Low-cost train operation control simulation system based on digital twinning and SSA-DBO algorithm

Abstract

The invention discloses a low-cost train operation control simulation system based on digital twinning and an SSA-DBO algorithm. The system is composed of hardware, software and a digital twinborn body. Firstly, a single chip microcomputer, a wireless communication module, a common cathode nixie tube, a yellow-green LED lamp, a motor driving module and the like are designed as hardware, and the modules form a physical entity; secondly, according to software design, input signals of a track circuit are monitored, and corresponding logic judgment and processing are carried out according to received information; and finally, the digital twin designs a virtual model based on a physical system, the virtual model comprises a train, a track, a sensor and other components, various state information of the physical system is collected in real time through a track circuit, the train sensor and the like of the solid model, the state information is uploaded to a server by utilizing the Internet of Things technology, and after data screening is carried out through an SSA-DBO algorithm, the state information of the physical system is obtained. The data and the model together drive the digital twin for monitoring, analysis, optimization, prediction and visualization.

Description

Technical Field

[0001] The present invention relates to the field of digital twin technology, and more particularly to a low-cost train operation control simulation system based on digital twin and SSA-DBO algorithm. Background Art

[0002] Although the existing fully automatic train control system is powerful, safe and reliable, its high cost makes it not suitable for current teaching simulation experiments or miniaturized applications while meeting the need for visualization. Therefore, the necessity of designing a low-cost, high-performance train operation control simulation system suitable for the above scenarios is becoming increasingly prominent.

[0003] Chinese Patent Application CN107818703A provides a virtual reality railway teaching device, which combines the physical object of the railway model and the three-dimensional holographic image of the railway model to vividly display the relevant information of the railway, increase the interest in the learning process, and improve the teaching quality. However, since this patent application does not mention hardware simulation, students cannot intuitively feel the train operation control. Chinese Patent CN219872616U proposes an intelligent training diagnosis device, which includes a simulated track and a camera module, and also includes a simulated train operation device; this patent application restores the installation dimensions of the trackside equipment on the operating line one by one, simulates the uniform running of the train, and uses the method of setting simulated train pictures on the simulated train operation device to simulate the train operation situation, realizing indoor linear array camera image acquisition, indoor fault simulation, indoor camera calibration, indoor standardized operation drill, etc., so as to effectively improve the technical business level of employees and achieve the purpose of improving work efficiency. However, this patent application cannot perform digital twin analysis and optimization on the data and predict faults. Chinese Patent CN115526021A proposes a method, medium and electronic device for simulating the operation of a train, and the method includes: after the simulated train completes positioning, when the current transponder is triggered, according to the fast transmission pulse, the acceleration a of the simulated train, the speed V of the simulated train, the state update period T of the simulated train and the transponder window D, calculate the current displacement error ΔS of the simulated train; use the current displacement error ΔS to correct the current displacement of the simulated train; this patent application can ensure that the simulated train runs stably for a certain period of time, and thus meet the stability test requirements. However, this patent application cannot be tested and verified for accuracy through an actual physical train. Summary of the Invention

[0004] Aiming at the deficiencies of the above-mentioned existing technologies, the present invention aims to provide a low-cost train operation control simulation system based on digital twin and SSA-DBO algorithm. Specifically, a single-chip microcomputer is used to implement the train operation control simulation system to ensure the accuracy of data acquisition and the immediacy of data processing; a complete set of simulation systems is designed, including hardware selection, software design, data communication, etc.; the present invention can realize the monitoring, data analysis optimization and prediction of trains by combining digital twins, and then display them on display devices such as screens. The system of the present invention not only achieves low cost, but also overall manages the module components in the entire system to ensure high performance. While being suitable for teaching simulation experiments or miniaturized applications, it also takes into account the visual demand scenarios, providing an intuitive operation interface and the function of displaying real-time data.

[0005] The technical solution of the present invention is specifically introduced as follows.

