A train fault-tolerant control method, device and equipment and readable storage medium

Abstract

The application discloses a kind of train fault-tolerant control method, device, equipment and readable storage medium, applied to train control field, comprising: according to train current operating condition and the equivalent guiding execution ability parameter after the equivalent guiding execution ability parameter of the quantitative of multiple source state information, determine optimal control parameter combination from parameter library;Based on optimal control parameter combination, using the extended state observation module in layered self-disturbance control framework, real-time observation value of guiding system state variable and comprehensive disturbance real-time observation value are obtained;Based on observation value, double-layer collaborative calculation is carried out through current loop self-disturbance control layer and gap layer self-disturbance control layer of layered self-disturbance control framework, dynamic disturbance compensation is carried out to control input, and the working current of guiding electromagnet is adjusted according to the control instruction generated by compensation, so that the guiding gap is stably set at preset target value.

Description

Technical Field

[0001] This invention relates to the field of train control, and in particular to a train fault-tolerant control method, apparatus, device, and readable storage medium. Background Technology

[0002] The guidance system of high-speed maglev trains uses guide electromagnets arranged on both sides of the track to generate lateral electromagnetic force, achieving stable guidance and attitude constraint for the train in straight and curved sections. As the operating speed of high-speed maglev trains continues to increase, the operation of the guidance system is susceptible to the combined effects of track irregularities, lateral aerodynamic forces, changes in structural parameters, and the aging of electromagnetic components. Under these conditions, if the guide electromagnets experience abnormal current, performance degradation, or partial failure, it can easily lead to an imbalance in the distribution of guiding force, causing rapid fluctuations in the guide gap, which in turn affects the lateral stability of the train. In severe cases, it may even cause system instability, threatening the safety of train operation.

[0003] Therefore, how to improve the control accuracy and robustness of high-speed maglev trains and ensure stable operation under conditions of guide electromagnet failure, high-speed operation, and strong external disturbances is a technical problem that urgently needs to be solved. Summary of the Invention

[0004] In view of this, the purpose of the present invention is to provide a train fault-tolerant control method, device, equipment and readable storage medium, which solves the problem in the prior art that the control accuracy and robustness of high-speed maglev trains are insufficient under conditions of guide electromagnet failure, high-speed operation and strong external disturbances, and thus cannot guarantee the stable operation of the train.

[0005] To solve the above-mentioned technical problems, the present invention provides a train fault-tolerant control method, comprising:

[0006] Based on the actual operating state of the guide electromagnet, multi-source status information is collected in real time, and the multi-source status information is quantified to output equivalent guide execution capability parameters.

[0007] Based on the current operating conditions of the train and the equivalent guidance execution capability parameters, the optimal combination of control parameters is obtained by searching and matching from a pre-built parameter library.

[0008] The optimal control parameter combination is input into the hierarchical active disturbance rejection control architecture, and the extended state observation module in the hierarchical active disturbance rejection control architecture is used to obtain the real-time observed values ​​of the guided system state variables and the real-time observed values ​​of the comprehensive disturbance in real time.

[0009] Based on the observed values ​​of the state variables of the guidance system and the real-time observed values ​​of the integrated disturbance, the current loop disturbance rejection control layer and the gap layer disturbance rejection control layer of the hierarchical disturbance rejection control architecture perform dual-layer collaborative calculation to dynamically compensate for the disturbances in the control input and generate control commands.

[0010] The operating current of the guide electromagnet is adjusted according to the control command to stabilize the guide gap at a preset target value.

[0011] Optionally, based on the actual operating state of the guide electromagnet, multi-source state information is collected in real time, and the multi-source state information is quantified to output equivalent guide execution capability parameters, including:

[0012] The multi-source status information of the guide electromagnet is collected in real time during actual operation; the multi-source status information includes the working current variation characteristics of the guide electromagnet, the guide gap response characteristics, and the gap deviation between the control command and the actual output of the guide electromagnet.

[0013] The multi-source state information is preprocessed, and based on a preset quantization representation algorithm, the preprocessed multi-source state information is fused and calculated to output the equivalent guided execution capability parameter.

[0014] Optionally, before inputting the optimal control parameter combination into the hierarchical active disturbance rejection control architecture, the method further includes:

[0015] The hierarchical active disturbance rejection control architecture is constructed, which includes a current loop active disturbance rejection control layer and a gap layer active disturbance rejection control layer.

[0016] The current loop active disturbance rejection control layer includes: a second-order extended state observation module;

[0017] The gap layer active disturbance rejection control layer includes: a second-order tracking differentiation module, a third-order extended state observation module, and a disturbance function module; the disturbance function module is:

[0018] ;

[0019] Where f0 represents the disturbance function; z3 represents the gap change; k z represents the known correlation gain coefficient of the guidance system; m represents the mass.

[0020] Optionally, before obtaining the optimal combination of control parameters by searching and matching from a pre-built parameter library based on the train's current operating conditions and the equivalent guidance execution capability parameters, the method further includes:

[0021] A multi-degree-of-freedom equivalent dynamic model of the guidance system is established. Based on the multi-degree-of-freedom equivalent dynamic model, the dynamic response characteristics of the guidance system under different train operating conditions and different equivalent guidance execution capability parameters are simulated, and multiple sets of simulation state data are obtained. The multiple sets of simulation state data are quadruple data of train operating conditions, equivalent guidance execution capability parameters, control parameters, and guidance system control performance indicators. The control parameters are the parameters in the hierarchical active disturbance rejection control architecture.

[0022] A nonlinear mapping model is constructed with control parameters, train operating conditions, and equivalent guidance execution capability parameters as inputs and guidance system control performance indicators as outputs. The nonlinear mapping model is trained using the multiple sets of simulation state data to obtain a trained nonlinear mapping model.

[0023] With the goal of optimizing the control performance of the guidance system, a particle swarm optimization algorithm is used in conjunction with the trained nonlinear mapping model to globally optimize the control parameters for each scenario, resulting in multiple sets of optimal control parameter combinations that adapt to different scenarios. The scenarios are those obtained by combining different train operating conditions and different equivalent guidance execution capability parameters.

[0024] All the optimal control parameters obtained through optimization are combined, organized and stored according to the scenario, and the parameter library is constructed.

[0025] Optionally, a multi-degree-of-freedom equivalent dynamic model of the guidance system is established, including:

[0026] The structural parameters, operating state parameters, and configuration information of the guide electromagnet of the guidance system are obtained. The guide electromagnet, the vehicle body, and their connecting structures are simplified by equivalent processing to construct the multi-degree-of-freedom equivalent dynamic model that can simultaneously reflect the lateral displacement and yaw motion characteristics of the guidance system.

[0027] Optionally, the control performance indicators of the guidance system include:

[0028] Clearance error index, settling time index, and control stability index;

[0029] The gap error index is used to characterize the steady-state accuracy and tracking performance of the guidance system, and is obtained by describing the statistical characteristics of the deviation between the target guidance gap and the actual guidance gap;

[0030] The adjustment time index is used to characterize the dynamic response speed of the guidance system and is determined based on the time it takes for the guidance gap response to first enter and remain within the allowable error range.

[0031] The control stability index is used to characterize the degree of oscillation and fluctuation in the response process of the guidance system. It is obtained by comprehensively characterizing the peak value change, overshoot characteristics, steady-state fluctuation amplitude, and statistical characteristics of the control quantity change characteristics of the guidance gap response curve.

[0032] Optionally, based on the train's current operating conditions and the equivalent guidance execution capability parameters, an optimal combination of control parameters is obtained by searching and matching from a pre-built parameter library, including:

[0033] Using the current train operating condition and the equivalent guidance execution capability parameter as dual search keywords, a preset search algorithm matches control parameter combinations adapted to the current scenario in the parameter library. The current train operating condition includes one or more of the following: real-time train speed, current track curvature, and track gradient. The parameter library is an optimal set of control parameters constructed offline based on a radial basis function neural network nonlinear mapping model and a particle swarm optimization algorithm. The parameter library stores optimal control parameters adapted to different operating scenarios and guidance electromagnet states according to the combination of train operating condition type and equivalent guidance execution capability parameter level.

