Multi-terminal visual control method and system in simulation environment

Abstract

The invention relates to the technical field of visual control, in particular to a multi-terminal visual control method and system in a simulation environment, and the method comprises the steps: building an urban traffic dynamic grid visual frame based on an urban traffic digital twin model; based on an urban traffic dynamic grid visualization framework, constructing a traffic flow prediction model to predict the traffic flow of the next period; according to the traffic flow prediction result of the next period, urban traffic road saturation calculation of the next period is carried out, and an urban traffic road saturation matrix of the next period is output; according to the urban traffic road saturation matrix of the next period, calculating a traffic flow offset threat value of the next period, presetting an offset value threshold value, comparing the traffic flow offset threat value with the offset value threshold value, extracting data greater than the offset value threshold value, performing visual display in a digital twin model, and performing ordered processing on threats.

Description

Background Art

[0002] With the development of smart cities and intelligent transportation, the use of digital twin technology to build urban traffic models has become a research hotspot. Traditionally, urban traffic management mainly relies on decentralized data collection and a single monitoring platform, which makes it difficult to fully reflect the real-time status of urban roads. The development of Internet of Things technology has made real-time traffic data collection, transmission and processing more popular, but how to effectively integrate these multi-source data and display them through intuitive visualization is still a problem that needs to be solved. Deep learning, especially the application of graph convolutional networks (GCN) and recurrent neural networks (such as LSTM) provides new ideas for traffic flow prediction, which can better capture spatial and temporal characteristics.

[0003] Disadvantages of existing technologies: Traditional systems often simply display real-time traffic flow data, lack the integration of road topology and real-time data, and it is difficult to build a comprehensive and dynamic traffic status display framework; most existing methods use simple time series or traditional statistical models for prediction, which cannot fully utilize the spatial correlation of urban roads, and the accuracy and real-time performance of the prediction results are poor; the existing system has deficiencies in visualization, and cannot achieve unified and intuitive dynamic display on multiple terminals, and lacks effective quantification and early warning processing of key indicators (such as road saturation and deviation threat value); traditional methods lack quantitative indicators for traffic deviation or abnormal conditions, and cannot quickly locate and deal with potential traffic risks, resulting in low emergency response efficiency. Summary of the invention

[0004] The main purpose of the present invention is to provide a multi-terminal visualization control method in a simulation environment, and further to provide a multi-terminal visualization control system in a simulation environment that can run and implement the above method, effectively solving the above problems mentioned in the background technology.

[0005] The technical solution of the present invention is as follows: In the first aspect, a multi-terminal visualization control method in a simulation environment is proposed, the method comprising the following steps: S1. Based on the urban traffic digital twin model, a dynamic grid visualization framework for urban traffic is constructed; S2. Based on the urban traffic dynamic grid visualization framework, a traffic flow prediction model is constructed to predict the traffic flow in the next cycle; S3, according to the traffic volume prediction result of the next cycle, calculate the saturation of the urban traffic roads of the next cycle, and output the saturation matrix of the urban traffic roads of the next cycle; S4. Calculate the traffic flow offset threat value of the next cycle according to the urban traffic road saturation matrix of the next cycle, preset the offset value threshold, compare the traffic flow offset threat value with the offset value threshold, extract data greater than the offset value threshold, and visualize it in the digital twin model to handle the threat in an orderly manner.

[0006] A further improvement of the present invention is that the construction of the urban traffic dynamic grid visualization framework in S1 includes the following specific steps: S11. Within the preset collection cycle, at 5-minute intervals, obtain the real-time traffic flow data of the monitoring nodes of M roads. , It represents the traffic flow data of road i at the nth statistics in this collection period. The value of i is 1-M, and the value of n is 1-N. N represents the number of statistics in this collection period. The real-time traffic flow matrix Q is output. ,in, The dimension is The real number space of ; S12, constructing the urban road topological connection relationship matrix A, ,in, The dimension is The real number space of the city road topological connection relationship matrix A is Indicates the connection relationship between road i and road j. When road i is directly connected to road j, ,otherwise , when i=j, The real-time traffic flow matrix Q and the urban road topology connection relationship matrix A together constitute a dynamic grid visualization framework for urban traffic.

