Rail transit system signal transmission method, electronic device and simulation system

Abstract

This application discloses a signal transmission method, electronic device, and simulation system for a rail transit system, relating to the field of rail transit engineering technology. The method includes determining, when a simulation model of a first subsystem of the rail transit system transmits a signal to a simulation model of a second subsystem, whether the original interface signal output by the simulation model of the first subsystem is a signal type with a known functional form, a signal type with sparse characteristics in the transform domain, a signal type constrained by physical laws, or a signal type interacting between heterogeneous models; reconstructing the interface signal according to the signal type into an interface signal that undistortsly represents the original physical process and matches the simulation step size of the receiving end, and inputting it into the simulation model of the second subsystem. This application can solve the problems of signal distortion, timing errors, and low computational efficiency in related technologies, enabling seamless interaction between subsystems at different time scales and ensuring the safe and efficient operation of the rail transit system.

Description

Technical Field

[0001] This application relates to the field of rail transit engineering technology, and in particular to a signal transmission method, electronic equipment and simulation system for rail transit systems. Background Technology

[0002] The rail transit system consists of multiple interdependent subsystems and has the characteristics of multiple physical domains and spanning time scales, such as microsecond-level train control signals and second-level passenger flow scheduling. Only through precise coordination among the subsystems can the safe, stable and efficient operation of the rail transit system be ensured.

[0003] Related technologies simulate the interaction process of different subsystems of a rail transit system through digital simulation technology. However, these technologies suffer from problems such as signal distortion, timing disorder, and low computational efficiency. They cannot accurately and realistically reflect the behavior of the physical system, thus affecting the safe and efficient operation of the rail transit system.

[0004] It should be noted that the information disclosed in the background section above is only used to enhance the understanding of the background of this application, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0005] This application provides a signal transmission method, electronic equipment, and simulation system for a rail transit system, which solves the problems of signal distortion, timing disorder, and low computational efficiency caused by the mismatch of simulation step sizes between subsystems, and enables seamless interaction between subsystems at different time scales, ensuring the safe and efficient operation of the rail transit system.

[0006] To solve the above-mentioned technical problems, this application provides the following technical solution: This application provides a signal transmission method for a rail transit system, including: When the simulation model of the first subsystem corresponding to the first subsystem of the rail transit system transmits a signal to the simulation model of the second subsystem corresponding to the second subsystem, the original interface signal output by the simulation model of the first subsystem is obtained. Determine the target signal type corresponding to the original interface signal; the target signal type is a first signal type with a known functional form, or a second signal type with sparse characteristics in the transform domain, or a third signal type constrained by physical laws, or a fourth signal type that interacts between heterogeneous models; Based on the target signal type, the interface signal is reconstructed into a target interface signal that undistorts the original physical process and matches the simulation step size of the second subsystem simulation model; The target interface signal is input into the second subsystem simulation model.

[0007] This application also provides an electronic device, including a memory and a processor, wherein the processor is used to implement the steps of the signal transmission method of the above-described rail transit system when executing a computer program stored in the memory.

[0008] Finally, this application also provides a simulation system for a rail transit system, including a first simulation device, a second simulation device, and a signal reconstruction processor; The first simulation device and the second simulation device are physically independent computing devices connected via a network, and both are connected to the signal reconstruction processor; The first simulation device runs a first subsystem model with a first simulation step size and a second subsystem model with a second simulation step size, and the second simulation device runs a third subsystem model with a third simulation step size; wherein the first simulation step size, the second simulation step size, and the third simulation step size are different from each other; The signal reconstruction processor is configured to implement the steps of the signal transmission method for the rail transit system when executing a computer program during signal transmission between the first subsystem model and the second subsystem model, and / or between the second subsystem model and the third subsystem model.

[0009] The advantage of the technical solution provided in this application lies in the fact that when the various digital subsystem models of the rail transit system interact with each other, they first identify the specific target signal type of the original interface signal of the transmitting end, adaptively select and apply the most matching high-fidelity reconstruction algorithm, and directly overcome the drawbacks of related technologies that use forced synchronization or simple interpolation methods through this intelligent reconstruction method based on signal type adaptation. It can perform the most matching high-fidelity signal reconstruction for different types of signals according to their own inherent characteristics, thereby fundamentally avoiding the loss and distortion of signal features caused by sampling rate mismatch or coarse processing, and ensuring simulation accuracy. At the same time, since each subsystem model can adopt the optimal simulation step according to its own dynamic requirements, Long-term operation eliminates the need for mandatory uniformity to a minimum step size, transforming the global and continuous high computational overhead into local and intelligent signal processing overhead at the interface, significantly saving computational resources and improving simulation efficiency. Furthermore, addressing the transmission delay issue in distributed simulation, the transmission of the original waveform, susceptible to delay, is transformed into robust parameter transmission and synchronous reconstruction, effectively suppressing timing errors and numerical instability caused by delay. This enhances the robustness and reliability of the entire collaborative simulation system, resolving the long-standing problems of signal distortion, timing errors, and low computational efficiency in multi-timescale collaborative simulation of rail transit. It enables seamless interaction between subsystems at different timescales, ensuring the safe and efficient operation of the rail transit system. In addition, the rail transit system and electronic equipment described in this application possess corresponding advantages.

[0010] The technical features mentioned above, those to be mentioned below, and those shown individually in the accompanying drawings can be arbitrarily combined, as long as the combined technical features are not contradictory. All feasible combinations of features are the technical content explicitly described in this application. Any one of the multiple sub-features contained in the same statement can be applied independently, without necessarily being applied together with other sub-features.

[0011] It should be understood that the above general description and the following detailed description are merely exemplary and do not limit this application. Attached Figure Description

[0012] To more clearly illustrate the technical solutions of this application or related technologies, the drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0013] Figure 1 A schematic flowchart of a signal transmission method for a rail transit system provided in this application; Figure 2 A schematic diagram of the original interface signals in the first exemplary application scenario provided in this application; Figure 3 A schematic diagram of the signal spectrum of the original interface signal in the first exemplary application scenario provided in this application; Figure 4 A schematic diagram of the target interface signal of the original interface signal in the first exemplary application scenario provided in this application; Figure 5 A schematic diagram of the original interface signals in the second exemplary application scenario provided in this application; Figure 6 A schematic diagram of the spectrum of the original interface signal in the second exemplary application scenario provided in this application; Figure 7 A schematic diagram of the target interface signal of the original interface signal in the second exemplary application scenario provided in this application; Figure 8 A schematic diagram of the signal reconstruction method for the third signal type provided in this application; Figure 9 A schematic diagram showing the spatiotemporal evolution comparison between the reconstructed temperature field and the actual temperature field provided in this application; Figure 10 A schematic diagram comparing the restored harmonics and signal harmonics in the fourth exemplary application scenario provided in this application; Figure 11A structural framework diagram of an exemplary embodiment of the simulation system for the rail transit system provided in this application; Figure 12 A schematic diagram of the original interface signals in the system application scenario provided in this application; Figure 13 A schematic diagram comparing the original interface signal and the destination interface signal in the first system application scenario provided in this application; Figure 14 A schematic diagram of the original interface signal and the target interface signal in the second system application scenario provided in this application. Detailed Implementation

[0014] To enable those skilled in the art to better understand the technical solutions of this application, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments. The terms "first," "second," "third," "fourth," etc., used in the specification and the aforementioned drawings are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. The term "exemplary" means "serving as an example, embodiment, or illustration." Any embodiment described herein as "exemplary" is not necessarily to be construed as superior to or better than other embodiments.

[0015] Rail transit systems are dynamic systems composed of multiple interdependent and interacting subsystems, such as track circuits, train control systems, and transport equipment. Their safe and efficient operation depends on the precise coordination of each subsystem. Field testing of rail transit systems requires substantial financial investment and may lead to safety accidents. Digital simulation technology, which simulates the behavior and performance of rail transit systems under various scenarios, such as fault tolerance and dynamic response, can identify potential problems. This not only avoids the costs and risks of physical experiments but also ensures the safe, stable, and efficient operation of actual rail transit systems, becoming a technical means for reliability verification and design optimization of rail transit systems.

