EPS emergency power supply control method and system based on rail transit

Abstract

The invention discloses an EPS emergency power supply control method and system based on rail transit, and relates to the technical field of EPS emergency power supply control, and the method specifically comprises the following steps: obtaining the load fluctuation information of a power grid when each charging device in a vehicle depot is started in real time, respectively generating an instantaneous power change coefficient and a starting load stability coefficient of each charging device; comprehensively analyzing the generated instantaneous power change coefficient and starting load stability coefficient of each charging device, and respectively generating a comprehensive instantaneous power change coefficient standard deviation and a comprehensive starting load stability coefficient standard deviation; and constructing a load fluctuation evaluation model for the generated standard deviation of the comprehensive instantaneous power change coefficient and the standard deviation of the comprehensive starting load stability coefficient, generating a load fluctuation evaluation coefficient, and performing analysis after generation. According to the method, the problem of misjudgment caused by load fluctuation when the metro depot charging equipment is started is solved, and accurate evaluation and dynamic adjustment of the power supply stability are realized.

Description

Technical Field

[0001] The present invention relates to the technical field of EPS emergency power supply control, and in particular to an EPS emergency power supply control method and system based on rail transit. Background Art

[0002] EPS Emergency Power Supply is a system that provides backup power for important equipment when the main power supply is interrupted, ensuring the safe operation of the rail transit system in emergency situations such as power outages. The purpose of controlling the EPS emergency power supply is to achieve seamless switching between the main power supply and the backup power supply, ensure power supply continuity, and reasonably manage load priorities, extend battery life, and avoid excessive discharge and unnecessary power consumption. The special feature of EPS emergency power supply control technology based on rail transit is that it must respond to power supply failures in a very short time to ensure the stable operation of key equipment such as signal systems and communication systems, and have intelligent load distribution and prediction capabilities to cope with power demand at different times and scenarios, thereby ensuring the safety of passengers and trains.

[0003] The existing rail transit-based EPS emergency power supply control technology monitors the voltage and frequency of the main power supply in real time. When an abnormality in the main power supply (such as power outage or voltage fluctuation) is detected, the control system immediately triggers the automatic switching device to switch the power supply from the main power supply to the EPS system, and starts the inverter to convert the DC power provided by the battery into AC power to supply rail transit equipment. In this process, the system prioritizes the loads according to the importance of different equipment to ensure stable power supply to key equipment such as signal systems and communication systems, while secondary equipment may take power restriction or power outage measures. In addition, the control system monitors the voltage, temperature and remaining capacity of the battery in real time, dynamically adjusts the inverter output power to prevent excessive discharge of the battery, and delays switching back to the main power supply after the main power supply is restored. At the same time, the battery is charged to ensure that it is in a good standby state at the next fault, thereby achieving rapid response and efficient management.

[0004] The prior art has the following deficiencies:

[0005] During the charging process after the subway depot is shut down at night, when multiple trains start charging equipment at the same time in a short period of time, there will be a significant load fluctuation phenomenon. Due to the instantaneous increase in charging load, the voltage of the depot power grid will fluctuate for a short time. This situation is easily misjudged as a power grid failure by the existing EPS emergency power control system, thereby triggering unnecessary EPS system switching. The existing technology lacks a real-time evaluation mechanism for charging load changes and cannot accurately distinguish normal charging load fluctuations from real power grid failures. This lack of predictive ability will cause the system to frequently switch to EPS power supply when the load suddenly increases. As a result, this frequent switching will not only affect the power supply stability of other facilities in the depot, but also accelerate the wear of EPS inverters and switching devices, increase the maintenance cost of the equipment, and in addition, if the charging plan is disrupted, it may cause delays in the charging of the train, which in turn affects the morning operation preparation time of the train, thereby having a negative impact on the normal operation of the subway system.

[0006] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not constitute the prior art that is already known to one of ordinary skill in the art. Summary of the invention

[0007] The purpose of the present invention is to provide a method and system for controlling an EPS emergency power supply based on rail transit to solve the problems in the above-mentioned background technology.

[0008] In order to achieve the above object, the present invention provides the following technical solution: a method for controlling EPS emergency power supply based on rail transit, specifically comprising the following steps:

[0009] When charging is carried out after the subway depot stops operating at night, the startup time and instantaneous charging current of each charging device are monitored in real time through the sensor network. Based on the load fluctuation when the charging device is started, a real-time monitoring framework for charging load fluctuation assessment and EPS response is established;

[0010] Obtain in real time the load fluctuation information on the power grid caused by the startup of each charging device in the depot, and analyze it after acquisition to generate the instantaneous power variation coefficient and startup load stability coefficient of each charging device;

[0011] Comprehensively analyze the generated instantaneous power variation coefficient and starting load stability coefficient of each charging device, and generate the standard deviation of the comprehensive instantaneous power variation coefficient and the standard deviation of the comprehensive starting load stability coefficient respectively;

[0012] A load fluctuation assessment model is constructed for the generated standard deviation of the comprehensive instantaneous power variation coefficient and the standard deviation of the comprehensive starting load stability coefficient, and a load fluctuation assessment coefficient is generated. After the generation, an analysis is performed to assess whether the load fluctuation on the power grid caused by the startup of all charging equipment in the depot is a normal charging load change, and the load fluctuation is divided into normal charging load fluctuation and abnormal load fluctuation according to the assessment results;

[0013] Based on the division results, the power supply switching strategy of the EPS system is dynamically adjusted according to different load fluctuation conditions;

[0014] Continuously monitor the load fluctuation of charging equipment to the power grid and the response effect of EPS system, adjust the load assessment model and switching strategy in real time, and optimize EPS control during the startup of charging equipment through feedback data.

[0015] Preferably, the load fluctuation information on the power grid caused by the startup of each charging device in the vehicle depot is obtained in real time, and analyzed after acquisition to generate the instantaneous power variation coefficient and the startup load stability coefficient of each charging device respectively, which specifically includes the following steps:

[0016] Real-time acquisition of load fluctuation information on the power grid caused by the startup of each charging device in the depot, and pre-processing after acquisition;

[0017] Extracting the instantaneous power change information and the starting load fluctuation information from the pre-processed load fluctuation information on the power grid caused by the starting of each charging device in the vehicle depot;

[0018] The extracted instantaneous power change information and starting load fluctuation information are analyzed to generate the instantaneous power change coefficient and starting load stability coefficient of each charging device respectively.

[0019] Preferably, the logic for obtaining the instantaneous power variation coefficient of each charging device is as follows:

[0020] Extract the instantaneous power change information from the load fluctuation information caused by the startup of each charging device in the depot, including the voltage and current values ​​at different times within a period of time after the startup of each charging device in the depot, and calibrate them as and represents the voltage value at time m within a period of time after the i-th charging device in the vehicle depot is started, DL i m represents the current value at time m within a period of time after the i-th charging device in the vehicle depot is started, i=1, 2, 3, ..., k, m=1, 2, 3, ..., g, k and g are both positive integers;

[0021] Calculate the actual power of each charging device in the depot at different times after it is started According to the formula:

[0022] Calculate the instantaneous power change coefficient of each charging device. The specific calculation formula is as follows:

[0023]

[0024] In the formula, IPVC i is the instantaneous power variation coefficient of the i-th charging device.

