Subway Electric Multiple Unit Emergency Operation Redundancy Control System and Method

Abstract

The present invention discloses an emergency operation redundant control system and method for a subway electric multiple unit, which relates to the technical field of emergency control. The system analyzes key components of a target subway electric multiple unit, constructs a main vehicle control network and a standby vehicle control network, diagnoses faults in the vehicle control data stream to obtain vehicle operation fault parameters. When the operation fault parameters reach a preset fault threshold, an emergency operation mode is obtained. Based on the emergency operation mode, operation strategy analysis is carried out to obtain vehicle emergency operation control strategy parameters, and emergency operation redundant control is performed on the target subway electric multiple unit. The present invention solves the technical problems in the prior art that the subway electric multiple unit lacks reliable redundant control, inaccurate fault diagnosis, and imperfect emergency operation strategies, resulting in poor safety and stability of emergency operation, and achieves the technical effects of realizing accurate fault diagnosis, optimizing emergency operation control strategies, improving resource utilization efficiency, and ensuring the safety and stability of vehicle emergency operation.

Description

Technical Field

[0001] The present invention relates to the technical field of emergency control, and particularly to an emergency operation redundant control system and method for subway electric multiple units. Background Art

[0002] In modern urban transportation, subway electric multiple units, as high-capacity public transportation vehicles, their safe and reliable operation is crucial. However, there are many problems in the current subway electric multiple unit control systems. On the one hand, most of the existing control systems lack effective redundancy design. On the other hand, there are obvious deficiencies in traditional fault diagnosis technologies. When facing complex passenger car operation data streams, the existing diagnostic methods often cannot accurately and timely identify faults. In addition, there are also defects in the emergency operation strategy. When a fault occurs and emergency treatment is required, the existing systems cannot fully consider the functional states of key components of the passenger car, the operating environment, and the mutual relationships between components. This makes the emergency operation strategy not optimized enough to reasonably utilize the available component resources while ensuring safety.

[0003] The prior art has technical problems such as the lack of reliable redundant control for subway electric multiple units, inaccurate fault diagnosis, and imperfect emergency operation strategies, resulting in poor safety and stability of emergency operation. Summary of the Invention

[0004] This application provides an emergency operation redundant control system and method for subway electric multiple units, which are used to solve the technical problems in the prior art that subway electric multiple units lack reliable redundant control, inaccurate fault diagnosis, and imperfect emergency operation strategies, resulting in poor safety and stability of emergency operation.

[0005] In view of the above problems, this application provides an emergency operation redundant control system and method for subway electric multiple units.

[0006] In the first aspect of this application, an emergency operation redundant control system for subway electric multiple units is provided. The system includes:

[0007] Bus key component set acquisition module, which is used to analyze the key components of the target subway electric vehicle, obtain the bus key component set, and equip the target subway electric vehicle with a redundant control unit, where the redundant control unit includes a main control unit and a standby control unit; a control network construction module, which is used to communicate and connect the main control unit and the standby control unit with the bus key component set respectively to construct a main bus control network and a standby bus control network; a bus operation fault parameter acquisition module, which is used to monitor and obtain the bus control data stream through the main bus control network, perform fault diagnosis on the bus control data stream, and obtain the bus operation fault parameters; an emergency operation mode acquisition module, which is used to trigger a redundant switching mechanism to switch and enable the standby bus control network when the bus operation fault parameters reach a preset fault threshold to obtain an emergency operation mode; a control strategy parameter acquisition module, which is used to perform operation strategy analysis on the bus operation fault parameters according to the emergency operation mode based on the standby bus control network, obtain the bus emergency operation control strategy parameters, and perform emergency operation redundancy control on the target subway electric vehicle based on the bus emergency operation control strategy parameters.

[0008] In the second aspect of the present application, a method for redundant control of emergency operation of subway electric vehicles is provided. The method includes:

[0009] Analyze the key components of the target subway electric vehicle to obtain the bus key component set, and equip the target subway electric vehicle with a redundant control unit, where the redundant control unit includes a main control unit and a standby control unit; communicate and connect the main control unit and the standby control unit with the bus key component set respectively to construct a main bus control network and a standby bus control network; monitor and obtain the bus control data stream through the main bus control network, perform fault diagnosis on the bus control data stream, and obtain the bus operation fault parameters; trigger a redundant switching mechanism to switch and enable the standby bus control network when the bus operation fault parameters reach a preset fault threshold to obtain an emergency operation mode; perform operation strategy analysis on the bus operation fault parameters according to the emergency operation mode based on the standby bus control network, obtain the bus emergency operation control strategy parameters, and perform emergency operation redundancy control on the target subway electric vehicle based on the bus emergency operation control strategy parameters.

[0010] One or more technical solutions provided in the present application have at least the following technical effects or advantages:

[0011] Analyze the key components of the target subway electric multiple unit (EMU) to obtain the set of key components of the EMU, and equip the target subway EMU with a redundant control unit; respectively communicate and connect the main control unit and the standby control unit with the set of key components of the EMU to construct the main control network and the standby control network of the EMU; monitor and obtain the control data stream of the EMU through the main control network of the EMU, perform fault diagnosis on the control data stream of the EMU to obtain the operation fault parameters of the EMU; when the operation fault parameters of the EMU reach the preset fault threshold, trigger the redundant switching mechanism to obtain the emergency operation mode; perform operation strategy analysis on the operation fault parameters of the EMU to obtain the emergency operation control strategy parameters of the EMU, and perform emergency operation redundancy control on the target subway EMU based on the emergency operation control strategy parameters of the EMU. It achieves the technical effects of realizing accurate fault diagnosis, optimizing the emergency operation control strategy, improving the resource utilization efficiency, and ensuring the safety and stability of the emergency operation of the EMU. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention, and those of ordinary skill in the art can obtain other drawings according to these drawings without creative efforts.

[0013] Figure 1 It is a schematic structural diagram of the emergency operation redundant control system for subway electric multiple units provided by the embodiment of the present application;

[0014] Figure 2 It is a schematic flow diagram of the emergency operation redundant control method for subway electric multiple units provided by the embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0015] The present application provides an emergency operation redundant control system and method for subway electric multiple units, which are used to solve the technical problems in the prior art that the subway electric multiple units lack reliable redundant control, inaccurate fault diagnosis, and imperfect emergency operation strategies, resulting in poor safety and stability of the emergency operation.

[0016] The following will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present application without creative efforts belong to the scope of protection of the present application.

[0017] Embodiment 1, as Figure 1 shown, the present application provides an emergency operation redundant control system for subway electric multiple units, and the system includes:

[0018] The key component set acquisition module 10 of the passenger car is used to analyze the key components of the target subway electric passenger car, obtain the key component set of the passenger car, and equip the target subway electric passenger car with a redundant control unit, where the redundant control unit includes a main control unit and a standby control unit.