[0007] A further improvement of the present invention is that S2 comprises the following specific steps: S21, extracting the real-time traffic flow matrix Q and the urban road topological connection relationship matrix A, wherein the nth column of the real-time traffic flow matrix Q is a vector , , The dimension is The real number space, It represents the traffic volume of M roads counted for the nth time in this cycle, and the value of n is 1-N; S22, construct the traffic flow matrix of this cycle as input , the output is the traffic flow matrix for the next cycle The traffic flow prediction model outputs the traffic flow matrix P of the next cycle.

[0008] A further improvement of the present invention is that the step S22 comprises the following specific steps: S221, the traffic flow prediction model includes a spatiotemporal graph convolutional neural network part and a recursive neural network part, wherein the spatiotemporal graph convolutional neural network part is composed of a spatial graph convolutional layer and a temporal convolutional layer; the real-time traffic flow matrix Q and the urban road topology connection relationship matrix A are input into the spatial graph convolutional layer of the spatiotemporal graph convolutional neural network, and the output result of the spatial graph convolutional layer is , , The dimension is The real number space, K represents the spatial feature dimension, Input the time convolution layer, which uses a one-dimensional convolution kernel for convolution, and the output result is a matrix V. , The dimension is The real number space, C is the time dimension feature, and the operation formula of the time convolution layer is , the nth column of the matrix V is the vector , , the value of n is 1-N; S222, the recursive neural network part adopts an LSTM network structure, including 5 LSTM network layers and a fully connected layer. Input LSTM network structure and output the corresponding hidden state , , The dimension is The working process of the LSTM network structure is: .

[0009] A further improvement of the present invention is that the loss function during the training of the vehicle flow prediction model is: ; in, represents the true value of the nth column of the traffic flow matrix in the next cycle, represents the predicted value of the nth column of the traffic flow matrix in the next cycle, represents the norm, represents the regularization coefficient, Represents the L1 norm of the road topology connection relationship matrix A.

[0010] A further improvement of the present invention is that S3 comprises the following specific steps: S31. Extract the predicted traffic flow matrix for the next cycle ; S32. Traffic flow matrix The nth column vector in , the vector It indicates the predicted traffic volume of M roads in the nth statistical period of the next cycle, where n is 1-N. The time period of the nth statistical period of the next cycle is obtained, and the historical traffic volume average of the traffic volume of M roads in the time period in the past thirty days is obtained to calculate the road saturation. The calculation formula of the road saturation is: ;in, It indicates the road saturation of road i in the nth statistics in the next cycle. represents the traffic volume of road i in the nth statistics in the next cycle, It represents the historical traffic flow average of road i in the time period of the nth statistics in the next cycle; S33, calculate in sequence and output the road saturation matrix S of the M roads in the next cycle N times. , The dimension is The real number space.

[0011] A further improvement of the present invention is that S4 includes the following specific steps: S41, extract the road saturation matrix S of M roads in the next cycle N times, where the single element in the matrix S is Indicates the road saturation of road i at the nth statistics in the next cycle; S42: Calculate the traffic flow deviation threat value of road i in the nth statistics in the next cycle , the calculation formula is: ; S43: Output the traffic flow deviation threat value matrix T of the M roads counted N times in the next cycle.

[0012] A further improvement of the present invention is that S4 also includes: presetting an offset value threshold t=0.25, extracting elements in the vehicle flow offset threat value matrix T that are greater than the offset value threshold t, and locating and visualizing them in the digital twin model to handle the threats in an orderly manner.

[0013] Secondly, a multi-terminal visual control system in a simulation environment is proposed, which includes: a visual framework construction module, a vehicle flow prediction module, a saturation calculation module, and a threat extraction module; The visualization framework construction module is used to construct a dynamic grid visualization framework for urban traffic based on the urban traffic digital twin model; The vehicle flow prediction module is used to construct a vehicle flow prediction model based on the urban traffic dynamic grid visualization framework to predict the vehicle flow of the next cycle; The saturation calculation module is used to calculate the saturation of urban traffic roads in the next cycle according to the traffic flow prediction result of the next cycle, and output the saturation matrix of urban traffic roads in the next cycle; The threat extraction module is used to calculate the vehicle flow offset threat value of the next cycle based on the urban traffic road saturation matrix of the next cycle, preset the offset value threshold, compare the vehicle flow offset threat value with the offset value threshold, and extract data greater than the offset value threshold for visual display in the digital twin model, so as to handle the threat in an orderly manner.