[0025] Preferably, the logic for obtaining the startup load stability coefficient of each charging device is as follows:

[0026] Extract the starting load fluctuation information from the pre-processed load fluctuation information caused by the startup of each charging device in the depot to the power grid, including the time required for each charging device in the depot to reach the power stable state from startup, and the current fluctuation amplitude, voltage fluctuation amplitude and current fluctuation frequency of each charging device during this time, and mark them as TS respectively. i IF i 、VF i and If i , TS i It represents the time required for the i-th charging device in the vehicle depot to reach the power stable state from startup, IF i Indicates the current fluctuation amplitude of the i-th charging device during this time, VF i Indicates the voltage fluctuation amplitude of the i-th charging device during this time, If i represents the current fluctuation frequency of the i-th charging device during the time, i=1, 2, 3, ..., k, k is a positive integer;

[0027] Calculate the starting load stability coefficient of each charging device. The specific calculation formula is as follows:

[0028]

[0029] Where, LSC i is the starting load stability factor of the i-th charging device.

[0030] Preferably, the acquisition logic of the comprehensive instantaneous power variation coefficient standard deviation and the comprehensive starting load stability coefficient standard deviation is as follows:

[0031] Extract the instantaneous power variation coefficient IPVC of each charging device i and starting load stability factor LSC i , i=1, 2, 3, …, k, k is a positive integer;

[0032] Calculate the instantaneous power variation coefficient IPVC of all charging devices i The average value of the starting load stability factor LSC of all charging equipment i The specific calculation formula is as follows:

[0033]

[0034] Where IPVC_ is the instantaneous power variation coefficient IPVC of all charging devices i The average value of LSC_ is the starting load stability factor LSC of all charging equipment. i The average value of

[0035] Calculate the standard deviation of the comprehensive instantaneous power variation coefficient and the standard deviation of the comprehensive starting load stability coefficient. The specific calculation formula is as follows:

[0036]

[0037] In the formula, σ IPVC is the standard deviation of the comprehensive instantaneous power variation coefficient, σ LSC is the standard deviation of the comprehensive starting load stability coefficient.

[0038] Preferably, a load fluctuation assessment model is constructed for the generated comprehensive instantaneous power variation coefficient standard deviation and the comprehensive starting load stability coefficient standard deviation to generate a load fluctuation assessment coefficient, which specifically includes the following steps:

[0039] Collect several comprehensive instantaneous power variation coefficient standard deviations, comprehensive starting load stability coefficient standard deviations and corresponding load fluctuation assessment coefficients generated in the past period of time, and calibrate them as and LFEC x , x represents the number of the integrated instantaneous power variation coefficient standard deviation, the integrated starting load stability coefficient standard deviation and the corresponding load fluctuation assessment coefficient generated in the past period of time, x = 3, 4, 5, ..., d, d is a positive integer, and the data collected in the past period of time is formed into a historical data set;

[0040] The multivariate regression model is selected as the load fluctuation assessment model, and is trained through historical data sets to determine the value of the regression coefficient according to the formula:

[0041]

[0042] Where β0, β1 and β2 are regression coefficients;

[0043] By minimizing the error between the predicted value and the actual value, the regression coefficient is optimized, and finally the values ​​of the regression coefficients β0, β1 and β2 are determined;

[0044] Using the final determined regression coefficient, the constructed load fluctuation assessment model is input with the real-time generated comprehensive instantaneous power variation coefficient standard deviation σ IPVC and the standard deviation of the comprehensive starting load stability factor σ LSC , generate load fluctuation assessment coefficient LFEC in real time.

[0045] Preferably, the generated load fluctuation evaluation coefficient LFEC is compared with a preset load fluctuation evaluation coefficient threshold LFEC. yuzhi Compare and evaluate whether the load fluctuation on the power grid caused by the startup of all charging equipment in the depot is a normal charging load change according to the comparison results, and divide it into normal charging load fluctuation and abnormal load fluctuation according to the evaluation results. The specific comparison analysis and classification are as follows:

[0046] If LFEC≤LFEC yuzhi , the load fluctuation on the power grid caused by the startup of all charging equipment in the depot belongs to normal charging load changes, so this situation is classified as normal charging load fluctuation;

[0047] If LFEC>LFEC yuzhi If the load fluctuation on the power grid caused by the startup of all charging equipment in the depot does not belong to the normal charging load change, then the situation is classified as abnormal load fluctuation.

[0048] Preferably, based on the division results, the power supply switching strategy of the EPS system is dynamically adjusted according to different load fluctuation conditions, specifically:

[0049] For load fluctuations that are normal charging load fluctuations, the power supply switching strategy for dynamically adjusting the EPS system is: maintain the current grid power supply, do not start EPS switching, and continue monitoring;

[0050] For abnormal load fluctuations, the power supply switching strategy for the EPS system is to dynamically adjust the power supply of the EPS to give priority to key equipment, adjust the power of the charging equipment and suspend some charging at the same time, and then gradually resume the charging plan after the load returns to normal.

[0051] Preferably, an EPS emergency power supply control system based on rail transit includes a real-time monitoring and data acquisition module, a load fluctuation data analysis module, a comprehensive fluctuation analysis module, a load fluctuation assessment modeling module, a dynamic power supply adjustment module, and an adaptive optimization and feedback module;

[0052] The real-time monitoring and data acquisition module monitors the startup time and instantaneous charging current of each charging device in real time through the sensor network when charging is carried out after the subway depot stops operating at night. It also establishes a real-time monitoring framework for charging load fluctuation assessment and EPS response based on the load fluctuation when the charging device is started.

[0053] The load fluctuation data analysis module obtains the load fluctuation information on the power grid caused by the startup of each charging device in the depot in real time, analyzes it after acquisition, and generates the instantaneous power variation coefficient and startup load stability coefficient of each charging device respectively;

[0054] The comprehensive fluctuation analysis module comprehensively analyzes the instantaneous power variation coefficient and the starting load stability coefficient of each charging device, and generates the standard deviation of the comprehensive instantaneous power variation coefficient and the standard deviation of the comprehensive starting load stability coefficient respectively;

[0055] The load fluctuation assessment modeling module constructs a load fluctuation assessment model for the generated standard deviation of the comprehensive instantaneous power variation coefficient and the standard deviation of the comprehensive starting load stability coefficient, generates a load fluctuation assessment coefficient, and analyzes it after generation to assess whether the load fluctuation caused by the startup of all charging equipment in the depot to the power grid is a normal charging load change, and divides it into normal charging load fluctuation and abnormal load fluctuation according to the assessment results;

[0056] Dynamic power supply adjustment module, based on the division results, dynamically adjusts the power supply switching strategy of the EPS system according to different load fluctuation conditions;

[0057] The adaptive optimization and feedback module continuously monitors the load fluctuation of the charging equipment to the power grid and the response effect of the EPS system, adjusts the load assessment model and switching strategy in real time, and optimizes the EPS control during the startup of the charging equipment through feedback data.