[0019] Specifically, the key component set acquisition module 10 of the passenger car plays a key role in ensuring the reliable operation of the subway electric passenger car. First, the Fault Tree Analysis (FTA) method is used to analyze the target subway electric passenger car from a system level, construct a logic tree that may cause train failures, and determine the key branches that affect normal operation. At the same time, combined with the Failure Mode and Effects Analysis (FMEA), for each component in subsystems such as the traction system, braking system, electrical control system, and door system of the train, its potential failure modes, causes, and consequences are evaluated in detail. Through comprehensive consideration of these analysis results, key components such as traction motors, brake discs, Controller Area Network (CAN) bus controllers, and door drive motors are identified, thus obtaining the key component set of the passenger car. After completing the key component analysis, this module equips the target subway electric passenger car with a redundant control unit. The main control unit uses a high-performance microprocessor, such as a multi-core processor based on the ARM architecture, and runs a Real-Time Operating System (RTOS) to ensure precise and fast control of key components. It is connected to key components through various communication interfaces, including but not limited to CAN bus, Ethernet interface, etc., receives component status information and sends control instructions. The standby control unit is a complete backup of the main control unit, also equipped with a high-performance microprocessor and corresponding communication interfaces in terms of hardware, and also runs the same RTOS and control programs in terms of software. The two maintain status synchronization through a heartbeat detection mechanism to ensure seamless takeover of control tasks when the main control unit fails, and to ensure the continuous and safe operation of the electric passenger car.

[0020] The control network construction module 20 is used to communicate and connect the main control unit and the standby control unit with the key component set of the passenger car respectively, and construct a main control network and a standby control network for the passenger car.

[0021] Specifically, when the control network construction module 20 constructs the network, for the connection between the main control unit and the set of key components of the passenger car, the industrial Ethernet technology with high bandwidth and low latency, such as the EtherCAT protocol, is first selected. This protocol can achieve microsecond-level synchronization accuracy. In terms of hardware, a specially designed Ethernet switch is adopted, and its ports have strong anti-interference ability and data forwarding ability to ensure the accuracy and stability of data during transmission. The switch is connected to the main control unit through optical fibers to ensure the signal quality during long-distance transmission. For the communication link between the main control unit and the key components, different types of Ethernet interfaces, such as RJ45 interfaces or optical fiber interfaces, are used according to the functions and data transmission requirements of different components. When constructing the backup control network of the passenger car, a similar method is also adopted, but in order to further improve the reliability, redundant design is added. In addition to using industrial Ethernet technology, the CAN bus is also combined as a backup communication method between the backup control unit and the set of key components. The CAN bus is connected using twisted pairs and has good anti-electromagnetic interference ability and error detection mechanism. In terms of network topology, a hybrid mode of star and bus is adopted to ensure that the backup control unit can still communicate with the key components when some links fail, thus constructing a stable and reliable backup control network for the passenger car, which together with the main control network of the passenger car provides guarantee for the safe operation of the subway electric passenger car.

[0022] The passenger car operation fault parameter acquisition module 30 is used to monitor and acquire the passenger car control data stream through the main control network of the passenger car, and perform fault diagnosis on the passenger car control data stream to obtain the passenger car operation fault parameters.

[0023] Specifically, the bus operation fault parameter acquisition module 30 monitors and acquires the bus control data stream through the main bus control network, and uses an algorithm combining wavelet analysis and neural network to perform fault diagnosis on it to obtain the bus operation fault parameters. In terms of data acquisition, high-precision sensors are deployed at key nodes in the main bus control network to collect various operation parameters of the bus in real time. These sensors include Hall sensors for measuring motor current and voltage, optoelectronic encoders for detecting vehicle speed, pressure sensors for monitoring braking pressure, etc. The acquired analog signals are converted into digital signals by a high-speed analog-to-digital converter after passing through a conditioning circuit, and then transmitted to the fault diagnosis unit through the network. In the fault diagnosis unit, wavelet analysis is first used to preprocess the bus control data stream. Wavelet transform can decompose the signal into wavelet coefficients of different scales, and these coefficients highlight the characteristics of the signal in different frequencies and time locals. For complex non-stationary signals such as the bus operation data stream, wavelet analysis can effectively remove noise interference and extract the signal components reflecting fault characteristics. For example, when there is an inter-turn short circuit fault in the traction motor, the motor current signal will have tiny fluctuations at a specific frequency, and wavelet analysis can enhance this weak fault characteristic signal. Subsequently, the feature data after wavelet processing is input into the neural network, and here a multi-layer feedforward neural network is adopted. The number of input layer nodes of the neural network is determined according to the number of features extracted after wavelet analysis, and each node corresponds to a feature value. The number of hidden layer nodes is determined through experiments and optimization, and the output layer nodes correspond to different fault types and fault severity levels. In the training stage, a large amount of historical fault data and normal operation data are used to train the neural network. By adjusting the weights and thresholds of the network, the network can learn the feature mapping relationships under different fault modes. In the actual diagnosis process, the neural network outputs the corresponding fault diagnosis results, that is, the bus operation fault parameters, according to the characteristics of the current bus control data stream input, accurately judging whether there is a fault, the fault type (such as electrical fault, mechanical fault, etc.) and the severity of the fault (minor, moderate, severe), providing a precise basis for subsequent emergency handling.

[0024] An emergency operation mode acquisition module 40, where the emergency operation mode acquisition module 40 is used to trigger a redundant switching mechanism to switch and enable the bus standby control network when the bus operation fault parameters reach a preset fault threshold, and obtain an emergency operation mode.

[0025] Specifically, the emergency operation mode acquisition module 40 is a key link to ensure the safe operation of the subway electric multiple unit in case of failure. When the failure parameters transmitted by the vehicle operation failure parameter acquisition module reach the preset failure threshold, this module is quickly activated. The internal failure monitoring sub-module compares the vehicle operation failure parameters with the preset thresholds in real time. These thresholds are determined based on a large amount of experimental data, simulation analysis, and actual operation experience, and are precisely set for different key components and failure types. Once it is detected that the failure parameters exceed the standard, the trigger signal is immediately transmitted to the redundant switching control sub-module. This sub-module is implemented based on a programmable logic controller (PLC), and there is a precise switching logic program preset in the PLC. First, it quickly evaluates the current failure situation to judge the scope and severity of the failure impact. Then, according to the preset switching strategy, the switching operation at the hardware level is executed through the relay array. For the key communication lines, high-speed electronic switches are used to implement the switching between the primary and backup networks. These electronic switches have a switching speed of nanoseconds, which can ensure the rapid conversion of the network connection. At the same time, at the software level, the vehicle backup control network is initialized and the configuration is updated, which involves loading the control parameters and algorithms optimized for the current failure scenario. These parameters and algorithms are stored in the non-volatile memory to ensure that the data will not be lost during the switching process. Through these technical means, the switching and enabling from the vehicle main control network to the vehicle backup control network are successfully completed, so that the subway electric multiple unit enters the emergency operation mode, ensuring that the train can still maintain the basic safe operation function in case of failure.

[0026] The control strategy parameter acquisition module 50, based on the vehicle backup control network according to the emergency operation mode, analyzes the operation strategy of the vehicle operation failure parameters to obtain the vehicle emergency operation control strategy parameters, and performs emergency operation redundancy control on the target subway electric multiple unit based on the vehicle emergency operation control strategy parameters.