[0014] The technical effects of the present invention are as follows: A multi-terminal visual control method in a simulation environment was constructed. This method uses the urban traffic digital twin model to build a real-time dynamic grid, which can effectively integrate monitoring data and urban road topology, so that the entire system can fully and real-time reflect the urban traffic situation; the spatial and temporal features are extracted from the real-time traffic data by combining the spatiotemporal graph convolutional neural network with LSTM, which greatly improves the prediction accuracy of the next cycle of traffic flow; by comparing with historical data to calculate the road saturation, and then calculating the traffic flow offset threat value according to the preset formula, the traffic status can be quantitatively evaluated, and abnormal data exceeding the threshold can be extracted to achieve early warning of potential risks. At the same time, this method locates and displays the prediction results and risk assessment data in the digital twin model, supports multi-terminal interaction, and enables traffic managers to intuitively grasp real-time information, so as to more effectively carry out emergency handling and decision support. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Other features, objects and advantages of the present invention will become more apparent from the detailed description of non-limiting embodiments made with reference to the following drawings: Figure 1 This is a flow chart of a multi-terminal visual control method in a simulation environment according to Embodiment 1 of the present invention; Figure 2 This is a schematic diagram of the structure of a multi-terminal visual control system in a simulation environment in Example 2 of the present invention. DETAILED DESCRIPTION

[0016] Example 1 This embodiment constructs a multi-terminal visualization control method in a simulation environment. The method uses the urban traffic digital twin model to construct a real-time dynamic grid, which can effectively integrate monitoring data and urban road topology, so that the entire system can fully and real-time reflect the urban traffic situation; the spatial and temporal features are extracted from the real-time traffic data by combining the spatiotemporal graph convolutional neural network with LSTM, which greatly improves the prediction accuracy of the next cycle of traffic flow; by comparing with historical data to calculate the road saturation, and then calculating the traffic flow offset threat value according to a preset formula, the traffic status can be quantitatively evaluated, and abnormal data exceeding the threshold can be extracted to achieve early warning of potential risks. At the same time, the method locates and displays the prediction results and risk assessment data in the digital twin model, supports multi-terminal interaction, and enables traffic managers to intuitively grasp real-time information, so as to more effectively carry out emergency handling and decision support.

[0017] A multi-terminal visual control method in a simulation environment, such as Figure 1 As shown, the specific steps include: S1. Based on the urban traffic digital twin model, a dynamic grid visualization framework for urban traffic is constructed; S2. Based on the urban traffic dynamic grid visualization framework, a traffic flow prediction model is constructed to predict the traffic flow in the next cycle; S3, according to the traffic volume prediction result of the next cycle, calculate the saturation of the urban traffic roads of the next cycle, and output the saturation matrix of the urban traffic roads of the next cycle; S4. Calculate the traffic flow offset threat value of the next cycle according to the urban traffic road saturation matrix of the next cycle, preset the offset value threshold, compare the traffic flow offset threat value with the offset value threshold, extract data greater than the offset value threshold, and visualize it in the digital twin model to handle the threat in an orderly manner.

[0018] In this embodiment, constructing a city traffic dynamic grid visualization framework in S1 includes the following specific steps: S11. Within the preset collection cycle, at 5-minute intervals, obtain the real-time traffic flow data of the monitoring nodes of M roads. , It represents the traffic flow data of road i at the nth statistics in this collection period. The value of i is 1-M, and the value of n is 1-N. N represents the number of statistics in this collection period. The real-time traffic flow matrix Q is output. ,in, The dimension is The real number space of ; S12, constructing the urban road topological connection relationship matrix A, ,in, The dimension is The real number space of the city road topological connection relationship matrix A is Indicates the connection relationship between road i and road j. When road i is directly connected to road j, ,otherwise , when i=j, The real-time traffic flow matrix Q and the urban road topology connection relationship matrix A together constitute a dynamic grid visualization framework for urban traffic.

[0019] In this embodiment, S2 includes the following specific steps: S21, extracting the real-time traffic flow matrix Q and the urban road topological connection relationship matrix A, wherein the nth column of the real-time traffic flow matrix Q is a vector , , The dimension is The real number space, It represents the traffic volume of M roads counted for the nth time in this cycle, and the value of n is 1-N; S22, construct the traffic flow matrix of this cycle as input , the output is the traffic flow matrix for the next cycle The traffic flow prediction model outputs the traffic flow matrix P of the next cycle.