[0016] The dynamic response speeds of various digital subsystems in rail transit systems vary greatly. For example, train control signals require microsecond-level simulation steps, while passenger flow scheduling can be completed in seconds. Multi-scale model collaborative simulation needs to address the coordination problem of asynchronous long subsystems, but this can lead to signal distortion across time scales (feature loss due to sampling rate mismatch during high- and low-frequency signal interaction) and timing discrepancies. Existing technologies using single-time-scale compatibility or simple interpolation cannot achieve distortion-free signal transmission and timing consistency between asynchronous long subsystems, while simultaneously conserving computational resources and maintaining simulation accuracy. Specifically, single-time-scale compatibility schemes force all subsystems (such as second-level passenger flow scheduling models) to operate in microsecond-level steps, resulting in a huge waste of computational resources. For low-frequency subsystems, their model states do not fundamentally change in most simulation steps, yet they need to be repeatedly calculated millions of times. This leads to massive and ineffective computational overhead, making simulation extremely inefficient and requiring extremely powerful computing hardware, thus violating the goal of conserving computational resources and resulting in high costs. While simple interpolation completion schemes allow each subsystem to operate at optimal efficiency, improper handling at the data interaction interface can introduce signal distortion and reduce simulation accuracy. Furthermore, simple interpolation completion schemes, such as coarse interpolation and extrapolation algorithms, can introduce numerical noise and delays into the co-simulation system. These non-physical numerical disturbances can propagate and feedback between different subsystems. In high-order, nonlinear, and strongly coupled dynamic systems, such as train operation and control loops, small numerical errors can be rapidly amplified, leading to problems such as simulation result divergence, non-physical oscillations, or non-repeatability dependent on simulation step size. This results in numerical instability of the entire simulation system, reducing the reliability of the simulation results, and exhibiting poor system stability, making it unsuitable as a basis for high reliability verification.

[0017] Therefore, it is evident that the relevant technologies suffer from signal distortion, timing discrepancies, and inefficiency due to simulation step size mismatches between subsystems. Signal distortion causes deformation of the waveform, amplitude, or phase of critical information transmitted between subsystems, such as train control commands, track circuit status, and sensor readings, making it impossible to accurately predict the behavior of the real system. Timing discrepancies lead to logical confusion, with different sub-models operating asynchronously in logical time. For example, the train model may have already moved forward, but the signal model's state may not have been updated, or there may be simulation delays in the trackside equipment's response to control center commands.

[0018] In view of this, in order to solve the problems of signal distortion caused by model simulation step size differences, low computational efficiency caused by mandatory uniform minimum step size, and system timing disorder and numerical instability caused by transmission delay and coarse interpolation when performing multi-model, distributed, cross-timescale collaborative simulation of complex large systems such as rail transit, this study aims to ensure seamless interaction of subsystem models at different time scales and improve simulation accuracy and efficiency.

[0019] In the process of multi-model collaborative simulation of rail transit systems, this invention first analyzes the characteristics of three types of models in the system—equipment, facilities, and information—to clarify their mixed characteristics of continuous dynamics and discrete events. Then, for ideal environments, it proposes matching techniques such as time-domain to frequency-domain conversion and interpolation calculation to solve the problems of insufficient sampling and aliasing. For physical environment transmission delays, it employs SOGI (Second-Order Generalized Integrator) or MSOGI (Multiple Second-Order Generalized Integrator) algorithms to process and analyze the types of models and their characteristics contained in the signal. Different modeling methods, modeling granularity, and model running time scales are implemented for different model types, and a systematic approach is used to achieve matching and collaboration between multiple systems. Based on prior knowledge of the interactive signals (such as known function forms, frequency domain sparsity, physical constraints, and anti-delay requirements), the optimal algorithm is dynamically selected to reconstruct the signals at the interfaces of different subsystem model interactions with high fidelity and high efficiency, thereby achieving seamless collaborative simulation across time scales.

[0020] The various non-limiting embodiments of this application are described in detail below with reference to the accompanying drawings and specific embodiments. First, please refer to... Figure 1 According to the signal transmission method for a rail transit system provided in this application, it can be implemented as a computer program product, installed and run on various node devices of a digital simulation platform supporting the rail transit system, such as user terminal devices and servers, for implementing digital simulation of the rail transit system. In some embodiments of the method, the method includes the following steps: S101: When the simulation model of the first subsystem corresponding to the first subsystem of the rail transit system transmits a signal to the simulation model of the second subsystem corresponding to the second subsystem, the original interface signal output by the simulation model of the first subsystem is obtained.

[0021] The rail transit system is a complex large system composed of multiple interdependent and dynamically interacting subsystems, including tracks, vehicles, signaling, power supply, and communication. The first subsystem simulation model is the digital simulation model corresponding to the subsystem initiating signal transmission within the rail transit system, such as a simulation model simulating the operating state of track circuits, used to output raw interface signals reflecting physical quantities such as track circuit current and voltage. The second subsystem simulation model is the digital simulation model corresponding to the subsystem receiving signals within the rail transit system, such as a simulation model simulating the train control system receiving track circuit signals. The raw interface signals are signals directly output by the first subsystem simulation model during operation, used for interaction between subsystems. Their form and content depend on the functional characteristics of the first subsystem and may include voltage waveforms, control command sequences, etc.

[0022] For example, when the simulation model of the first subsystem corresponding to the first subsystem in the rail transit system (e.g., passenger flow scheduling model A with a simulation step size of 10ms) needs to transmit signals to the simulation model of the second subsystem corresponding to the second subsystem (e.g., train control core logic model B with a simulation step size of 200us), the original interface signal generated by it is obtained from the output port of the simulation model of the first subsystem. This original interface signal is a discretized representation of the original physical process (e.g., changes in track circuit current) at the time scale of the first model.

[0023] S102: Determine the target signal type corresponding to the original interface signal.

[0024] The target signal type refers to the category categorized based on the inherent characteristics of the original interface signal. This includes the first signal type, whose functional form is known; the second signal type, which exhibits sparse characteristics in the transform domain; the third signal type, constrained by physical laws; and the fourth signal type, representing interactions between heterogeneous models. The reconstruction methods for different signal types differ. The first signal type refers to a signal whose mathematical model (e.g., sine function, impulse function) can be known in advance. The second signal type refers to a signal whose energy or information is concentrated in a few components within a certain mathematical transform domain (e.g., frequency domain, wavelet domain). The third signal type refers to a signal whose generation and changes follow specific physical laws (e.g., dynamic equations, energy conservation). The fourth signal type refers to signals transmitted between simulation models that are physically distributed, have different computing capabilities, or different clock sources, and are susceptible to network latency.

[0025] S103: Based on the target signal type, reconstruct the interface signal into a target interface signal that undistorts the original physical process and matches the simulation step size of the second subsystem simulation model.

[0026] The target interface signal, after reconstruction, is a signal that can represent the original physical process without distortion and matches the simulation step size of the second subsystem simulation model; it is the effective input signal for the second subsystem simulation model. The first signal type can represent the original physical process without distortion, for example, by ensuring waveform distortion-free waveform transmission through parameter passing. The second signal type can represent the original physical process without distortion, for example, by accurately reconstructing it from a small amount of data, ensuring information distortion-free information. The third signal type can represent the original physical process without distortion by ensuring the signal conforms to physical laws; its essence is also distortion-free. The fourth signal type can represent the original physical process without distortion, for example, by using phase-locked loop (PLL) to compensate for delays, ensuring timing and waveform distortion-free waveforms.

[0027] In this embodiment, based on the target signal type determined in step S102, a dedicated signal reconstruction algorithm matching that type is invoked to intelligently reconstruct the original interface signal bound to the first model step size into a completely new target interface signal. This target interface signal must meet two requirements: 1) Distortion-free representation of the original physical process: The reconstructed signal must retain all the key physical characteristics represented by the original signal (such as amplitude, frequency, phase, waveform shape, and the underlying physical laws); 2) Matching the simulation step size of the second subsystem simulation model: The interval between data points (i.e., the equivalent sampling rate) of the reconstructed signal sequence must be adapted to the simulation step size of the second model so that the second model can directly and correctly receive and process it.

[0028] S104: Input the target interface signal into the simulation model of the second subsystem.

[0029] The high-fidelity target interface signal with an appropriate step size generated in step S103 is input into the simulation model of the second subsystem as its driving input or state feedback, thereby completing this cross-time scale signal interaction.