[0027] Specifically, when the passenger car enters the emergency operation mode, this module starts to work based on the passenger car's standby control network. First, it receives the current passenger car operation fault parameters from the passenger car's standby control network. These parameters cover electrical system fault information (such as abnormal motor current, voltage fluctuation), mechanical drive system faults (such as too high gearbox oil temperature, abnormal vibration), and the fault states of other key subsystems. When analyzing the operation strategy, a model predictive control (MPC) algorithm combined with an expert system is used. On the one hand, the MPC algorithm establishes a system state equation according to the dynamic model, electric traction model, etc. of the subway electric passenger car, uses the current fault parameters as the initial conditions, and predicts the future operation state of the passenger car under different control strategies. It solves an optimal control problem within a finite time domain through online optimization, considering multiple objectives such as speed control, traction force distribution, braking distance, etc., and finds the optimal control sequence that satisfies system constraints (such as motor power limit, braking safety distance). On the other hand, a large number of rules formed by domain expert experience and historical fault cases are stored in the expert system. For example, when a certain type of specific electrical fault occurs and the mechanical system is partially normal, the expert system will give corresponding control suggestions, such as adjusting the motor control strategy, switching to the standby power supply, etc. These rules are integrated with the results of the MPC algorithm, and through methods such as weighted decision-making for comprehensive analysis, the emergency operation control strategy parameters of the passenger car are finally obtained. Finally, based on these accurate emergency operation control strategy parameters of the passenger car, control instructions are sent to each key component through the passenger car's standby control network. For motor control, vector control technology is adopted to adjust the torque and speed of the motor according to the strategy parameters; for the braking system, the braking pressure and braking force distribution are accurately controlled. Thus, the emergency operation redundancy control of the target subway electric passenger car is realized, ensuring that the passenger car can operate in the safest and most stable manner under fault conditions.

[0028] In a possible implementation manner, the passenger car operation fault parameter acquisition module 30 is further configured to perform the following steps:

[0029] Collect and obtain the subway electric passenger car control data set, where the subway electric passenger car control data set includes the subway electric passenger car operation data set and the control unit working data set; successively perform associated feature extraction on the subway electric passenger car operation data set according to the passenger car key component set to obtain the passenger car component associated operation feature set, and at the same time extract the control unit working feature set of the control unit working data set; perform dimensionality reduction and recognition training on the passenger car component associated operation feature set and the control unit working feature set to generate a passenger car fault adaptive diagnosis network; perform fault diagnosis on the passenger car control data stream based on the passenger car fault adaptive diagnosis network, and output the passenger car operation fault parameters.

[0030] Specifically, in the operating environment of subway electric multiple units, a control data set of subway electric multiple units is collected through a variety of high-precision sensors and data acquisition devices, which comprehensively cover all key parts and control systems of the electric multiple units, thereby collecting a rich control data set of subway electric multiple units. Among them, the operating data set of subway electric multiple units covers many parameters reflecting the train operation status. For example, a speed sensor installed near the wheel continuously and accurately measures the running speed of the train and records the real-time change value of the speed; a load sensor located at the carbody connection can obtain the real-time load data of different cars to reflect the load condition of the vehicle, and these data are crucial for understanding the working status of the motor. At the same time, the working data set of the control unit focuses on the working conditions of the control unit. From the input signals of the control unit, including the command signals from the driver operation console and the real-time status signals feedback by various sensors, to the execution parameters of each algorithm module inside the control unit, such as the proportional, integral, and differential parameters in the speed regulation algorithm, and the signals output by the control unit to each actuator, as well as the status information of the communication module during communication with other subsystems, such as communication delay and data packet loss rate, are all completely recorded, jointly constituting a complete control data set of subway electric multiple units.

[0031] When performing feature extraction on the subway electric train operation dataset, various specific means are used based on the set of key components of the train. For the traction motor, a key component, the Apriori algorithm is first used to mine the association rules between data such as motor current, voltage, and speed. The support is set to 0.6 and the confidence is set to 0.8 to find the frequent item sets and determine the stable associations of these parameters under different working conditions. At the same time, wavelet analysis is used, and a suitable wavelet basis function (such as db4) is selected to decompose the data. The wavelet coefficient changes of the motor during the starting, accelerating, constant speed, and decelerating stages are analyzed at 3 to 5 scales to extract the features that can reflect the motor state. For the braking system, the K-Means clustering algorithm is used to cluster data such as braking pressure, braking distance, and brake pad temperature into 5 categories, and the central values and data distributions of each category are analyzed to obtain the association features of these parameters under different braking intensities. For the door system, the relationships between parameters such as door opening time, closing time, current change, and anti-pinch pressure threshold are mined through association rules, with the support set to 0.5 and the confidence set to 0.7, so as to obtain the association features of door operation. Finally, the associated operation feature set of the train components is integrated. For the control unit working dataset, spectral analysis is used to perform a fast Fourier transform on the input signal, with the sampling frequency set to 1000 Hz, and the spectrum in the 0 to 500 Hz frequency band is obtained to analyze features such as the main frequency and harmonic components. For the control algorithm parameters, their mean, variance, and standard deviation are calculated within a 10-minute sliding window. For example, for the proportional parameter of speed control, the change range of its mean in different operation stages is observed. For the output signal, the level change and the duty cycle change of the PWM signal are converted into logical expressions, and the logical relationship between the duty cycle change from 20% to 80% and motor drive is analyzed. The communication data packets are captured using Wireshark, and the data transmission delay, packet loss rate, and retransmission times are counted every 5 seconds, and the communication protocol fields are parsed to extract the control unit working feature set.

[0032] In the process of building a bus fault diagnosis system, dimensionality reduction identification training is first carried out on the associated operation feature set of bus components and the working feature set of control units. The principal component analysis (PCA) algorithm is used to perform dimensionality reduction on the feature set. By calculating the eigenvalues and eigenvectors of the covariance matrix, the first several principal components are selected according to the contribution rate to remove redundant information. For example, the originally high-dimensional motor operation features (including multi-dimensional information such as current, voltage, temperature, vibration frequency, etc.) are reduced to several main comprehensive feature dimensions. At the same time, linear discriminant analysis (LDA) is combined to further enhance the class separability, making it easier to distinguish different fault types and normal states in the low-dimensional space. For the dimensionality-reduced features, a deep neural network (DNN) is used for identification training. A neural network architecture including an input layer, multiple hidden layers, and an output layer is constructed. The number of nodes in the input layer matches the number of dimensionality-reduced features. In the training stage, a large number of sample data with fault labels and normal operation labels are input into the network, and the weights and biases of the network are adjusted through the backpropagation algorithm, enabling the network to learn the mapping relationship between features and fault types. As the training continues, the network parameters are continuously optimized, and finally a bus fault adaptive diagnosis network is generated.

[0033] During actual operation, based on this bus fault adaptive diagnosis network, fault diagnosis is carried out on the bus control data stream. After the preprocessed bus control data stream collected in real time is input into the diagnosis network. The network analyzes and judges the features in the data stream according to the learned patterns and weights. If the features in the data stream match a certain fault pattern during the training process, the network will output the corresponding bus operation fault parameters, which describe in detail the type of fault (such as electrical fault, mechanical fault), the location of the fault (such as a certain motor, a certain door), and the severity of the fault (slight, moderate, severe), providing key basis for subsequent fault repair and emergency handling.