[0034] Determine whether the matched control parameter combination meets the parameter adjustment range of the hierarchical active disturbance rejection control architecture and the real-time operation requirements of the guidance system;

[0035] If the conditions are met, the combination of control parameters obtained from the matching is output as the optimal control parameters for the current scenario.

[0036] The present invention also provides a train fault-tolerant control device, comprising:

[0037] The quantization module is used to collect multi-source status information in real time according to the actual operating state of the guide electromagnet, quantize the multi-source status information, and output equivalent guide execution capability parameters.

[0038] The retrieval module is used to search and match from a pre-built parameter library based on the current operating conditions of the train and the equivalent guidance execution capability parameters to obtain the optimal combination of control parameters.

[0039] The observation acquisition module is used to input the optimal control parameter combination into the hierarchical active disturbance rejection control architecture, and to use the extended state observation module in the hierarchical active disturbance rejection control architecture to acquire the observation values ​​of the guiding system state variables and the real-time observation values ​​of the comprehensive disturbance in real time.

[0040] The compensation module is used to perform dynamic disturbance compensation on the control input and generate control commands based on the observed values ​​of the state variables of the guidance system and the real-time observed values ​​of the comprehensive disturbances, through two-layer collaborative calculations of the current loop active disturbance rejection control layer and the gap layer active disturbance rejection control layer of the hierarchical active disturbance rejection control architecture.

[0041] The adjustment and control module is used to adjust the working current of the guide electromagnet according to the control command, so that the guide gap is stabilized at a preset target value.

[0042] The present invention also provides a train fault-tolerant control device, comprising:

[0043] Memory, used to store computer programs;

[0044] A processor is used to implement the train fault-tolerant control method described above when executing the computer program.

[0045] The present invention also provides a computer-readable storage medium storing computer-executable instructions, which, when loaded and executed by a processor, implement the train fault-tolerant control method described above.

[0046] As can be seen, this invention collects multi-source state information in real time based on the actual operating state of the guide electromagnet, quantifies the multi-source state information, and outputs equivalent guide execution capability parameters. Based on the current train operating conditions and the equivalent guide execution capability parameters, it retrieves and matches parameters from a pre-built parameter library to obtain the optimal control parameter combination. This optimal control parameter combination is input into a hierarchical active disturbance rejection control architecture. Using the extended state observation module within the hierarchical architecture, it acquires real-time observations of the guide system state variables and the comprehensive disturbance. Based on these observations, the current loop and gap layers of the hierarchical architecture perform dual-layer collaborative calculations to dynamically compensate for disturbances in the control input and generate control commands. The operating current of the guide electromagnet is adjusted according to the control commands to stabilize the guide gap at a preset target value. The beneficial effects of this invention are as follows: it adapts control parameters to different train operating conditions and the execution capability of the guide electromagnet through offline optimization and online rapid matching of the parameter library; and it effectively offsets the effects of various disturbances such as system model uncertainty, guide electromagnet performance degradation / failure, track irregularities, and external wind disturbances by using real-time observation and dynamic compensation of hierarchical active disturbance rejection control. This significantly improves the fault tolerance and control stability of the train guidance system, ensuring that the guide gap remains stable at the preset target value. It effectively solves the problems of poor adaptability of fixed parameters and untimely disturbance compensation in traditional control methods, which lead to low control accuracy and easy system instability. It provides a reliable control guarantee for the safe and stable operation of the train under different operating conditions and throughout the entire life cycle of the electromagnet.

[0047] In addition, the present invention also provides a train fault-tolerant control device, equipment and readable storage medium, which also have the above-mentioned beneficial effects. Attached Figure Description

[0048] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0049] Figure 1 This is a schematic diagram of the principle of a high-speed maglev system;

[0050] Figure 2 A flowchart of a train fault-tolerant control method provided in an embodiment of the present invention;

[0051] Figure 3 A flowchart illustrating a quantization method provided in an embodiment of the present invention;

[0052] Figure 4 A flowchart illustrating a control and global parameter optimization process provided in an embodiment of the present invention;

[0053] Figure 5 An example diagram of a current loop active disturbance rejection control layer provided in an embodiment of the present invention;

[0054] Figure 6 An example diagram of a gap layer self-disturbance rejection control layer provided in an embodiment of the present invention;

[0055] Figure 7 A flowchart illustrating a train fault-tolerant control architecture provided in an embodiment of the present invention;

[0056] Figure 8 This is a schematic diagram of the structure of a train fault-tolerant control device provided in an embodiment of the present invention;

[0057] Figure 9 This is a schematic diagram of the structure of a train fault-tolerant control device provided in an embodiment of the present invention. Detailed Implementation

[0058] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0059] like Figure 1 As shown, Figure 1This is a schematic diagram of a high-speed maglev system. The high-speed maglev train guidance system uses guide electromagnets arranged on both sides of the track to generate lateral electromagnetic force, achieving stable guidance and attitude constraint for the train in straight and curved sections. With the continuous increase in the operating speed of high-speed maglev trains, the operation of the guidance system is easily affected by factors such as track irregularities, lateral aerodynamic forces, changes in structural parameters, and aging of electromagnetic components. Under these conditions, if the guide electromagnet experiences abnormal current, performance degradation, or partial failure, it can easily lead to an imbalance in the distribution of guiding force, causing rapid fluctuations in the guide gap, thus affecting the lateral stability of the train. In severe cases, it may even lead to system instability, threatening the safety of train operation. Therefore, given the multi-source disturbances and significantly increased uncertainty in component performance faced by the high-speed maglev train guidance system, there is an urgent need for a fault-tolerant control method that can automatically adjust control parameters according to changes in the system's operating state when different types of abnormalities or failures occur in the guide electromagnets, and maintain stable operation of the guidance system under complex conditions, in order to improve the operational safety and engineering reliability of the high-speed maglev train guidance system.

[0060] This invention proposes a fault-tolerant control method for the guidance system of high-speed maglev trains based on active disturbance rejection control and intelligent parameter tuning. This method addresses the problems of decreased control accuracy, insufficient robustness, and difficulty in parameter tuning in existing technologies under conditions of guide electromagnet failure, high-speed operation, and strong external disturbances. Please refer to the following for details. Figure 2 , Figure 2 A flowchart illustrating a train fault-tolerant control method provided in an embodiment of the present invention. The method may include:

[0061] S101: Based on the actual operating state of the guide electromagnet, collect multi-source status information in real time, quantify the multi-source status information, and output equivalent guide execution capability parameters.

[0062] The execution subject in this embodiment is a terminal. This embodiment does not limit the type of terminal, as long as it can perform the operation of the train fault-tolerant control method. This step realizes a quantitative description of the health status and working capacity of the guiding actuator (guiding electromagnet), transforming the abstract health level into calculable and feedback-capable equivalent guiding actuator capability parameters, providing real-time, quantitative status input basis for subsequent fault status identification and adaptive matching of control parameters.

[0063] It should be noted that the actual operating state in this embodiment refers to the collective term for all real-time changing state quantities, environmental quantities, and equipment state quantities during train operation. The multi-source state information in this embodiment refers to various monitoring data that reflect the actual operating state of the guide electromagnet, serving as the basis for quantifying the electromagnet's health. This embodiment does not specifically limit the multi-source state information. For example, the multi-source state information in this embodiment may include at least the operating current variation characteristics of the guide electromagnet, real-time response data of the guide gap, and the gap deviation value between the control command and the actual output of the electromagnet. The equivalent guide execution capability parameter in this embodiment is used to quantitatively characterize the change in the effective guiding capability of the guide electromagnet under different performance degradation and failure states. This allows the guide electromagnet to be described using a unified parameter form under different operating conditions such as normal operation, partial degradation, and severe failure, thereby achieving a unified characterization of the fault state of the guide actuator. The equivalent guide execution capability parameter is a single characteristic parameter or a multi-dimensional characteristic parameter group. Its value is positively correlated with the health and guide execution capability of the guide electromagnet, used to accurately characterize the current actual operating state and fault degradation level of the guide electromagnet, providing core state basis for subsequent optimal control parameter retrieval and matching. For example, a higher parameter value indicates a stronger actuation capability and a better health condition of the guide electromagnet; or a higher parameter value level indicates a stronger actuation capability and a better health condition of the guide electromagnet.

