove validity of our RSU deployment scheme.

With the goal of quick response to fault handling, research on emergency handling methods for urban rail transit equipment failures. Based on the analysis of the current situation of emergency handling of urban rail transit equipment failures, a method for rapid response to equipment failure handling is proposed from the two dimensions of fault diagnosis and troubleshooting; Integrate decision tree and case-based reasoning methods to diagnose equipment faults, and use decision matrix analysis to make intelligent decisions on fault handling related resources. The research results of this paper provide a reference for emergency response methods of urban rail transit.

The Demand response bus is a useful complement to the conventional bus that can make up the defects of service inefficiency and lines easy to become ageing for the conventional bus. The key problem of the research on the demand response bus is to solve the contradiction between the service efficiency and the operating cost, which is embodied in the problem of bus route scheduling. However, the existing researches have not been perfect enough to solve the problem at the theoretical and the practical level. The actual subjects of the demand response bus are considered in this manuscript. Through the discussion of overall cost and efficiency, the optimization model of the demand response bus in urban area is constructed. The features and model solutions are analyzed, and the semi-dynamic approximation solution of the model that considering the actual subjects is proposed. Finally, a region of Ningbo is taking for the case study. The results show that the optimization scheme can improve the comprehensive travel efficiency by 42.2% compared with the traditional bus. The results prove the feasibility and optimization of the method, and provide a new way of thinking and mode for the practice of the demand response bus.

Bearing is an important component of automobile system, and its mechanical properties have become a key factor that directly affects the safety performance of automobiles. Through reasonable and accurate analysis of various mechanical properties of hub bearing, the design of hub bearing can be optimized, the strength requirement can be guaranteed, and its service life can be predicted. Taking the bearing input shaft of I3 model of an automobile as the research object, by analyzing various loads, the equilibrium equation of force system under steady-state conditions is established, and the bearing input bearing strength and service life are judged by calculation.

With the impact of COVID-19, more people are choosing to travel by private cars, which will cause problems such as traffic congestion. It is essential for traffic engineers to have real-time traffic volume, speed, and individual vehicle length. In this study, the ACC7350 millimeter-wave radar was tested, and its advantages and disadvantages were analyzed in vehicle speed, distance from the radar, and vehicle trajectory. The speed detection error between MWR and GPS was within ±6%, and the distance detection error was ±20%. Then the traffic flow detection results of the camera and millimeter-wave radar were compared and analyzed. Results show that the mistakes of traffic flow detection based on vision and MWR are ±4% and ±13%, respectively. Finally, we proposed a traffic data processing method combined with a camera-based target tracking algorithm.

At present, the transformer fault diagnosis methods based on clustering and neural network generally ignore the influence of unbalanced data sets, which could lead the model fall into local optimum. In order to improve the accuracy of transformer fault diagnosis with unbalanced data as samples, a decision network based on improved fuzzy clustering and deep neural network is proposed. A distance correction term is introduced to modify the fuzzy membership function, improve the membership relationship of the boundary data to majority class cluster, and reduce the influence of the wrong division of the boundary data on the cluster center of a minority one, so that the cluster center tends to be stable in the ideal position, and the imbalanced data set is effectively divided, the resolution of data set is improved. In addition, referring to the process of human learning and voting activities, a DNN adjudication network is established after the IFCM. Through learning the partitioned data sets to different degrees, the common network and expert network are used for joint voting, so as to obtain more reliable prediction results. The experimental results show that this method can effectively improve the accuracy of transformer fault diagnosis under unbalanced data sets.