Dual-mode variable frequency power supply system and intelligent switching control method for rail transit vehicle

Abstract

The application discloses a rail transit vehicle dual-mode variable frequency power supply system and an intelligent switching control method, relates to the technical field of rail transit, and comprises the following steps: in the process that a frequency converter is switched from a normal ventilation mode to an emergency ventilation mode, a frequency scanning mechanism is executed, frequency response analysis is performed on electromagnetic resonance response characteristics of key electrical components in an electrical system of a rail transit vehicle, corresponding resonance frequencies and amplitude information thereof are extracted, and an electromagnetic resonance sensitive frequency band mapping table is constructed; and a dynamic characteristic vector of a variable frequency output signal is generated by collecting current dynamic operation parameters of a variable frequency power supply in real time. Through frequency scanning and dynamic characteristic fusion, the application realizes feedforward identification and intelligent evaluation of electromagnetic interference, drives a tunable damping loop to dynamically inhibit resonance interference based on risk scoring, improves electromagnetic compatibility and operation safety of a rail transit system in a complex environment, and realizes a change from passive protection to active intervention.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of rail transit technology, in particular to a dual-mode variable frequency power supply system for rail transit vehicles and an intelligent switching control method. BACKGROUND

[0002] Intelligent switching of a dual-mode variable frequency power supply for rail transit vehicles refers to a technology that enables the vehicle to intelligently judge and automatically switch between two different operating mode variable frequency power supplies in different operating scenarios (such as ground section and underground section, traction mode and parking standby mode) according to real-time power demand and changes in the power supply environment. The system usually integrates multiple sensors and control algorithms to monitor the power grid state, load type, energy consumption level, and operating conditions, dynamically select the appropriate variable frequency power supply mode (such as high-frequency high-power traction mode and low-frequency low-power maintenance mode), and thus achieve energy efficiency optimization, reduce switching delay, and improve vehicle operating stability and safety. This switching process has features such as non-human intervention, adaptive control, and seamless switching, and is an important part of an intelligent rail power supply system.

[0003] In the operation of rail transit vehicles, intelligent switching of a dual-mode variable frequency power supply can achieve dynamic response and mode switching in two states: normal ventilation and emergency ventilation. In the normal ventilation state, the system defaults to energy-saving mode, which meets the basic ventilation and passenger comfort requirements through low-frequency and low-voltage output, achieving minimal energy consumption. When a fire, smoke, or sudden accident triggers the emergency ventilation mechanism, the system automatically switches to high-performance emergency mode, and the variable frequency power supply increases in frequency and voltage to drive high-power exhaust and supply equipment to run at high speed, rapidly exhausting harmful gases and delivering fresh air to the evacuation area. This process relies on real-time monitoring of multiple parameters (such as temperature, gas concentration, and current fluctuation) and intelligent discrimination algorithms to achieve seamless switching between modes and stable power supply, ensuring that the ventilation system responds promptly and reliably in extreme situations.

[0004] The existing technology has the following disadvantages:

[0005] In the operation of rail transit vehicles, the existing dual-mode variable frequency power supply needs to complete the frequency and voltage increase operation in a very short time to drive high-power ventilation equipment when switching from normal ventilation mode to emergency ventilation mode. However, during this switching process, the sudden change in frequency of the variable frequency output signal may form electromagnetic resonance coupling under certain conditions with ventilation motors, onboard cables, and related auxiliary electrical equipment, thereby significantly amplifying the internal electromagnetic interference effect of the system. Such interference may not only cause transient distortion in the train control system, signal transmission link, and driving instruction system, but also may cause short-term failure of the communication module, especially in complex signal environments such as tunnels and underground sections, which may easily cause abnormal control of train operation safety and pose a significant safety hazard.

[0006] The above information disclosed in the background section is only for the purpose of enhancing the understanding of the background of the present disclosure, and therefore it can include information that does not constitute the prior art known to those of ordinary skill in the art. SUMMARY

[0007] The purpose of the present application is to provide a rail transit vehicle dual-mode variable frequency power supply system and an intelligent switching control method. By constructing a sensitive frequency band mapping through frequency scanning, the system realizes the feedforward identification of potential interference points. In combination with a multi-dimensional dynamic feature vector, a coupling risk score is constructed, so that the system has intelligent judgment ability for interference trends. Through an adjustable damping loop driven by the risk score, electromagnetic energy is absorbed on demand to realize immediate energy regulation response under high risk. This closed-loop mechanism effectively avoids the problems of communication interruption and control signal distortion caused by resonance in the traditional variable frequency system during high-frequency mutation. It is especially suitable for complex scenes such as tunnels and underground sections in the rail transit environment with extremely high requirements for safety and reliability, realizes the transition from "interference passive response" to "risk perception driven active intervention", and has good engineering implementability and popularization value to solve the problems in the background technology.