[0064] Furthermore, the above-mentioned method of collecting multi-source status information in real time based on the actual operating state of the guide electromagnet, quantifying the multi-source status information, and outputting equivalent guide execution capability parameters can specifically include the following steps: real-time collection of multi-source status information of the guide electromagnet in its actual operating state; the multi-source status information includes the operating current variation characteristics of the guide electromagnet, the guide gap response characteristics, and the gap deviation between the control command and the actual output of the guide electromagnet; preprocessing the multi-source status information, and based on a preset quantization characterization algorithm, fusing and calculating the preprocessed multi-source status information to output equivalent guide execution capability parameters.

[0065] This embodiment does not impose specific limitations on preprocessing, such as removing abnormal noise data and performing standardization and normalization to unify data dimensions and units. This embodiment also does not impose specific limitations on the quantization representation algorithm, as long as it can transform the health status, performance degradation degree, and guiding execution capability of the guide electromagnet into quantifiable and searchable feature parameters. (See reference...) Figure 3 , Figure 3This is a flowchart illustrating a quantification method provided in an embodiment of the present invention. Based on the actual operating state of the guide electromagnet, the characteristics of its operating current variation, guide gap response, and gap deviation between the control command and the actual output of the guide electromagnet are collected and analyzed in real time. The aforementioned multi-source state information reflecting the current working capacity of the guide electromagnet is comprehensively processed, thereby converting the health status of the guide actuator into an equivalent guide execution capability parameter.

[0066] S102: Based on the current operating conditions of the train and the equivalent guidance execution capability parameters, the optimal combination of control parameters is obtained by searching and matching from the pre-built parameter library.

[0067] In this embodiment, the current operating condition of the train refers to the real-time environment and operating conditions of the train. Core parameters include the train's real-time operating speed, the current track curvature, and the track gradient, which are crucial for matching optimal control parameters (different operating conditions correspond to different optimal control parameters). The parameter library in this embodiment is a set of optimal control parameters pre-calculated using offline modeling and swarm intelligence optimization algorithms and stored in the onboard controller. The database categorizes and stores corresponding optimal control parameters according to combinations of train operating conditions and equivalent guidance execution capability parameters, supporting real-time retrieval and matching. The optimal control parameters (combinations) in this embodiment are retrieved from the parameter library and are adapted to the train's current operating conditions and the actual execution capability of the guidance electromagnet. This combination of control parameters is the core adjustable parameter of the hierarchical active disturbance rejection control architecture; that is, the control parameters are parameters within the hierarchical active disturbance rejection control architecture, which can be understood as parameters of the active disturbance rejection controller. For example, it may include extended state observer gain, nonlinear state error feedback parameters, current / gap control layer adjustment coefficients, etc., generally a set of synergistic parameters. Furthermore, when the current operating conditions of the train and / or the equivalent guidance execution capability parameters change, the control parameter matching process is retried, so that the parameters of the self-disturbance immunity controller are adjusted accordingly with the changes in system state, thereby ensuring that the guidance control system can maintain good dynamic response performance and operational stability under different fault conditions and various operating conditions.

[0068] Furthermore, based on the train's current operating conditions and equivalent guidance execution capability parameters, the optimal control parameter combination is obtained by searching and matching from a pre-built parameter library. Specifically, this may include: using the train's current operating conditions and equivalent guidance execution capability parameters as dual search keywords, and using a preset search algorithm to match control parameter combinations adapted to the current scenario in the parameter library; determining whether the matched control parameter combination meets the parameter adjustment range of the hierarchical active disturbance rejection control architecture and the real-time operation requirements of the guidance system; if it does, outputting the matched control parameter combination as the optimal control parameters for the current scenario.

[0069] In this embodiment, the current train operating condition may include at least one or more of the following: real-time train speed, current track curvature, and track gradient. This embodiment does not limit the retrieval algorithm. For example, hash table retrieval algorithms and linear lookup table retrieval algorithms are acceptable. The parameter library in this embodiment is an optimal control parameter set constructed offline based on an RBF (Radial Basis Function) neural network nonlinear mapping model combined with a PSO (Particle Swarm Optimization) swarm intelligence optimization algorithm. It can be understood that the parameter library categorizes and stores optimal control parameters adapted to different operating scenarios and guide electromagnet states according to the combination of train operating condition type and equivalent guidance execution capability parameter level. This embodiment also performs dual verification on the matched control parameter combinations, checking both the adjustment range of the hierarchical active disturbance rejection control architecture parameters and the real-time operating requirements of the guidance system. This effectively avoids the risks of control command failure, abnormal system response, or even instability caused by parameters exceeding the range or not matching the real-time operating state. It ensures that the optimal output control parameters not only meet the hardware and algorithm constraints of the control architecture but also adapt to the current train operating conditions and the execution capability of the guidance electromagnet, providing a safe and reliable parameter guarantee for the accurate execution of subsequent hierarchical active disturbance rejection control.

[0070] Furthermore, before obtaining the optimal combination of control parameters by searching and matching from a pre-built parameter library based on the train's current operating conditions and equivalent guidance execution capability parameters, the following additional steps may be included:

[0071] Step 11: Establish a multi-degree-of-freedom equivalent dynamic model of the guidance system, and simulate the dynamic response characteristics of the guidance system under different train operating conditions and different equivalent guidance execution capability parameters based on the multi-degree-of-freedom equivalent dynamic model to obtain multiple sets of simulation state data.

[0072] This step establishes a multi-degree-of-freedom equivalent dynamic model of the train guidance system to realistically reproduce the dynamic coupling relationship between the guide electromagnet, the car body, and the track. Using this model, simulations are performed under different train operating conditions and different equivalent guidance execution capability parameters to obtain dynamic response data of the guidance system under corresponding conditions. This provides a sufficient and standardized simulation data foundation for subsequent nonlinear mapping model training. The multiple sets of simulation state data in this embodiment are quadruple data, specifically including train operating conditions, equivalent guidance execution capability parameters, control parameters, and guidance system control performance indicators. For example, the multi-degree-of-freedom equivalent dynamic model is used to simulate 350km / h straight / 200km / h curved tracks (different operating conditions), guide electromagnet health score of 100 / degradation score of 70 (different equivalent guidance execution capability parameters), and different combinations of layered active disturbance rejection control parameters. Simulations are performed under scenarios such as track irregularities / external strong winds (different disturbances), obtaining corresponding guidance gap stability accuracy, response time, disturbance rejection capability, and other guidance system control performance indicators, thus forming complete quadruple simulation data including operating conditions, execution capability, control parameters, and performance indicators.

[0073] Furthermore, the aforementioned establishment of a multi-degree-of-freedom equivalent dynamic model for the guidance system can specifically include: acquiring the structural parameters, operating state parameters, and configuration information of the guidance electromagnets; performing equivalent simplification on the guidance electromagnets, vehicle body, and their connecting structures; and constructing a multi-degree-of-freedom equivalent dynamic model that can simultaneously reflect the lateral displacement and yaw motion characteristics of the guidance system. Specifically, a multi-degree-of-freedom equivalent dynamic model can be constructed under the condition of neglecting the dynamic influence of the suspension system on the guidance system, to describe the dynamic response behavior of the guidance system under normal and fault conditions. This multi-degree-of-freedom equivalent dynamic model effectively supports offline optimization and online matching of optimal control parameters under different operating conditions and different execution capabilities, improving the feasibility and control accuracy of the entire fault-tolerant control scheme.

[0074] Step 12: Construct a nonlinear mapping model with control parameters, train operating conditions and equivalent guidance execution capability parameters as inputs and guidance system control performance indicators as outputs. Use multiple sets of simulation state data to train the nonlinear mapping model to obtain a trained nonlinear mapping model.