[0006] The present invention provides a low-cost train operation control simulation system based on digital twin and SSA-DBO algorithm, which includes a physical entity and a digital twin; the physical entity and the digital twin are connected through a communication module for real-time data exchange, so that the digital twin can perform dynamic simulation, monitoring, analysis, optimization, and feedback the results of the optimal vehicle speed and the optimal inter-vehicle distance after analysis and optimization prediction of the digital twin to the physical entity, so that the train in the physical entity adjusts the vehicle speed or the inter-vehicle distance according to the feedback data to form a complete control system; where:

[0007] The physical entity includes a track circuit, a controller, a digital tube display module, a signal lamp module, and a motor drive module; the track circuit is used to detect the running state and position of the train, and transmit the train occupancy information to the controller in the form of an electrical signal; the controller makes corresponding logical judgments and processes according to the received information: controls the signal lamp module to turn on a red light when there is a train on the track, and controls the signal lamp module to turn on a green light when there is no train; uses the digital tube display module to display the section where the train is located; calculates the distance between the sections where two trains are located, and uses the motor drive module to control the start and stop of the train;

[0008] The digital twin is a virtual model designed based on the physical entity. It receives the data transmitted from the track circuit to the controller and the real-time data of the train wirelessly transmitted to the digital twin database server, and further realizes the prediction of the train position and train speed through the combination of the SSA-DBO algorithm and the prediction model, as well as optimizing the controller parameters for the monitoring, analysis, optimization, prediction and visualization of the physical entity.

[0009] In the present invention, the communication module adopts a Bluetooth wireless communication module.

[0010] In the present invention, the digital twin can extract data such as speed, acceleration, and train spacing from historical train operation data, and then use the SSA-DBO algorithm to optimize the above-mentioned data such as speed, acceleration, and train spacing. The optimized data is transmitted to the long short-term memory network (LSTM). The LSTM combines historical data to construct a prediction model to ensure that the train motion state can be accurately captured.

[0011] In the present invention, the digital twin can comprehensively analyze various factors that affect the change of train speed. These factors include but are not limited to the total track mileage, the number of trains on the track, the formation mode, time, location, weather conditions, and natural disasters, etc. After performing feature engineering processing on these factors, the SSA-DBO algorithm is used to optimize the dataset of the prediction model. Then, using the long short-term memory network (LSTM), combined with historical data, the data such as the speed, acceleration, and train spacing of the train are predicted. This method can not only effectively capture the dynamic changes during the train operation, but also provide reliable prediction results under complex environmental conditions, thus ensuring the safety and efficiency of the train operation.

[0012] In the present invention, in the following coefficient parameter of the SSA-DBO algorithm, 0.8 is taken, and the rolling step size is a dynamically adjusted step size, that is, as the number of iterations increases, the step size is reduced to perform accurate data selection for predicting position and speed.

[0013] In the SSA algorithm, the following coefficient refers to the degree of dependence of the followers in the sparrow algorithm on the position of the discoverer when updating their own positions. A higher following coefficient will make the followers more inclined to follow the discoverer, which can enhance the local search ability; while a lower following coefficient will cause the followers to be more independent, thus enhancing the global search ability.

[0014] In the present invention, the following coefficient is taken as 0.8, and the speed of other trains changes around the speed of the first train, that is, the optimal solution of the speed of the other trains sought is around the speed of the first train, so as to ensure that the speeds of each train are maintained within the same level of speed range, preventing the rear train from having too high a speed and causing a rear-end collision with the front train or the rear train having too low a speed resulting in low efficiency of the entire system.

[0015] In the DBO algorithm, the rolling step size refers to the distance that controls the movement of the dung beetle during each position update. The parameter of the rolling step size is dynamically adjusted, and the rolling step size gradually decreases as the number of iterations increases. This can balance the global search and local development capabilities. A larger step size in the early stage allows the dung beetle to quickly cover a wider search space, which helps to discover potential high-quality solutions; a smaller rolling step size in the later stage enables the dung beetle to perform more detailed searches to accurately locate the position of the optimal solution.