[0008] In order to achieve the above-mentioned purpose, the present application provides the following technical scheme: a rail transit vehicle dual-mode variable frequency power supply intelligent switching control method, comprising the following steps:

[0009] In the process of switching the frequency converter from the normal ventilation mode to the emergency ventilation mode, a frequency scanning mechanism is executed to analyze the frequency response of the electromagnetic resonance response characteristics of the key electrical components in the electrical system of the rail transit vehicle, extract the corresponding resonance frequency and its amplitude information, and construct an electromagnetic resonance sensitive frequency band mapping table;

[0010] Real-time acquisition of the current dynamic operation parameters of the variable frequency power supply generates a dynamic feature vector of the variable frequency output signal, which is used as the basic data for judging the electromagnetic coupling trend under the current variable frequency power supply system operation state;

[0011] The frequency value of the collected variable frequency output signal is matched with the frequency interval in the electromagnetic resonance sensitive frequency band mapping table, the frequency closeness factor is calculated, and the amplitude of each index in the dynamic feature vector is combined to construct a comprehensive coupling risk score through a multi-feature fusion algorithm. Based on the comprehensive coupling risk score, it is determined whether there is a potential electromagnetic resonance coupling risk in the current variable frequency output frequency;

[0012] When the comprehensive coupling risk score exceeds the preset risk score threshold, an equivalent damping loop with adjustable damping characteristics is automatically connected to absorb electromagnetic energy and relieve electromagnetic resonance. The damping strength of the damping loop is dynamically adjusted according to the comprehensive coupling risk score to improve the absorption capacity of the variable frequency power supply system to interference energy.

[0013] Preferably, the step of constructing the electromagnetic resonance sensitive frequency band mapping table by performing the frequency scanning mechanism comprises:

[0014] The preset frequency modulation control unit is used to drive the frequency converter output signal to cover the pre-defined frequency range in a step-by-step incremental manner, and the voltage and current response signals of the key electrical component nodes are collected under the excitation of each target frequency point. The response signals are subjected to frequency spectrum analysis by fast Fourier transform, and the response peak value of current resonance amplification at the frequency point is extracted.

[0015] Based on the resonance peak frequency statistical results of multiple measuring points, the frequency interval with resonance characteristics is identified by combining the amplitude average value and the change trend of multiple scanning data, and the electromagnetic resonance sensitive frequency band mapping table containing the resonance center frequency, the amplitude threshold and the frequency bandwidth is constructed.

[0016] Preferably, the step of collecting the output parameters of the frequency conversion power supply in real time and generating a dynamic feature vector comprises:

[0017] A multi-channel high-precision sensor array is arranged at the output end of the frequency conversion power supply and the key load node, and a plurality of electrical operation indexes of the frequency converter are collected in real time. The sampling rate of the sensor is not less than 10 kHz.

[0018] The multi-dimensional operation parameters collected in real time are subjected to standardization preprocessing, and a dynamic feature vector matrix containing a time stamp index is constructed by vectorization, which is used for electromagnetic coupling trend judgment of the current state.

[0019] Preferably, the step of matching the frequency value of the frequency conversion output with the frequency interval in the electromagnetic resonance sensitive frequency band mapping table and calculating the frequency proximity factor comprises:

[0020] The sensitive frequency intervals recorded in the electromagnetic resonance sensitive frequency band mapping table are extracted, and the center frequency thereof is taken as a representative frequency point to construct a frequency reference set;

[0021] Based on the current output frequency of the frequency converter, the Euclidean distance between the current output frequency and each sensitive frequency point is calculated, and the Euclidean distance is subjected to reverse normalization processing to obtain a basic frequency proximity, which reflects the proximity degree of the current frequency and the sensitive point;

[0022] A weight factor is introduced as a weighting parameter, the weight coefficient of each frequency band is set according to the resonance amplitude value size of each sensitive frequency band in the previous scanning, and the weight coefficient is multiplied by the basic frequency proximity to calculate a weighted frequency proximity factor, which is one of the important input features for subsequent coupling risk score calculation.

[0023] Preferably, the step of constructing a comprehensive coupling risk score by a multi-feature fusion algorithm comprises:

[0024] The input feature set is constructed, all input features are normalized and a vector sequence is formed according to the time stamp;

[0025] The historical data is trained by using a support vector regression model to form a mapping model between the features and the interference level. The model output is a continuous score value between 0 and 1, indicating the risk degree of electromagnetic resonance coupling at the current moment. The score result is used as the final coupling risk judgment basis to drive the subsequent response mechanism.

[0026] Preferably, the frequency proximity factor is fused with the amplitude of each feature in the dynamic feature vector, and a step of generating a comprehensive coupling risk score includes:

[0027] A multi-dimensional feature association matrix is established by obtaining the frequency proximity factor of the current variable frequency and each sensitive frequency band in the mapping table, and a dynamic coupling response coefficient is assigned to each dynamic feature index. The dynamic coupling response coefficient is used to represent the influence weight of the feature on the electromagnetic state under different proximity degrees;

[0028] After multiplying each feature value by its corresponding coupling response coefficient, weighted accumulation is performed, and the change gradient of the frequency proximity factor is extracted as an acceleration degree index reflecting the frequency proximity sensitive area, embedded in the dynamic scoring function as an adjustment factor, to establish a dynamic scoring function, output a comprehensive coupling risk score value; wherein the dynamic scoring function introduces a time decay factor, which increases the risk value weight for the case of continuously approaching the sensitive frequency band, to realize the "time + amplitude + proximity" three-factor coupling modeling.

[0029] Preferably, after obtaining the comprehensive coupling risk score at the current moment, it is compared with a preset risk score threshold to calculate a damping strength control factor for guiding the subsequent strength adjustment behavior of the damping circuit. The damping strength control factor is used to represent the degree to which the current risk score exceeds the threshold, and is amplified by a nonlinear enhancement function for high-risk response. The calculation formula is as follows: , wherein: is the comprehensive coupling risk score, indicating the risk intensity of the current frequency coupling interference; is the system preset risk score threshold, used to delimit the risk critical value; is the damping strength control factor, controlling the access strength of the equivalent damping; is the damping coefficient upper limit control parameter, determining the maximum accessible damping level; is a nonlinear amplification index, used to improve the sensitivity of the high-frequency coupling interference risk response.