[0058] In the above technical solution, the technical effects and advantages provided by the present invention are:

[0059] 1. The present invention greatly improves the ability to perceive load fluctuations when multi-point charging equipment in subway depots is started by introducing a mechanism of real-time monitoring and data analysis. Specifically, the start-up time and instantaneous charging current of each charging device are collected in real time through a sensor network, and the instantaneous power change coefficient and the starting load stability coefficient are generated. These detailed data provide an accurate basis for evaluating load fluctuations. Compared with traditional methods, the present invention can quickly reflect the changes in grid load when the equipment is started, thereby avoiding misjudgments caused by information lags. This efficient data analysis method makes load fluctuation assessment more accurate and provides a reliable basis for subsequent decision-making.

[0060] 2. The comprehensive analysis and multiple regression model used in the present invention constructs a precise load fluctuation assessment model that can effectively distinguish between normal and abnormal load fluctuations. This method calculates the standard deviation of the instantaneous power change coefficient and the starting load stability coefficient of each device, combines historical data for regression analysis, generates a load fluctuation assessment coefficient, and compares it with the preset threshold. Compared with the prior art, this evaluation method not only improves the accuracy of load fluctuation classification, but also enables the system to automatically adapt to different charging scenarios and dynamically adjust the parameters of the evaluation model. As a result, the system can respond more flexibly to the load changes caused by the simultaneous startup of multiple charging devices, thereby avoiding unnecessary EPS system switching and improving the reliability and stability of power supply.

[0061] 3. The present invention significantly improves the system's ability to cope with complex load fluctuations by dynamically adjusting the EPS system's power supply strategy and continuous adaptive optimization mechanism. When the evaluation results show that the load fluctuation is normal, the system maintains the existing power supply state, reduces the number of EPS switching times, and reduces equipment wear and energy consumption; when the evaluation results show abnormal fluctuations, the system immediately starts EPS power supply to ensure the stable operation of key equipment. This differentiated response method ensures that the power supply system can make flexible and accurate adjustments under variable charging load conditions, ensuring the safety of the charging process in the subway depot and the continuity of the charging plan. At the same time, through continuous monitoring and adjustment of the system response effect, the system can continuously optimize itself during long-term operation, making the overall control strategy more efficient and intelligent, and achieving reduced equipment operating costs and extended service life. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0063] Figure 1 The present invention is a flow chart of a method and system for controlling an EPS emergency power supply based on rail transit.

[0064] Figure 2 The present invention is a module schematic diagram of an EPS emergency power supply control method and system based on rail transit. DETAILED DESCRIPTION

[0065] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as limited to the examples set forth herein; rather, these example embodiments are provided so that the description of the present disclosure will be more comprehensive and complete, and the concept of the example embodiments will be fully conveyed to those skilled in the art.

[0066] The present invention provides Figure 1 The EPS emergency power supply control method based on rail transit shown in the figure specifically includes the following steps:

[0067] When charging is carried out after the subway depot stops operating at night, the startup time and instantaneous charging current of each charging device are monitored in real time through the sensor network. Based on the load fluctuation when the charging device is started, a real-time monitoring framework for charging load fluctuation assessment and EPS response is established;

[0068] High-precision current sensors and time recording modules are installed on each charging device in the subway depot. The current sensors monitor the current changes of the charging equipment during startup and operation in real time, collect instantaneous current data at a frequency of one second or higher, and transmit it to the data processing center. The time recording module synchronously records the time when the charging device starts to supply power, and accurately obtains the startup time of the device. All sensors are connected to the central data processing system through wireless or wired networks (such as LoRaWAN, Zigbee or industrial Ethernet) to form a unified sensor network, which can collect the status data of the charging equipment in real time with low latency and high precision, and provide the necessary data basis for subsequent analysis.

[0069] In the data processing system, the instantaneous current data transmitted by each charging device is analyzed through real-time data analysis software to establish a monitoring framework for charging load fluctuation evaluation and EPS response. The analysis software uses the set threshold and data processing algorithms (such as real-time fluctuation detection algorithm and data smoothing algorithm) to automatically identify the load fluctuation characteristics of the charging equipment at startup, and analyzes the frequency characteristics of the current change through fast Fourier transform (FFT) to identify the amplitude and duration of the fluctuation. Based on these data, the system will calculate the charging load coefficient and load fluctuation index, and compare the results with the preset EPS response strategy to determine in real time whether the load fluctuation is normal. When the load fluctuation exceeds the preset range, the system issues an early warning and adjusts the EPS response strategy to achieve accurate monitoring and response to charging load fluctuations.

[0070] This method is designed to solve the problem of the EPS system misjudging a power grid failure due to a sudden increase in load when multiple trains in a subway depot start charging equipment at the same time. The sensor network monitors the start-up time and instantaneous current of each charging device in real time to ensure that the load changes during the charging process can be accurately captured, which provides an accurate data basis for real-time analysis and evaluation of charging load fluctuations. Subsequently, the data analysis system can quickly evaluate the nature of the load fluctuations and establish a correlation between the charging load and the EPS response, thereby avoiding unnecessary EPS switching due to normal charging load fluctuations. This precise load management method not only improves the stability of the power supply, but also reduces the wear and maintenance requirements of the EPS system, thereby optimizing the overall operating efficiency of the subway system.

[0071] Obtain in real time the load fluctuation information on the power grid caused by the startup of each charging device in the depot, and analyze it after acquisition to generate the instantaneous power variation coefficient and startup load stability coefficient of each charging device;

[0072] In this embodiment, the load fluctuation information on the power grid caused by the startup of each charging device in the vehicle depot is obtained in real time, and analyzed after acquisition to generate the instantaneous power variation coefficient and the startup load stability coefficient of each charging device respectively, which specifically includes the following steps:

[0073] Real-time acquisition of load fluctuation information on the power grid caused by the startup of each charging device in the depot, and pre-processing after acquisition;

[0074] Real-time acquisition of load fluctuation information on the power grid caused by the startup of each charging device in the vehicle depot can be achieved by installing a high-precision sensor network on each charging device. Each charging device is equipped with a current sensor and a voltage sensor, which are used to monitor the instantaneous current and voltage changes of the device in real time. These sensors transmit data to the central data processing system at a high frequency through wireless transmission protocols (such as LoRaWAN or Zigbee) or wired industrial Ethernet. The data processing system timestamps the data transmitted by the sensor through a time synchronization mechanism to ensure the timing and consistency of data acquisition, thereby fully recording the load changes of each device during the startup process. This data acquisition process continues during the startup of the charging device to ensure that the system can obtain the dynamic impact of the charging process on the power grid in real time.