[0075] This step involves constructing a nonlinear mapping model, which describes the impact of control parameter variations on the control performance of the guidance system. In this embodiment, gap error, settling time, and control stability can be selected as the guidance system's control performance indicators. Of course, to better construct the nonlinear mapping model, other indicators can be added to improve the control effect of the trained control parameters. Specifically, the gap error indicator characterizes the system's steady-state accuracy and tracking performance, and can be obtained by describing the statistical characteristics of the deviation between the target guidance gap and the actual guidance gap. The settling time characterizes the system's dynamic response speed, and can be determined based on the time it takes for the guidance gap response to first enter and remain within the allowable error range. Control stability characterizes the degree of oscillation and fluctuation during the system's response process, and can be obtained by comprehensively characterizing the statistical characteristics of the guidance gap response curve, such as peak value variation, overshoot characteristics, steady-state fluctuation amplitude, and control quantity variation characteristics.

[0076] Specifically, by using control parameters, train operating conditions, and equivalent guidance execution capability parameters as input features of the nonlinear mapping model, and using gap error, settling time, and control stability as output labels, an input-output sample set is formed for training the nonlinear mapping model. Based on this sample set, an RBF neural network can be used to establish the mapping relationship between control parameters and guidance system control performance indicators. Through the training process, the deviation between the prediction results of the nonlinear mapping model and the actual guidance system control performance indicators corresponding to the samples is minimized, thus obtaining a nonlinear mapping model that can characterize the intrinsic relationship between changes in control parameters and changes in guidance system control performance. Using the nonlinear mapping model, the control performance of the guidance system under different combinations of control parameters can be predicted and evaluated without directly affecting the actual guidance system.

[0077] Step 13: With the goal of optimizing the control performance of the guidance system, the particle swarm optimization algorithm is used in combination with the trained nonlinear mapping model to perform global optimization of the control parameters corresponding to each scenario, so as to obtain multiple sets of optimal control parameter combinations that are suitable for different scenarios; the scenarios are obtained by different train operating conditions and different equivalent guidance execution capability parameter combinations.

[0078] This step involves the global optimization of control parameters based on a swarm intelligence optimization algorithm. Building upon the aforementioned nonlinear mapping model, the swarm intelligence optimization algorithm (PSO) is introduced to perform a global search and iterative optimization of the active disturbance rejection controller parameters (i.e., control parameters). The optimization process can be found by referring to [reference needed]. Figure 4 , Figure 4This is a flowchart illustrating a global optimization process for control and parameters, provided by an embodiment of the present invention. By constructing a fitness evaluation function corresponding to the control performance of the guidance system, the performance under different combinations of control parameters is evaluated, and candidate solutions are continuously updated during the swarm search process. Through iterative search and filtering, the optimal combination of control parameters that maximizes system performance under the corresponding train operating conditions and equivalent guidance execution capability parameters is obtained, thereby providing optimized parameter configurations for subsequent guidance control.

[0079] Step 14: Combine all the optimal control parameters obtained through optimization, organize and store them according to the scenario, and build a parameter library.

[0080] This step involves classifying, organizing, and structurally storing multiple sets of optimal control parameters obtained through global optimization that are adapted to different train operating conditions and equivalent guidance execution capability parameters, according to operating condition type and equivalent guidance execution capability level. This forms a parameter library that can be quickly retrieved and called, enabling the train to quickly match the optimal control parameters based on the current operating conditions and the actual execution capability of the guidance electromagnet during real-time operation. This provides stable, reliable, and scenario-adaptive parameter support for the subsequent hierarchical active disturbance rejection control architecture.

[0081] S103: Input the optimal control parameter combination into the hierarchical active disturbance rejection control architecture, and use the extended state observation module in the hierarchical active disturbance rejection control architecture to obtain the real-time observation values ​​of the guided system state variables and the integrated disturbance in real time.

[0082] The hierarchical active disturbance rejection control architecture in this embodiment is a two-level control system designed for the guidance system of high-speed maglev trains. Its core principle is to treat model uncertainty, external disturbances, and actuator (guide electromagnet) failures as a unified comprehensive disturbance, achieving stable control without precise model dependence through online estimation and compensation. The extended state observation module in this embodiment is the core functional module of the hierarchical active disturbance rejection control architecture. Its function is to perceive and estimate the operating state and disturbances of the guidance system in real time, directly outputting the observed values ​​of the guidance system state variables and the real-time observed values ​​of the comprehensive disturbance, providing core data support for control command generation. The observed values ​​of the system state variables are parameter values ​​that reflect the overall operating state of the guidance system, including directly measurable parameters such as guide clearance and vehicle yaw angle, as well as non-measurable state quantities such as vehicle lateral acceleration and guide electromagnetic force change rate estimated by algorithms. The real-time observed value of the comprehensive disturbance is a unified disturbance value obtained by the extended state observation module after fusing and estimating various disturbance factors in the operation of the guidance system. These disturbance factors include, but are not limited to, system model parameter uncertainties, guide electromagnet performance degradation / failure, track irregularities, and external environmental interference, serving as the core basis for dynamic disturbance compensation.

[0083] S104: Based on the observed values ​​of the state variables of the guidance system and the real-time observed values ​​of the integrated disturbances, the current loop disturbance rejection control layer and the gap layer disturbance rejection control layer of the hierarchical active disturbance rejection control architecture are used to perform dual-layer collaborative calculations to dynamically compensate for the disturbances in the control input and generate control commands.

[0084] The hierarchical active disturbance rejection (ADNR) architecture in this embodiment consists of a current loop ADNR layer and a gap layer ADNR layer. These layers are responsible for the rapid tracking of the guide electromagnet current and the stable adjustment of the guide gap, respectively, working together to achieve the control objective. Specifically, the current loop ADNR layer, as the bottom-level execution control, is responsible for accurately and quickly tracking current commands and providing real-time compensation for electromagnet drive circuitry and load disturbances. The gap layer ADNR layer, as the upper-level decision control, is responsible for maintaining the guide gap stable at a preset target value and actively compensating for system-level disturbances.

[0085] The purpose of steps S103 and S104 is to achieve online disturbance estimation and active compensation. After obtaining the observation results (i.e., the observed values ​​of the guidance system state variables and the real-time observed values ​​of the comprehensive disturbance), the guidance control system performs dynamic disturbance compensation on the control input based on the observation results. This mitigates the impact of disturbances introduced by the guidance actuator on the system's operating state in a timely manner when performance degradation or failure occurs. The two-layer collaborative calculation in this embodiment refers to the collaborative calculation process completed by the current loop active disturbance rejection control layer and the gap layer active disturbance rejection control layer in the hierarchical active disturbance rejection control architecture. The gap layer calculates the adjustment requirement based on the preset target value of the guidance gap, and the current layer converts this requirement into an electromagnet current adjustment command. Simultaneously, it combines the comprehensive disturbance value to complete the compensation calculation. The two layers communicate data and collaboratively output control results. The dynamic disturbance compensation in this embodiment refers to the proactive calculation of compensation amount added during the control input calculation process for the real-time observed value of the comprehensive disturbance. The core is to offset the impact of disturbances on the guidance system by correcting the control command in advance (rather than correcting after the disturbance occurs), achieving feedforward compensation and ensuring system stability.