[0030] In the technical solution provided in this application embodiment, when the digital subsystem models of the rail transit system interact with each other, the specific target signal type of the original interface signal of the transmitting end is first identified, and the most matching high-fidelity reconstruction algorithm is adaptively selected and applied. This intelligent reconstruction method based on signal type adaptation directly overcomes the drawbacks of related technologies that use forced synchronization or simple interpolation methods. It can perform the most matching high-fidelity signal reconstruction for different types of signals according to their own inherent characteristics, thereby fundamentally avoiding signal feature loss and distortion caused by sampling rate mismatch or coarse processing, and ensuring simulation accuracy. At the same time, since each subsystem model can adopt the optimal simulation step according to its own dynamic requirements, Long-term operation eliminates the need for mandatory uniformity to a minimum step size, transforming the global and continuous high computational overhead into local and intelligent signal processing overhead at the interface, greatly saving computing resources and improving simulation efficiency. In addition, addressing the transmission delay problem in distributed simulation, the transmission of the original waveform, which is susceptible to delay, is transformed into robust parameter transmission and synchronous reconstruction, effectively suppressing timing errors and numerical instability caused by delay. This enhances the robustness and reliability of the entire collaborative simulation system, solving the long-standing problems of signal distortion, timing errors, and low computational efficiency in multi-timescale collaborative simulation of rail transit. It enables seamless interaction between subsystems at different time scales, ensuring the safe and efficient operation of the rail transit system.

[0031] Considering that inaccurate determination of the original interface signal type can directly lead to improper selection of subsequent reconstruction methods, resulting in signal distortion and low reconstruction efficiency, in order to ensure accurate determination of the target signal type based on the inherent characteristics of the signal and the transmission scenario, this embodiment also provides an implementation method for automatically or semi-automatically determining the target signal type based on the signal source and characteristics, which may include the following: If the original interface signal is a physical quantity or control command with standard waveform characteristics in a rail transit system, then the target signal type is the first signal type; if the energy of the original interface signal in the frequency domain is concentrated on a finite number of discrete frequency components, then the target signal type is the second signal type; if the original interface signal is the direct output of the physical system and its variation is governed by physical laws, then the target signal type is the third signal type; if the original interface signal is transmitted between simulation models that are distributed and have different processing capabilities, then the target signal type is the fourth signal type.

[0032] Standard waveform characteristics refer to signals with fixed and predictable waveform shapes, such as sine waves, square waves, and pulse waves. Their variation patterns conform to clear mathematical models or industry standards and are commonly found in standardized signals in rail transit systems. Physical quantities are measurable parameters used to describe the physical state of a rail transit system, such as current, voltage, pressure, and speed; they are data exchanged between subsystems. Control commands are instruction signals used in rail transit systems to achieve coordinated control between subsystems, such as train acceleration commands and turnout switching commands; they have clear functional orientations and standard formats. Frequency domain energy refers to the energy distribution of a signal along the frequency dimension; the energy magnitude of different frequency components reflects the spectral characteristics of the signal. Discrete frequency components refer to frequency points in the signal spectrum that appear as isolated peaks; signal energy is concentrated at these finite frequency points rather than being continuously distributed. A physical system refers to the actual hardware equipment or physical structures in a rail transit system, such as tracks, trains, and sensors; its operating state and output signals are constrained by objective physical laws. Physical laws are objective laws governing the operation of physical systems and signal changes, such as Newton's laws of motion, the law of electromagnetic induction, and the law of heat conduction; they are the intrinsic basis for signal changes. Distributed deployment refers to the deployment of multiple subsystem simulation models on different physical computing devices, which interact with each other via a network. These devices can be geographically dispersed and physically independent. Differences in processing capabilities mean that the computing devices hosting the different subsystem simulation models vary in terms of computing speed, storage capacity, and data processing efficiency, resulting in differences in the signal processing and response speeds of the models.

[0033] In this embodiment, the following rule knowledge base can be established: If the original interface signal is a physical quantity or control command with standard waveform characteristics in the rail transit system, such as power frequency (50Hz) track circuit current, PWM (pulse width modulation) drive signal, standard communication protocol frame, etc., the waveform function of these signals is determined by industry standards or circuit design, and is therefore determined to be the first signal type. If the energy of the original interface signal in the frequency domain is concentrated on a finite number of discrete frequency components, and if only a few obvious peaks are found in its spectrum (such as several natural vibration frequencies of the train bogie, power line harmonics), it is determined to be the second signal type. If the original interface signal is the direct output of the physical system and its variation is governed by physical laws, such as the axle sensor signal being constrained by wheelset kinematics, the motor torque signal being constrained by electromagnetic equations, and the battery charging and discharging current being constrained by electrochemical equations, this type of signal is determined to be the third signal type. If the original interface signal is transmitted between simulation models with distributed deployment and different processing capabilities, the judgment is made based on the architecture information of the simulation system. If the signal sender and receiver models are running on different physical servers (i.e., simulation devices), or if their simulation clock sources are independent, it is determined to be the fourth signal type, and the anti-delay processing procedure is initiated.

[0034] As can be seen from the above, this embodiment achieves accurate and rapid identification of the original interface signal type by clarifying the judgment criteria for various target signal types, avoiding the selection of the reconstruction method due to misjudgment of the type; accurate signal type identification provides a prerequisite for selecting the optimal reconstruction algorithm for different signal characteristics, ensuring that the reconstruction process is more targeted, improving both the accuracy and efficiency of signal reconstruction, and laying the foundation for seamless signal transmission across time scales and scenarios.

[0035] In the above embodiments, there is no limitation on how to perform step S103. This embodiment provides an implementation method for reconstructing the target interface signal according to the target signal type, which may include the following steps: If the original interface signal is of the first signal type, the discrete data points of the original interface signal are reconstructed into a continuous and smooth complete signal waveform on the target time scale corresponding to the simulation model of the second subsystem, which is used as the target interface signal; if the original interface signal is of the second signal type, a signal consistent with the original high-dimensional signal in terms of time-domain waveform and frequency-domain components is restored from the original interface signal, which is used as the target interface signal; if the original interface signal is of the third signal type, the original interface signal is fitted under the corresponding physical rules as physical constraints to obtain the target interface signal; if the original interface signal is of the fourth signal type, the amplitude, frequency and phase characteristic parameters of the original interface signal are extracted, and a signal synchronized with the original interface signal is regenerated, which is used as the target interface signal.

[0036] Among them, physical rules refer to the physical laws that are subject to constraints, and the original high-dimensional signal is a theoretical reference benchmark, which refers to the complete, high-data-volume real physical signal under ideal conditions (i.e., without distortion or downsampling). The original interface signal is the actual processing object, which is the signal that is actually received at the interface after being output from the first subsystem model in the simulation system, and is often downsampled or distorted.

[0037] In this embodiment, if the original interface signal is of the first signal type (with a known function form), this type of signal is highly dependent on the matching degree of the prior model, and the approximate function form of the signal needs to be determined in advance. It is suitable for distortion repair and downsampling restoration of scene signals with known function forms, such as the classic 50Hz signal of a track circuit and typical power electronic signals. At this time, the goal of reconstruction is to restore the signal, which is roughly represented by discrete data points, into a complete signal waveform that is continuous and smooth on the second model time scale. For example, for a track circuit current signal that is known to be a 50Hz sine wave, the reconstruction algorithm does not simply perform linear interpolation on sparse points, but generates a standard sine wave that continuously varies on the target time scale based on the sine function model, and uses this as the target interface signal to ensure the integrity of the waveform features. If the original interface signal is of the second signal type (sparse in the transform domain), the goal of reconstruction is to restore a signal that is consistent with the original high-dimensional signal in terms of time domain waveform and frequency domain components from the signal that may lose information due to the reduction in sampling rate. This method leverages its sparse prior knowledge in the frequency domain to infer the complete signal with high precision from limited data using mathematical methods (such as compressed sensing), rather than simply performing data interpolation. It is suitable for simulations of strongly coupled, multi-scale mixed rail transit systems, and is particularly advantageous in scenarios where the behavior of the digital twin must be consistent with that of the physical system. If the original interface signal is of the third signal type (constrained by physical laws), the reconstruction process must treat the physical rules describing the signal as inviolable constraints. When fitting or reconstructing the signal, the reconstruction algorithm forces the generated target interface signal to satisfy these physical laws (e.g., the axle counting signal must match the wheelset size and speed), thus obtaining a target interface signal that both fits the data and conforms to physical common sense. If the original interface signal is of the fourth signal type (interaction between heterogeneous models), the timing asynchrony problem caused by transmission delay needs to be addressed. The reconstruction method extracts core characteristic parameters such as amplitude, frequency, and phase from the original signal, and then, at the receiving end, uses these parameters to regenerate a new signal synchronized with the original signal as the target interface signal. This method transforms the transmission of waveform streams, which are susceptible to latency, into the transmission of parameter packets, which are relatively insensitive to latency, and then accurately reproduces the waveform locally.