[0075] Preprocessing is performed to ensure the accuracy, completeness and efficiency of data analysis, and to reduce the impact of noise and outliers in the data on the subsequent analysis process. Preprocessing mainly includes three aspects: data denoising, outlier detection and elimination, and data format standardization. First, the system uses filtering algorithms (such as Kalman filtering) to denoise the real-time transmitted data and smooth out high-frequency fluctuations caused by sensor noise or environmental interference; second, outlier detection is performed by setting a threshold range to identify and eliminate obvious anomalies in the data, such as current or voltage spikes caused by transient interference; finally, the data format and time stamp collected by each sensor are unified to ensure that data from different devices can be aligned and compared, thereby providing standardized data input for subsequent load fluctuation information extraction and coefficient calculation. These preprocessing steps can improve the analysis accuracy and stability of the data and lay the foundation for subsequent load fluctuation assessment.

[0076] Extracting the instantaneous power change information and the starting load fluctuation information from the pre-processed load fluctuation information on the power grid caused by the starting of each charging device in the vehicle depot;

[0077] Extracting the instantaneous power change information and starting load fluctuation information caused by the preprocessed charging equipment in the vehicle depot when starting the power grid can be achieved by combining feature extraction algorithms with data screening rules. First, the instantaneous power change information of each charging device during the startup period is obtained by calculating the instantaneous power at each time point through the preprocessed real-time current and voltage data. Next, the system extracts the instantaneous power data sequence within this time range according to the set time window (such as the first 10 seconds or 20 seconds after the device is started), analyzes the power change rate and amplitude, and summarizes the characteristic information of the instantaneous power change. At the same time, in order to extract the starting load fluctuation information, the system calculates the maximum and minimum differences between the current and voltage changes in the same time window as a quantitative indicator of the current fluctuation amplitude and the voltage fluctuation amplitude. By setting rules, the system can identify the time periods with large load fluctuations during the startup process, extract the key characteristic data of the load fluctuations from these time periods, and organize them into the starting load fluctuation information for subsequent analysis. This method ensures that the extracted characteristic data is highly timely and representative, thereby more accurately reflecting the impact of each charging device on the power grid at startup.

[0078] The extracted instantaneous power change information and starting load fluctuation information are analyzed to generate the instantaneous power change coefficient and starting load stability coefficient of each charging device respectively.

[0079] In this embodiment, the logic for obtaining the instantaneous power variation coefficient of each charging device is as follows:

[0080] Extract the instantaneous power change information from the load fluctuation information caused by the startup of each charging device in the depot, including the voltage and current values ​​at different times within a period of time after the startup of each charging device in the depot, and calibrate them as and It represents the voltage value at time m within a period of time after the i-th charging device in the vehicle depot is started. represents the current value at time m within a period of time after the i-th charging device in the vehicle depot is started, i=1, 2, 3, ..., k, m=1, 2, 3, ..., g, k and g are both positive integers;

[0081] Extracting the instantaneous power change information from the load fluctuation information caused by the startup of each charging device in the vehicle depot can be achieved by extracting features from the real-time collected data. First, the voltage and current value data collected in real time by the sensor network are used. These data have been de-noised by the filtering algorithm during the preprocessing process, and the instantaneous spikes have been eliminated by abnormal value detection. The specific extraction process is that the system extracts the voltage and current values ​​at each time point point by point within the time range by setting a time window (such as the time interval from 10 seconds to 20 seconds after the device is started). The system stores each set of voltage and current data in the time window in pairs to ensure the timing consistency and synchronization of the data. Subsequently, the system calculates the instantaneous power at each moment based on the extracted voltage and current values ​​to obtain the instantaneous power change information within a period of time after the charging device is started. This extraction method ensures that all data are accurately associated with the device startup time point, and the power calculation at each time point is based on the synchronized voltage and current data, thereby providing complete and accurate data input for the subsequent power change coefficient calculation.

[0082] Calculate the actual power of each charging device in the depot at different times after it is started According to the formula:

[0083] Calculate the instantaneous power change coefficient of each charging device. The specific calculation formula is as follows:

[0084]

[0085] In the formula, IPVC i is the instantaneous power variation coefficient of the i-th charging device.

[0086] This formula is used to calculate the instantaneous power variation coefficient IPVC of each charging device iThe purpose of this method is to measure the power fluctuation amplitude of the charging equipment during startup, so as to evaluate its instantaneous load impact on the power grid. The calculation process first calculates the absolute value of the difference in instantaneous power between adjacent moments. To obtain the power change between each time point; the use of absolute value is to ignore the direction of power increase or decrease, and only focus on the size of the change, so as to fully reflect the intensity of the fluctuation. These changes are summed one by one and then divided by g-1 (the total number of sampling points minus one) to calculate the average change in the entire startup process, standardize the degree of fluctuation, and make it comparable between different time periods and different charging devices. This method ensures that the instantaneous power change coefficient can not only capture the severity of power fluctuations, but also quantify the overall impact of such fluctuations on the power grid, thereby providing a basis for the response adjustment of the EPS system.

[0087] Instantaneous power variation coefficient IPVC of the i-th charging device i The size of directly reflects the severity of the power fluctuation of the equipment during the startup process, thus affecting the assessment of the overall load fluctuation in the vehicle section. When the instantaneous power variation coefficient is large, it means that the power variation of the charging equipment is large when it is started, which may cause a strong instantaneous load impact on the power grid, thereby causing more obvious grid voltage fluctuations; on the contrary, a smaller instantaneous power variation coefficient means that the power change of the charging equipment is relatively stable when it is started, and the impact on the power grid is small. When the instantaneous power variation coefficients of all charging devices are comprehensively analyzed, the fluctuation of the overall charging load in the vehicle section can be obtained. By comparing this overall change with the preset normal load fluctuation range, it can be determined whether the current load fluctuation in the vehicle section belongs to the normal charging process, or whether the EPS system needs to be started to respond due to abnormal fluctuations. Therefore, the size of the instantaneous power variation coefficient is an important indicator for evaluating whether the charging load fluctuation is within the normal range.

[0088] In this embodiment, the logic for obtaining the startup load stability coefficient of each charging device is as follows:

[0089] Extract the starting load fluctuation information from the pre-processed load fluctuation information caused by the startup of each charging device in the depot to the power grid, including the time required for each charging device in the depot to reach the power stable state from startup, and the current fluctuation amplitude, voltage fluctuation amplitude and current fluctuation frequency of each charging device during this time, and mark them as TS respectively. i IF i 、VF i and If i , TS i It represents the time required for the i-th charging device in the vehicle depot to reach the power stable state from startup, IF iIndicates the current fluctuation amplitude of the i-th charging device during this time, VF i Indicates the voltage fluctuation amplitude of the i-th charging device during this time, If i represents the current fluctuation frequency of the i-th charging device during the time, i=1, 2, 3, ..., k, k is a positive integer;

[0090] Extracting the information of the starting load fluctuation caused to the power grid by the start-up of each charging device in the pre-processed vehicle depot can be achieved through feature recognition and data screening algorithms. First, the system uses the pre-processed real-time current and voltage data, in conjunction with the time synchronization mechanism, to determine the time period from the start of the charging device to the power stable state. By setting the power fluctuation threshold, the system identifies the starting and ending points of the power stable state, and then determines the start duration. Next, during this time period, the system calculates the difference between the maximum and minimum values ​​of the current and voltage data sequences by statistical methods, thereby extracting the current fluctuation amplitude and voltage fluctuation amplitude. At the same time, the system uses spectrum analysis (such as fast Fourier transform, FFT) to perform frequency analysis on the current data and calculate the current fluctuation frequency, that is, the number of occurrences of the current fluctuation peak in this time period. These data are automatically extracted from the pre-processed original current and voltage data by software analysis tools to ensure the accuracy of time synchronization and the integrity of feature data, providing accurate input for the subsequent calculation of the starting load stability factor.