[0034] In a possible implementation manner, the bus operation fault parameter acquisition module 30 is further configured to perform the following steps:

[0035] Perform fault correlation analysis on each feature information in the associated operation feature set of bus components and the working feature set of control units respectively to obtain an operation feature correlation coefficient set and a working feature correlation coefficient set; perform feature dimensionality reduction screening on the operation feature correlation coefficient set and the working feature correlation coefficient set according to a preset correlation coefficient threshold to obtain an available associated operation feature set and an available working feature set; use a deep learning network to perform fault identification training on the available associated operation feature set and the available working feature set respectively to obtain a bus component fault diagnosis network and a control unit fault diagnosis network; perform equal-weight fusion on the bus component fault diagnosis network and the control unit fault diagnosis network to generate the bus fault adaptive diagnosis network.

[0036] Specifically, fault correlation analysis is carried out for each feature information in the bus component associated operation feature set and the control unit working feature set. For the bus component associated operation feature set, for example, when analyzing the features related to the traction motor, by calculating the correlation between the current feature and the temperature feature, the correlation between the speed feature and the torque feature, and so on. By analogy, all feature combinations involved in each component are analyzed to obtain the operation feature correlation coefficient set. Similarly, in the control unit working feature set, the correlation between the control signal frequency feature and the instruction execution delay feature, the correlation between the communication status feature and the algorithm parameter adjustment feature, etc. are analyzed, and then the working feature correlation coefficient set is obtained. These correlation coefficients accurately reflect the degree of tightness of the association between each feature in the case of a fault.

[0037] Next, feature dimensionality reduction screening is performed on the operation feature correlation coefficient set and the working feature correlation coefficient set according to a preset correlation coefficient threshold. This preset threshold is determined through a large amount of historical fault data and experimental simulation analysis. If the correlation coefficient of a certain feature with other features is lower than the threshold, it indicates that the contribution of this feature in fault diagnosis is small and can be screened out. After such a screening process, an available associated operation feature set is obtained from the bus component associated operation feature set, and an available working feature set is obtained from the control unit working feature set. Such dimensionality reduction operations can effectively reduce the complexity and redundancy of the data while retaining the information most valuable for fault diagnosis.

[0038] Then, deep learning networks are used to perform fault identification training on the available associated operation feature set and the available working feature set respectively. For the available associated operation feature set, a convolutional neural network (CNN) is used. The data of the available associated operation feature set is used as input, local features are extracted through the convolutional layer, and feature compression is performed by the pooling layer. After processing through multiple hidden layers, the fault identification is finally output at the output layer. During the training process, a large number of sample data with labeled fault types are used. By adjusting the weights and biases of the network, the network can accurately identify the fault type according to the input features, thereby obtaining the bus component fault diagnosis network. For the available working feature set, a similar deep learning algorithm is also used, and a deep belief network (DBN) is used for training to obtain the control unit fault diagnosis network.

[0039] Finally, the bus component fault diagnosis network and the control unit fault diagnosis network are fused with equal weights, that is, the same weight is assigned to the two networks during the fusion process, and the output results of them are comprehensively processed. For example, when the diagnosis probabilities of the two networks for a certain fault are both certain values, the two probability values are added and averaged to generate an adaptive bus fault diagnosis network. This fused network can make full use of the information from both the bus components and the control unit to diagnose the faults during the bus operation more comprehensively and accurately.

[0040] In a possible implementation manner, the control strategy parameter acquisition module 50 is further configured to perform the following steps:

[0041] According to the emergency operation mode, perform function limit screening on the set of key bus components to obtain a set of available bus components; based on the bus standby control network, perform operation strategy analysis on the bus operation fault parameters to obtain a bus operation control strategy; according to the component emergency rule, perform priority analysis on the set of available bus components to obtain the emergency priority of the bus components; monitor and obtain the current bus operation data stream through the bus standby control network, and based on the bus operation control strategy, perform analysis and optimization on the emergency priority of the bus components and the current bus operation data stream to obtain the bus emergency operation control strategy parameters.

[0042] Specifically, in the emergency operation state, first, perform function limit screening on the set of key bus components according to the established emergency operation mode. This process requires in-depth analysis of the specific requirements and limiting conditions of the emergency operation mode. For example, in the case of partial function limitation caused by a specific fault, if the power supply system is abnormal, those comfort components with high power requirements and non-critical functions (such as some lighting systems, some function modules of the air conditioner, etc.) will be identified. Through this refined screening based on the emergency operation mode, the components that can still work normally or can work with limited functions under the current emergency situation are distinguished from the set of key bus components, so as to obtain a set of available bus components.

[0043] Based on the bus backup control network, in case of emergency, the bus backup control network becomes the key to ensuring operation. It receives bus operation fault parameters, which cover fault information in multiple aspects such as from the electrical system to the mechanical transmission. Based on the intelligent processing module in the bus backup control network, first, a fault analysis algorithm is used to classify and quantify the bus operation fault parameters. For example, for electrical fault parameters, it is judged whether it is a short circuit, an open circuit, or a voltage anomaly, and the specific value of the fault severity is determined. Combining the dynamic model of the bus and the emergency operation safety standard, a rule-based policy library is constructed to generate a policy model. If the traction motor fault parameter shows insufficient motor output power, considering the current operating state information such as the speed and load of the bus at the same time, by searching in the policy library, a policy is generated to adjust the power output of other normal motors and change the transmission ratio to maintain the speed. For the brake system fault parameter, if it shows uneven braking force, according to the safety standard and the dynamic model, policies such as redistributing the braking force and enabling the auxiliary braking device will be formulated. In addition, the relevance and mutual influence between various components will also be considered. For example, when there are slight fault parameters in the steering system, the impact on driving stability and the force changes on other suspension and transmission components will be comprehensively considered, so as to formulate a series of comprehensive policies including adjusting the speed limit and optimizing the suspension parameters, so as to obtain a comprehensive and accurate bus operation control strategy to ensure the bus runs as safely and stably as possible in case of failure.

[0044] According to the component emergency rules, a priority analysis is carried out on the set of available bus components. These component emergency rules are formulated by comprehensively considering various factors such as the importance of the component to the safe operation of the bus, its replaceability after failure, and its impact on other components. For key components that directly affect driving safety, such as the core components of the brake system and the bogie, the highest priority is given; for components that can maintain the basic operation function of the bus to a certain extent, such as some communication modules and auxiliary power supply equipment, the priorities are determined in turn according to their roles in the emergency situation, and then the bus component emergency priorities are obtained.

[0045] Finally, the current bus operation data stream is continuously monitored and obtained through the bus backup control network. This data stream contains the actual operation state information of each component of the bus at the current moment. Based on the previously obtained bus operation control strategy, the bus component emergency priorities and the current bus operation data stream are analyzed and optimized. In this process, an optimization algorithm is used to comprehensively consider the component priorities and information such as speed, load, and working parameters of each component in the real-time data stream, and find the combination of control parameters that can make the bus run most stably and safely in the current emergency state. Finally, the bus emergency operation control strategy parameters are obtained. These parameters will guide the specific operation of the bus in case of emergency and ensure the safety of the bus and passengers.