[0030] Preferably, when the damping strength control factor is obtained, it is used to dynamically control the adjustment element in the equivalent damping circuit to adjust the effective impedance of the damping circuit. The adjustment formula is as follows: ,in: The effective impedance of the damping circuit at the current moment; This represents the minimum impedance value that the system can connect to, and thus its maximum energy absorption capacity. The maximum impedance value that the variable frequency power supply system is connected to by default represents the minimum energy absorption; t is a time variable that supports dynamic adjustment and response tracking.

[0031] The dual-mode variable frequency power supply system for rail transit vehicles includes a resonant characteristic modeling module, an operation status sensing module, a risk identification and assessment module, and an intelligent damping control module.

[0032] The resonance characteristic modeling module executes a frequency scanning mechanism during the process of the frequency converter switching from normal ventilation mode to emergency ventilation mode. It performs frequency response analysis on the electromagnetic resonance response characteristics of key electrical components in the electrical system of rail transit vehicles, extracts the corresponding resonance frequency and its amplitude information, and constructs an electromagnetic resonance sensitive frequency band mapping table.

[0033] The operation status sensing module collects the current dynamic operating parameters of the frequency converter in real time and generates a dynamic feature vector of the frequency converter output signal, which serves as the basic data for judging the electromagnetic coupling trend of the current frequency converter system operation status.

[0034] The risk identification and assessment module performs neighborhood matching operations on the frequency values ​​of the collected inverter output signal and the frequency ranges in the electromagnetic resonance sensitive frequency band mapping table to calculate the frequency proximity factor. Combined with the amplitude of each indicator in the dynamic feature vector, a comprehensive coupling risk score is constructed through a multi-feature fusion algorithm. Based on the comprehensive coupling risk score, it is determined whether there is a potential electromagnetic resonance coupling risk at the current inverter output frequency.

[0035] The intelligent damping control module automatically connects to an equivalent damping circuit with adjustable damping characteristics when the comprehensive coupling risk score exceeds the preset risk score threshold, in order to absorb electromagnetic energy and alleviate electromagnetic resonance phenomena; and dynamically adjusts the damping strength of the damping circuit according to the comprehensive coupling risk score to improve the frequency converter power supply system's ability to absorb interference energy.

[0036] The technical effects and advantages provided by the present invention in the above technical solution are as follows:

[0037] The application realizes the feedforward identification of potential interference points of the system by constructing a sensitive frequency band mapping through frequency scanning, and combines a multi-dimensional dynamic feature vector to construct a coupling risk score, so that the system has intelligent judgment ability on the interference trend. Through the adjustable damping loop driven by the risk score, the electromagnetic energy is absorbed on demand to realize the real-time energy regulation response under high risk. This closed-loop mechanism effectively avoids the problems of communication interruption and control signal distortion caused by resonance in the traditional variable frequency system during the high-frequency mutation process, and is especially suitable for the rail transit environment with extremely high requirements for safety and reliability in complex scenes such as tunnels and underground sections, realizes the change from 'interference passive response' to 'risk perception driven active intervention', and has good engineering implementability and popularization value. BRIEF DESCRIPTION OF DRAWINGS

[0038] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed in the embodiments will be briefly introduced. Obviously, the drawings in the following description are only some embodiments described in the present application, and other drawings can also be obtained by those skilled in the art based on these drawings.

[0039] Figure 1 The method flow chart of the intelligent switching control method of the dual-mode variable frequency power supply of the rail transit vehicle of the present application.

[0040] Figure 2 The module schematic diagram of the dual-mode variable frequency power supply system of the rail transit vehicle of the present application. DETAILED DESCRIPTION

[0041] Example implementations will now be described more fully with reference to the accompanying drawings. Example implementations may, however, be implemented in many different forms and should not be construed as limited to the examples set forth herein; rather, these example implementations are provided so that this disclosure will be thorough and complete, and will fully convey the scope of example implementations to those skilled in the art.

[0042] The present application provides an intelligent switching control method of a dual-mode variable frequency power supply of a rail transit vehicle as shown in Figure 1 The method flow chart of the intelligent switching control method of the dual-mode variable frequency power supply of the rail transit vehicle of the present application.

[0043] In the process of switching the frequency converter from the normal ventilation mode to the emergency ventilation mode, the frequency scanning mechanism is executed to analyze the frequency response of the electromagnetic resonance response characteristics of the key electrical components such as the ventilation motor, the on-board cable and the filter in the electrical system of the rail transit vehicle, extract the corresponding resonance frequency and amplitude information, and construct an electromagnetic resonance sensitive frequency band mapping table;

[0044] The step of executing the frequency scanning mechanism to construct the electromagnetic resonance sensitive frequency band mapping table includes:

[0045] Using a preset frequency modulation control unit, the inverter output signal is driven to scan and cover a predefined frequency range in a progressively increasing manner. Under the excitation of each target frequency point, the voltage and current response signals of key electrical component nodes such as ventilation motors, vehicle cables and filters are collected. The response signals are then subjected to spectrum analysis by Fast Fourier Transform (FFT) to extract the response peaks where current resonance amplification occurs at the frequency points.