[0086] Specifically, the optimal control parameters are loaded into the hierarchical active disturbance rejection control architecture, completing the initial configuration of the core adjustable parameters of the architecture; the guide gap control layer uses the preset target value of the guide gap as the core reference, and combines the real-time value of the guide gap and key state variables such as the vehicle yaw angle in the observed state variables of the guide system to calculate the real-time deviation and deviation change rate of the guide gap, thereby determining the guide electromagnetic force adjustment requirements required to achieve gap stability, forming the control reference command of the current control layer; the current loop active disturbance rejection control layer receives the control reference command issued by the gap layer active disturbance rejection control layer, combines the real-time observation value of the comprehensive disturbance, and performs precise calculations based on the calculation rules set by the optimal control parameters, simultaneously initiating the dynamic disturbance compensation process: according to the magnitude, direction and... The changing trend is analyzed, and the corresponding disturbance compensation is calculated. This compensation is then added to the basic calculation results of the current loop active disturbance rejection control layer to offset the impact of various disturbances such as system model uncertainty, electromagnet performance degradation, and track irregularities on the control effect. Through the two-layer collaborative linkage of the hierarchical active disturbance rejection control architecture, the stability requirements of the guide gap are transformed into current adjustment commands that the guide electromagnet can directly execute. The generated control commands are subjected to amplitude limiting and smoothing filtering to avoid sudden jumps in commands that cause sudden changes in the guide electromagnet current and fluctuations in electromagnetic force, ensuring the safety, stability, and accuracy of the control commands. Finally, control commands adapted to the current train operating conditions, electromagnet execution capabilities, and real-time disturbance status are output, providing a clear execution basis for subsequent adjustment of the guide electromagnet operating current.

[0087] Furthermore, before inputting the optimal control parameter combination into the hierarchical active disturbance rejection control architecture, the above may specifically include: constructing a hierarchical active disturbance rejection control architecture, which includes a current loop active disturbance rejection control layer and a gap layer active disturbance rejection control layer; the current loop active disturbance rejection control layer includes a second-order extended state observation module; the gap layer active disturbance rejection control layer includes a second-order tracking differential module, a third-order extended state observation module, and a disturbance function module.

[0088] Specifically, considering the characteristics of high-speed maglev train guidance systems—fast response, complex operating conditions, and susceptibility to disturbances and faults—a hierarchical active disturbance rejection control architecture is constructed. For example... Figure 5 and Figure 6 As shown, Figure 5 An example diagram of a current loop active disturbance rejection control layer provided in an embodiment of the present invention; Figure 6 This is an example diagram of a gap-layer active disturbance rejection control layer provided in an embodiment of the present invention. In the current loop active disturbance rejection framework, the second-order extended state observer (ESO) can calculate a fast tracking of the output current. And the precise observation of the total disturbance, z2, the basic control variable A linear feedback algorithm is used. Indicates input voltage The output of the second-order extended state observer ESO difference, It is the proportional coefficient of linear feedback. It is the gain coefficient of the disturbance and control quantities. This represents unknown external disturbances. Represents resistance. Represents the capacitor. In the gap-loop active disturbance rejection framework, the second-order tracking differentiator (TD) outputs the reference signal. Tracking signal Simultaneously output reference signal Differential signal In the system, it plays the role of arranging the transient process and providing the differential of the reference signal. The third-order extended state observer (ESO) can calculate the system's observation of the unknown disturbance by compensating for the two different directions of u through the output gap of the controlled object (guide electromagnet). z3 represents the gap change, z4 represents the gap change rate, and z5 represents the unknown disturbance to the system. The disturbance compensation control law consists of two parts: a basic control variable and a disturbance compensation variable. It can be designed using various control algorithms. Disturbance function module. and This represents the known model information in the guidance system. The perturbation function module is as follows: f0 represents the perturbation function; k z The known relevant gain coefficient of the guidance system is represented by ; m represents the mass. The current loop active disturbance rejection layer of this embodiment can accelerate response and adjustment speed. The gap layer active disturbance rejection layer of this embodiment can better compensate for errors caused by disturbances, making the gap more stable and preventing accidents within permissible limits.

[0089] S105: Adjust the working current of the guide electromagnet according to the control command to stabilize the guide gap at the preset target value.

[0090] This embodiment adjusts the working current of the guide electromagnet according to control commands, changing the output of the guide electromagnetic force to correct the guide gap of the guide system in real time, stabilizing the guide gap at a preset target value. The control commands in this embodiment are precise operation commands generated by a layered active disturbance rejection control architecture through two-layer collaborative calculation and dynamic disturbance compensation, and can be directly issued to the execution unit. Essentially, they are working current adjustment commands for the guide electromagnet, clearly specifying the target current value to which the electromagnet needs to be adjusted. The guide electromagnetic force output in this embodiment is the electromagnetic force generated when the guide electromagnet is supplied with working current. It is the core driving force for adjusting the guide gap. By changing the magnitude and direction of the electromagnetic force, the lateral and yaw movements of the vehicle body can be constrained, achieving the correction and stabilization of the guide gap. The guide gap is the physical gap between the guide electromagnet and the track side beam, and is the core control object of the guide system. Its stability directly determines the smoothness and safety of train operation. The preset target value in this embodiment is the optimal reference value for the guide gap, calibrated through simulation and experimentation based on the structural safety requirements, operational stability requirements, and hardware adaptation requirements of the high-speed maglev train guide system. For example, the control accuracy of the guide gap is ±0.2~0.5mm of the preset target value.

[0091] For example, by modifying the drive current command of the guide electromagnet, adjusting the guide gap control reference, or adaptively modifying the control gain, the adverse effects introduced by insufficient guide force, response lag, or external disturbances can be actively offset when the guide actuator experiences performance degradation or failure. This can promptly weaken the impact of fault disturbances on the system's operating state and prevent the guide system from experiencing severe fluctuations, control saturation, or instability due to sudden faults or parameter changes.

[0092] Furthermore, the effectiveness of the aforementioned fault-tolerant control method can be verified under multiple operating conditions. For example, the guidance control method can be simulated or tested under different train speeds, different track curvature conditions, and various combinations of guide actuator failures. The focus is on evaluating the guide clearance stability, control accuracy, and disturbance rejection performance of the guidance system under different operating conditions. By comparing and analyzing the operating effects of the guidance system under normal and fault conditions, the ability of the method to maintain stable system operation under high-speed, complex disturbances, and guide actuator failures is verified, thereby confirming the effectiveness and engineering feasibility of the method in high-speed maglev train guidance systems.

[0093] Applying the train fault-tolerant control method provided in this embodiment of the invention, S101: Based on the actual operating state of the guide electromagnet, multi-source state information is collected in real time, and the multi-source state information is quantified to output equivalent guide execution capability parameters; S102: Based on the current operating condition of the train and the equivalent guide execution capability parameters, the optimal control parameter combination is obtained by searching and matching from a pre-built parameter library; S103: The optimal control parameter combination is input into the hierarchical active disturbance rejection control architecture, and the extended state observation module in the hierarchical active disturbance rejection control architecture is used to obtain the observed values ​​of the guide system state variables and the real-time observed values ​​of the comprehensive disturbance in real time; S104: Based on the observed values ​​of the guide system state variables and the real-time observed values ​​of the comprehensive disturbance, the current loop active disturbance rejection control layer and the gap active disturbance rejection control layer of the hierarchical active disturbance rejection control architecture are used for dual-layer collaborative calculation to perform dynamic disturbance compensation on the control input and generate control commands; S105: The operating current of the guide electromagnet is adjusted according to the control commands to stabilize the guide gap at a preset target value. This method not only adapts control parameters to different train operating conditions and the performance status of the guide electromagnets through offline optimization and rapid online matching of parameter libraries, but also effectively offsets the effects of various disturbances such as system model uncertainty, guide electromagnet performance degradation / failure, track irregularities, and external wind disturbances by using real-time observation and dynamic compensation of hierarchical active disturbance rejection control. This significantly improves the fault tolerance and control stability of the train guidance system, ensuring that the guide gap remains stable at the preset target value. It effectively solves the problems of poor adaptability of fixed parameters, untimely disturbance compensation leading to low control accuracy and easy system instability in traditional control methods, and provides a reliable control guarantee for the safe and stable operation of trains under different operating conditions and throughout the entire life cycle of the electromagnets.

[0094] The key to this invention lies in constructing a hierarchical active disturbance rejection and fault-tolerant control method for high-speed maglev train guidance systems. This method unifies model uncertainties, external disturbances, and actuator degradation into extended disturbances for online estimation and compensation. Combined with actuator capability characterization and parameter self-optimization mechanisms, it achieves stable control and highly robust operation of the guidance system under complex operating conditions and multiple fault conditions. Please refer to [link / reference] for details. Figure 7 , Figure 7 A flowchart illustrating a train fault-tolerant control architecture provided in this embodiment of the invention may specifically include:

[0095] (1) System modeling and performance characterization.

