In order to solve the problems of low efficiency and danger in the existing power maintenance training methods, a power maintenance evaluation system based on immersive virtual reality technology is developed in this paper. Based on the equipment 3D accurate model and virtual reality simulation technology, the system imports the power plant scene and equipment accurate model through MakeRea13D platform for content development. Using key technologies such as model lightweight, 3D UI display and VR multi means interaction, an immersive virtual maintenance and virtual scene operation simulation platform is established. At the same time, five led-cave, mobile virtual platform and helmet immersive simulation environment are built, and three environments can be used for rendering display. According to the field environment of the actual operation and the real structure of the equipment, the system establishes three-dimensional virtual scenes such as maintenance and operation required for the evaluation, and supplements the accurate model data required for the evaluation. The actual working environment of the power station is reconstructed according to the requirements of the evaluation content, which helps to improve the professional technicians’ sense of substitution and realism of the operation scene.
This paper analyzes the relevant data of domestic agricultural products cold chain logistics enterprises,explores the reasons that affect the performance of agricultural products cold chain logistics enterprises,and establishes an index system for the performance evaluation of agricultural products cold chain logistics enterprises.We collect the actual data of relevant com-panies,quantify and normalize the index values corresponding to the index system,form the corresponding sample training set,and use GA(Genetic Algorithm)to optimize BP artificial neural network.The application model based on GA neural network algorithm is established,and simulation training is carried out to determine the optimal number of hidden layers.At the same time,we compare the results of various evaluation methods to prove the effectiveness and relative accuracy of this method.The application methods and prospects of this method in the decision-making and management system of agricultural products cold chain logistics enterprises are described.
With the real-time changes of flight taking-off and landing,how to reasonably allocate flight support equipment and workers has always been a problem to be solved in the airport.By analyzing the problems and challenges faced by the traditional scheduling mode,a flight support task allocation model based on mixed integer programming algorithm is proposed.Model 1 calculates the minimum number of resources required for flight support according to the task type and resource working time range,and Model 2 can deduce all support task lists in real time according to the changing flight execution.Combined with the list of available resources,it can quickly calculate the resource allocation method to obtain the optimal matching result.By ana-lyzing the practical application effect of the algorithm,the proposed resource allocation method realizes the optimal scheduling and improves the utilization efficiency of flight support resources.
In order to improve the accuracy of power outage type judgment,a fast judgment method of power outage type based on transient disturbance characteristics of electric energy meter is proposed.In the transient disturbance characteristic sampling signal of electric energy meter,the instantaneous fundamental amplitude is extracted by Hilbert transform method,the noise of the sampling signal is suppressed by sliding singular value decomposition method,and the characteristic waveform of singular value is obtained.The main frequency points in the sampling signal are collected by envelope extremum algorithm to obtain the frequency spectrum.The characteristic quantity of transient disturbance of electric energy meter is extracted and normalized to judge the type of power outage.Experiments show that the proposed method can accurately judge the type of power outage,and has good anti-noise performance.
In order to improve the utilization rate of mobile learning resources and optimize the navigation effect and operation performance of learning resources,taking English mobile learning resources as the research object,a navigation method based on multi-objective optimization is proposed.The English mobile learning resources are preprocessed by word segmentation and invalid word removal,the storage location of the target resources is determined by parallel search,and the learning resource re-trieval path is generated by multi-objective optimization algorithm,so as to obtain the final navigation results of English mobile learning resources.The experimental tests are carried out from two aspects:single resource navigation task and multiple re-source navigation tasks.The results show that the proposed method is less affected by the navigation task scene,the navigation success rate and precision rate are increased by 1.0％and 0.5％,respectively,and the memory occupation and time expendi-ture in the operation process are smaller.
In order to reduce the positioning error and avoid the inspection risk caused by complex environment,the intelligent automatic path planning technology of transmission line UAV inspection based on PID algorithm is studied.Using RTK posi-tioning technology,the location information of inspection mark points of transmission line UAV is collected.Using the path generation method,the optimal geometric path of UAV from the starting point to the target point is simulated,a dynamic mod-el of UAV height and attitude directly controlled by PID parameters is created,and the path planning tracking controller based on PID algorithm is used to control the UAV to fly to the target point according to the optimal path.The intelligent automatic path planning of transmission line UAV inspection is realized.The experimental results show that the positioning error of the mark points to be inspected is always less than 0.25 × 10-4.The UAV inspection path of transmission line planned by this technology is short,which can effectively avoid the inspection risk caused by complex environment.