[0038] As can be seen from the above, this embodiment provides specific signal reconstruction methods for different signal types. It restores the continuous waveform of regular signals, restores sparse signals using their structural characteristics, ensures that physical signals conform to natural laws, and resists delays through parameter synchronization for network transmission signals. This targeted signal reconstruction processing, compared to single interpolation or forced synchronization methods, can achieve higher-quality signal restoration at the interface, thereby directly improving simulation accuracy and system robustness, while maintaining the flexibility of independent operation of each model.

[0039] Furthermore, this application also provides an exemplary reconstruction method for the target interface signal corresponding to a signal with a known functional form in a rail transit system. This information can be obtained by extracting and standardizing signal parameters, and then dynamically adjusting the parameters according to the target time scale to achieve cross-scale signal distortion-free reconstruction. The process may include the following: Using the simulation model of the first subsystem as the transmitting end, a signal reconstruction function is determined based on the original interface signal to characterize the amplitude variation law of the original interface signal over time. Based on the signal reconstruction function, corresponding target feature parameters are extracted from each discrete data point of the original interface signal. The target feature parameters are standardized and sent to the simulation model of the second subsystem. The simulation model of the second subsystem acts as the receiving end, receiving and parsing the standardized target feature parameters. Using the signal reconstruction function and the target feature parameters, a continuous and smooth complete signal waveform on the target time scale is reconstructed as the target interface signal.

[0040] Discrete data points are discrete data samples acquired at regular time intervals during the acquisition process of the original interface signal. Each data point contains the signal amplitude information at the corresponding moment. The target time scale is the time interval standard used by the simulation model of the second subsystem for simulation calculations, i.e., the time dimension corresponding to the simulation step size, such as 200µs, 1ms, etc. A continuous and smooth complete signal waveform can fully reflect the change law of the original signal, without data breaks or abrupt changes. The continuous waveform curve with derivatives can accurately reproduce the signal changes of the original physical process. The signal reconstruction function is a mathematical expression used to describe the change law of the amplitude of the original interface signal over time. For example, it is a function containing parameters such as amplitude, frequency, phase, and attenuation coefficient, and is the core basis for signal reconstruction. Target feature parameters are parameters extracted from the discrete data points of the original interface signal that can characterize the core characteristics of the signal, such as signal amplitude, frequency, phase, pulse width, etc. Standardization processing refers to the process of converting the extracted target feature parameters into a unified format and numerical range, which facilitates transmission and parsing between different subsystem simulation models and eliminates parsing errors caused by differences in parameter formats.

[0041] In this embodiment, signals such as track circuit current, train control command pulses, and turnout operating voltages in a hybrid rail transit system need to be transmitted between models at multiple time scales. Due to significant differences in simulation step sizes among the models (e.g., a microsecond-level model with a sampling rate of 1MHz, while a second-level model has a sampling rate of only 1Hz), directly transmitting the original signals can easily lead to signal distortion. For signals with known function forms (such as a 50Hz sine wave), a unified model can be established based on the parameterized characteristics of the signal. Key characteristic parameters of signals such as track circuit current (e.g., amplitude, frequency, phase, pulse width, rise time) are extracted and standardized from the original time scale. Then, according to the requirements of the target time scale, the parameters are dynamically adjusted to reconstruct a suitable signal form, ultimately achieving the goal of distortion-free signal transmission across time scales and high consistency between simulation results and the actual system. Taking a 50Hz track circuit current sinusoidal signal as an example, the first established signal reconstruction function relationship is: Here, s(t) is the time-varying signal, the dependent variable of the function, representing the signal amplitude at time t (e.g., voltage, displacement, sound pressure, etc.). A is the initial amplitude of the signal (amplitude at t=0), representing the maximum amplitude of the signal at the initial moment. e is the base of the natural logarithm, a constant, approximately equal to 2.71828. α is the attenuation coefficient (α>0), determining the rate of amplitude decay. The larger α is, the faster the amplitude decays; the smaller α is, the slower the decay. Time t is the independent variable of the function, usually in seconds (s). cos represents the cosine function, describing the oscillation characteristics of the signal. f is the oscillation frequency of the signal, in Hertz (Hz), representing the number of oscillation cycles completed per second. 2πf represents the angular frequency (ω), representing the oscillation radians per second, ω=2πf. Φ is the initial phase or phase offset, in radians (rad), determining the initial position of the cosine wave at t=0. Then, the signal sampling rate was reduced to 50 times, 150 times, and 250 times, respectively, equivalent to the sinusoidal signal of the track circuit current being input from the first subsystem simulation model A with a simulation step size of 1ms to other subsystem simulation models B, C, and D with simulation step sizes of 200us, 600us, and 4us. To verify the effectiveness of the signal reconstruction method in this embodiment, the above method was used, and to more closely approximate the real signal, a 100Hz harmonic signal was added before reconstructing the downsampled signal, as shown below. Figure 2 The number of sampling points obtained by the original signal through the simulation model A of the first subsystem, using the simulation step size as the sampling rate. Figure 3 It is the signal spectrum extracted through an algorithm. Figure 4 It reconstructs the signal by restoring it to model D using the signal spectrum and prior information, and can adapt to a simulation step size of 4µs.

[0042] As can be seen from the above, this embodiment avoids the bandwidth occupation and transmission delay problems caused by the transmission of a large number of discrete data points by extracting signal feature parameters and standardizing transmission. Based on the signal reconstruction function relationship and target feature parameters, the reconstruction ensures that the function form of the reconstructed signal is consistent with that of the original signal. It can accurately restore the waveform characteristics of the original signal on the target time scale and reflect the original physical process without distortion. This solves the problem of transmission distortion of signals with known function forms across time scales and improves the efficiency and accuracy of signal transmission.

[0043] Furthermore, this application also presents the process of accurately reconstructing the original signal in a rail transit system for a sparse signal in a certain transform domain (time domain, frequency domain, etc.) (such as train vibration waveform, train control communication signal) by collecting a small number of non-adaptive linear projection values ​​and using a reconstruction algorithm: Compressed sensing measurements are performed on the original interface signal to obtain the corresponding linear projection value. Based on the sparsity characteristics of the original interface signal in the frequency domain, the corresponding sparse transform basis and observation matrix are determined. Using the sparse transform basis and observation matrix, a sensing matrix for signal reconstruction is generated. Based on the sparse reconstruction signal restoration process, the target interface signal that is consistent with the original high-dimensional signal in terms of time domain waveform and frequency domain components is reconstructed according to the sensing matrix and linear projection value.

[0044] Compressed sensing measurement, based on compressed sensing theory, involves non-adaptive linear projection sampling of the original interface signal (e.g., a high-dimensional vibration signal) using an observation matrix independent of the signal (such as a Bernoulli random matrix). This allows for reconstruction of the original signal by acquiring only a small number of projection values, reducing the amount of data collected and transmitted. The linear projection values ​​are a set of low-dimensional data values ​​obtained after compressed sensing measurement, representing the projection of the original high-dimensional signal onto the observation matrix and containing key feature information of the original signal. Sparse transform bases are used to transform the original signal to orthogonal basis functions in the sparse domain, such as discrete Fourier transform bases or wavelet transform bases, making the signal exhibit sparse characteristics in this transform domain (i.e., most coefficients are zero or close to zero). The observation matrix is ​​used in compressed sensing measurement to perform linear projection of the original signal. Its construction must satisfy the finite isometry condition to ensure accurate reconstruction of the original signal from the linear projection values. The sensing matrix, formed by combining the sparse transform base and the observation matrix, is the core computational matrix in the compressed sensing signal reconstruction process, used to establish the correlation between the linear projection values ​​and the original signal. The original high-dimensional signal is the complete, high-data-volume real physical signal of the original interface signal under ideal conditions (i.e., distortion-free and downsampling-free). The sparse reconstruction signal restoration process is based on the perceptual matrix and linear projection values. It involves solving for the sparse coefficients through optimization algorithms to restore the original high-dimensional signal. In other words, given the linear projection values ​​obtained from the reconstructed signal, and using the perceptual matrix as the transformation relationship, an optimization problem for sparse reconstruction signal restoration is constructed and solved (e.g., minimizing the L1 norm to find the sparsest solution). Solving this problem allows the reconstruction of the target interface signal that is highly consistent with the original high-dimensional signal in terms of time-domain waveform and frequency-domain components.