[0016] In the present invention, at the beginning of each iteration, a round of position update of the SSA algorithm is first executed, and then based on the updated result, the position adjustment of the DBO algorithm is performed. By combining the different role behaviors in the two algorithms, the leaders and followers in the SSA algorithm and the behavior models such as dung beetle ball rolling, dung beetle dancing, dung beetle reproduction, small dung beetle foraging, and dung beetle stealing in the DBO algorithm are selected to optimize the data.

[0017] In the present invention, an initialization strategy in the SSA algorithm is used to generate an initial population. Since the SSA algorithm has good randomness and diversity when exploring the search space, the SSA algorithm is used for global search to quickly locate the potential optimal solution region. The optimal solution of the SSA is used as the initial population of the DBO, and the local development ability of the DBO is utilized for fine search.

[0018] In the global search stage, the SSA algorithm is adopted, which helps to quickly cover a larger search space. In the local search stage, the DBO algorithm is adopted, which helps to refine the quality of the solution and avoid falling into local optima.

[0019] In the present invention, after data optimization, a long short-term memory network (LSTM) is used to predict data such as the speed, acceleration, and train-to-train distance of the train. As a powerful recurrent neural network, the LSTM is good at processing time series data and can effectively capture the dynamic changes during the train operation. Through the training of a large amount of historical train operation data, the prediction model based on the LSTM can learn the temporal patterns and rules during the train operation, can handle various fluctuations and non-linear relationships during the train operation, and then predict data such as speed, acceleration, and train-to-train distance in the future for a period of time. The LSTM also has memory and forgetting mechanisms, can maintain the memory of historical data for a long time, and can adjust the model in a timely manner according to new input data to maintain the accuracy and robustness of the prediction.

[0020] In the present invention, the prediction model gives the predicted position, speed, train-to-train distance, etc. of the train. The digital twin can adjust the train control system through wireless communication according to the optimized prediction parameters. For example, it can adjust the vehicle speed and control the distance between trains.

[0021] As described above, the present invention integrates hardware parts including a single-chip microcomputer control, wireless communication, track circuit detection and display, and other hardware modules. The physical system exchanges real-time data with the digital twin through wireless communication. The digital twin can perform dynamic simulation, monitoring, analysis, and optimization based on data such as speed, acceleration, and train-to-train distance transmitted by the sensors of the physical system, and feedback the results to the physical system, thus forming a complete control system. Compared with the prior art, the present invention has the following advantages:

[0022] 1. The present invention simplifies the system structure by integrating multiple functional modules and improves the overall reliability and stability.

[0023] 2. Through lightweight design and modular structure, the present invention enhances environmental adaptability and portability, enabling it to be directly used in environments such as classroom podiums and desk tops in scenarios like railway engineering education, scientific research, and miniaturized applications.

[0024] 3. The present invention has high scalability and flexibility. The system has good scalability, and additional modules such as tracks, trains, and signal lights can be added to the physical entity according to actual needs.

[0025] 4. The present invention utilizes digital twin technology to update the virtual model in real time through sensor data, enabling the monitoring and visualization of the operation of the physical entity on a computer. At the same time, researchers can optimize train operation through virtual simulation analysis and conduct fault prediction and health management.

[0026] 5. The present invention utilizes the SSA-DBO algorithm, which uses the Sparrow Search Algorithm (SSA) for global search and combines the local optimization of the Dung Beetle Optimization Algorithm (DBO), improving the data optimization effect of the train digital twin model.

[0027] 6. The present invention utilizes LSTM (Long Short-Term Memory Network), which can excellently process time series data, predict trends based on historical data, and can adapt to various different types of time series data, having a strong fitting ability for non-linear relationships and complex dynamic change sequence data. Description of the Drawings

[0028] Figure 1 For the overall hardware design.

[0029] Figure 2 For Figure 1 The circuit design diagram of the main control device.

[0030] Figure 3 For Figure 1 The circuit design diagram of the on-vehicle device.

[0031] Figure 4 For the control program flow chart.

[0032] Figure 5 For the hardware system test Figure 1 .