Insufficient code writing will lead to loopholes in the automatic code generation of the applet.In order to ensure the security of the automatic code generation of the applet,it is necessary to design a real-time detection system for the automatic code generation of the applet vulnerability.A method of real-time detection system for small program vulnerabilities is de-signed.In this method,software is designed in the system,and the vulnerability data in the applet are collected by the software to facilitate subsequent detection;at the same time,different hardware function modules are designed.Among them the task management function module plays an important role,the module can effectively allocate detection tasks,which greatly im-proves the detection efficiency of the system.Based on the designed software and hardware,the overall design of the detection system is realized.The experimental results show that the method has strong accuracy and high reliability by testing the num-ber of loopholes detected,time-consuming test and number of missed detections.
In order to ensure the stable operation of the machine room,a system of data collection and distribution is designed to monitor equipment operating information.The system mainly uses real-time information to manage the working state of ma-chine room.C8051F020 microcontroller is used as the core controller of the system,and RTL8019AS chip is used as the Ether-net controller of the system,through the chip communication with the microcontroller can be realized,which is conducive to the user of remote monitoring and management of machine room.The system has good expansibility,reliability and security,and can be used in many fields.
There are long text data of power customer service tickets,which is a challenge to the construction of the model to classification power customer service tickets.Therefore,this paper proposes a classification model based on hierarchical infor-mation fusion to improve the analysis ability of long text.The Word2vec method is used to process the words in the sentences,and then the word vector and sentence matrix are obtained.The bidirectional long-term and short-term memory network(BiL-STM)is used to learn the dependence between words,and the TextCNN is used to learn the correlation between sentences.The multi-layer perceptron(MLP)is used to extract the deep semantic features,which are learned at all levels to achieve fea-ture layer fusion.The proposed model is tested on a dataset containing thirty thousand real power customer service ticket sam-ples,the average classification accuracy of the five types of service tickets is 0.921,and the average macro-F1 score is 0.901.The results show that compared with TextCNN,BiLSTM,and deep belief network(DBN),the recognition accuracy of the proposed method is improved by 1.9％,5.3％,and 13.5％,respectively,which can give an outstanding performance on the classification of power customer service tickets.
By analyzing the current problems of data transmission in data center,a dynamic transmission policy based on data prioritization is designed using deep reinforcement learning.The model calculates the priority of each data stream based on three characteristics,and aims at reducing the average waiting time to achieve the dynamic transmission.The comparison experi-ments with traditional algorithms show that the strategy model can effectively reduce the delay of data transmission and improve the efficiency of data center resource usage.
A large scale of unregulated charging of electric vehicles can lead to excessive fluctuation in grid load and significant voltage deviation.Therefore,it is necessary to optimize the charging behavior of electric vehicles in a multi-objective manner.By analyzing the driving behavior and charging habits of electric vehicle drivers,an electric vehicle charging load model is estab-lished.Based on the particle swarm optimization algorithm,binary coding is introduced to solve the optimization function and a-chieve the optimization of orderly charging strategies.Tests show that by utilizing the proposed method to optimize the char-ging behavior of vehicles in an orderly manner,the load on charging stations fluctuates within the range of 2000 kW to 2500 kW,with a tendency towards smoother fluctuations,the voltage deviation remains within 3.00％,and the network loss rate is around 2％,which effectively reduces the loss rate of the power grid and provides assurance for the stable operation of the power system.
The low accuracy of aerial image road traffic sign recognition may lead to the untimely receipt of road sign informa-tion,which threatens the life safety of drivers and passengers.Therefore,an aerial image road traffic sign recognition method based on multi-scale deep learning is proposed.A 12-position camera is used to collect the road traffic sign image,and the col-lected road sign image is preprocessed by image color standardization and image component thresholding.Combined with DAE network and EDFCC algorithm,a multi-scale deep learning clustering model is constructed.The preprocessed road traffic sign image is input into the model,and the identification of road traffic signs in aerial images is realized according to the output re-sults of the model.The experimental results show that the image collected by the proposed method has high signal-to-noise ra-tio,high definition,high recall rate and high accuracy.
Aiming at the problems of low accuracy and poor processing efficiency of the existing urban user electricity cost as-sessment model,an information integration processing method is proposed based on artificial intelligence technology,and an in-telligent system platform is constructed.IP network is used to complete the optimization design,the layout of electricity cost monitoring by establishing a multicarrier system is rationalized,the coverage of urban user monitoring is maximized,and the SX1276 processing chip is developed by Semtech and the XD1129 chip fusion is used to strengthen the intelligent system plat-form processing power.This study also uses the TOPSIS ideal solution algorithm to reasonably evaluate the electricity cost in-formation of multiple users in a certain community.The experimental results show that the evaluation algorithm model has high accuracy,and the error is only 0.4％in the environment of 1000 user measurement points.