[0020] In this embodiment, S22 includes the following specific steps: S221, the traffic flow prediction model includes a spatiotemporal graph convolutional neural network part and a recursive neural network part, wherein the spatiotemporal graph convolutional neural network part is composed of a spatial graph convolutional layer and a temporal convolutional layer; the real-time traffic flow matrix Q and the urban road topology connection relationship matrix A are input into the spatial graph convolutional layer of the spatiotemporal graph convolutional neural network, and the output result of the spatial graph convolutional layer is , , The dimension is The real number space, K represents the spatial feature dimension, Input the time convolution layer, which uses a one-dimensional convolution kernel for convolution, and the output result is a matrix V. , The dimension is The real number space, C is the time dimension feature, and the operation formula of the time convolution layer is , the nth column of the matrix V is the vector , , the value of n is 1-N; S222, the recursive neural network part adopts an LSTM network structure, including 5 LSTM network layers and a fully connected layer. Input LSTM network structure and output the corresponding hidden state , , The dimension is The working process of the LSTM network structure is: .

[0021] In this embodiment, the loss function of the vehicle flow prediction model during training is: ; in, represents the true value of the nth column of the traffic flow matrix in the next cycle, represents the predicted value of the nth column of the traffic flow matrix in the next cycle, represents the norm, represents the regularization coefficient, Represents the L1 norm of the road topology connection relationship matrix A.

[0022] In this embodiment, S3 includes the following specific steps: S31. Extract the predicted traffic flow matrix for the next cycle ; S32. Traffic flow matrix The nth column vector in , the vector It indicates the predicted traffic volume of M roads in the nth statistical period of the next cycle, where n is 1-N. The time period of the nth statistical period of the next cycle is obtained, and the historical traffic volume average of the traffic volume of M roads in the time period in the past thirty days is obtained to calculate the road saturation. The calculation formula of the road saturation is: ;in, It indicates the road saturation of road i in the nth statistics in the next cycle. represents the traffic volume of road i in the nth statistics in the next cycle, It represents the historical traffic flow average of road i in the time period of the nth statistics in the next cycle; S33, calculate in sequence and output the road saturation matrix S of the M roads in the next cycle N times. , The dimension is The real number space.

[0023] In this embodiment, S4 includes the following specific steps: S41, extract the road saturation matrix S of M roads in the next cycle N times, where the single element in the matrix S is Indicates the road saturation of road i at the nth statistics in the next cycle; S42: Calculate the traffic flow deviation threat value of road i in the nth statistics in the next cycle , the calculation formula is: ; S43: Output the traffic flow deviation threat value matrix T of the M roads counted N times in the next cycle.

[0024] In this embodiment, S4 also includes: presetting an offset value threshold t=0.25, extracting elements in the vehicle flow offset threat value matrix T that are greater than the offset value threshold t, and locating and visualizing them in the digital twin model to handle the threats in an orderly manner.

[0025] Example 2 This embodiment proposes a multi-terminal visual control system in a simulation environment. Figure 2 As shown, it includes: visualization framework construction module, traffic flow prediction module, saturation calculation module, threat extraction module; The visualization framework construction module is used to construct a dynamic grid visualization framework for urban traffic based on the urban traffic digital twin model; The vehicle flow prediction module is used to construct a vehicle flow prediction model based on the urban traffic dynamic grid visualization framework to predict the vehicle flow of the next cycle; The saturation calculation module is used to calculate the saturation of urban traffic roads in the next cycle according to the traffic flow prediction result of the next cycle, and output the saturation matrix of urban traffic roads in the next cycle; The threat extraction module is used to calculate the vehicle flow offset threat value of the next cycle based on the urban traffic road saturation matrix of the next cycle, preset the offset value threshold, compare the vehicle flow offset threat value with the offset value threshold, and extract data greater than the offset value threshold for visual display in the digital twin model, so as to handle the threat in an orderly manner.