[0046] Based on the statistical results of the resonant peak frequency at multiple measurement points, and combined with the amplitude average and variation trend of multiple scan data, frequency ranges with resonance characteristics are identified. An electromagnetic resonance sensitive frequency band mapping table containing the resonant center frequency, amplitude threshold, and frequency bandwidth is constructed. The mapping table has updatable properties and supports subsequent dynamic learning and correction.

[0047] This step aims to provide precise and forward-looking electromagnetic resonance risk identification for rail transit vehicles during frequency converter power mode switching, thereby achieving proactive prevention and dynamic adaptive control of electromagnetic interference at the system level. In rail transit electrical systems, equipment such as ventilation motors, on-board cables, and filters are prone to forming resonant paths within specific frequency ranges due to their structural characteristics and distributed parameters under high-frequency switching conditions, creating high-amplitude electromagnetic coupling channels at the system level. During the switch from normal ventilation mode to emergency ventilation mode, the output frequency often needs to increase rapidly within a very short time. This sudden frequency change easily falls into these "electromagnetically sensitive frequency bands," causing serious electromagnetic interference problems and affecting the operational stability and safety of critical subsystems such as train control, communication, and sensing.

[0048] By implementing a frequency scanning mechanism, the system can inject excitation signals into the target frequency range in a controlled manner under non-emergency conditions. It analyzes the voltage and current response amplitudes of each electrical component at specific frequencies, identifying the system's inherent or combined resonance characteristics. The resonant center frequency, amplification factor, and bandwidth obtained in this process constitute the electrical system's "electromagnetic sensitive frequency fingerprint," which is ultimately organized into an electromagnetic resonance sensitive frequency band mapping table, providing data support for subsequent risk identification and response decisions. By constructing this mapping table, the system possesses the ability to statically identify resonance risks, issuing warnings before the inverter's dynamic output frequency approaches the sensitive frequency band, thereby activating relevant suppression measures to prevent electromagnetic interference events and significantly improving the electromagnetic compatibility and operational safety of the rail transit vehicle frequency conversion switching system.

[0049] Real-time acquisition of dynamic operating parameters such as the current output frequency, amplitude, and current fluctuation of the variable frequency power supply, generating dynamic feature vectors of the variable frequency output signal, which serve as the basis for judging the electromagnetic coupling trend under the current system operating state;

[0050] The steps for real-time acquisition of variable frequency power supply output parameters and generation of dynamic feature vectors include:

[0051] By setting a multi-channel high-precision sensor array at the output end of the variable frequency power supply and the key load node, the output frequency, output voltage effective value, output current fluctuation rate, waveform distortion factor (THD), and power factor of the variable frequency drive are collected in real time, and the sensor sampling rate is not less than 10 kHz to ensure that the high dynamic characteristic changes are captured during the frequency rapid transition process.

[0052] The real-time collected multi-dimensional operating parameters are standardized and preprocessed, and a dynamic feature vector matrix containing a time stamp index is constructed by vectorization, which is used for the electromagnetic coupling trend judgment of the current state.

[0053] The purpose of this step is to provide high-precision, time-sequenced, and quantifiable operating state perception for the variable frequency power supply of the rail transit vehicle during the high dynamic switching process, so as to accurately judge the electromagnetic coupling trend and dynamically identify the interference risk. When the variable frequency power supply switches from the normal ventilation mode to the emergency ventilation mode, the output frequency, voltage, current, and other operating parameters will change dramatically. These dynamic changes not only relate to the load response, but also directly affect the electromagnetic stability of the entire vehicle electrical system. Especially during the frequency rapid transition process, if the frequency or current waveform of the output signal is close to the resonant frequency or coupling path of certain systems, electromagnetic resonance phenomenon may be excited, causing interference amplification effect. Therefore, the system must have high-resolution and real-time acquisition capability for these key operating parameters.

[0054] By deploying high-precision sensors at the output end of the variable frequency drive and the key nodes of the system, real-time collection of electrical characteristic parameters such as output frequency, amplitude, current fluctuation rate, power factor, and waveform distortion (such as THD) is achieved, and these raw signals are processed into structured dynamic feature vectors, which can not only accurately describe the current system operating state, but also be used as input for coupling risk scoring models, anomaly detection algorithms, and other modules. The existence of feature vectors allows the changes in electromagnetic state to be quantified and modeled in mathematical form, enabling the entire system to have intelligent decision-making capabilities such as state visualization, trend predictability, and response controllability. This step essentially maps complex electrical operating behavior into a data-driven intelligent perception portal and is the premise and foundation of the entire resonant interference recognition and control system.

[0055] The collected variable frequency output signal frequency value is matched with the frequency interval in the electromagnetic resonance sensitive frequency band mapping table, the frequency proximity factor is calculated, and the amplitude of each index in the dynamic feature vector is combined to construct a comprehensive coupling risk score through a multi-feature fusion algorithm; based on the comprehensive coupling risk score, it is determined whether there is a potential electromagnetic resonance coupling risk in the current variable frequency output frequency;

[0056] The step of performing neighborhood matching of the variable frequency output frequency value with the frequency intervals in the electromagnetic resonance sensitive frequency band mapping table and calculating the frequency closeness factor comprises:

[0057] The step of extracting all recorded sensitive frequency intervals from the electromagnetic resonance sensitive frequency band mapping table, taking the center frequency as the representative frequency point, constructing a frequency reference set, and then calculating the Euclidean distance between the current output frequency of the frequency converter and each sensitive frequency point, and performing reverse normalization processing on the distance value to obtain a basic frequency closeness reflecting the closeness of the current frequency to the sensitive point;

[0058] In order to consider the difference in historical interference intensity of different sensitive frequency bands, a weight factor is introduced as a weighting parameter, the weight coefficient of each frequency band is set according to the resonance amplitude value in the previous scanning, and the weighted frequency closeness factor is calculated by multiplying the basic frequency closeness, which is one of the important input features for subsequent coupling risk score calculation. This method improves the expression ability of the closeness factor to the interference risk contribution.