[0046] In a possible implementation manner, the control strategy parameter acquisition module 50 is further configured to perform the following steps:

[0047] According to the component emergency rule, determine the component priority evaluation index set, and based on the component priority evaluation index set, construct a multi-dimensional evaluation polar coordinate system; divide the numerical range of each index information in the component priority evaluation index set to obtain an index numerical division threshold set; map the index numerical division threshold set to the multi-dimensional evaluation polar coordinate system for interval level marking to obtain a priority evaluation polar coordinate system; based on the priority evaluation polar coordinate system, perform multi-dimensional quantitative evaluation on each component information in the available passenger car component set to generate a passenger car component coordinate interval atlas; perform priority sorting on the available passenger car component set according to the area level of the passenger car component coordinate interval atlas to obtain the passenger car component emergency priority.

[0048] Specifically, in the emergency handling process, first, according to the component emergency rule, clarify the component priority evaluation index set, which covers multiple key dimensions, including the degree of impact of the component on the running safety of the passenger car, the frequency of failure occurrence, the replaceability of the component, the chain impact of the failure on other components, and the difficulty of maintenance, etc. Based on this rich evaluation index set, construct a multi-dimensional evaluation polar coordinate system, with each evaluation index as a dimension. For example, with the degree of safety impact as the polar axis direction, other indexes expand around it to form a multi-dimensional spatial structure, providing a comprehensive framework for subsequent evaluation.

[0049] For each index information in the component priority evaluation index set, perform a detailed division of the numerical range, which requires a large amount of historical data, experimental results, and industry experience. For example, for the index of failure occurrence frequency, according to past statistics, it is divided into three ranges: low (less than 5 failures per year), medium (5 - 10 failures per year), and high (more than 10 failures per year); for replaceability, it is divided into very easy to replace (there are ready-made spare parts and no complex operation is required for replacement), relatively easy to replace (specific spare parts are required and there is a certain difficulty in replacement), difficult to replace (no spare parts or special customization is required), etc. Through such division, an index numerical division threshold set is obtained.

[0050] In the key steps of constructing the priority evaluation system, it is necessary to accurately map the threshold set of index values to the multi-dimensional evaluation polar coordinate system. For each threshold of the index value, find its corresponding position and range in the polar coordinate system. For example, in a three-dimensional evaluation polar coordinate system, taking the importance of components to safety, failure frequency, and maintenance difficulty as the three coordinate axes, for the threshold division of the safety importance index, if the value is within the threshold range of 0 to 3 (low importance), mark the corresponding interval on the corresponding coordinate axis. After all the threshold values of the index values are mapped to the multi-dimensional evaluation polar coordinate system, perform priority division identification according to the area of the interval enclosed by the threshold values of each index value level. This interval area has a profound meaning, which comprehensively reflects the performance of the component on each evaluation index. Taking a simple two-dimensional case as an example, if a component has high numerical levels on both the safety importance and failure frequency indicators, the area of the rectangular interval enclosed by these two thresholds will be large. In the multi-dimensional case, this relationship is more complex, but the principle is the same. The larger the area, the more important the component occupies in multiple key evaluation indicators, which also means that the priority level of this component is higher. Through such detailed mapping and division, a priority evaluation polar coordinate system that can accurately reflect the priority of components is successfully obtained, providing a strong basis for the accurate evaluation and ranking of bus components in the future.

[0051] Based on this priority evaluation polar coordinate system, conduct multi-dimensional quantitative evaluation on the information of each component in the set of available bus components. For each component, determine its position in the polar coordinate system according to its specific values on each evaluation index, and generate an atlas of coordinate intervals of bus components. For example, a certain key braking component, due to its extremely high impact on safety, low failure frequency, poor replaceability, etc., its coordinate interval in the polar coordinate system is in the core and specific area.

[0052] Finally, perform priority ranking on the set of available bus components according to the area level of the atlas of coordinate intervals of bus components. The area level comprehensively reflects the priority situation of components in each dimension. The larger the area (indicating outstanding performance in multiple high-priority dimensions), the higher its priority. In this way, the emergency priority of bus components is accurately obtained, providing a scientific basis for resource allocation and operation sequence in emergency operation.

[0053] In a possible implementation manner, the control strategy parameter acquisition module 50 is further configured to execute the following steps:

[0054] Extract the operating characteristics from the current bus operation data stream to obtain a bus operation status feature set, which includes component status, driving environment, and bus operation status; perform data association mining based on the bus operation control strategy to construct a bus operation control solution space, and perform constraint partitioning on the bus operation control solution space according to the bus operation status feature set to obtain a bus operation control parameter library; determine the first emergency bus control component according to the emergency priority of the bus components; perform strategy equilibrium optimization on the bus operation control parameter library based on the first emergency bus control component to determine the bus emergency operation control strategy parameters.

[0055] Specifically, when processing the current bus operation data stream, first, for the component status, detailed information is obtained through the sensor data installed on each key component. For example, the real-time temperature value, current magnitude, and rotational speed data of the motor are extracted from the temperature sensor, current transformer, and rotational speed encoder of the traction motor, so as to judge whether the motor is in a normal working state, whether there is overheating, overload, or abnormal vibration; the braking pressure data is obtained from the pressure sensor of the braking system, and combined with the feedback of the brake pad wear sensor to understand the effectiveness and wear degree of the braking system; the position sensor and force sensor of the door system can reflect the opening and closing state of the door, whether it is jammed, and whether the anti-pinch function is normal. These component-related data together constitute the component status information. For the driving environment, a variety of environment perception technologies are used. The track slope and flatness information are obtained through the track detection sensors installed at the bottom of the bus to judge whether the bus is climbing, descending, or running on a flat track; the weather sensor at the front of the bus can detect weather conditions, such as the impact of rain, snow, fog, etc. on visibility and track friction, and at the same time, combined with the humidity sensor on the track, accurately evaluate the impact of weather and track humidity on the bus driving; in addition, cameras and image processing technologies are used to identify the environmental characteristics around the track, such as the curvature of the curve and whether there are foreign objects invading, etc. These data completely depict the driving environment where the bus is located. In terms of the bus operation status, it is the extraction of the overall motion parameters of the bus. The real-time driving speed of the bus is obtained from the speed sensor, and combined with the data of the acceleration sensor to determine whether the bus is accelerating, decelerating, or moving at a constant speed; the driving direction of the bus and the attitude change of the bus during driving are obtained through the gyroscope and direction sensor. These data accurately reflect the current operation status of the bus. Through such a comprehensive and in-depth extraction process, a bus operation status feature set including component status, driving environment, and bus operation status is finally formed.

[0056] Based on the obtained bus operation control strategy, data correlation mining is carried out. By analyzing various rules, parameters in the control strategy and their mutual relationships, a bus operation control solution space is constructed using data mining algorithms. This solution space contains various possible control schemes, but these schemes need to be further screened and optimized. The bus operation control solution space is constrained and divided according to the bus operation state feature set. For example, if the component status shows that a certain motor has a minor fault, the driving environment is on a slippery track and the bus operation state is high-speed driving, then these conditions will limit the solution space and eliminate those control schemes that do not conform to the current situation. After such screening, a more targeted bus operation control parameter library is obtained.

[0057] Determine the first emergency bus control component according to the emergency priority of bus components. This priority is obtained through complex evaluation before. In case of emergency, the high-priority component ranked at the top is the first emergency bus control component. For example, when there is a serious fault in the braking system, the key braking valve in the braking system is the first emergency bus control component because it is crucial for the safe parking of the bus.