[0096] A multi-degree-of-freedom coupled dynamic model of the guidance system is established to accurately reproduce its dynamic characteristics. Based on multi-source state information, a unified quantitative characterization method for the guidance capability of the guidance electromagnet is constructed to achieve accurate assessment of its health / degradation / failure status.

[0097] (2) Control architecture design.

[0098] Design a hierarchical active disturbance rejection control architecture to adapt to the multi-stage control requirements of the guidance system; realize online estimation of system state and comprehensive disturbance, and offset various uncertainties through active compensation.

[0099] (3) Parameter mapping and global optimization.

[0100] Construct a nonlinear mapping model (such as an RBF neural network) between control parameters and system performance to achieve rapid prediction of parameters and performance; use a swarm intelligence optimization algorithm (such as the PSO algorithm) in combination with the nonlinear mapping model to perform global optimization of control parameters and obtain the optimal combination of control parameters suitable for different scenarios.

[0101] (4) Adaptive matching and verification.

[0102] It achieves adaptive matching of control parameters for fault / degradation states, and quickly retrieves the optimal combination of control parameters based on real-time operating conditions and the state of the guide electromagnet; it completes the verification of fault-tolerant control effect under multiple operating conditions, and ensures the stable and reliable operation of the system in all scenarios and throughout its entire life cycle.

[0103] The method of this invention has four main advantages:

[0104] (1) Strong robustness. In view of the characteristics of strong nonlinearity, strong coupling and susceptibility to external disturbances and actuator failures in the guidance system of high-speed maglev train, a hierarchical guidance control system based on active disturbance rejection control is constructed. The model uncertainty, external disturbances and performance degradation of the guidance electromagnet are uniformly regarded as extended disturbances for online estimation and compensation. Without relying on an accurate mathematical model of the system, it can still effectively suppress the influence of disturbances and significantly improve the operational stability and robustness of the guidance system under complex working conditions and fault conditions. It is particularly suitable for high-speed operation scenarios.

[0105] (2) Strong adaptability. By introducing an equivalent execution capability representation method for the health status of the guide actuator, a unified description of the action capability of the guide electromagnet under different degradation and failure states is realized. Combined with the extended state observation and adaptive matching mechanism of control parameters, the guide control system can dynamically adjust the control strategy according to the actual operating state and fault characteristics. Even when one or more guide electromagnets are abnormal, the system can still maintain controllability and stability, which significantly enhances the fault tolerance and engineering adaptability of the guide system.

[0106] (3) Intelligentization. A nonlinear mapping model between control parameters and system performance was constructed, and a swarm intelligence optimization algorithm was introduced to globally optimize the key parameters of the active disturbance rejection controller, avoiding the traditional method of repeatedly adjusting parameters based on manual experience. Through this method, the optimal combination of control parameters can be quickly obtained under different operating speeds and fault modes, enabling the control system to have the ability to self-optimize and adaptively adjust parameters, which meets the needs of the intelligent development of the high-speed maglev train control system;

[0107] (4) Strong engineering practicality. It is mainly based on the existing operating status information and control structure of the guidance system, without the need to add complex hardware or make major modifications to the original system, which is convenient for integration with the existing high-speed maglev train guidance control system. At the same time, through simulation or actual operation verification under multiple working conditions, it is proved that the method can maintain good guidance gap control accuracy and anti-disturbance performance under different speeds, different track conditions and multiple fault combinations, which has high engineering practical value and prospects for promotion and application.

[0108] The train fault-tolerant control device provided in the embodiments of the present invention will be described below. The train fault-tolerant control device described below can be referred to in correspondence with the train fault-tolerant control method described above.

[0109] Please refer to the details. Figure 8 , Figure 8 A schematic diagram of a train fault-tolerant control device provided in an embodiment of the present invention may include:

[0110] The quantization module 100 is used to collect multi-source status information in real time according to the actual operating state of the guide electromagnet, quantize the multi-source status information, and output equivalent guide execution capability parameters.

[0111] The retrieval module 200 is used to retrieve and match from a pre-built parameter library based on the current operating conditions of the train and the equivalent guidance execution capability parameters to obtain the optimal combination of control parameters.

[0112] The observation acquisition module 300 is used to input the optimal control parameter combination into the hierarchical active disturbance rejection control architecture, and to use the extended state observation module in the hierarchical active disturbance rejection control architecture to acquire the observation values ​​of the guiding system state variables and the real-time observation values ​​of the comprehensive disturbance in real time.

[0113] The compensation module 400 is used to perform dynamic disturbance compensation on the control input and generate control commands based on the observed values ​​of the state variables of the guidance system and the real-time observed values ​​of the comprehensive disturbances, through two-layer collaborative calculations of the current loop active disturbance rejection control layer and the gap layer active disturbance rejection control layer of the hierarchical active disturbance rejection control architecture.

[0114] The adjustment control module 500 is used to adjust the working current of the guide electromagnet according to the control command, so that the guide gap is stabilized at a preset target value.

[0115] Furthermore, based on the above embodiments, the quantization module 100 may include:

[0116] A multi-source status information acquisition unit is used to acquire the multi-source status information of the guide electromagnet in real time during actual operation; the multi-source status information includes the working current variation characteristics of the guide electromagnet, the guide gap response characteristics, and the gap deviation between the control command and the actual output of the guide electromagnet.

[0117] The quantization unit is used to preprocess the multi-source state information and, based on a preset quantization representation algorithm, fuse and calculate the preprocessed multi-source state information to output the equivalent guided execution capability parameter.

[0118] Furthermore, based on the above embodiments, the train fault-tolerant control device may further include:

[0119] A hierarchical active disturbance rejection control architecture construction module is used to construct the hierarchical active disturbance rejection control architecture before inputting the optimal control parameter combination into the hierarchical active disturbance rejection control architecture. The hierarchical active disturbance rejection control architecture includes a current loop active disturbance rejection control layer and a gap layer active disturbance rejection control layer.

[0120] The current loop active disturbance rejection control layer includes: a second-order extended state observation module;

[0121] The gap layer active disturbance rejection control layer includes: a second-order tracking differentiation module, a third-order extended state observation module, and a disturbance function module; the disturbance function module is:

[0122] ;

[0123] Where f0 represents the disturbance function; z3 represents the gap change; k z represents the known correlation gain coefficient of the guidance system; m represents the mass.

[0124] Furthermore, based on the above embodiments, the train fault-tolerant control device may further include:

[0125] The equivalent dynamic model construction module is used to establish a multi-degree-of-freedom equivalent dynamic model of the guidance system. Based on the multi-degree-of-freedom equivalent dynamic model, the module simulates the dynamic response characteristics of the guidance system under different train operating conditions and different equivalent guidance execution capability parameters, and obtains multiple sets of simulation state data. The multiple sets of simulation state data are quadruple data of train operating conditions, equivalent guidance execution capability parameters, control parameters, and guidance system control performance indicators. The control parameters are the parameters in the hierarchical active disturbance rejection control architecture.

[0126] The nonlinear mapping model construction module is used to construct a nonlinear mapping model with control parameters, train operating conditions and equivalent guidance execution capability parameters as inputs and guidance system control performance indicators as outputs. The nonlinear mapping model is trained using the multiple sets of simulation state data to obtain a trained nonlinear mapping model.

[0127] The parameter optimization module is used to optimize the control performance of the guidance system by employing a particle swarm optimization algorithm in combination with the trained nonlinear mapping model to globally optimize the control parameters for each scenario, thereby obtaining multiple sets of optimal control parameter combinations that adapt to different scenarios. The scenarios are obtained by combining different train operating conditions and different equivalent guidance execution capability parameters.

[0128] The parameter library construction module is used to combine all the optimal control parameters obtained through optimization, organize and store them according to the scenario, and construct the parameter library.