[0091] Calculate the starting load stability coefficient of each charging device. The specific calculation formula is as follows:

[0092]

[0093] Where, LSC i is the starting load stability factor of the i-th charging device.

[0094] This formula is used to calculate the starting load stability factor LSC of the i-th charging device. i The purpose is to comprehensively evaluate the impact of charging equipment on grid load stability during startup. i * Combined with the current fluctuation amplitude IF i , voltage fluctuation range VF i and current fluctuation frequency Among them, IF i and VF i The product of represents the comprehensive intensity of current and voltage fluctuations, reflecting the overall severity of fluctuations when the equipment starts; multiplied by In order to properly consider the impact of fluctuation frequency, the square root form smoothes the impact of high-frequency fluctuations to prevent excessive impact on the coefficient when the frequency is too high. i *ln(1+IFi +VF i ) Normalize these fluctuations in time and amplitude, TS i Indicates the time required for the equipment to reach a stable power state. The longer the time, the more diluted the impact of load fluctuations on the power grid will be. Logarithmic function ln(1+IF i +VF i ) makes the extremely large fluctuation amplitude smoothed, ensuring that the calculation results are not overly affected by extreme values, so that they will not be ignored when the fluctuation is small. This calculation method comprehensively considers the overall impact of the fluctuation amplitude, frequency and time on the power grid, making the calculated starting load stability factor more stable and adaptable, providing accurate indicators for subsequent evaluation.

[0095] Starting load stability factor LSC of the i-th charging device i The size of LSC directly reflects the intensity and stability of the load fluctuation during the startup process of the device, thus affecting whether the overall load fluctuation is evaluated as a normal charging process. i A small value means that the current and voltage fluctuations of the device are more severe, with higher frequency and shorter duration when starting, which will cause a larger instantaneous impact on the power grid and may make the overall load fluctuation look more unstable. On the contrary, a smaller LSC i A value of 0 indicates that the load changes smoothly when the device is started, which has a smaller impact on the power grid and is more likely to be regarded as normal charging fluctuations. i After comprehensive analysis of the values, it is possible to accurately determine whether the load fluctuation of the entire vehicle section when the charging equipment is started is within the preset normal range, thereby deciding whether it is necessary to adjust the power supply strategy of the EPS system to avoid unnecessary system switching.

[0096] Comprehensively analyze the generated instantaneous power variation coefficient and starting load stability coefficient of each charging device, and generate the standard deviation of the comprehensive instantaneous power variation coefficient and the standard deviation of the comprehensive starting load stability coefficient respectively;

[0097] In this embodiment, the acquisition logic of the comprehensive instantaneous power variation coefficient standard deviation and the comprehensive starting load stability coefficient standard deviation is as follows:

[0098] Extract the instantaneous power variation coefficient IPVC of each charging device i and starting load stability factor LSC i , i=1, 2, 3, …, k, k is a positive integer;

[0099] Extracting the instantaneous power change coefficient and the starting load stability coefficient of each charging device can be achieved by combining data analysis software with a preset feature extraction algorithm. First, the system collects and preprocesses the voltage and current data in real time during the startup of the charging device, and stores these data in time series. Subsequently, the system uses a sliding window analysis method to divide the preprocessed data into time segments of each charging device in the startup phase, and calculates the instantaneous power change value at each time point in these segments. According to the change of these power data, the system uses the characteristic difference summation method to calculate the instantaneous power change coefficient of each charging device. For the extraction of the starting load stability coefficient, the system detects the current fluctuation amplitude, voltage fluctuation amplitude, and current fluctuation frequency in the startup phase, and uses a complex formula to calculate the starting load stability coefficient in combination with the time required for the power to reach a stable state. During the entire extraction process, the data analysis software automatically calculates and matches these key parameters, summarizes the final coefficient results, and outputs them for subsequent comprehensive analysis. This process relies on the software's automatic analysis of data and formula application, thereby ensuring the efficiency and accuracy of the extraction process.

[0100] Calculate the instantaneous power variation coefficient IPVC of all charging devices i The average value of the starting load stability factor LSC of all charging equipment i The specific calculation formula is as follows:

[0101]

[0102] Where IPVC_ is the instantaneous power variation coefficient IPVC of all charging devices i The average value of LSC_ is the starting load stability factor LSC of all charging equipment. i The average value of

[0103] Calculate the standard deviation of the comprehensive instantaneous power variation coefficient and the standard deviation of the comprehensive starting load stability coefficient. The specific calculation formula is as follows:

[0104]

[0105] In the formula, σ IPVC is the standard deviation of the comprehensive instantaneous power variation coefficient, σ LSC is the standard deviation of the comprehensive starting load stability coefficient.

[0106] Standard deviation of comprehensive instantaneous power variation coefficient σ IPVC The size of directly reflects the consistency of instantaneous power changes of all charging devices during the startup process. IPVCIt means that the instantaneous power changes of each charging device are relatively consistent, indicating that the load impact of the equipment on the power grid at startup is relatively uniform and in line with expectations, which is usually evaluated as a normal charging load fluctuation. IPVC This indicates that different charging devices have large differences in power fluctuations at startup, and some of them may have abnormally drastic power changes, causing uneven load fluctuations on the power grid. This may be evaluated as an abnormal situation and require further adjustment of the EPS system to maintain power supply stability.

[0107] Standard deviation of comprehensive starting load stability factor σ LSC The size of reflects the consistency of the load fluctuation stability of the charging equipment during the startup process. LSC It means that the startup load fluctuations of each charging device are relatively stable and the differences between them are small, which means that the load impact of the charging device on the power grid during the startup process is regular and predictable, which helps to be evaluated as a normal charging load change. On the contrary, a larger σ LSC This means that the startup load fluctuation stability of each device varies greatly. Some devices may have large load instability during startup, causing irregular fluctuations in the power grid. This may be regarded as an abnormal fluctuation, and the power supply strategy or equipment operating parameters need to be adjusted to ensure the overall stability of the power grid.