[0058] When determining the parameters of the bus emergency operation control strategy, in-depth analysis is carried out based on the first emergency bus control component, and the functional characteristics and current state of the first emergency bus control component are analyzed in detail. For example, if it is the key braking valve of the braking system, it is necessary to clarify the current value of its valve opening, the feedback of braking pressure, and the possible scope of fault influence, etc. Then, based on this, the bus operation control parameter library is screened to eliminate obviously inapplicable parameter combinations. The multi-objective optimization algorithm is used to perform strategy balance optimization. Considering various factors such as the safe, stable and efficient operation of the bus, taking the braking system as an example, it is necessary to ensure that the braking distance is within the safe range, and at the same time avoid unstable phenomena such as vehicle skidding and tail-swing during braking, and also take into account the impact on other normally operating components. Search for the simulation results under different parameter combinations in the parameter library. By establishing the dynamic model and control model of the bus operation, calculate the performance indicators corresponding to each parameter combination. In this process, combined with the actual bus operation state feature set, such as the current driving speed, load, track conditions, etc., the parameter combination is further constrained and optimized. If the bus is in a high-speed driving and fully loaded state, and the track has a certain slope, it is more inclined to select parameters that can ensure smooth and effective braking under such complex conditions. After multiple iterative calculations and comparative analyses, the parameters of the bus emergency operation control strategy that best conform to the current emergency situation are finally determined to ensure that the bus can operate as safely as possible in the fault emergency state.

[0059] In a possible implementation manner, the control strategy parameter acquisition module 50 is further configured to perform the following steps:

[0060] According to the emergency operation requirements of the passenger car and the emergency operation balance target of the passenger car, an emergency operation fitness function is constructed. The strategy parameters of the passenger car operation control parameter library are evaluated by using the emergency operation fitness function to obtain an emergency operation strategy fitness set; the passenger car operation control parameter library is mapped and associated with the available passenger car component set to obtain a component operation control strategy parameter set; based on the emergency operation strategy fitness set, the component operation control strategy parameter set is compared and optimized to determine the first emergency operation strategy parameter of the first emergency passenger car control component; based on the first emergency operation strategy parameter, a Nash equilibrium optimization is performed on the component operation control strategy parameter set to determine the passenger car emergency operation control strategy parameter.

[0061] Specifically, first, according to the emergency operation requirements and balance target of the passenger car, each key performance index is clarified. For example, the goals in the emergency state may include stability (maintaining the continuity of the passenger car operation), safety (reducing or avoiding critical failures), response efficiency (rapid adjustment in case of emergencies), etc. These goals form the basis for the design of the fitness function. To achieve the above balance target, the factors that may affect the emergency operation of the passenger car are screened, and the main input parameters of the fitness function are selected. These parameters include the current operation state of the passenger car, the health state of key components, environmental conditions, emergency performance requirements, etc. Each parameter is assigned a weight according to its influence degree on the emergency operation target. For example, the health state of key components has a greater impact on system safety, so its weight will be higher. According to these input parameters and weights, an emergency operation fitness function is constructed. The fitness function can be expressed as a comprehensive evaluation formula, such as: F = w1×P1 + w2×P2 + w3×P3, where P1, P2, P3 are each key parameter, w1, w2, w3 are their corresponding weights, and F represents the comprehensive fitness score. The fitness function is used to evaluate each strategy parameter in the passenger car operation control parameter library one by one. For each control strategy parameter, it is substituted into the fitness function for calculation in turn, so as to obtain the fitness score of this strategy. The higher the fitness score, the closer the performance of this strategy to the target requirements in the emergency situation. After the one-by-one evaluation of the fitness function, the fitness scores of all strategy parameters are summarized to generate an emergency operation strategy fitness set. This set contains the fitness scores of each strategy parameter, providing a data basis for the subsequent strategy selection and optimization.

[0062] In the process of formulating the emergency operation plan, it is necessary to map and associate the bus operation control parameter library with the set of available bus components. The bus operation control parameter library contains various parameters affecting bus operation, such as speed control parameters, braking pressure parameters, power distribution parameters, etc. The set of available bus components covers components that can work properly or with limitations under the current emergency situation, such as key traction motors, braking system components, bogies, door systems, etc. Taking the traction motor as an example, in the process of mapping and association, parameters related to the motor are screened out from the bus operation control parameter library, including the starting current parameter of the motor, the operating speed control parameter, the torque adjustment parameter, etc. These parameters correspond to the available bus component of the traction motor and constitute a set of operation control strategy parameters for the traction motor. For the braking system components, parameters such as the adjustment parameter of the braking pressure, the braking response time parameter, and the braking force distribution parameter under different braking levels are found from the parameter library and associated with the braking components to form a set of control strategy parameters for the braking system. Similarly, parameters related to the bogie, such as the suspension stiffness adjustment parameter and the shock absorber damping parameter, are mapped with the bogie components, and the switch speed parameter and the anti-pinch force parameter of the door system are associated with the door components. Through such a mapping and association method, a set of optimization ranges for the control strategy parameters of each available bus component, that is, the component operation control strategy parameter set, is determined, which lays a solid foundation for subsequent precise control and strategy optimization.

[0063] The emergency operation strategy fitness set and the component operation control strategy parameter set are the key bases. Each element in the emergency operation strategy fitness set represents the comprehensive fitness degree of a set of passenger car operation control parameters in the emergency operation scenario. It is a quantitative result obtained from multi-dimensional considerations such as safety, stability, and component protection. The component operation control strategy parameter set details the possible control strategy parameter ranges for each available passenger car component. Based on the emergency operation strategy fitness set, a comparison and optimization of the component operation control strategy parameter set are carried out. For the control strategy parameter subset of each available passenger car component, its corresponding fitness value in the emergency operation strategy fitness set is matched and analyzed. Among all the components, the first emergency passenger car control component has the highest priority and plays a crucial role in the safety and stability of the passenger car's emergency operation. Carefully compare the fitness values corresponding to the various control strategy parameters of the first emergency passenger car control component, and select the parameter combination with the maximum fitness. For example, if the first emergency passenger car control component is a key valve in the braking system, then different valve opening speeds, pressure control parameters, etc. constitute its control strategy parameter set. Through the fitness evaluation in the emergency operation strategy fitness set regarding aspects such as braking stability, braking distance, and the overall dynamic impact on the vehicle, select the parameters that can make the braking system perform optimally in an emergency, that is, the first emergency operation strategy parameters, so as to ensure that the key components of the passenger car can operate in the best way in an emergency and guarantee the safety of the passenger car and passengers.