[0033] Figure 6 For the hardware system test Figure 2 .

[0034] Figure 7 For the data exchange flow chart between the physical entity and the virtual entity.

[0035] Figure 8 For the flow chart of the SSA-DBO algorithm and the LSTM long short-term neural network model. Specific Embodiments

[0036] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions in the embodiments of this application will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of this application. Obviously, the described embodiments are some, but not all, of the embodiments of this application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in this application without creative efforts fall within the scope of protection of this application.

[0037] The present invention provides a low-cost train operation control simulation platform based on digital twin and SSA-DBO algorithm, including hardware design, software design, and digital twin body design.

[0038] The hardware design of the present invention is divided into two parts: a master control device and an on-vehicle device. The two devices in the hardware design of the present invention both use the STC89C52 single-chip microcomputer as the control core to handle logical operations, communication tasks, and interactions with external devices; the master control device uses a DC track circuit to detect the running state and position of the train, that is, when a train enters a track section, the corresponding input port of the single-chip microcomputer is triggered by changing the intensity of the track current; the master control device uses a common-cathode digital tube to realize the digital information display function and adopts dynamic scanning technology to improve the utilization rate of hardware resources; the master control device uses red and green LED lights as an intuitive operation interface indicator, with green indicating that the train track section is not occupied and red indicating that the track section is occupied; the on-vehicle device uses a motor drive module to start and stop the train.

[0039] The software design of the present invention is that the single-chip microcomputer monitors the input signal of the track circuit and makes corresponding logical judgments and processing based on the received information. For example, if there is a train on the track, the red light is on, if there is no train, the green light is on, and if two trains are adjacent, the following train is stopped using wireless communication for logical judgment to ensure that the current state can be quickly and accurately understood.

[0040] The digital twin body of the present invention designs a virtual model based on the physical system. The virtual model includes components such as trains, tracks, and sensors. Various state information of the physical system is collected in real time through sensors and uploaded to the server through Internet of Things technology for monitoring, analysis, optimization, prediction, and visualization.

[0041] The following further elaborates on the embodiments of the present invention in conjunction with the accompanying drawings.

[0042] As Figure 1 shown, the overall hardware design of the system in this embodiment includes a track circuit, a Bluetooth wireless communication module, a digital tube display module, a signal lamp module, a single-chip microcomputer (controller), and a control module (motor drive module), etc.

[0043] AsFigure 2 As shown in the figure, the circuit design diagram of the system master control device in this embodiment includes four parts: the track circuit and the single-chip microcomputer module, the communication module and the single-chip microcomputer module, the digital tube and the single-chip microcomputer module, and the power supply module and each module. Among them:

[0044] The track circuit and the single-chip microcomputer module in this embodiment: The track circuit detects the occupancy of the train and transmits it to the single-chip microcomputer module in the form of an electrical signal. The single-chip microcomputer module controls the start of the following train according to the occupancy of the train, so as to realize the following train tracking the preceding train and keeping the distance within a certain range.

[0045] The communication module and the single-chip microcomputer module in this embodiment: The single-chip microcomputer calculates the distance between the two trains in the section according to the information transmitted by the track circuit. If the distance is too close, the single-chip microcomputer sends information to the communication module, and then transmits it to the following train through the communication module to stop the following train; if the distance is too far, the single-chip microcomputer sends information to the communication module, and then transmits it to the following train through the communication module to make the following train run.

[0046] The digital tube and the single-chip microcomputer module in this embodiment: The single-chip microcomputer uses the digital tube to display the sections where the two trains are located according to the information transmitted by the track circuit.

[0047] The power supply module and each module in this embodiment: The power supply module provides stable power for the entire system and protects each module from the influence of power noise and interference through power filter capacitors.

[0048] The interface allocation of the master control device in this embodiment is shown in Table 1.