[0108] A load fluctuation assessment model is constructed for the generated standard deviation of the comprehensive instantaneous power variation coefficient and the standard deviation of the comprehensive starting load stability coefficient, and a load fluctuation assessment coefficient is generated. After the generation, an analysis is performed to assess whether the load fluctuation on the power grid caused by the startup of all charging equipment in the depot is a normal charging load change, and the load fluctuation is divided into normal charging load fluctuation and abnormal load fluctuation according to the assessment results;

[0109] In this embodiment, a load fluctuation assessment model is constructed based on the generated comprehensive instantaneous power variation coefficient standard deviation and the comprehensive starting load stability coefficient standard deviation to generate a load fluctuation assessment coefficient, which specifically includes the following steps:

[0110] Collect several comprehensive instantaneous power variation coefficient standard deviations, comprehensive starting load stability coefficient standard deviations and corresponding load fluctuation assessment coefficients generated in the past period of time, and calibrate them as and LFEC x , x represents the number of the integrated instantaneous power variation coefficient standard deviation, the integrated starting load stability coefficient standard deviation and the corresponding load fluctuation assessment coefficient generated in the past period of time, x = 3, 4, 5, ..., d, d is a positive integer, and the data collected in the past period of time is formed into a historical data set;

[0111] The collection of several standard deviations of the comprehensive instantaneous power change coefficient, the standard deviation of the comprehensive startup load stability coefficient, and the corresponding load fluctuation assessment coefficient generated in the past period of time can be achieved through the data storage and automated log management system. First, after calculating the standard deviation of the comprehensive instantaneous power change coefficient, the standard deviation of the comprehensive startup load stability coefficient, and generating the corresponding load fluctuation assessment coefficient each time, the system automatically records these results in the database with the timestamp as the index. The data storage system uses a relational database or a time series database (such as MySQL, InfluxDB, etc.), which can effectively manage a large amount of time series data. The system configures scheduled tasks or automatic scripts to regularly archive these standard deviations and assessment coefficient data in chronological order every day or every week to form a historical data set. When model training or analysis is required, the data analysis software can directly retrieve data for a specific time period from the database, and filter and summarize it according to the set time range (such as the past month or a quarter). This method ensures the integrity and traceability of the data, and ensures that the data collection process does not rely on human intervention, so that the load fluctuation assessment model can quickly call historical data for training and optimization when needed.

[0112] The reason why x is limited to a positive integer greater than or equal to 3 is that three regression coefficients β0, β1 and β2 need to be calculated in the regression model, while there is only one target equation. Therefore, at least three sets of data are required to form three independent equations to construct a solvable system of equations. Each set of data (i.e., each x value) corresponds to a set of historical standard deviations of the comprehensive instantaneous power variation coefficient, the standard deviation of the comprehensive starting load stability coefficient, and the generated load fluctuation assessment coefficient, which are used to determine the relationship between these coefficients. Only when x is greater than or equal to 3 can it be guaranteed that the system of equations has enough independent equations to solve, and then accurately determine the values ​​of the three regression coefficients. If x is less than 3, there are not enough data points, and the regression system of equations will become an underdetermined problem, and the regression coefficients cannot be uniquely determined, which will cause the model to be unable to be effectively trained and lose its predictive ability. Therefore, the limitation of at least three data points ensures the mathematical solvability and training effect of the model.

[0113] The multivariate regression model is selected as the load fluctuation assessment model, and is trained through historical data sets to determine the value of the regression coefficient according to the formula:

[0114]

[0115] Where β0, β1 and β2 are regression coefficients;

[0116] The multiple regression model is a statistical model used to analyze the linear relationship between multiple independent variables and a dependent variable, and to predict the value of the dependent variable by establishing an equation. The multiple regression model is selected as the load fluctuation assessment model because it can simultaneously consider multiple influencing factors (here, the standard deviation of the comprehensive instantaneous power variation coefficient and standard deviation of comprehensive starting load stability factor ) Load fluctuation evaluation factor LFEC x The regression model can be used to determine the different contributions of these factors to load fluctuations, thereby improving the accuracy of prediction. The regression coefficients β0, β1, and β2 in the model represent different meanings: β0 is the intercept, which means and LFEC when all are zero x value, which reflects the basic load fluctuation level; β1 is The regression coefficient represents the standard deviation of the instantaneous power variation coefficient on LFEC x The influence of instantaneous power fluctuation on load fluctuation; and β2 is The regression coefficient of x The impact of these factors can be accurately quantified through the multivariate regression model, thus providing a scientific basis for evaluating and adjusting the load fluctuation of charging equipment.

[0117] By minimizing the error between the predicted value and the actual value, the regression coefficient is optimized, and finally the values ​​of the regression coefficients β0, β1 and β2 are determined;

[0118] The purpose of optimizing the regression coefficients by minimizing the error between the predicted value and the actual value is to improve the prediction accuracy of the model and make the multivariate regression model more accurate in evaluating load fluctuations in practical applications. The specific approach is to use the Least Squares Method, which is an optimization method widely used in regression analysis. The software system compares the actual load fluctuation evaluation coefficients in the historical data with the values ​​predicted by the model, calculates the square of the error between each data point, and sums all the squared error values ​​to form a loss function. The system continuously adjusts the values ​​of the regression coefficients β0, β1, and β2 through iterative optimization algorithms (such as the gradient descent method) to minimize the value of the loss function. This process is automatically performed in the data analysis software. When the loss function reaches the minimum value, the regression coefficient values ​​determined by the model can minimize the error between the predicted value and the actual value, thereby ensuring that the model has the best prediction effect on new data. The result of this is a set of optimal regression coefficients, which enables the model to more accurately reflect the actual impact of equipment startup on the grid load in future load fluctuation assessments.

[0119] Using the final determined regression coefficient, the constructed load fluctuation assessment model is input with the real-time generated comprehensive instantaneous power variation coefficient standard deviation σ IPVC and the standard deviation of the comprehensive starting load stability factor σ LSC , generate load fluctuation assessment coefficient LFEC in real time.

[0120] In this embodiment, the generated load fluctuation evaluation coefficient LFEC is compared with the preset load fluctuation evaluation coefficient threshold LFEC. yuzhi Compare and evaluate whether the load fluctuation on the power grid caused by the startup of all charging equipment in the depot is a normal charging load change according to the comparison results, and divide it into normal charging load fluctuation and abnormal load fluctuation according to the evaluation results. The specific comparison analysis and classification are as follows:

[0121] If LFEC≤LFEC yuzhi , the load fluctuation on the power grid caused by the startup of all charging equipment in the depot belongs to normal charging load changes, so this situation is classified as normal charging load fluctuation;

[0122] This situation means that the load fluctuations on the power grid caused by the startup of all charging equipment in the vehicle depot are within the normal range. That is, the instantaneous power changes and load stability fluctuations when each device is started will not have too much impact on the power grid, and the power grid can withstand such fluctuations normally without triggering the power supply switching of the EPS system. In this case, the system will classify the current charging status as "normal charging load fluctuations", the charging equipment can continue to work as planned, the power supply of the power grid remains stable, and the EPS equipment does not need to intervene, reducing the wear and tear of frequent switching of equipment. This division ensures that the system will not make misjudgments when facing normal load fluctuations, avoids unnecessary energy switching and operational intervention, and maintains the efficient operation of the power supply system in the vehicle depot.

[0123] If LFEC>LFEC yuzhi If the load fluctuation on the power grid caused by the startup of all charging equipment in the depot does not belong to the normal charging load change, then the situation is classified as abnormal load fluctuation.