[0064] In the context of emergency operation, the first emergency operation strategy parameters provide crucial anchor points for the entire optimization process. Based on this, performing a Nash equilibrium optimization on the set of component operation control strategy parameters is a complex but crucial step. Nash equilibrium is a stable state of strategy combination, in which, given the strategies of other participants, the strategies of each component of the passenger car are optimal. For a passenger car, the first emergency operation strategy parameters of the first emergency passenger car control component have been determined, which is the most important control basis in the current emergency situation. Taking this parameter as the core, start analyzing the parameters of other components in the set of component operation control strategy parameters. For example, if the first emergency operation strategy parameter is the optimal parameter of a key component of the braking system, during the optimization process, it is necessary to analyze how to adjust the parameters of other components such as the traction system, steering system, and suspension system to ensure that while the braking system performs at its best, the entire passenger car is in a balanced and stable operating state. In the traction system, parameters such as the power and speed of the motor need to cooperate with the braking system to avoid braking failure or vehicle out of control caused by unreasonable power output; the steering angle and power assist of the steering system should adapt to the vehicle's attitude change during braking to prevent sideslip or fishtailing; the stiffness and damping parameters of the suspension system also need to be adjusted accordingly to ensure passenger comfort and vehicle stability during emergency braking or other emergency operations. Through continuous simulation and calculation, use the Nash equilibrium algorithm to search in the vast space of the set of component operation control strategy parameters to find such a set of parameters that when the first emergency passenger car control component operates according to the optimal strategy, the strategies of other components also reach a balanced state, thereby determining the emergency operation control strategy parameters of the passenger car and ensuring the safe and stable operation of the passenger car in an emergency.

[0065] Embodiment 2, based on the same inventive concept as the emergency operation redundancy control system of the subway electric passenger car in the foregoing embodiment, as Figure 2 shown, this application provides a method for emergency operation redundancy control of a subway electric passenger car. The method in the embodiments of this application and the system embodiments are based on the same inventive concept. Among them, the method includes:

[0066] Step S100: Analyze the key components of the target subway electric passenger car to obtain a set of passenger car key components, and equip the target subway electric passenger car with a redundancy control unit, where the redundancy control unit includes a main control unit and a standby control unit.

[0067] Step S200: Communicate and connect the main control unit and the standby control unit with the set of passenger car key components respectively to construct a main passenger car control network and a standby passenger car control network.

[0068] Step S300: Monitor and obtain the passenger car control data stream through the main passenger car control network, perform fault diagnosis on the passenger car control data stream, and obtain passenger car operation fault parameters.

[0069] Step S400: When the bus operation fault parameters reach the preset fault threshold, trigger the redundant switching mechanism to switch and enable the bus standby control network, and obtain the emergency operation mode.

[0070] Step S500: Based on the bus standby control network and according to the emergency operation mode, analyze the operation strategy of the bus operation fault parameters, obtain the bus emergency operation control strategy parameters, and perform emergency operation redundancy control on the target subway electric vehicle based on the bus emergency operation control strategy parameters.

[0071] Further, step S300 further includes:

[0072] Step S310: Collect and obtain the subway electric vehicle control data set, where the subway electric vehicle control data set includes the subway electric vehicle operation data set and the control unit working data set.

[0073] Step S320: Sequentially perform associated feature extraction on the subway electric vehicle operation data set according to the bus key component set to obtain the bus component associated operation feature set, and at the same time extract the control unit working feature set of the control unit working data set.

[0074] Step S330: Perform dimensionality reduction identification training on the bus component associated operation feature set and the control unit working feature set to generate a bus fault adaptive diagnosis network.

[0075] Step S340: Perform fault diagnosis on the bus control data stream based on the bus fault adaptive diagnosis network, and output the bus operation fault parameters.

[0076] Further, step S330 further includes:

[0077] Step S331: Respectively perform fault correlation analysis on each feature information in the bus component associated operation feature set and the control unit working feature set to obtain the operation feature correlation coefficient set and the working feature correlation coefficient set.

[0078] Step S332: Perform feature dimensionality reduction screening on the operation feature correlation coefficient set and the working feature correlation coefficient set according to the preset correlation coefficient threshold to obtain the available associated operation feature set and the available working feature set.

[0079] Step S333: Use a deep learning network to perform fault identification training on the available associated operation feature set and the available working feature set respectively to obtain a bus component fault diagnosis network and a control unit fault diagnosis network.

[0080] Step S334: Perform weighted fusion with equal weights on the bus component fault diagnosis network and the control unit fault diagnosis network to generate the bus fault adaptive diagnosis network.

[0081] Further, step S500 further includes:

[0082] Step S510: Perform function limit screening on the set of key bus components according to the emergency operation mode to obtain a set of available bus components.

[0083] Step S520: Analyze the operation fault parameters of the bus based on the bus standby control network to obtain a bus operation control strategy.

[0084] Step S530: Analyze the priority of the set of available bus components according to the component emergency rules to obtain the emergency priority of the bus components.

[0085] Step S540: Monitor and obtain the current bus operation data stream through the bus standby control network, and analyze and optimize the emergency priority of the bus components and the current bus operation data stream based on the bus operation control strategy to obtain the emergency operation control strategy parameters of the bus.

[0086] Further, step S530 further includes:

[0087] Step S531: According to the component emergency rules, determine a set of component priority evaluation indicators, and based on the set of component priority evaluation indicators, construct a multi-dimensional evaluation polar coordinate system.

[0088] Step S532: Divide the numerical range of each index information in the set of component priority evaluation indicators to obtain a set of index numerical division thresholds.

[0089] Step S533: Map the set of index numerical division thresholds into the multi-dimensional evaluation polar coordinate system for interval level marking to obtain a priority evaluation polar coordinate system.

[0090] Step S534: Perform multi-dimensional quantitative evaluation on each component information in the set of available bus components based on the priority evaluation polar coordinate system to generate a bus component coordinate interval atlas.

[0091] Step S535: Sort the set of available bus components according to the area level of the bus component coordinate interval atlas to obtain the emergency priority of the bus components.

[0092] Further, step S540 further includes:

[0093] Step S541: Extract the operation characteristics from the current bus operation data stream to obtain a bus operation state feature set, where the bus operation state feature set includes component states, driving environments, and the bus operation state.

[0094] Step S542: Perform data association mining based on the bus operation control strategy to construct a bus operation control solution space, and constraint partition the bus operation control solution space according to the bus operation state feature set to obtain a bus operation control parameter library.

[0095] Step S543: Determine the first emergency bus control component according to the emergency priority of the bus components.

[0096] Step S544: Perform policy equilibrium optimization on the bus operation control parameter library based on the first emergency bus control component to determine the bus emergency operation control strategy parameters.

[0097] Further, step S544 further includes:

[0098] Step S5441: Construct an emergency operation fitness function according to the bus emergency operation requirements and the bus emergency operation equilibrium target, and use the emergency operation fitness function to evaluate the policy parameters of the bus operation control parameter library to obtain an emergency operation policy fitness set.

[0099] Step S5442: Map and associate the bus operation control parameter library with the available bus component set to obtain a component operation control policy parameter set.

[0100] Step S5443: Compare and optimize the component operation control policy parameter set based on the emergency operation policy fitness set to determine the first emergency operation policy parameters of the first emergency bus control component.

[0101] Step S5444: Perform Nash equilibrium optimization on the component operation control policy parameter set based on the first emergency operation policy parameters to determine the bus emergency operation control strategy parameters.

[0102] It should be noted that the above sequence of the embodiments of the present application is only for description and does not represent the advantages or disadvantages of the embodiments. And the above describes specific embodiments of the present specification. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0103] The above are only the preferred embodiments of the present application and are not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

[0104] This specification and the accompanying drawings are merely illustrative of the present application and are considered to cover any and all modifications, variations, combinations, or equivalents within the scope of the present application. Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalent technologies, the present application is intended to include these changes and modifications.