[0045] In this embodiment, when train vibration waveforms, train control communication signals, and other signals are transmitted between multiple time-scale models in a hybrid rail transit system, the time-scale matching problem also exists, requiring reconstruction of the transmitted signals to prevent distortion. For this type of signal reconstruction problem—where the signal is compressible or sparse in a certain transform domain (time domain, frequency domain, wavelet domain, etc.)—compressed sensing theory can be used. Compressed sensing theory states that when a signal is compressible or sparse in a certain transform domain, by collecting a small number of non-adaptive linear projection values ​​of the signal, establishing a mathematical reconstruction model, and solving it using a reconstruction algorithm, the original signal can be reconstructed relatively accurately. Taking the reconstruction of a frequency-sparse signal as an example, firstly, a multi-frequency sinusoidal signal containing random phase and amplitude is generated, a Bernoulli observation matrix Φ is constructed, and then y=Φ... The compressed measurement value is obtained by signal', y is the low-dimensional measurement value obtained by linearly projecting the original signal signal through the observation matrix Φ, and signal' is the original high-dimensional signal. The sensing matrix is ​​then constructed based on the DFT (Discrete Fourier Transform) with 4x oversampling. An improved OMP (Orthogonal Matching Pursuit) algorithm with SVD (Singular Value Decomposition) regularization is run. Finally, the frequency and phase can be extracted by peak detection and interpolation, and amplitude correction and error analysis are performed.

[0046] To ensure the effectiveness of the method provided in this embodiment, this embodiment takes a signal length of 2560 as an example, and the original interface signal is a sparse frequency signal, such as... Figure 5 As shown, the frequency components of the original interface signal are as follows: Figure 6 As shown, the original sparse signal was accurately reconstructed using far fewer than 2560 measurements (800). The comparison between the reconstructed signal and the original signal is as follows: Figure 7 As shown, the mean error is 0.48%, which meets the expected signal reconstruction requirements.

[0047] As can be seen from the above, this embodiment converts high-dimensional sparse signals into low-dimensional linear projection values ​​for transmission through compressed sensing measurement, which significantly reduces the bandwidth occupation and transmission delay of signal transmission and improves transmission efficiency. Based on the reconstruction process of sparse transformation basis, observation matrix and sensing matrix, the time domain and frequency domain characteristics of the original high-dimensional signal can be accurately restored, avoiding signal distortion caused by sampling rate mismatch, ensuring the accurate transmission of sparse characteristic signals between subsystems across time scales, and providing reliable signal input for subsequent simulation calculations.

[0048] Furthermore, this application also provides an exemplary signal reconstruction method for signals in rail transit systems that need to satisfy specific physical laws (such as dynamic equations), which may include the following: The physical laws describing the original interface signal are converted into the physical constraints corresponding to the original interface signal; a mathematical model for signal reconstruction is established with the goal of minimizing the error between the reconstructed signal and the original interface signal and satisfying the physical constraints; the signal reconstruction model is calculated, and the target interface signal that satisfies the physical laws is output.

[0049] Among them, objective physical laws or rules governing the generation and changes of the original interface signal, such as the law of heat conduction, dynamic equations, and physical constraints on wheelset motion, determine the boundary and inherent logic of signal changes. Physical constraints transform physical laws into mathematical constraints that can be used for signal reconstruction, such as the temperature change rate constraint determined by the law of heat conduction and the axle-counting signal pulse interval constraint determined by the physical dimensions of the wheelset. The mathematical model for signal reconstruction aims to minimize the signal reconstruction error while incorporating mathematical expressions established by physical constraints. It guides the calculation of the signal reconstruction process, ensuring that the reconstruction result is both close to the original signal and conforms to physical laws. The error between the reconstructed signal and the original interface signal is an indicator such as the amplitude difference and waveform similarity between the reconstructed target interface signal and the original interface signal at corresponding moments. The smaller the error, the closer the reconstructed target interface signal is to the original interface signal.

[0050] In this embodiment, axle counting signals, interlocking signals, and block signals in a mixed rail transit system are all subject to certain physical constraints. For example, the physical dimensions of the wheelsets and their operating speeds all impose constraints on the axle counting signal. When these signals interact across time scales, to avoid signal deviations, they can be reconstructed under enhanced physical constraints, such as... Figure 8 As shown, prior physical rules can be transformed into constraints for the reconstruction algorithm, upgrading from unconstrained data-driven reconstruction to precise reconstruction guided by physical constraints, thus solving problems such as noise interference, signal loss, and distortion. First, the constrained module undergoes parameter calibration and raw signal acquisition in a real-world environment. Constraint functions are extracted and the module is modeled. The model data is reconstructed by embedding constraint functions. Finally, the simulation model is deployed on FPGA boards and compared with the actual module. Through parameter iteration, the model accuracy is continuously optimized. To verify the feasibility of signal reconstruction based on the physical constraint enhancement algorithm, the physical laws of heat conduction can be integrated into the thermal field reconstruction process and tested. The spatiotemporal evolution of the reconstructed temperature field is compared with the actual temperature field. Figure 9 As shown.

[0051] In this embodiment, physical constraint-enhanced reconstruction of the original interface signal of the third signal type incorporates physical laws into the compressed sensing algorithm, thereby ensuring the convergence of the compressed sensing technology. This method embeds prior knowledge of the physical system (such as dynamic equations, boundary conditions, sensor characteristics, etc.) as constraints into the reconstruction model, significantly improving reconstruction accuracy and robustness. For example, the optimization problem of physical constraint-enhanced reconstruction, i.e., the signal reconstruction mathematical model, can be expressed as the following formula: .

[0052] in, Let χ be the signal to be reconstructed in physical space, R represent the real number field, and N be the dimension of the signal to be reconstructed. Let y be the measurement vector, and M be the dimension of the measurement vector y. N, For the measurement matrix, For sparse regularization, The physical constraint regularization term is used to force the reconstruction result to satisfy physical laws. Its specific form depends on the application scenario. λ1 is the weight coefficient of the sparse regularization term, and λ2 is the weight coefficient of the physical constraint regularization term. Solving this optimization model that incorporates physical constraints, the final output is a target interface signal that both matches the observed data and strictly conforms to physical laws.

[0053] Furthermore, this application also provides an exemplary signal reconstruction method for signals in rail transit systems that need to satisfy specific physical laws (such as dynamic equations), applicable to distributed simulation scenarios with network transmission delays, and may include the following: A phase-locked loop is used to extract the amplitude, frequency, and phase characteristics of the original interface signal in real time. The amplitude, frequency, and phase characteristics are used as initial conditions to configure the center frequency and phase of the adaptive filter. The adaptive filter is then driven to track and reconstruct the received components of the original interface signal to generate a target interface signal that is synchronized with the original interface signal in terms of amplitude, frequency, and phase.

[0054] Phase-locked loop (PLL) feature extraction refers to the real-time tracking of the original interface signal at the signal transmitting or receiving end using PLL technology to accurately extract its instantaneous amplitude, frequency, and phase characteristic parameters. Configuring the adaptive filter involves using the extracted frequency and phase characteristic parameters as initial conditions to configure the center resonant frequency and initial phase of the adaptive filter (such as SOGI or MSOGI), enabling it to quickly lock onto the input signal. Tracking and reconstruction drive the adaptive filter to track the received signal (which may be distorted due to delay). The filter effectively suppresses jitter caused by noise and delay, generating a clean target interface signal that is strictly synchronized with the original signal in amplitude, frequency, and phase. MSOGI, based on SOGI, implements multi-band collaborative processing. This method achieves enhanced system stability through three dimensions: decoupling the extraction of fundamental and harmonic components, significantly improved resistance to complex interference, and enhanced system stability. For example, this embodiment addresses the need for high-speed interaction of signals between digital systems in a heterogeneous computing unit and multi-machine distributed environment. To resolve the time scale mismatch caused by transmission delay, a signal reconstruction method is provided below, which can restore signal accuracy and prevent signal deviation due to transmission: A1: The amplitude and frequency of the original interface signal output by the first subsystem are phase-locked and transmitted.