[0026] In this embodiment, the construction of the urban traffic dynamic grid visualization framework includes the following specific steps: First, within the preset collection cycle, at a time interval of 5 minutes, obtain the real-time traffic flow data of the monitoring nodes belonging to M roads. , It represents the traffic flow data of road i at the nth statistics in this collection period. The value of i is 1-M, and the value of n is 1-N. N represents the number of statistics in this collection period. The real-time traffic flow matrix Q is output. ,in, The dimension is Then, the urban road topological connection relationship matrix A is constructed. ,in, The dimension is The real number space of the city road topological connection relationship matrix A is Indicates the connection relationship between road i and road j. When road i is directly connected to road j, ,otherwise , when i=j, The real-time traffic flow matrix Q and the urban road topology connection relationship matrix A together constitute a dynamic grid visualization framework for urban traffic.

[0027] In this embodiment, the implementation of the traffic flow prediction module includes the following specific steps: First, extract the real-time traffic flow matrix Q and the urban road topology connection relationship matrix A, where the nth column of the real-time traffic flow matrix Q is a vector , , The dimension is The real number space, It represents the traffic flow of M roads counted for the nth time in this cycle, and the value of n is 1-N; then, the input is constructed as the traffic flow matrix of this cycle , the output is the traffic flow matrix for the next cycle The traffic flow prediction model outputs the traffic flow matrix P of the next cycle.

[0028] In this embodiment, the traffic flow prediction model includes a spatiotemporal graph convolutional neural network part and a recursive neural network part, wherein the spatiotemporal graph convolutional neural network part is composed of a spatial graph convolutional layer and a temporal convolutional layer; the real-time traffic flow matrix Q and the urban road topological connection relationship matrix A are input into the spatial graph convolutional layer of the spatiotemporal graph convolutional neural network, and the output result of the spatial graph convolutional layer is , , The dimension is The real number space, K represents the spatial feature dimension, Input the time convolution layer, which uses a one-dimensional convolution kernel for convolution, and the output result is a matrix V. , The dimension is The real number space, C is the time dimension feature, and the operation formula of the time convolution layer is , the nth column of the matrix V is the vector , , n is 1-N; the recursive neural network part adopts LSTM network structure, including 5 LSTM network layers and a fully connected layer. Input LSTM network structure and output the corresponding hidden state , , The dimension is The working process of the LSTM network structure is: .

[0029] In this embodiment, the loss function during the training of the traffic flow prediction model is: ; in, represents the true value of the nth column of the traffic flow matrix in the next cycle, represents the predicted value of the nth column of the traffic flow matrix in the next cycle, represents the norm, represents the regularization coefficient, Represents the L1 norm of the road topology connection relationship matrix A.

[0030] In this embodiment, the implementation of the saturation calculation module includes the following specific steps: first extract the predicted traffic flow matrix for the next cycle ; Then for the traffic flow matrix The nth column vector in , the vector It indicates the predicted traffic volume of M roads in the nth statistical period of the next cycle, where n is 1-N. The time period of the nth statistical period of the next cycle is obtained, and the historical traffic volume average of the traffic volume of M roads in the time period in the past thirty days is obtained to calculate the road saturation. The calculation formula of the road saturation is: ;in, It indicates the road saturation of road i in the nth statistics in the next cycle. represents the traffic volume of road i in the nth statistics in the next cycle, represents the historical traffic flow mean of road i in the time period of the nth statistical period in the next cycle; finally, the calculations are performed in sequence to output the road saturation matrix S of M roads in the Nth statistical period in the next cycle. , The dimension is The real number space.

[0031] In this embodiment, the implementation of the threat extraction module includes the following specific steps: first extract the road saturation matrix S of M roads in the next cycle N times, where a single element in the matrix S represents the road saturation of road i at the nth statistical time in the next cycle; next, calculate the traffic flow deviation threat value of road i at the nth statistical time in the next cycle , the calculation formula is: ; Finally, the traffic flow offset threat value matrix T of the M roads counted N times in the next cycle is output.

[0032] In this embodiment, the implementation of the threat extraction module also includes: presetting an offset value threshold t=0.25, extracting elements in the vehicle flow offset threat value matrix T that are greater than the offset value threshold t, and locating and visualizing them in the digital twin model to handle the threats in an orderly manner.

[0033] The above-mentioned parameters and steps for each unit module to realize corresponding functions in a multi-terminal visualization control system under a simulation environment of the present invention can refer to the parameters and steps in the embodiment of a multi-terminal visualization control method under a simulation environment in Example 1 above.