Claims

1. Emergency operation redundancy control system for subway electric multiple units, characterized in that, The system includes: A key component set acquisition module for passenger cars, which is used to analyze the key components of the target subway electric passenger car, obtain the key component set of the passenger car, and equip the target subway electric passenger car with a redundant control unit. Among them, the redundant control unit includes a main control unit and a standby control unit; A control network construction module, which is used to communicate and connect the main control unit and the standby control unit with the key component set of the passenger car respectively to construct a main control network and a standby control network for the passenger car; A passenger car operation fault parameter acquisition module, which is used to monitor and obtain the passenger car control data stream through the main control network of the passenger car, and perform fault diagnosis on the passenger car control data stream to obtain the passenger car operation fault parameters; An emergency operation mode acquisition module, which is used to trigger a redundant switching mechanism to switch and enable the standby control network of the passenger car when the passenger car operation fault parameters reach a preset fault threshold, and obtain an emergency operation mode; A control strategy parameter acquisition module, which is based on the standby control network of the passenger car and according to the emergency operation mode, analyzes the operation strategy of the passenger car operation fault parameters, obtains the emergency operation control strategy parameters of the passenger car, and performs emergency operation redundancy control on the target subway electric passenger car based on the emergency operation control strategy parameters of the passenger car; Among them, the control strategy parameter acquisition module includes: Perform function limit screening on the key component set of the passenger car according to the emergency operation mode to obtain an available passenger car component set; Based on the standby control network of the passenger car, analyze the operation strategy of the passenger car operation fault parameters to obtain the passenger car operation control strategy; Perform priority analysis on the available passenger car component set according to the component emergency rule to obtain the passenger car component emergency priority; Monitor and obtain the current passenger car operation data stream through the standby control network of the passenger car, and analyze and optimize the passenger car component emergency priority and the current passenger car operation data stream based on the passenger car operation control strategy to obtain the emergency operation control strategy parameters of the passenger car; Among them, the control strategy parameter acquisition module includes: According to the component emergency rule, determine the component priority evaluation index set, and construct a multi-dimensional evaluation polar coordinate system based on the component priority evaluation index set; Divide the numerical range of each index information in the component priority evaluation index set to obtain an index numerical division threshold set; Map the index numerical division threshold set to the multi-dimensional evaluation polar coordinate system for interval level marking to obtain a priority evaluation polar coordinate system; Perform multi-dimensional quantitative evaluation on each component information in the available passenger car component set based on the priority evaluation polar coordinate system to generate a passenger car component coordinate interval atlas; Sort the available passenger car component set according to the area level of the passenger car component coordinate interval atlas to obtain the passenger car component emergency priority.

2. The redundant control system for emergency operation of subway electric multiple units according to claim 1, characterized in that, The passenger car operation fault parameter acquisition module includes: Collect and obtain the subway electric multiple unit control data set, where the subway electric multiple unit control data set includes the subway electric multiple unit operation data set and the control unit working data set; According to the set of key components of the multiple unit, perform associated feature extraction on the subway electric multiple unit operation data set in sequence to obtain the associated operation feature set of the multiple unit components, and at the same time extract the control unit working feature set of the control unit working data set; Perform dimensionality reduction and identification training on the associated operation feature set of the multiple unit components and the control unit working feature set to generate a multiple unit fault adaptive diagnosis network; Based on the multiple unit fault adaptive diagnosis network, perform fault diagnosis on the multiple unit control data stream, and output the multiple unit operation fault parameters.

3. The redundant control system for emergency operation of subway electric multiple units according to claim 2, characterized in that, The multiple unit operation fault parameter acquisition module includes: Perform fault correlation analysis on each feature information in the associated operation feature set of the multiple unit components and the control unit working feature set respectively to obtain the operation feature correlation coefficient set and the working feature correlation coefficient set; According to the preset correlation coefficient threshold, perform feature dimensionality reduction screening on the operation feature correlation coefficient set and the working feature correlation coefficient set to obtain the available associated operation feature set and the available working feature set; Use deep learning networks to perform fault identification training on the available associated operation feature set and the available working feature set respectively to obtain the multiple unit component fault diagnosis network and the control unit fault diagnosis network; Perform equal-weight fusion on the multiple unit component fault diagnosis network and the control unit fault diagnosis network to generate the multiple unit fault adaptive diagnosis network.

4. The redundant control system for emergency operation of subway electric multiple units according to claim 1, characterized in that, The control strategy parameter acquisition module includes: Perform operation feature extraction on the current multiple unit operation data stream to obtain the multiple unit operation state feature set, where the multiple unit operation state feature set includes component state, driving environment, and multiple unit operation state; Based on the multiple unit operation control strategy, perform data association mining to construct the multiple unit operation control solution space, and perform constraint partitioning on the multiple unit operation control solution space according to the multiple unit operation state feature set to obtain the multiple unit operation control parameter library; According to the emergency priority of the multiple unit components, determine the first emergency multiple unit control component; Based on the first emergency multiple unit control component, perform strategy equilibrium optimization on the multiple unit operation control parameter library to determine the multiple unit emergency operation control strategy parameters.

5. The emergency operation redundancy control system for subway electric multiple units according to claim 4, characterized in that, The control strategy parameter acquisition module includes: According to the multiple unit emergency operation requirements and the multiple unit emergency operation equilibrium target, construct an emergency operation fitness function, and use the emergency operation fitness function to evaluate the strategy parameters of the multiple unit operation control parameter library to obtain the emergency operation strategy fitness set; Map and associate the multiple unit operation control parameter library with the available multiple unit component set to obtain the component operation control strategy parameter set; Based on the emergency operation strategy fitness set, compare and optimize the component operation control strategy parameter set to determine the first emergency operation strategy parameters of the first emergency multiple unit control component; Based on the first emergency operation strategy parameters, perform Nash equilibrium optimization on the component operation control strategy parameter set to determine the multiple unit emergency operation control strategy parameters.

6. Emergency operation redundancy control method for subway electric multiple units, characterized in that, The method is applied to the system according to any one of claims 1-5, and the method includes: Performing key component analysis on the target subway electric multiple unit to obtain a set of key components of the multiple unit, and equipping the target subway electric multiple unit with a redundant control unit, where the redundant control unit includes a main control unit and a standby control unit; Communicatively connecting the main control unit and the standby control unit to the set of key components of the multiple unit respectively to construct a main control network and a standby control network of the multiple unit; Monitoring and acquiring the control data stream of the multiple unit through the main control network of the multiple unit, and performing fault diagnosis on the control data stream of the multiple unit to obtain the operation fault parameters of the multiple unit; When the operation fault parameters of the multiple unit reach a preset fault threshold, triggering a redundancy switching mechanism to switch and enable the standby control network of the multiple unit to obtain an emergency operation mode; Based on the standby control network of the multiple unit, parsing the operation strategy of the operation fault parameters according to the emergency operation mode to obtain the emergency operation control strategy parameters of the multiple unit, and performing emergency operation redundancy control on the target subway electric multiple unit based on the emergency operation control strategy parameters of the multiple unit.

Images