[0049] Table 1 Interface Allocation of the Master Control Device Single-Chip Microcomputer

[0050]

[0051] As Figure 3 shown, the circuit design diagram of the system vehicle-mounted device in this embodiment includes two parts: the Bluetooth wireless communication module and the train operation control module. The vehicle-mounted device is a core control system designed around a single-chip microcomputer, integrating power management, reset mechanism, clock generation, serial communication and general input / output functions. A reliable reset circuit is realized by means of R17 and C3 components to ensure the initialization process of the system. A 12Mhz crystal oscillator is used in cooperation with a capacitor to provide a clock signal for the single-chip microcomputer. In terms of communication, the P3.0 TXD and P3.1 RXD pins are connected to the Bluetooth wireless communication module to support data transmission. The P1.0 port is connected to the motor drive module to control the start and stop of the train motor.

[0052] The interface allocation of the vehicle-mounted device in this embodiment is shown in Table 2.

[0053] Table 2 Circuit Design of Vehicle-Mounted Device and Single-Chip Microcomputer Interface

[0054]

[0055] As shown Figure 4 in the figure, the software program running steps of this embodiment are as follows:

[0056] The first step: Initialize each module;

[0057] The second step: The single-chip microcomputer in the master control device judges whether there is voltage in the circuit;

[0058] The third step: If there is no voltage, send a signal to the single-chip microcomputer in the master control device indicating that the track is not occupied by a vehicle, control the signal lamp to turn on the green light, and then return to the second step; if there is voltage, send a signal to the single-chip microcomputer in the master control device indicating that the track is occupied by a vehicle, control the signal lamp to turn on the red light.

[0059] The fourth step: The single-chip microcomputer in the master control device judges whether two trains are in adjacent track sections;

[0060] The fifth step: If they are not in adjacent sections, the single-chip microcomputer in the master control device sends a driving signal to the train through the signal sending module, and then returns to the second step; if they are in adjacent sections, the single-chip microcomputer in the master control device sends a stop signal to the train through the signal sending module, and then returns to the second step.

[0061] As shown Figure 5 in the figure, after testing, the digital tube of the master control device of this embodiment can display the track section where the vehicle is located (section 1 in the figure), and in the track section where the vehicle is located, the signal lamp of the master control device turns on the red light.

[0062] As shown Figure 6 in the figure, when there are two trains on the physical entity platform of this embodiment, the digital tube of the master control device can display the track sections where the two trains are located (section 1 and section 3 in the figure), and in the track section where the vehicle is located, the signal lamp of the master control device turns on the red light.

[0063] The digital twin model of this embodiment builds the UI based on RealityCapture 1.5 and Unreal Engine 5.5. The UI includes the real-time speed of the train, the predicted speed of the train, the running state of the train, and the running position of the train. Using the digital twin model of this embodiment for dynamic simulation can select factors such as the total track mileage, the number of trains on the track, the formation mode, time, location, weather, natural disasters, etc. to predict faults.

[0064] The digital twin model of this embodiment uses the SSA-DBO algorithm for data analysis and then inputs it to the LSTM long short-term memory network to predict the train position and train speed.

[0065] As shown Figure 7As shown, in the interaction between the virtual model and the physical model, after the single-chip microcomputer of the master control device in this embodiment receives the track circuit signal, it sends the track circuit signal data to the server database while judging whether there is a train in the subsequent track section and whether the occupied sections are adjacent, and sends a control signal. Subsequently, the server database optimizes the data through the SSA-DBO algorithm to find the optimal vehicle speed and optimal position, and then forms a digital twin model through the fusion of data-driven and model-driven of the virtual model. Subsequently, monitoring, dynamic simulation, analysis and optimization of the simulated data are carried out, and then the data is sent to the train through wireless communication and displayed on the display device together with the virtual model for students to refer to.

[0066] As Figure 8 In the flow chart of the SSA-DBO algorithm and the LSTM long short-term neural network model of this embodiment as shown, at the beginning of each iteration, first perform a round of position update of the SSA algorithm, and then perform position adjustment of the DBO algorithm based on the updated result. Combine the different role behaviors in the two algorithms to select the leader and follower in the SSA algorithm and the behavior models such as dung beetle ball rolling, dung beetle dancing, dung beetle reproduction, small dung beetle foraging, and dung beetle stealing in the DBO algorithm to optimize the data. In the SSA algorithm, initialize the strategy to generate the initial population for global search, quickly locate the potential optimal solution area, use the optimal solution of the SSA as the initial population of the DBO, and use the local development ability of the DBO for fine search. After data optimization, use the long short-term memory network LSTM to predict the data such as the speed, acceleration, and train spacing of the train.