[0059] The step of constructing a comprehensive coupling risk score through a multi-feature fusion algorithm comprises:

[0060] An input feature set is constructed, including the weighted frequency closeness factor, the output current fluctuation rate, the output voltage distortion rate (THD), the instantaneous power change rate, the power factor and the frequency variation amplitude in the sampling period, and all input features are normalized and formed into a vector sequence according to the time stamp;

[0061] Support vector regression (SVR) or decision tree regression model is used to train the historical data to form a mapping model between features and interference levels. The model output is a continuous score value between 0 and 1, representing the risk degree of electromagnetic resonance coupling at the current time. The higher the score value, the greater the risk. The score result is used as the basis for the final coupling risk judgment to drive the subsequent response mechanism. The fusion model has the ability to process multi-dimensional nonlinear data and can comprehensively reflect the sensitivity of the current complex state of the system to the coupling interference, improving the accuracy and noise immunity of the overall judgment.

[0062] The step of fusing the frequency closeness factor with the amplitude of each feature in the dynamic feature vector and generating a comprehensive coupling risk score comprises:

[0063] A multi-dimensional feature association matrix is established by obtaining the frequency closeness factor of the current variable frequency and each sensitive frequency band in the mapping table, and a dynamic coupling response coefficient is assigned to each dynamic feature index (such as current fluctuation rate, voltage distortion, power fluctuation). The dynamic coupling response coefficient is used to represent the influence weight of the feature on the electromagnetic state under different closeness;

[0064] The characteristic values are multiplied by their corresponding coupling response coefficients, and then weighted and accumulated, and the change gradient of the frequency proximity factor is extracted as an acceleration degree index reflecting the frequency proximity sensitive area, embedded in a dynamic scoring function as a regulating factor, to establish a dynamic scoring function, and output a comprehensive coupling risk score value; wherein the dynamic scoring function introduces a time decay factor, which increases the risk value weight for the case of continuously approaching the sensitive frequency band, thereby realizing the "time + amplitude + proximity" three-factor coupling modeling, effectively improving the timeliness and forward-looking prediction ability of the risk score.

[0065] The core role of this step is to realize real-time intelligent evaluation and judgment of the electromagnetic resonance coupling risk that may be generated in the dynamic switching process of the dual-mode variable frequency power supply of the rail transit vehicle. The goal is to use a data-driven approach to identify and quantify potential risks of resonance interference under the current variable frequency output state in advance, thereby providing decision-making basis for subsequent damping control and protection response. Because when the frequency converter switches from normal ventilation mode to emergency ventilation mode, the output frequency needs to rise sharply or change in a short time, and it is easy to cross multiple electromagnetic sensitive frequency bands during this process, and these frequency bands may form unstable resonance paths in the vehicle electrical system. Without effective identification mechanism, transient electromagnetic interference is easy to occur, which interferes with the train control system and communication link, and even threatens the safety of train operation.

[0066] To this end, the current output frequency of the frequency converter is first matched with the pre-constructed "electromagnetic resonance sensitive frequency band mapping table". By defining the "neighborhood matching" method, it can be identified whether the frequency is near the sensitive frequency interval, and then a "frequency proximity factor" is generated through distance calculation algorithm, which is a quantitative index indicating how close the current frequency is to the sensitive frequency band. The higher the proximity, the greater the resonance possibility. However, proximity alone is not enough to completely determine the interference risk, so this step further introduces multi-dimensional dynamic features such as current fluctuation rate, voltage distortion rate (THD), frequency change slope, power factor, etc. to build a complete "dynamic feature vector". These parameters reflect the electrical stability and volatility of the system at this frequency. Then, by fusing the "frequency proximity factor" with the above feature vector, a "comprehensive coupling risk score" is generated using multi-feature fusion algorithms (such as weighted linear model, support vector machine, decision tree, etc.). This score is a quantitative judgment result of the coupling interference risk at the current time, and the higher the score, the more likely the system is to produce electromagnetic resonance amplification effect due to the frequency being in the sensitive area. The essence of this step is to establish a risk identification model that fuses frequency features and dynamic response features in the context of complex electrical system frequency and time domain interaction, so that the variable frequency system has the ability of early warning, high sensitivity and fast response, thereby enhancing the electromagnetic compatibility, operation reliability and rail transit operation safety of the variable frequency power supply system.

[0067] When the comprehensive coupling risk score exceeds the preset risk score threshold, an equivalent damping circuit with adjustable damping characteristics is automatically accessed to absorb electromagnetic energy and alleviate electromagnetic resonance; and the damping strength of the damping circuit is dynamically adjusted according to the comprehensive coupling risk score to improve the absorption capacity of the system to interference energy and reduce the electromagnetic coupling risk.