[0034] Example 3 This embodiment provides an electronic device, including: a processor and a memory, wherein the memory stores a computer program that can be called by the processor; the processor executes the above-mentioned multi-terminal visualization control method in a simulation environment by calling the computer program stored in the memory.

[0035] The electronic device may have relatively large differences due to different configurations or performances, and may include one or more processors (Central Processing Units, CPU) and one or more memories, wherein the memory stores at least one computer program, and the computer program is loaded and executed by the processor to implement a multi-terminal visualization control method in a simulation environment provided by the above method embodiment. The electronic device may also include other components for implementing the functions of the device, for example, the electronic device may also have components such as a wired or wireless network interface and an input and output interface to input and output data. This embodiment will not be described in detail here.

[0036] Those skilled in the art know that the present invention can be implemented as a system, method or computer program product. Therefore, the present disclosure can be specifically implemented in the following forms, namely: it can be complete hardware, it can be complete software (including firmware, resident software, microcode, etc.), or it can be a combination of hardware and software, which is generally referred to as a "circuit", "module" or "system" herein. In addition, in some embodiments, the present invention can also be implemented in the form of a computer program product in one or more computer-readable media, and the computer-readable medium contains computer-readable program code.

[0037] Any combination of one or more computer-readable media may be used. A computer-readable medium may be a computer-readable signal medium or a computer-readable storage medium. A computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or device, or any combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In this document, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in conjunction with an instruction execution system, device, or device.

[0038] The present invention is described with reference to flowcharts and block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process or block in the flowchart and block diagram, as well as the combination of processes and blocks in the flowchart or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts. Figure 1 Process or multiple processes and boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0039] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 Process or multiple processes and boxes Figure 1 The steps for the functions specified in one or more boxes.

[0040] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the enlightenment of the present invention, ordinary technicians in this field can also make many forms without departing from the scope of protection of the purpose of the present invention and the claims, which all fall within the protection of the present invention.

Claims

1. A multi-terminal visual control method in a simulation environment, characterized in that: The specific steps include: S1. Based on the urban traffic digital twin model, a dynamic grid visualization framework for urban traffic is constructed; S2. Based on the urban traffic dynamic grid visualization framework, a traffic flow prediction model is constructed to predict the traffic flow in the next cycle; S3, according to the traffic volume prediction result of the next cycle, calculate the saturation of the urban traffic roads of the next cycle, and output the saturation matrix of the urban traffic roads of the next cycle; S4. Calculate the traffic flow offset threat value of the next cycle according to the urban traffic road saturation matrix of the next cycle, preset the offset value threshold, compare the traffic flow offset threat value with the offset value threshold, extract data greater than the offset value threshold, and visualize it in the digital twin model to handle the threat in an orderly manner.

2. The multi-terminal visualization control method in a simulation environment according to claim 1, characterized in that: The construction of the urban traffic dynamic grid visualization framework in S1 includes the following specific steps: S11. Within the preset collection cycle, at 5-minute intervals, obtain the real-time traffic flow data of the monitoring nodes of M roads. , It represents the traffic flow data of road i at the nth statistics in this collection period. The value of i is 1-M, and the value of n is 1-N. N represents the number of statistics in this collection period. The real-time traffic flow matrix Q is output. ,in, The dimension is The real number space of ; S12, constructing the urban road topological connection relationship matrix A, ,in, The dimension is The real number space of the city road topological connection relationship matrix A is Indicates the connection relationship between road i and road j. When road i is directly connected to road j, ,otherwise , when i=j, The real-time traffic flow matrix Q and the urban road topology connection relationship matrix A together constitute a dynamic grid visualization framework for urban traffic.

3. The multi-terminal visualization control method in a simulation environment according to claim 2, characterized in that: The S2 comprises the following specific steps: S21, extracting the real-time traffic flow matrix Q and the urban road topological connection relationship matrix A, wherein the nth column of the real-time traffic flow matrix Q is a vector , , The dimension is The real number space, It represents the traffic volume of M roads counted for the nth time in this cycle, and the value of n is 1-N; S22, construct the traffic flow matrix of this cycle as input , the output is the traffic flow matrix for the next cycle The traffic flow prediction model outputs the traffic flow matrix P of the next cycle.