[0067] It should be noted that in this application, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover a non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not expressly listed, or also includes elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "including a..." does not exclude the existence of additional identical elements in the process, method, article or device including the element.

[0068] The above are only specific embodiments of the present application, enabling those skilled in the art to understand or implement the present application. Various modifications to these embodiments will be obvious to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to these embodiments shown herein, but rather to the broadest scope consistent with the principles and novel features claimed herein.

Claims

1. A low-cost train operation control simulation system based on digital twin and SSA-DBO algorithm, characterized by: It includes a physical entity and a digital twin; the physical entity and the digital twin are connected through a communication module for real-time data exchange, so that the digital twin can perform dynamic simulation and monitoring, analysis, optimization, and feed back the results of the optimal speed and optimal inter-vehicle distance predicted by the digital twin analysis and optimization to the physical entity, so that the train in the physical entity can adjust the speed or inter-vehicle distance according to the feedback data, forming a complete control system; wherein: The physical entity includes a track circuit, a controller, a digital tube display module, a signal light module and a motor drive module; the track circuit is used to detect the running state and position of the train, and transmit the train occupancy information to the controller in the form of an electrical signal; the controller performs corresponding logical judgment and processing according to the received information: when there is a car on the track, the signal light module is controlled to light a red light, and when there is no car, the signal light module is controlled to light a green light; the digital tube display module is used to display the section where the train is located; the distance between the sections where the two trains are located is calculated, and the motor drive module is used to control the start and stop of the train; The digital twin is a virtual model designed based on a physical entity. It transmits the data on the track section occupied by the train transmitted to the controller by the received track circuit and the real-time data including the train speed, acceleration, and whether the train is moving obtained by the train sensors to the digital twin database server, and further realizes the prediction of train position and train speed through the prediction model, so as to monitor, analyze, optimize, predict and visualize the physical entity.

2. The system according to claim 1, characterized in that The communication module adopts Bluetooth wireless communication module.

3. The system according to claim 1, characterized in that The digital twin extracts speed, acceleration and train spacing information from historical driving data, and then uses the SSA-DBO algorithm to optimize the speed, acceleration and train spacing data, and passes the optimized data to the long short-term memory network LSTM. LSTM combines historical data to build a prediction model to ensure that the train movement status can be accurately captured.

4. The system according to claim 1, characterized in that After the digital twin performs feature engineering on various factors that affect train speed changes, the SSA-DBO algorithm is used to optimize the parameter configuration of the prediction model, and then the long short-term memory network LSTM is used to calculate the predicted speed and position; the details are as follows: At the beginning of each iteration, a round of position update of the SSA algorithm is executed first, and then the position of the DBO algorithm is adjusted based on the updated result. The leaders and followers in the SSA algorithm and the dung beetle rolling, dancing, breeding, foraging and stealing behavior models in the DBO algorithm are selected to optimize the data including speed, acceleration and distance between trains. In the SSA algorithm, the strategy is initialized to generate the initial population for global search, and the potential optimal solution area is quickly located. The optimal solution of SSA is used as the initial population of DBO, and the local development capability of DBO is used for fine search. After data optimization, the long short-term memory network LSTM is used to predict the speed, acceleration and train spacing data of the train.

5. The system according to claim 1, characterized in that The following coefficient parameter of the SSA-DBO algorithm is set to 0.8, that is, the first train acts as the leader to ensure that the following trains closely follow the first train in speed changes to prevent rear-end collisions; the rolling step size is a dynamically adjusted step size, that is, as the number of iterations increases, the step size is reduced to select accurate data for predicting position and speed.

Images