[0068] After obtaining the comprehensive coupling risk score at the current time, it is first compared with the preset risk score threshold to calculate the damping strength control factor for guiding the subsequent strength adjustment behavior of the damping circuit. The damping strength control factor is used to represent the degree to which the current risk score exceeds the threshold, and is amplified by a nonlinear enhancement function to respond to high risk. The calculation formula is as follows: , wherein: is the comprehensive coupling risk score, representing the risk strength of the current frequency coupling interference; is the system preset risk score threshold, used to determine the risk critical value; is the damping strength control factor, controlling the access strength of the equivalent damping; is the damping coefficient upper limit control parameter, determining the maximum accessible damping level; is a nonlinear amplification index, used to improve the sensitivity of the high-frequency coupling interference risk response;

[0069] The role of this step is to convert the fuzzy risk score into a quantifiable damping strength control factor, so that the system can make differentiated and moderate responses under different risk levels, effectively avoiding overcompensation or reaction lag.

[0070] When the damping strength control factor is obtained, it is used to dynamically control the adjusting element (such as a variable impedance device, a MOSFET array, or a digital controllable resistance network) in the equivalent damping circuit to adjust the effective impedance of the damping circuit. The adjustment formula is as follows: , wherein: is the effective impedance of the damping circuit at the current time; is the minimum impedance value that the system can access, representing the maximum energy absorption capacity; is the maximum impedance value that the system accesses by default, representing the minimum energy absorption; t is a time variable, supporting dynamic adjustment and response tracking;

[0071] The purpose of this step is to adjust the equivalent impedance of the energy absorption path in the damping circuit in a continuously adjustable manner according to the current interference risk level faced by the system. This achieves a balanced control strategy that rapidly absorbs electromagnetic energy under high-risk conditions and maintains optimal energy efficiency under low-risk conditions. This mechanism ensures that the system has adaptive, nonlinear, and refined response capabilities when dealing with different levels of interference, avoiding problems such as "response delay," "energy waste," or "over-suppression" that exist in traditional fixed damping mechanisms.

[0072] The core function of this step is to achieve dynamic interference suppression and energy absorption control of the frequency converter power supply system of rail transit vehicles under electromagnetic resonance risk conditions. It is a key execution link of the risk response mechanism in the entire intelligent switching control method. During the switching process of the frequency converter from normal ventilation mode to emergency ventilation mode, the frequency output frequency often rises rapidly in a short period of time, which is very likely to fall into the resonant sensitive frequency band corresponding to the electrical equipment of the system (such as ventilation motors, on-board cables, filters, etc.), thereby causing adverse phenomena such as severe electromagnetic energy resonance, harmonic amplification, and even interference propagation. Once electromagnetic resonance is formed, it may cause problems such as signal distortion, interruption, or false triggering in communication links and train control signal systems, directly endangering the safety of train operation. Therefore, after real-time assessment that the system has entered a high-risk state of electromagnetic coupling, the system must quickly take effective measures to interrupt the resonant channel, absorb excessive electromagnetic energy, and block the interference propagation path.

[0073] This step involves setting a risk score threshold. When the overall system coupling risk score exceeds this threshold, an adjustable damping circuit is automatically activated. Unlike traditional passive filtering devices, this damping circuit has adjustable damping characteristics, specifically reflected in its equivalent impedance dynamically changing according to the risk level. Initially, the system uses minimal damping to suppress mild coupling. When the risk score rises significantly, the system dynamically adjusts the parameters of key components in the circuit (such as MOSFET-controlled variable resistors and capacitor networks) to significantly improve absorption capacity and enhance buffering and dissipation capabilities against high-amplitude electromagnetic energy. Through this real-time sensing and execution linkage mechanism, the system can proactively take countermeasures before interference causes functional damage, achieving a "risk-driven control response."

[0074] In addition, another important role of this step is to achieve adaptive balance of energy efficiency and suppression capability. The traditional fixed damping structure has two problems: waste of damping power consumption in non-risk state, and insufficient damping in high-risk state. This step dynamically adjusts the impedance level through the function mapping between damping strength and risk score, so that the system maintains the best interference absorption efficiency under different working conditions, not only enhancing the anti-interference ability, but also improving the overall energy efficiency and operation economy of the system. In summary, this step converts the risk identification result into precise control behavior, and is the key node of the whole intelligent switching control method to realize closed-loop regulation and prevent problems from occurring.

[0075] Through the above-mentioned intelligent switching control method of the dual-mode variable frequency power supply of the rail transit vehicle, active identification, real-time warning and dynamic suppression of electromagnetic resonance coupling risk can be realized, and the electromagnetic compatibility and operation stability of the variable frequency power supply system under high dynamic working conditions are significantly improved. The scheme realizes the feedforward identification of potential interference points by constructing a sensitive frequency band mapping through frequency scanning; combined with the construction of a coupling risk score based on a multi-dimensional dynamic feature vector, the system has intelligent judgment ability for interference trend; further, through the adjustable damping loop driven based on the risk score, the electromagnetic energy is absorbed as needed, realizing the "immediate energy regulation response under high risk". This closed-loop mechanism effectively avoids the problems of communication interruption and control signal distortion caused by resonance in the traditional variable frequency system during high frequency mutation, and is especially suitable for complex scenes such as tunnels and underground sections in the rail transit environment with extremely high requirements for safety and reliability. Overall, this scheme realizes the transformation from "passive response to interference" to "risk perception driven active intervention", and has good engineering implementability and promotion value.

[0076] The present application provides a dual-mode variable frequency power supply system for rail transit vehicles as shown in Figure 2 The system comprises a resonance characteristic modeling module, an operating state sensing module, a risk identification and evaluation module, and an intelligent damping control module.