4. The multi-terminal visualization control method in a simulation environment according to claim 3, characterized in that: The S22 comprises the following specific steps: S221, the traffic flow prediction model includes a spatiotemporal graph convolutional neural network part and a recursive neural network part, wherein the spatiotemporal graph convolutional neural network part is composed of a spatial graph convolutional layer and a temporal convolutional layer; the real-time traffic flow matrix Q and the urban road topology connection relationship matrix A are input into the spatial graph convolutional layer of the spatiotemporal graph convolutional neural network, and the output result of the spatial graph convolutional layer is , , The dimension is The real number space, K represents the spatial feature dimension, Input the time convolution layer, which uses a one-dimensional convolution kernel for convolution, and the output result is a matrix V. , The dimension is The real number space, C is the time dimension feature, and the operation formula of the time convolution layer is , the nth column of the matrix V is the vector , , the value of n is 1-N; S222, the recursive neural network part adopts an LSTM network structure, including 5 LSTM network layers and a fully connected layer. Input LSTM network structure and output the corresponding hidden state , , The dimension is The working process of the LSTM network structure is: .

5. The multi-terminal visualization control method in a simulation environment according to claim 4 is characterized in that: The loss function of the traffic flow prediction model during training is: ； in, represents the true value of the nth column of the traffic flow matrix in the next cycle, represents the predicted value of the nth column of the traffic flow matrix in the next cycle, represents the norm, represents the regularization coefficient, Represents the L1 norm of the road topology connection relationship matrix A.

6. The multi-terminal visualization control method in a simulation environment according to claim 5, characterized in that: The S3 includes the following specific steps: S31. Extract the predicted traffic flow matrix for the next cycle ; S32. Traffic flow matrix The nth column vector in , the vector It indicates the predicted traffic volume of M roads in the nth statistical period of the next cycle, where n is 1-N. The time period of the nth statistical period of the next cycle is obtained, and the historical traffic volume average of the traffic volume of M roads in the time period in the past thirty days is obtained to calculate the road saturation. The calculation formula of the road saturation is: ;in, It indicates the road saturation of road i in the nth statistics in the next cycle. represents the traffic volume of road i in the nth statistics in the next cycle, It represents the historical traffic flow average of road i in the time period of the nth statistics in the next cycle; S33, calculate in sequence and output the road saturation matrix S of the M roads in the next cycle N times. , The dimension is The real number space.

7. The multi-terminal visualization control method in a simulation environment according to claim 6, characterized in that: The S4 includes the following specific steps: S41, extract the road saturation matrix S of M roads in the next cycle N times, where the single element in the matrix S is Indicates the road saturation of road i at the nth statistics in the next cycle; S42: Calculate the traffic flow deviation threat value of road i in the nth statistics in the next cycle , the calculation formula is: ； S43: Output the traffic flow deviation threat value matrix T of the M roads counted N times in the next cycle.

8. The multi-terminal visualization control method in a simulation environment according to claim 7, characterized in that: The S4 also includes: presetting an offset value threshold t=0.25, extracting elements greater than the offset value threshold t in the vehicle flow offset threat value matrix T, locating and visualizing them in the digital twin model, and processing the threats in an orderly manner.

9. A multi-terminal visualization control system in a simulation environment, which is implemented based on a multi-terminal visualization control method in a simulation environment according to any one of claims 1 to 8, characterized in that: The system includes: a visualization framework construction module, a vehicle flow prediction module, a saturation calculation module, and a threat extraction module; The visualization framework construction module is used to construct a dynamic grid visualization framework for urban traffic based on the urban traffic digital twin model; The vehicle flow prediction module is used to construct a vehicle flow prediction model based on the urban traffic dynamic grid visualization framework to predict the vehicle flow of the next cycle; The saturation calculation module is used to calculate the saturation of urban traffic roads in the next cycle according to the traffic flow prediction result of the next cycle, and output the saturation matrix of urban traffic roads in the next cycle; The threat extraction module is used to calculate the vehicle flow offset threat value of the next cycle based on the urban traffic road saturation matrix of the next cycle, preset the offset value threshold, compare the vehicle flow offset threat value with the offset value threshold, and extract data greater than the offset value threshold for visual display in the digital twin model, so as to handle the threat in an orderly manner.

Images