The poor invulnerability of communication network affects the communication connectivity and security of AC/DC hy-brid power system,an optimization algorithm of invulnerability of short wave communication network in AC/DC hybrid power system is proposed.This paper defines the network invulnerability and analyzes its influencing factors.On this basis,it builds the invulnerability model of short wave communication network based on hot theory,determines the invulnerability optimization function of short wave communication network,and elaborates the constraint conditions for solving the optimization function.Based on the construction optimization function,the tabu search algorithm is introduced to solve the construction optimization function.The output result of the algorithm is the invulnerability optimization result of short wave communication network in AC/DC hybrid power system.The experimental data show that the maximum network connectivity and network survivability obtained by the proposed algorithm are large,which fully proves that the proposed algorithm has better network invulnerability optimization performance.
The highway emergency resource scheduling plan lacks of considering the traffic flow characteristics of bottleneck sections,which leads to longer scheduling time,hence,a bi-objective scheduling method for highway emergency resources which considers the traffic flow characteristics of bottleneck sections is studied.Using the cusp mutation theoretical model,this paper analyzes the traffic flow characteristics of the bottleneck section,sets the bi-objective function with the shortest path and the least time as the objectives,and uses the hybrid leapfrog algorithm to solve the objective function and output the highway e-mergency resource scheduling plan.The test results show that the path of the scheduling plan solved by the studied method is relatively short under the premise of guaranteeing the total time used is the minimum,and the total length of the path is rela-tively short,which has better practical application effect.
Aiming at the problem of unbalanced load management caused by the access of distributed power sources such as charging piles,photovoltaic power generation and wind power generation,this research applies edge computing technology to design a multi-load management system,and uses edge devices and edge servers to complete load data calculation and calcula-tion.For data analysis,the load metering module uses the SOC RN2026A64 chip,and uses multi-channel high-precision syn-chronous sampling to collect power load data,and control the distributed power and energy storage units to complete the opti-mal operation.The experimental results show that the system has higher computing efficiency,the maximum average comple-tion time of business requests does not exceed 0.6s,and the power load does not exceed 5 kW after optimal scheduling.
Due to the lack of overall consideration of customer satisfaction,cost and environmental factors in the research of net-work optimization of reverse logistics,logistics cost,customer service time,service coverage of recycling stations and carbon e-mission are taken as the optimization objectives of electronic products recycling reverse logistics to build a multi-objective opti-mization model.An improved two-archive algorithm(Two-archive 2)is employed for the problem of unbalanced between con-vergence and diversity in this model.The simulation results show that the introduced algorithm can guarantee the diversity and convergence of the optimal solution,and reduce the complexity at the same time.
In order to avoid that the settlement exceeds the budget in the construction process of power transmission and trans-formation project,the whole process cost risk modeling and evaluation method of power transmission and transformation pro-ject based on grey theory is proposed.The characteristic mapping model of cost risk identification in the whole process of power transmission and transformation project is established.The analysis sequence,the analysis matrix,the grey correlation degree,the grey theory to calculate the estimate and settlement cost information respectively,the risk function and probability density function,and the cost risk of the whole process of power transmission and transformation project according to the expected val-ue and variance of the risk function.Simulation experiments show that this evaluation method conforms to the law of risk,can correctly identify sensitive factors,and is helpful to better control the cost of engineering construction.
In order to alleviate parking difficulties existing in urban road traffic,a smart three-dimensional parking garage is de-signed in this study,which is designed to transform the traditional parking garage.The safety alarm function is used to detect the ambient temperature of the parking garage to prevent fires.Temperature sensing fiber,infrared pyroelectric sensor and temperature sensor complete the temperature collection of the entire parking garage.The system allocates reasonable parking spaces to users according to the user's initial position and parking time requirements,and a heuristic search algorithm is used to calculate the optimal path,which could reasonably schedule and allocate parking space resources in the parking garage.The ex-perimental results show that the parking space turnover rate of the system in this study is 8.1％,and the average parking path length of each parking task is at least 251.1 m.
Electricity,water,gas and other industries have problems such as inability to share information,meter reading and leakage,and present the characteristics of heterogeneous and massive data.How to deal with the centralized management and cleaning of data is one of the challenges.This paper designs a multi-source data platform based on a non-relational database.The proposed data platform includes data transmission,data governance and data release,and studies heterogeneous data fu-sion,batch computing capability,high availability,platform security and reliability 4 key technologies.Through the effective processing of complex data,it shows that the data platform can support the functional requirements in practical applications.