[0077] The resonance characteristic modeling module performs frequency scanning mechanism during the process of switching the frequency converter from normal ventilation mode to emergency ventilation mode, performs frequency response analysis on the electromagnetic resonance response characteristics of the key electrical components in the electrical system of the rail transit vehicle, extracts the resonance frequency and its amplitude information, and constructs an electromagnetic resonance sensitive frequency band mapping table.

[0078] The operating state sensing module collects the current dynamic operating parameters of the variable frequency power supply in real time to generate a dynamic feature vector of the variable frequency output signal, which is used as basic data for judging the electromagnetic coupling trend under the current variable frequency power supply system operating state.

[0079] The risk identification and evaluation module performs neighborhood matching operation on the collected variable frequency output signal frequency value and the frequency interval in the electromagnetic resonance sensitive frequency band mapping table, calculates the frequency closeness factor, and combines the amplitude of each index in the dynamic characteristic vector to construct a comprehensive coupling risk score through a multi-feature fusion algorithm; whether the current variable frequency output frequency has a potential electromagnetic resonance coupling risk is determined based on the comprehensive coupling risk score;

[0080] The intelligent damping control module automatically accesses an equivalent damping circuit with adjustable damping characteristics to absorb electromagnetic energy and relieve electromagnetic resonance when the comprehensive coupling risk score exceeds the preset risk score threshold; and the damping strength of the damping circuit is dynamically adjusted according to the comprehensive coupling risk score to improve the absorption capacity of the variable frequency power supply system to interference energy.

[0081] The track vehicle dual-mode variable frequency power supply intelligent switching control method provided by the embodiment of the application is realized through the track vehicle dual-mode variable frequency power supply system, and the specific method and process of the track vehicle dual-mode variable frequency power supply system are described in the embodiment of the track vehicle dual-mode variable frequency power supply intelligent switching control method, which will not be repeated here.

[0082] The above formulas are dimensionless numerical calculations, and the formulas are obtained by software simulation of a large amount of data to obtain a formula closest to the actual situation, and the preset parameters in the formula are set by a person skilled in the art according to the actual situation.

[0083] The above only describes some exemplary embodiments of the application by way of illustration, and it is needless to say that those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the application. Therefore, the above drawings and descriptions are illustrative in nature and should not be understood as limiting the scope of protection of the claims of the application.

[0084] The above is only a specific embodiment of the present application, but the protection scope of the present application is not limited thereto, and any person skilled in the art can easily think of changes or replacements within the technical range disclosed in the present application, which should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A method for intelligent switching control of dual-mode variable frequency power supply for rail transit vehicles, characterized in that, The method comprises the following steps: During the process of switching the frequency converter from the normal ventilation mode to the emergency ventilation mode, a frequency scanning mechanism is executed to analyze the electromagnetic resonance response characteristics of key electrical components in the electrical system of the rail transit vehicle, extract the corresponding resonance frequency and amplitude information, and construct an electromagnetic resonance sensitive frequency band mapping table; Real-time acquisition of current dynamic operation parameters of the variable frequency power supply generates a dynamic characteristic vector of the variable frequency output signal, which is used as basic data for judging the electromagnetic coupling trend under the current variable frequency power supply system operation state; The collected variable frequency output signal frequency value is matched with the frequency interval in the electromagnetic resonance sensitive frequency band mapping table, the frequency closeness factor is calculated, and the amplitude of each index in the dynamic characteristic vector is combined to construct a comprehensive coupling risk score through a multi-feature fusion algorithm; Based on the comprehensive coupling risk score, it is determined whether there is a potential electromagnetic resonance coupling risk in the current variable frequency output frequency; When the comprehensive coupling risk score exceeds the preset risk score threshold, an equivalent damping circuit with adjustable damping characteristics is automatically connected to absorb electromagnetic energy and relieve electromagnetic resonance; The damping strength of the damping circuit is dynamically adjusted according to the comprehensive coupling risk score to improve the absorption capacity of the variable frequency power supply system to interference energy; The steps of matching the variable frequency output frequency value with the frequency interval in the electromagnetic resonance sensitive frequency band mapping table and calculating the frequency closeness factor include: Extract all recorded sensitive frequency intervals from the electromagnetic resonance sensitive frequency band mapping table, take the center frequency as the representative frequency point, and construct a frequency reference set; Based on the current output frequency of the frequency converter, the Euclidean distance between each sensitive frequency point is calculated, and the Euclidean distance is inversely normalized to obtain the basic frequency closeness, which reflects the closeness of the current frequency to the sensitive point; A weight factor is introduced as a weighting parameter, the weight coefficient of each frequency band is set according to the resonance amplitude in the previous scanning, and the basic frequency closeness is multiplied by the weight coefficient to calculate the weighted frequency closeness factor, which is one of the important input features for subsequent coupling risk score calculation; The steps of fusing the frequency closeness factor with the amplitude of each feature in the dynamic characteristic vector and generating the comprehensive coupling risk score include: A multi-dimensional feature association matrix is established by obtaining the frequency closeness factor of the current variable frequency and each sensitive frequency band in the mapping table, and a dynamic coupling response coefficient is assigned to each dynamic characteristic index, which is used to represent the influence weight of the feature on the electromagnetic state under different closeness; After multiplying each feature value by its corresponding coupling response coefficient, the weighted cumulative value is obtained, and the change gradient of the frequency closeness factor is extracted as an acceleration degree index reflecting the frequency closeness sensitive area, which is embedded into the dynamic scoring function as an adjustment factor to establish a dynamic scoring function and output the comprehensive coupling risk score value; The dynamic scoring function introduces a time decay factor to increase the risk value weight for the case of continuously approaching the sensitive frequency band; After obtaining the comprehensive coupling risk score of the current moment, it is compared with the preset risk score threshold to calculate the damping strength regulation factor for guiding the subsequent strength adjustment behavior of the damping loop. The damping strength regulation factor is used to represent the degree of exceeding the threshold of the current risk score, and the high risk response is amplified through a nonlinear enhancement function. The calculation formula is as follows: Wherein: is the comprehensive coupling risk score, indicating the risk strength of the current frequency coupling interference; is the system preset risk score threshold, used to delimit the risk critical value; is the damping strength regulation factor, controlling the access strength of the equivalent damping; is the damping coefficient upper limit control parameter, determining the maximum accessible damping level; is a nonlinear amplification index, used to improve the sensitivity of the high frequency coupling interference risk response.