[0129] Furthermore, based on the above embodiments, the equivalent dynamic model construction module may include:

[0130] The equivalent dynamic model construction unit is used to obtain the structural parameters, operating state parameters and configuration information of the guide system and the guide electromagnet, and to perform equivalent simplification processing on the guide electromagnet, the vehicle body and its connecting structure to construct the multi-degree-of-freedom equivalent dynamic model that can simultaneously reflect the lateral displacement and yaw motion characteristics of the guide system.

[0131] Furthermore, based on the above embodiments, the control performance indicators of the guidance system may include at least: a gap error indicator, a settling time indicator, and a control stability indicator; wherein, the gap error indicator is used to characterize the steady-state accuracy and tracking performance of the guidance system, and is obtained by describing the statistical characteristics of the deviation between the target guidance gap and the actual guidance gap; the settling time indicator is used to characterize the dynamic response speed of the guidance system, and is determined based on the time it takes for the guidance gap response to first enter and remain within the allowable error range; the control stability indicator is used to characterize the degree of oscillation and fluctuation during the response process of the guidance system, and is obtained by comprehensively characterizing the peak value change, overshoot characteristics, steady-state fluctuation amplitude, and statistical characteristics of the control quantity change characteristics of the guidance gap response curve.

[0132] Furthermore, based on the above embodiments, the retrieval module 200 may include:

[0133] The matching unit is used to match a combination of control parameters adapted to the current scenario in the parameter library using the current operating condition of the train and the equivalent guidance execution capability parameter as dual search keywords and through a preset search algorithm. The current operating condition of the train includes one or more of the following: real-time train speed, current track curvature, and track gradient. The parameter library is an optimal set of control parameters constructed offline based on a radial basis function neural network nonlinear mapping model and a particle swarm optimization algorithm. The parameter library stores the optimal control parameters adapted to different operating scenarios and guidance electromagnet states according to the combination of train operating condition type and equivalent guidance execution capability parameter level.

[0134] The judgment unit is used to determine whether the matched combination of control parameters meets the parameter adjustment range of the hierarchical active disturbance rejection control architecture and the real-time operation requirements of the guidance system.

[0135] The result unit is used to output the matched combination of control parameters as the optimal control parameters for the current scenario if the conditions are met.

[0136] It should be noted that the order of the modules and units in the above-mentioned train fault-tolerant control device can be changed without affecting the logic.

[0137] The train fault-tolerant control device provided in this embodiment of the invention comprises a quantization module 100, which collects multi-source state information in real time based on the actual operating state of the guide electromagnet, quantizes the multi-source state information, and outputs equivalent guide execution capability parameters; a retrieval module 200, which retrieves and matches the optimal control parameter combination from a pre-built parameter library based on the current operating condition of the train and the equivalent guide execution capability parameters; an observation value acquisition module 300, which inputs the optimal control parameter combination into a hierarchical active disturbance rejection control architecture, and uses the extended state observation module in the hierarchical active disturbance rejection control architecture to acquire real-time observation values ​​of the guide system state variables and real-time observation values ​​of the comprehensive disturbance; a compensation module 400, which performs dynamic disturbance compensation on the control input based on the observation values ​​of the guide system state variables and the real-time observation values ​​of the comprehensive disturbance, through a two-layer collaborative calculation using the current loop active disturbance rejection control layer and the gap active disturbance rejection control layer of the hierarchical active disturbance rejection control architecture, and generates control commands; and an adjustment control module 500, which adjusts the operating current of the guide electromagnet according to the control commands to stabilize the guide gap at a preset target value. This device adapts control parameters to different train operating conditions and the performance status of the guide electromagnets through offline optimization and rapid online matching of parameter libraries. It also effectively offsets the effects of various disturbances such as system model uncertainty, guide electromagnet performance degradation / failure, track irregularities, and external wind disturbances by using real-time observation and dynamic compensation of hierarchical active disturbance rejection control. This significantly improves the fault tolerance and control stability of the train guidance system, ensuring that the guide gap remains stable at the preset target value. It effectively solves the problems of poor adaptability of fixed parameters, low control accuracy, and easy system instability caused by untimely disturbance compensation in traditional control methods, and provides a reliable control guarantee for the safe and stable operation of trains under different operating conditions and throughout the entire life cycle of the electromagnets.

[0138] The train fault-tolerant control device provided in the embodiments of the present invention will be described below. The train fault-tolerant control device described below can be referred to in correspondence with the train fault-tolerant control method described above.

[0139] Please refer to Figure 9 , Figure 9 A schematic diagram of a train fault-tolerant control device provided in an embodiment of the present invention may include:

[0140] Memory 10 is used to store computer programs;

[0141] The processor 20 is used to execute computer programs to implement the above-described train fault-tolerant control method.

[0142] The memory 10, processor 20, and communication interface 31 all communicate with each other through the communication bus 32.

[0143] In this embodiment of the invention, the memory 10 is used to store one or more programs. The programs may include program code, which includes computer operation instructions. In this embodiment of the invention, the memory 10 may store programs for implementing the following functions:

[0144] Based on the actual operating status of the guide electromagnet, multi-source status information is collected in real time, and the multi-source status information is quantified to output equivalent guide execution capability parameters.

[0145] Based on the current operating conditions of the train and the equivalent guidance execution capability parameters, the optimal combination of control parameters is obtained by searching and matching from a pre-built parameter library.

[0146] The optimal combination of control parameters is input into the hierarchical active disturbance rejection control architecture, and the extended state observation module in the hierarchical active disturbance rejection control architecture is used to obtain the real-time observation values ​​of the guided system state variables and the real-time observation values ​​of the comprehensive disturbance.

[0147] Based on the observed values ​​of the state variables of the guidance system and the real-time observed values ​​of the integrated disturbances, a two-layer collaborative calculation is performed through the current loop active disturbance rejection control layer and the gap layer active disturbance rejection control layer of the hierarchical active disturbance rejection control architecture to perform dynamic disturbance compensation on the control input and generate control commands;

[0148] Adjust the working current of the guide electromagnet according to the control command to stabilize the guide gap at the preset target value.

[0149] In one possible implementation, the memory 10 may include a program storage area and a data storage area, wherein the program storage area may store the operating system and applications required for at least one function; and the data storage area may store data created during use.

[0150] Furthermore, memory 10 may include read-only memory and random access memory, providing instructions and data to the processor. A portion of the memory may also include NVRAM. The memory stores operating systems and operating instructions, executable modules, or data structures, or subsets thereof, or extended sets thereof, wherein the operating instructions may include various operating instructions for implementing various operations. The operating system may include various system programs for implementing various basic tasks and handling hardware-based tasks.

[0151] Processor 20 can be a central processing unit (CPU), an application-specific integrated circuit, a digital signal processor, a field-programmable gate array, or other programmable logic device. Processor 20 can be a microprocessor or any conventional processor. Processor 20 can call programs stored in memory 10.

[0152] Communication interface 31 can be an interface for the communication module, used to connect with other devices or systems.

[0153] Of course, it should be noted that, Figure 9 The structure shown does not constitute a limitation on the train fault-tolerant control device in the embodiments of the present invention. In practical applications, the train fault-tolerant control device may include more than Figure 9 More or fewer components as shown, or combinations of certain components.

[0154] The computer-readable storage medium provided in the embodiments of the present invention is described below. The computer-readable storage medium described below can be referred to in correspondence with the train fault-tolerant control method described above.

[0155] The present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the above-described train fault-tolerant control method.

[0156] The computer-readable storage medium may include various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0157] 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. For the apparatus disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to in the method section.

[0158] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or 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, but such implementations should not be considered beyond the scope of this invention.