[0055] In this step, the raw signal acquisition device B (trackside axle counter) collects the pulse signals generated when the train wheelsets pass over the track. The raw signal is a continuous high-dimensional sequence (e.g., 10 seconds × 1kHz sampling rate, 10,000 sampling points), containing pulses from the passing wheelsets, track vibration noise, electromagnetic interference, etc. Phase-locked loop (PLL) feature extraction uses a built-in PLL to perform phase-locking processing on the raw pulse signal, extracting and locking three feature parameters: amplitude A: pulse peak value (e.g., 3V), reflecting the signal strength of the wheelsets passing the sensor; frequency f: pulse repetition frequency (e.g., 3.2Hz), directly corresponding to the speed of the passing wheelsets; phase θ: the initial phase of the pulse (e.g., π / 3), reflecting the timing start of the passing wheelsets. To reduce transmission bandwidth, sparse sampling projects the high-dimensional raw signal into a low-dimensional observation vector. For example, of the original 1000 data points, only 10% of the effective sampling points are retained. The phase-locking parameters (A, f, θ) and the low-dimensional observation data are transmitted to the control center via a dedicated rail transit data communication network (e.g., LTE-M).

[0056] A2: Reconstruct the original interface signal based on the transmitted parameters to achieve signal matching.

[0057] The system receives the phase-locked loop (PLL) parameters (amplitude A, frequency f, phase θ) and low-dimensional observation data transmitted from terminal B. The PLL parameters can be used as the initial input to the MSOGI or SOGI algorithm, driving the filter to converge to the target frequency and reconstructing the complete signal. Initialization of the SOGI or SOGI filter: The frequency f obtained from PLL is set as the resonant frequency of the SOGI or SOGI algorithm, and amplitude A is used as the reference for the output amplitude. The MSOGI or SOGI algorithm reconstructs the complete pulse sequence from the low-dimensional observation data through resonant filtering, automatically filtering out noise introduced during transmission. Phase calibration of the reconstructed signal is performed based on the phase θ obtained from PLL, ensuring complete timing alignment with the original signal. The reconstructed signal is then matched and verified against the PLL parameters transmitted from terminal B: Amplitude matching: The error between the peak value of the reconstructed signal and the PLL amplitude A is ≤0.1V; Frequency matching: The error between the pulse frequency of the reconstructed signal and the PLL frequency f is ≤0.01Hz; Waveform matching: The timing of the reconstructed pulse is completely aligned with the phase θ of the original signal, without offset. After passing the above verifications, the reconstructed signal is used as the input signal for the next small-step simulation model. To verify the feasibility of the reconstruction method, a sinusoidal signal injected with harmonics was reconstructed using the method described above. The fundamental frequency was reconstructed with high accuracy, and the test results for the harmonic reconstruction are as follows: Figure 10 As shown, the signal synchronization error reconstructed by the algorithm compensation model is controlled within an acceptable range for engineering applications.

[0058] MSOGI is composed of multiple SOGI connected in parallel, and the state equation of SOGI is as follows: ; When the input signal u = Usin(ωt + Φ), the output vector solution is: .

[0059] In the formula, It is the time derivative of the state vector, which describes the dynamic rate of change of the internal state of SOGI and is used to track the frequency and phase of the input signal; The in-phase state components eventually converge to the in-phase component of the input signal; The orthogonal state components eventually converge to the orthogonal components of the input signal; and They are A is the time derivative; A is the state matrix, a 2×2 system matrix describing the dynamic changes of the internal state, with elements determined by the damping coefficient k and the adaptive angular frequency ω′; B is the input matrix, a 2×1 input matrix describing the driving effect of the input signal on the state; k is the algorithm parameter, the damping coefficient; ω′ is the algorithm parameter, the adaptive tracking angular frequency (unit: rad / s), updated in real time through a nonlinear law; y is the output vector, a 2×1 output vector containing in-phase output u′ and quadrature output qu′; the u′ output component is in-phase, in phase with the input signal u; the qu′ output component is quadrature, with a 90° phase difference (90° lag) from the input signal u, where q is the quadrature component identifier; C is a 2×2 output matrix that maps the state vector to the output vector, representing the internal state of SOGI. , It is converted into a directly usable in-phase / quadrature output signal. U is the signal parameter, which is the amplitude (peak value) of the input sine signal; ω is the signal parameter, which is the angular frequency of the input sine signal (ω=2πf, where f is the frequency, unit: Hz). These are signal parameters, representing the initial phase (in rad) of the input sinusoidal signal.

[0060] As can be seen from the above, this embodiment extracts signal feature parameters in real time through a phase-locked loop, ensuring accurate capture of the original signal characteristics; the adaptive filter is configured based on the feature parameters and tracks the signal, effectively compensating for the transmission delay and signal offset caused by the differences in processing capabilities of distributed deployment and heterogeneous models, realizing the synchronous reconstruction of signal amplitude, frequency and phase, solving the problem of timing disorder in signal transmission between heterogeneous models, and improving the stability and accuracy of signal transmission.

[0061] It should be noted that there is no strict order of execution for the steps in this application. As long as they conform to a logical order, these steps can be executed simultaneously or in a certain preset order. Figure 1This is just an illustrative example and does not mean that this is the only possible execution order.

[0062] This application also provides a corresponding apparatus for the signal transmission method of a rail transit system, further enhancing the practicality of the method. The apparatus can be described from both a functional module perspective and a hardware perspective. The signal transmission apparatus for a rail transit system provided in this application is described below. This apparatus is used to implement the signal transmission method for a rail transit system provided in this application. In this embodiment, the signal transmission apparatus for a rail transit system may include or be divided into one or more program modules. These one or more program modules are stored in a storage medium and executed by one or more processors to complete the signal transmission method for a rail transit system disclosed in Embodiment 1. The program module referred to in this embodiment refers to a series of computer program instruction segments capable of performing specific functions, which are more suitable than the program itself for describing the execution process of the signal transmission apparatus for a rail transit system in the storage medium. The following description will specifically introduce the functions of each program module in this embodiment. The signal transmission apparatus for a rail transit system described below can be referred to in correspondence with the signal transmission method for a rail transit system described above.

[0063] From the perspective of functional modules, the signal transmission device for the rail transit system provided in this embodiment may include: The raw signal acquisition module is used to acquire the raw interface signal output by the simulation model of the first subsystem when the simulation model of the first subsystem corresponding to the first subsystem of the rail transit system transmits a signal to the simulation model of the second subsystem corresponding to the second subsystem. The signal type identification module is used to determine the target signal type corresponding to the original interface signal. The target signal type is either a first signal type with a known functional form, or a second signal type with sparse characteristics in the transform domain, or a third signal type constrained by physical laws, or a fourth signal type that interacts between heterogeneous models. The signal reconstruction module is used to reconstruct the interface signal into a target interface signal that accurately represents the original physical process and matches the simulation step size of the second subsystem simulation model, based on the target signal type; and input the target interface signal into the second subsystem simulation model.

[0064] For example, in some embodiments of this example, the signal reconstruction module described above may further be used for: If the original interface signal is of the first signal type, the discrete data points of the original interface signal are reconstructed into a continuous and smooth complete signal waveform on the target time scale corresponding to the simulation model of the second subsystem, which is used as the target interface signal. If the original interface signal is of the second signal type, a signal that is consistent with the original high-dimensional signal in terms of time domain waveform and frequency domain composition is recovered from the original interface signal and used as the target interface signal. If the original interface signal is of the third type, the original interface signal is fitted under the condition that the corresponding physical rules are used as physical constraints to obtain the target interface signal; If the original interface signal is of the fourth signal type, extract the amplitude, frequency and phase characteristic parameters of the original interface signal, and regenerate a signal synchronized with the original interface signal as the target interface signal.