2. The intelligent switching control method for dual-mode variable frequency power supply of rail transit vehicle according to claim 1, characterized in that, The steps of executing the frequency scanning mechanism to construct the electromagnetic resonance sensitive frequency band mapping table include: The preset frequency modulation control unit is used to drive the frequency converter output signal to scan and cover the predefined frequency range in a step-by-step incremental manner, and the voltage and current response signals of the key electrical component nodes are collected under the excitation of each target frequency point. The response signals are subjected to frequency spectrum analysis through fast Fourier transform, and the response peak value of current resonance amplification at the frequency point is extracted. Based on the statistical results of the resonance peak frequency of multiple measuring points, the amplitude average value and the change trend of multiple scanning data are combined to identify the frequency interval with resonance characteristics, and a mapping table of electromagnetic resonance sensitive frequency band is constructed, which includes the resonance center frequency, the amplitude threshold and the frequency bandwidth.

3. The intelligent switching control method for dual-mode variable frequency power supply of rail transit vehicle according to claim 1, characterized in that, The step of collecting the output parameters of the variable frequency power supply in real time and generating a dynamic feature vector includes: A multi-channel high-precision sensor array is arranged at the output end of the variable frequency power supply and the key load nodes, and a plurality of electrical operation indexes of the frequency converter are collected in real time. The sampling rate of the sensor is not less than 10 kHz; The multi-dimensional operation parameters collected in real time are standardized and preprocessed, and a dynamic feature vector matrix containing a time stamp index is constructed through vectorization, which is used for the judgment of the electromagnetic coupling trend of the current state.

4. The intelligent switching control method for dual-mode variable frequency power supply of rail transit vehicle according to claim 1, characterized in that, The step of constructing a comprehensive coupling risk score through a multi-feature fusion algorithm includes: An input feature set is constructed, all input features are normalized and formed into a vector sequence according to the time stamp; A support vector regression model is used to train the historical data to form a mapping model between the features and the interference level. The model output is a continuous score value between 0 and 1, representing the risk degree of electromagnetic resonance coupling at the current time. The score result is used as the basis for the final coupling risk judgment to drive the subsequent response mechanism.

5. The intelligent switching control method for dual-mode variable frequency power supply of rail transit vehicle according to claim 1, characterized in that, When getting the damping strength regulation factor Then it is used to dynamically control the adjusting element in the equivalent damping circuit, adjust the effective impedance of the damping circuit, and the adjustment formula is as follows: Wherein: The effective impedance of the damping circuit at the current time; The minimum impedance value that the variable frequency power supply system can access, representing the maximum energy absorption capacity; The maximum impedance value that the variable frequency power supply system defaults to access, representing the minimum energy absorption; t time variable, supporting dynamic adjustment and response tracking.

6. A dual-mode variable frequency power supply system for a rail transit vehicle, for implementing the intelligent switching control method of a dual-mode variable frequency power supply for a rail transit vehicle according to any one of claims 1-5, characterized in that, It includes a resonance characteristic modeling module, an operation state sensing module, a risk identification and evaluation module, and an intelligent damping control module. The resonance characteristic modeling module performs a frequency scanning mechanism during the switching of the frequency converter from the normal ventilation mode to the emergency ventilation mode, performs frequency response analysis on the electromagnetic resonance response characteristics of the key electrical components in the electrical system of the rail transit vehicle, extracts the resonance frequency and its amplitude information, and constructs an electromagnetic resonance sensitive frequency band mapping table. The operation state sensing module collects the current dynamic operation parameters of the variable frequency power supply to generate a dynamic feature vector of the variable frequency output signal, which is used as the basis data for judging the electromagnetic coupling trend under the current variable frequency power supply system operation state. The risk identification and evaluation module performs neighborhood matching operation on the collected variable frequency output signal frequency value and the frequency interval in the electromagnetic resonance sensitive frequency band mapping table, calculates the frequency proximity factor, and combines the amplitudes of the indicators in the dynamic feature vector to construct a comprehensive coupling risk score through a multi-feature fusion algorithm. Based on the comprehensive coupling risk score, it is determined whether there is a potential electromagnetic resonance coupling risk in the current variable frequency output frequency. When the comprehensive coupling risk score exceeds the preset risk score threshold, the intelligent damping control module automatically accesses an equivalent damping circuit with adjustable damping characteristics to absorb electromagnetic energy and relieve electromagnetic resonance. And according to the comprehensive coupling risk score, the damping strength of the damping circuit is dynamically adjusted to improve the absorption capacity of the variable frequency power supply system to the interference energy.

Images