[0124] This situation means that the load fluctuation caused by the charging equipment in the vehicle depot when it is started exceeds the normal range and is an abnormal load fluctuation. At this time, the instantaneous power changes and load instability generated by the charging equipment during the startup process have a large impact on the power grid, which may cause grid voltage fluctuations and affect the normal power supply of other facilities. In this case, the system will classify the current state as "abnormal charging load fluctuations" and may trigger corresponding protection or adjustment mechanisms, such as scheduling the EPS system to switch power supply to reduce pressure on the power grid, or adjusting the startup sequence of charging equipment to alleviate the impact of load fluctuations. This division ensures that the system can respond quickly to potential abnormal grid load conditions and prevent greater impact on grid stability, thereby improving the power supply safety and reliability of the entire vehicle depot.

[0125] Pre-set load fluctuation evaluation factor threshold LFEC yuzhi It can be determined through historical data analysis and statistical modeling, and the specific operations can be performed through software tools. First, the system will retrieve the historical data of the standard deviation of the comprehensive instantaneous power variation coefficient and the standard deviation of the comprehensive starting load stability coefficient of the charging equipment in the vehicle depot in the past certain period of time and the corresponding generated LFEC from the database. Through data analysis software (such as Python's data analysis library or dedicated statistical software), these historical data are statistically analyzed, and common statistical methods such as percentile method, normal distribution fitting or cluster analysis are used to find out the range of values ​​in which the load fluctuation in the historical data is normal. For example, the mean and standard deviation of the LFEC in the historical data can be calculated, and an appropriate confidence interval (such as 95% or 99%) can be selected as the range of normal fluctuations; or cluster analysis can be used to automatically classify the data into two categories: normal fluctuations and abnormal fluctuations, and the boundary value between the two categories is used as the threshold. After determining the preliminary LFEC yuzhi After that, the system can be further optimized through machine learning algorithms, continuously learning new data to dynamically adjust the threshold to ensure that it is consistent with the actual operation. Through these methods, the system can automatically analyze a large amount of historical data and ultimately determine a suitable load fluctuation assessment coefficient threshold, making the load fluctuation assessment of charging equipment more accurate.

[0126] Based on the division results, the power supply switching strategy of the EPS system is dynamically adjusted according to different load fluctuation conditions;

[0127] In this embodiment, based on the division results, the power supply switching strategy of the EPS system is dynamically adjusted according to different load fluctuation conditions, specifically:

[0128] For load fluctuations that are normal charging load fluctuations, the power supply switching strategy for dynamically adjusting the EPS system is: maintain the current grid power supply, do not start EPS switching, and continue monitoring;

[0129] For load fluctuations that are normal charging load fluctuations, the power supply switching strategy for dynamically adjusting the EPS system can be achieved by combining the real-time monitoring system with the automation control logic. First, the system will continuously collect the current and voltage data of the charging equipment, as well as the overall load fluctuation of the power grid, and determine whether the load fluctuation is within the normal range through the pre-set threshold. When the load fluctuation assessment model determines that the fluctuation is normal, the software control system will automatically maintain the current power supply status of the power grid and will not trigger the EPS power supply switching. This judgment process relies on real-time data analysis and the calculation of the load fluctuation assessment coefficient. The results are automatically stored in the control database for dispatchers to review at any time. The system will also be equipped with a monitoring alarm function. Once the load fluctuation is detected to be close to the preset threshold upper limit, it will be warned in advance for further observation and adjustment. The implementation of this method ensures that under normal load fluctuations, equipment wear and energy loss caused by frequent switching of power supply sources are avoided, while maintaining the continuity and stability of power supply, providing a stable power grid environment for the charging process in the vehicle depot.

[0130] For abnormal load fluctuations, the power supply switching strategy for the EPS system is to dynamically adjust the power supply of the EPS to give priority to key equipment, adjust the power of the charging equipment and suspend some charging at the same time, and then gradually resume the charging plan after the load returns to normal.

[0131] For load fluctuations that are abnormal, the power supply switching strategy for dynamically adjusting the EPS system can be achieved by combining the real-time monitoring system, intelligent control algorithm and automated execution module. First, the system monitors the load fluctuation evaluation coefficient LFEC in real time through data collection and analysis. Once it is detected that LFEC exceeds the preset threshold and is determined to be an abnormal load fluctuation, the control system will immediately issue an instruction to start EPS power supply to ensure the continuous power supply of key equipment and avoid affecting the operation of important equipment due to unstable power grid. At the same time, the system will automatically adjust the power output of the charging equipment, reduce the power of non-critical charging equipment or completely suspend the charging of some equipment to reduce the instantaneous load pressure of the power grid. This process is controlled by the intelligent power management algorithm, and the software system can gradually execute the power adjustment and suspension instructions of the charging equipment based on the importance priority of the equipment. The monitoring system will continue to detect the load status of the power grid. When the load fluctuation returns to the normal range, the system will gradually restore the charging equipment that was previously suspended or reduced in power, and restore the charging plan one by one according to the priority. This method can respond quickly when the power grid load exceeds the normal range, avoid causing greater impact on the power grid, and ensure the power supply stability of important equipment, which helps to reduce equipment failures and extend the service life of the EPS system and the power grid.

[0132] Continuously monitor the load fluctuation of charging equipment to the power grid and the response effect of EPS system, adjust the load assessment model and switching strategy in real time, and optimize EPS control during the startup of charging equipment through feedback data.

[0133] To achieve "continuously monitoring the load fluctuations of charging equipment to the power grid and the response effect of the EPS system, adjusting the load assessment model and switching strategy in real time, and optimizing the EPS control during the startup of the charging equipment through feedback data", this can be achieved through the combination of real-time data acquisition system, intelligent algorithm analysis, machine learning model optimization and automatic control module.

[0134] First, the system will collect instantaneous power during charging, grid voltage changes, and response data during EPS system switching in real time through current and voltage sensors deployed at key nodes of charging equipment and power grids. The collected data is transmitted to the central data processing platform in real time through a data transmission network (such as industrial Ethernet or wireless network). The data processing platform uses data analysis software to analyze these data in real time, identify the load fluctuations caused to the power grid during the startup of the charging equipment, and the power supply switching effect of the EPS system in response to these fluctuations. The analysis results will be intuitively displayed in the control center in the form of charts or dashboards to facilitate real-time monitoring by dispatchers.

[0135] Next, the system uses intelligent algorithms to analyze the load fluctuation data and EPS response data, and continuously corrects and optimizes the load assessment model through these data. Specifically, the system will compare the real-time collected data with the historical data, find the difference between the model evaluation results and the actual situation, and fine-tune the model parameters through machine learning algorithms (such as regression analysis or neural networks) to make the load fluctuation assessment model more in line with the actual situation. In this way, the system can automatically adjust the model parameters when the load fluctuation pattern of the charging equipment changes, so that the evaluation results are more accurate. In addition, the system will dynamically adjust the switching strategy according to the actual EPS switching effect (such as switching speed and voltage stability after switching), such as optimizing the switching timing of EPS or adjusting the startup order of charging equipment to ensure a more sensitive response in similar fluctuations in the future.