[0065] For example, in some other embodiments of this example, the signal reconstruction module described above may further be used for: Using the simulation model of the first subsystem as the transmitting end, a signal reconstruction function relationship representing the amplitude variation law of the original interface signal over time is determined based on the original interface signal; based on the signal reconstruction function relationship, corresponding target feature parameters are extracted from each discrete data point of the original interface signal; each target feature parameter is standardized and sent to the simulation model of the second subsystem; The simulation model of the second subsystem serves as the receiving end, receiving and parsing the standardized target feature parameters. Using the signal reconstruction function relationship and the target feature parameters, a continuous and smooth complete signal waveform on the target time scale is reconstructed as the target interface signal.

[0066] For example, in some other embodiments of this example, the signal reconstruction module described above may further be used for: Compressed sensing measurements are performed on the original interface signal to obtain the linear projection value corresponding to the original interface signal; Based on the sparse characteristics of the original interface signal in the frequency domain, the corresponding sparse transform basis and observation matrix are determined. A sensing matrix for signal reconstruction is generated using a sparse transform basis and an observation matrix. Based on the sparse reconstruction signal restoration process, the target interface signal that is consistent with the original high-dimensional signal in terms of time-domain waveform and frequency-domain components is reconstructed according to the perception matrix and linear projection value.

[0067] For example, in some other embodiments of this example, the signal reconstruction module described above may further be used for: The physical laws describing the original interface signals are converted into the physical constraints corresponding to the original interface signals. A mathematical model for signal reconstruction is established with the goal of minimizing the error between the reconstructed signal and the original interface signal, while satisfying physical constraints. The signal reconstruction model is calculated to output the target interface signal that satisfies the physical laws.

[0068] For example, the signal reconstruction mathematical model of the above embodiments may be as follows: ; in, Let χ be the signal to be reconstructed in physical space, R represent the real number field, and N be the dimension of the signal to be reconstructed. Let y be the measurement vector, and M be the dimension of the measurement vector y. N, For the measurement matrix, For sparse regularization, λ1 is the weight coefficient of the sparse regularization term, and λ2 is the weight coefficient of the physical constraint regularization term.

[0069] For example, in some other embodiments of this example, the signal reconstruction module described above may further be used for: A phase-locked loop is used to extract the amplitude, frequency, and phase characteristic parameters of the original interface signal in real time. The center frequency and phase of the adaptive filter are configured using the amplitude characteristic parameter, frequency characteristic parameter, and phase characteristic parameter as initial conditions. The adaptive filter is driven to track and reconstruct the received original interface signal components, generating a target interface signal that is synchronized with the original interface signal in terms of amplitude, frequency, and phase.

[0070] For example, in some other embodiments of this example, the signal type identification module described above may further be used for: If the original interface signal is a physical quantity or control command with standard waveform characteristics in the rail transit system, then the target signal type is the first signal type. If the energy of the original interface signal in the frequency domain is concentrated on a finite number of discrete frequency components, then the target signal type is the second signal type. If the original interface signal is the direct output of the physical system and its variation is governed by physical laws, then the target signal type is the third signal type. If the original interface signal is transmitted between simulation models that are distributed and have different processing capabilities, then the target signal type is the fourth signal type.

[0071] The signal transmission device for the rail transit system mentioned above is described from the perspective of functional modules. Furthermore, this application also provides an electronic device, which is described from the perspective of hardware. Figure 9 This is a schematic diagram of the structure of an electronic device provided in one embodiment of this application. The electronic device includes a memory and a processor. The memory stores a computer program, and the processor is configured to run the computer program to perform the steps in any of the above-described embodiments of the signal transmission method for a rail transit system.

[0072] It is understood that if the signal transmission method of the rail transit system in the above embodiments is implemented as a software functional unit and sold or used as an independent product, it can be stored in a non-volatile storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the related technology, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and executes all or part of the steps of the methods in the various embodiments of this application. The aforementioned storage medium includes, but is not limited to, various media capable of storing program code, such as: USB flash drive, mobile hard disk, read-only memory (ROM), random access memory (RAM), electrically erasable programmable ROM, register, hard disk, multimedia card, card-type memory (e.g., SD or DX memory), magnetic memory, removable disk, CD-ROM, magnetic disk, or optical disk. Based on this, this application also provides a non-volatile storage medium storing a computer program, which, when executed by a processor, performs the steps of the signal transmission method of the rail transit system as described in any of the above embodiments.

[0073] It is understood that if the signal transmission method of the rail transit system in the above embodiments is implemented as a software functional unit and sold or used as an independent product, the computer software product may not need to be stored in a physical storage medium. For example, it can be directly transmitted to a computer or other device with information processing capabilities via a wired or wireless network to execute all or part of the steps of the methods in the various embodiments of this application. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the related technology, or all or part of the technical solution, can be embodied in the form of a software product. Based on this, this application also provides a computer program product, which stores a computer program. When the computer program is executed by a processor, it performs the steps of the signal transmission method of the rail transit system as described in any of the above embodiments.

[0074] Finally, this application also provides a simulation system for a rail transit system; please refer to [link / reference]. Figure 11The system may include a first simulation device 111, a second simulation device 112, and a signal reconstruction processor 113. The first and second simulation devices are physically independent computing devices connected via a network and are both connected to the signal reconstruction processor. The first simulation device runs a first subsystem model with a first simulation step size and a second subsystem model with a second simulation step size, and the second simulation device runs a third subsystem model with a third simulation step size. The first, second, and third simulation step sizes are different from each other. The signal reconstruction processor is configured to execute a computer program to implement the steps of the signal transmission method for a rail transit system as described in any of the above embodiments when signal transmission occurs between the first and second subsystem models and / or between the second and third subsystem models.

[0075] The simulation system for rail transit systems is based on the various subsystems of the complex rail transit system, establishing a technical foundation for multi-timescale analysis, cross-scale variable reconstruction, and synchronization mechanisms. This foundation supports real-time coupled simulation of multi-scale, multi-granularity models. The simulation system employs a combination of theoretical analysis and multi-scenario adaptation verification. From the perspective of system simulation equipment theory, it studies the impact of the diversity of rail transit system characteristics, Shannon sampling theorem constraints, computational resource conservation, simulation accuracy, and other indicators on the simulation timescale. It determines the conditions for determining the simulation timescale that meet the requirements of digital experimental verification, forming the technical foundation for multi-timescale simulation of complex rail transit systems. Furthermore, the simulation system establishes a multi-scenario, multi-timescale model matching technology system and conducts corresponding feasibility verification, supporting the implementation of multi-scale model collaborative simulation of rail transit transport equipment.

[0076] For example, in the digital simulation operation of the entire rail transit system, a simulation system is built across simulation devices: Model A and Model B are deployed in simulation device 111, and subsystem model C is deployed in simulation device 112. The simulation step size for A is 1ms, for B it is 20us, and for C it is 10us. In simulation device 111, model A transmits its output signal to B, while model B transmits its output signal to model C deployed in simulation device 112. This scenario presents two types of step size inconsistency problems: one is within simulation device 111, where models A and B, due to differences in functional requirements, exhibit a mismatch in step size. Since model A's simulation step size of 10ms is equivalent to a reduced sampling rate of the output signal, transmitting the signal to model B with a simulation step size of 200us is equivalent to an increased sampling rate. Assuming that the correlation variables between A and B have sparse frequency characteristics, compressed sensing theory is used in model B to reconstruct the interface signals between A and B, thereby eliminating the step size mismatch problem and achieving step size alignment during data interaction between models A and B. Figure 12These are the interface signals (100 points) output by model A at a simulation step size of 10ms. Figure 13 The first type involves comparing the reconstructed interface signal in model B with the original signal (5000 points), unaffected by the step size of model A, and finding an error of 0.48%. The second type involves signal transmission between model B and model C. Since the two models are deployed on different simulation devices, there are not only step size mismatch issues but also communication transmission delays. To address this, the MSOGI algorithm reconstruction technique is used. This involves phase-locked transmission of the amplitude and frequency of the interface signal output from model B. Model C reconstructs the interface signal based on the transmitted parameters, achieving signal matching. Assuming the interface signals of models B and C are 50Hz sinusoidal signals, the comparison results between the original interface signal and the reconstructed interface signal in model C are as follows: Figure 14 As shown, the amplitude error is 0.1%. Synchronous correction ensures the timing consistency of the cross-machine collaborative model.