[0159] Finally, it should be noted that in this document, relationships such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0160] The foregoing has provided a detailed description of a train fault-tolerant control method, apparatus, device, and computer-readable storage medium provided by the present invention. Specific examples have been used to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. At the same time, for those skilled in the art, there will be changes in specific implementation methods and application scope based on the ideas of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A train fault-tolerant control method, characterized in that, include: Based on the actual operating status of the guide electromagnet, multi-source status information is collected in real time, and the multi-source status information is quantified to output an equivalent guide execution capability parameter. The equivalent guide execution capability parameter is used to quantitatively characterize the change in the effective guide capability of the guide electromagnet under different performance degradation and failure states. The equivalent guide execution capability parameter is a single feature parameter or a multi-dimensional feature parameter group, and its value is positively correlated with the health level and guide execution capability of the guide electromagnet, and is used to accurately characterize the current actual working status and fault degradation level of the guide electromagnet. Based on the current operating conditions of the train and the equivalent guidance execution capability parameters, the optimal combination of control parameters is obtained by searching and matching from a pre-built parameter library. The optimal control parameter combination is input into the hierarchical active disturbance rejection control architecture, and the extended state observation module in the hierarchical active disturbance rejection control architecture is used to obtain the real-time observed values ​​of the guided system state variables and the real-time observed values ​​of the comprehensive disturbance in real time. Based on the observed values ​​of the state variables of the guidance system and the real-time observed values ​​of the integrated disturbance, the current loop disturbance rejection control layer and the gap layer disturbance rejection control layer of the hierarchical disturbance rejection control architecture perform dual-layer collaborative calculation to dynamically compensate for the disturbances in the control input and generate control commands. The operating current of the guide electromagnet is adjusted according to the control command to stabilize the guide gap at a preset target value.

2. The train fault-tolerant control method according to claim 1, characterized in that, Based on the actual operating state of the guide electromagnet, multi-source status information is collected in real time, and the multi-source status information is quantified to output equivalent guide execution capability parameters, including: The multi-source status information of the guide electromagnet is collected in real time during actual operation; the multi-source status information includes the working current variation characteristics of the guide electromagnet, the guide gap response characteristics, and the gap deviation between the control command and the actual output of the guide electromagnet. The multi-source state information is preprocessed, and based on a preset quantization representation algorithm, the preprocessed multi-source state information is fused and calculated to output the equivalent guided execution capability parameter.

3. The train fault-tolerant control method according to claim 1, characterized in that, Before inputting the optimal control parameter combination into the hierarchical active disturbance rejection control architecture, the following is also included: The hierarchical active disturbance rejection control architecture is constructed, which includes a current loop active disturbance rejection control layer and a gap layer active disturbance rejection control layer. The current loop active disturbance rejection control layer includes: a second-order extended state observation module; The gap layer active disturbance rejection control layer includes: a second-order tracking differentiation module, a third-order extended state observation module, and a disturbance function module; the disturbance function module is: ； Where f0 represents the disturbance function; z3 represents the gap change; k z represents the known correlation gain coefficient of the guidance system; m represents the mass.

4. The train fault-tolerant control method according to claim 1, characterized in that, Before obtaining the optimal combination of control parameters by searching and matching from a pre-built parameter library based on the train's current operating conditions and the equivalent guidance execution capability parameters, the process also includes: A multi-degree-of-freedom equivalent dynamic model of the guidance system is established. Based on the multi-degree-of-freedom equivalent dynamic model, the dynamic response characteristics of the guidance system under different train operating conditions and different equivalent guidance execution capability parameters are simulated, and multiple sets of simulation state data are obtained. The multiple sets of simulation state data are quadruple data of train operating conditions, equivalent guidance execution capability parameters, control parameters, and guidance system control performance indicators. The control parameters are the parameters in the hierarchical active disturbance rejection control architecture. A nonlinear mapping model is constructed with control parameters, train operating conditions, and equivalent guidance execution capability parameters as inputs and guidance system control performance indicators as outputs. The nonlinear mapping model is trained using the multiple sets of simulation state data to obtain a trained nonlinear mapping model. With the goal of optimizing the control performance of the guidance system, a particle swarm optimization algorithm is used in conjunction with the trained nonlinear mapping model to globally optimize the control parameters for each scenario, resulting in multiple sets of optimal control parameter combinations that adapt to different scenarios. The scenarios are those obtained by combining different train operating conditions and different equivalent guidance execution capability parameters. All the optimal control parameters obtained through optimization are combined, organized and stored according to the scenario, and the parameter library is constructed.

5. The train fault-tolerant control method according to claim 4, characterized in that, Establish a multi-degree-of-freedom equivalent dynamic model of the guidance system, including: The structural parameters, operating state parameters, and configuration information of the guide electromagnet of the guidance system are obtained. The guide electromagnet, the vehicle body, and their connecting structures are simplified by equivalent processing to construct the multi-degree-of-freedom equivalent dynamic model that can simultaneously reflect the lateral displacement and yaw motion characteristics of the guidance system.

6. The train fault-tolerant control method according to claim 4, characterized in that, The control performance indicators of the guidance system include: clearance error indicator, settling time indicator, and control stability indicator; The gap error index is used to characterize the steady-state accuracy and tracking performance of the guidance system, and is obtained by describing the statistical characteristics of the deviation between the target guidance gap and the actual guidance gap; The adjustment time index is used to characterize the dynamic response speed of the guidance system and is determined based on the time it takes for the guidance gap response to first enter and remain within the allowable error range. The control stability index is used to characterize the degree of oscillation and fluctuation in the response process of the guidance system. It is obtained by comprehensively characterizing the peak value change, overshoot characteristics, steady-state fluctuation amplitude, and statistical characteristics of the control quantity change characteristics of the guidance gap response curve.

7. The train fault-tolerant control method according to claim 1, characterized in that, Based on the current operating conditions of the train and the equivalent guidance execution capability parameters, the optimal combination of control parameters is obtained by searching and matching from a pre-built parameter library, including: Using the current train operating condition and the equivalent guidance execution capability parameter as dual search keywords, a preset search algorithm matches control parameter combinations adapted to the current scenario in the parameter library. The current train operating condition includes one or more of the following: real-time train speed, current track curvature, and track gradient. The parameter library is an optimal set of control parameters constructed offline based on a radial basis function neural network nonlinear mapping model and a particle swarm optimization algorithm. The parameter library stores optimal control parameters adapted to different operating scenarios and guidance electromagnet states according to the combination of train operating condition type and equivalent guidance execution capability parameter level. Determine whether the matched control parameter combination meets the parameter adjustment range of the hierarchical active disturbance rejection control architecture and the real-time operation requirements of the guidance system; If the conditions are met, the combination of control parameters obtained from the matching is output as the optimal control parameters for the current scenario.

8. A train fault-tolerant control device, characterized in that, include: The quantization module is used to collect multi-source status information in real time based on the actual operating state of the guide electromagnet, quantify the multi-source status information, and output equivalent guide execution capability parameters. The equivalent guide execution capability parameters are used to quantitatively characterize the changes in the effective guiding capability of the guide electromagnet under different performance degradation and failure states. The equivalent guide execution capability parameters are single feature parameters or multi-dimensional feature parameter groups, and their values ​​are positively correlated with the health level and guide execution capability of the guide electromagnet, and are used to accurately characterize the current actual working state and fault degradation level of the guide electromagnet. The retrieval module is used to search and match from a pre-built parameter library based on the current operating conditions of the train and the equivalent guidance execution capability parameters to obtain the optimal combination of control parameters. The observation acquisition module is used to input the optimal control parameter combination into the hierarchical active disturbance rejection control architecture, and to use the extended state observation module in the hierarchical active disturbance rejection control architecture to acquire the observation values ​​of the guiding system state variables and the real-time observation values ​​of the comprehensive disturbance in real time. The compensation module is used to perform dynamic disturbance compensation on the control input and generate control commands based on the observed values ​​of the state variables of the guidance system and the real-time observed values ​​of the comprehensive disturbances, through two-layer collaborative calculations of the current loop active disturbance rejection control layer and the gap layer active disturbance rejection control layer of the hierarchical active disturbance rejection control architecture. The adjustment and control module is used to adjust the working current of the guide electromagnet according to the control command, so that the guide gap is stabilized at a preset target value.

9. A train fault-tolerant control device, characterized in that, include: Memory, used to store computer programs; A processor for executing the computer program to implement the train fault-tolerant control method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when loaded and executed by a processor, implement the train fault-tolerant control method as described in any one of claims 1 to 7.

Images