[0077] As can be seen from the above, the simulation system provided in this embodiment for rail transit systems can solve the traditional contradiction between accuracy, efficiency, and stability in multi-timescale collaborative simulation. It performs intelligent reconstruction matching different signal types, fundamentally avoiding signal distortion: parameterized reconstruction is used for regular signals (such as a 50Hz sine wave), transmitting signal genes rather than data points; compressed sensing reconstruction is used for sparse frequency domain signals, restoring waveforms with high precision using a small number of sampling points. Mechanistically, this ensures the integrity of signal characteristics, resulting in simulation accuracy significantly higher than traditional methods. The simulation system collaborates on demand, achieving precise allocation of computing resources. It allows each subsystem to use the optimal simulation step size according to its own dynamic requirements, without forcibly unifying to the minimum clock. It transforms the massive, continuous global computational overhead into local, controllable signal processing overhead at the interface, greatly saving computation time and hardware costs, making large-scale, real-time simulation of the entire system feasible. Addressing the network latency problem in distributed simulation, parameter phase-locked synchronization effectively suppresses the impact of transmission delay, transforming latency-sensitive waveform transmission into robust parameter transmission and local reconstruction, effectively compensating for latency and ensuring the timing consistency and stability of the system across simulation devices and heterogeneous computing environments. By leveraging a modular technology system to cover multiple scenarios, it provides various signal reconstruction methods such as parametric reconstruction, compressed sensing, physical constraint enhancement, and phase-locked loop reconstruction. It allows for flexible selection of the optimal signal reconstruction method based on prior signal knowledge (known models, frequency domain sparsity, physical constraints, anti-delay, etc.), covering multiple types of signals in rail transit and offering better practicality.

[0078] The foregoing has provided a detailed description of a signal transmission method, electronic equipment, and simulation system for a rail transit system provided in this application. The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. Whether the units and algorithm steps of the various examples described in the disclosed embodiments are executed by electronic hardware or computer software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, and such implementations should not be considered beyond the scope of this application. Several improvements and modifications can be made to this application without departing from the principles of this application, and these improvements and modifications also fall within the protection scope of this application.

Claims

1. A signal transmission method for a rail transit system, characterized in that, include: When the simulation model of the first subsystem corresponding to the first subsystem of the rail transit system transmits a signal to the simulation model of the second subsystem corresponding to the second subsystem, the original interface signal output by the simulation model of the first subsystem is obtained. Determine the target signal type corresponding to the original interface signal; the target signal type is a first signal type with a known functional form, or a second signal type with sparse characteristics in the transform domain, or a third signal type constrained by physical laws, or a fourth signal type that interacts between heterogeneous models; Based on the target signal type, the interface signal is reconstructed into a target interface signal that undistorts the original physical process and matches the simulation step size of the second subsystem simulation model; The target interface signal is input into the second subsystem simulation model.

2. The signal transmission method for a rail transit system according to claim 1, characterized in that, Based on the target signal type, the interface signal is reconstructed into a target interface signal that undistortsly represents the original physical process and matches the simulation step size of the second subsystem simulation model, including: If the original interface signal is of the first signal type, each discrete data point of the original interface signal is reconstructed into a continuous and smooth complete signal waveform on the target time scale corresponding to the second subsystem simulation model, so as to serve as the target interface signal; If the original interface signal is of the second signal type, a signal that is consistent with the original high-dimensional signal in terms of time-domain waveform and frequency-domain components is recovered from the original interface signal and used as the target interface signal. If the original interface signal is a third type of signal, the original interface signal is fitted under the condition that the corresponding physical rules are used as physical constraints to obtain the target interface signal; If the original interface signal is of the fourth signal type, the amplitude, frequency and phase characteristic parameters of the original interface signal are extracted, and a signal synchronized with the original interface signal is regenerated as the target interface signal.

3. The signal transmission method for a rail transit system according to claim 1, characterized in that, If the original interface signal is of the first signal type, the interface signal is reconstructed into a target interface signal that undistorts the original physical process and matches the simulation step size of the second subsystem simulation model, including: Using the first subsystem simulation model as the transmitting end, a signal reconstruction function relationship characterizing the amplitude variation law of the original interface signal over time is determined based on the original interface signal; based on the signal reconstruction function relationship, corresponding target feature parameters are extracted from each discrete data point of the original interface signal; each target feature parameter is standardized and sent to the second subsystem simulation model; The second subsystem simulation model serves as the receiving end, receiving and parsing the standardized target feature parameters. Using the signal reconstruction function and the target feature parameters, a continuous and smooth complete signal waveform on the target time scale is reconstructed as the target interface signal.

4. The signal transmission method for a rail transit system according to claim 1, characterized in that, If the original interface signal is of the second signal type, the interface signal is reconstructed into a target interface signal that undistorts the original physical process and matches the simulation step size of the second subsystem simulation model, including: Compressed sensing measurement is performed on the original interface signal to obtain the linear projection value corresponding to the original interface signal; Based on the sparse characteristics of the original interface signal in the frequency domain, the corresponding sparse transform basis and observation matrix are determined. Using the sparse transform basis and the observation matrix, a sensing matrix for signal reconstruction is generated; Based on the sparse reconstruction signal restoration process, the target interface signal that is consistent with the original high-dimensional signal in terms of time domain waveform and frequency domain components is reconstructed according to the perception matrix and the linear projection value.

5. The signal transmission method for a rail transit system according to claim 1, characterized in that, If the original interface signal is of the third signal type, the interface signal is reconstructed into a target interface signal that undistorts the original physical process and matches the simulation step size of the second subsystem simulation model, including: The physical laws describing the original interface signal are converted into the physical constraints corresponding to the original interface signal. A mathematical model for signal reconstruction is established with the goal of minimizing the error between the reconstructed signal and the original interface signal, while satisfying the physical constraints. The signal reconstruction model is calculated to output a target interface signal that satisfies physical laws.

6. The signal transmission method for a rail transit system according to claim 5, characterized in that, The mathematical model for signal reconstruction is as follows: ； in, Let χ be the signal to be reconstructed in physical space, R represent the real number field, and N be the dimension of the signal to be reconstructed. Let y be the measurement vector, and M be the dimension of the measurement vector y. N, For the measurement matrix, For sparse regularization, λ1 is the weight coefficient of the sparse regularization term, and λ2 is the weight coefficient of the physical constraint regularization term.

7. The signal transmission method for a rail transit system according to claim 1, characterized in that, If the original interface signal is of the fourth signal type, the interface signal is reconstructed into a target interface signal that undistorts the original physical process and matches the simulation step size of the second subsystem simulation model, including: The amplitude characteristic parameters, frequency characteristic parameters, and phase characteristic parameters of the original interface signal are extracted from the original interface signal in real time using a phase-locked loop. Using the amplitude characteristic parameter, the frequency characteristic parameter, and the phase characteristic parameter as initial conditions, configure the center frequency and phase of the adaptive filter; The adaptive filter is driven to track and reconstruct the received original interface signal components, generating a target interface signal that is synchronized with the original interface signal in terms of amplitude, frequency, and phase.

8. The signal transmission method for a rail transit system according to any one of claims 1 to 7, characterized in that, Determining the target signal type corresponding to the original interface signal includes: If the original interface signal is a physical quantity or control command with standard waveform characteristics in the rail transit system, then the target signal type is the first signal type; If the energy of the original interface signal in the frequency domain is concentrated on a finite number of discrete frequency components, then the target signal type is the second signal type. If the original interface signal is the direct output of the physical system and its variation is governed by physical laws, then the target signal type is the third signal type. If the original interface signal is transmitted between simulation models that are distributed and have different processing capabilities, then the target signal type is the fourth signal type.

9. An electronic device, characterized in that, include: Memory, used to store computer programs; A processor, configured to execute the computer program to implement the steps of the signal transmission method for a rail transit system as described in any one of claims 1 to 8.

10. A simulation system for a rail transit system, characterized in that, It includes a first simulation device, a second simulation device, and a signal reconstruction processor; The first simulation device and the second simulation device are physically independent computing devices connected via a network, and both are connected to the signal reconstruction processor; The first simulation device runs a first subsystem model with a first simulation step size and a second subsystem model with a second simulation step size, and the second simulation device runs a third subsystem model with a third simulation step size; wherein the first simulation step size, the second simulation step size, and the third simulation step size are different from each other; The signal reconstruction processor is configured to, when executing a computer program, implement the steps of the signal transmission method for a rail transit system as described in any one of claims 1 to 8, during signal transmission between the first subsystem model and the second subsystem model, and / or between the second subsystem model and the third subsystem model.

Images