[0136] Finally, the system will store all analysis results and adjustment records as feedback data in the historical database for long-term performance optimization. The system can regularly review these feedback data, conduct more in-depth data mining and trend analysis, and identify potential problems in the charging equipment startup and EPS system response process. For example, if certain charging devices often trigger EPS switching under specific load conditions, the system can optimize the load distribution during the charging process and reduce the impact on the power grid by adjusting the startup parameters or power configuration of these devices. In this way, the system can gradually improve its adaptability to grid load fluctuations and significantly improve the stability and efficiency of EPS control.

[0137] This approach can ensure that the system remains flexible and efficient in long-term operation, adapt to the changing grid load and charging equipment characteristics, and avoid power supply instability caused by evaluation model deviation or improper switching strategy. At the same time, real-time monitoring and model optimization can also reduce equipment wear and energy waste, increase the operating life of the entire charging system and EPS system, and ensure safe and reliable power supply to the subway depot during charging.

[0138] like Figure 2 The EPS emergency power supply control system based on rail transit shown in the figure includes a real-time monitoring and data acquisition module, a load fluctuation data analysis module, a comprehensive fluctuation analysis module, a load fluctuation assessment modeling module, a dynamic power supply adjustment module and an adaptive optimization and feedback module;

[0139] The real-time monitoring and data acquisition module monitors the startup time and instantaneous charging current of each charging device in real time through the sensor network when charging is carried out after the subway depot stops operating at night. It also establishes a real-time monitoring framework for charging load fluctuation assessment and EPS response based on the load fluctuation when the charging device is started.

[0140] The load fluctuation data analysis module obtains the load fluctuation information on the power grid caused by the startup of each charging device in the depot in real time, analyzes it after acquisition, and generates the instantaneous power variation coefficient and startup load stability coefficient of each charging device respectively;

[0141] The comprehensive fluctuation analysis module comprehensively analyzes the instantaneous power variation coefficient and the starting load stability coefficient of each charging device, and generates the standard deviation of the comprehensive instantaneous power variation coefficient and the standard deviation of the comprehensive starting load stability coefficient respectively;

[0142] The load fluctuation assessment modeling module constructs a load fluctuation assessment model for the generated standard deviation of the comprehensive instantaneous power variation coefficient and the standard deviation of the comprehensive starting load stability coefficient, generates a load fluctuation assessment coefficient, and analyzes it after generation to assess whether the load fluctuation caused by the startup of all charging equipment in the depot to the power grid is a normal charging load change, and divides it into normal charging load fluctuation and abnormal load fluctuation according to the assessment results;

[0143] Dynamic power supply adjustment module, based on the division results, dynamically adjusts the power supply switching strategy of the EPS system according to different load fluctuation conditions;

[0144] The adaptive optimization and feedback module continuously monitors the load fluctuation of the charging equipment to the power grid and the response effect of the EPS system, adjusts the load assessment model and switching strategy in real time, and optimizes the EPS control during the startup of the charging equipment through feedback data.

[0145] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0146] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center containing one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0147] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0148] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0149] In the several embodiments provided in the present application, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0150] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0151] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0152] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A method for controlling EPS emergency power supply based on rail transit, characterized in that: The specific steps include: When charging is carried out after the subway depot stops operating at night, the startup time and instantaneous charging current of each charging device are monitored in real time through the sensor network. Based on the load fluctuation when the charging device is started, a real-time monitoring framework for charging load fluctuation assessment and EPS response is established; Obtain in real time the load fluctuation information on the power grid caused by the startup of each charging device in the depot, and analyze it after acquisition to generate the instantaneous power variation coefficient and startup load stability coefficient of each charging device; Comprehensively analyze the generated instantaneous power variation coefficient and starting load stability coefficient of each charging device, and generate the standard deviation of the comprehensive instantaneous power variation coefficient and the standard deviation of the comprehensive starting load stability coefficient respectively; A load fluctuation assessment model is constructed for the generated standard deviation of the comprehensive instantaneous power variation coefficient and the standard deviation of the comprehensive starting load stability coefficient, and a load fluctuation assessment coefficient is generated. After the generation, an analysis is performed to assess whether the load fluctuation on the power grid caused by the startup of all charging equipment in the depot is a normal charging load change, and the load fluctuation is divided into normal charging load fluctuation and abnormal load fluctuation according to the assessment results; Based on the division results, the power supply switching strategy of the EPS system is dynamically adjusted according to different load fluctuation conditions; Continuously monitor the load fluctuation of charging equipment to the power grid and the response effect of EPS system, adjust the load assessment model and switching strategy in real time, and optimize EPS control during the startup of charging equipment through feedback data.

2. According to the EPS emergency power supply control method based on rail transit in claim 1, it is characterized in that: The load fluctuation information on the power grid caused by the startup of each charging device in the depot is obtained in real time, and analyzed after acquisition to generate the instantaneous power variation coefficient and startup load stability coefficient of each charging device respectively, which specifically includes the following steps: Real-time acquisition of load fluctuation information on the power grid caused by the startup of each charging device in the depot, and pre-processing after acquisition; Extracting the instantaneous power change information and the starting load fluctuation information from the pre-processed load fluctuation information on the power grid caused by the starting of each charging device in the vehicle depot; The extracted instantaneous power change information and starting load fluctuation information are analyzed to generate the instantaneous power change coefficient and starting load stability coefficient of each charging device respectively.

3. The EPS emergency power supply control method based on rail transit according to claim 2 is characterized in that: The logic for obtaining the instantaneous power variation coefficient of each charging device is as follows: Extract the instantaneous power change information from the load fluctuation information caused by the startup of each charging device in the depot, including the voltage and current values ​​at different times within a period of time after the startup of each charging device in the depot, and calibrate them as and It represents the voltage value at time m within a period of time after the i-th charging device in the vehicle depot is started. represents the current value at time m within a period of time after the i-th charging device in the vehicle depot is started, i=1, 2, 3, ..., k, m=1, 2, 3, ..., g, k and g are both positive integers; Calculate the actual power of each charging device in the depot at different times after it is started According to the formula: Calculate the instantaneous power change coefficient of each charging device. The specific calculation formula is as follows: In the formula, IPVC i is the instantaneous power variation coefficient of the i-th charging device.

4. The EPS emergency power supply control method based on rail transit according to claim 3 is characterized in that: The logic for obtaining the startup load stability coefficient of each charging device is as follows: Extract the starting load fluctuation information from the pre-processed load fluctuation information caused by the startup of each charging device in the depot to the power grid, including the time required for each charging device in the depot to reach the power stable state from startup, and the current fluctuation amplitude, voltage fluctuation amplitude and current fluctuation frequency of each charging device during this time, and mark them as TS respectively. i IF i 、VF i and If i , TS i It represents the time required for the i-th charging device in the vehicle depot to reach the power stable state from startup, IF i Indicates the current fluctuation amplitude of the i-th charging device during this time, VF i Indicates the voltage fluctuation amplitude of the i-th charging device during this time, If i represents the current fluctuation frequency of the i-th charging device during the time, i=1, 2, 3, ..., k, k is a positive integer; Calculate the starting load stability coefficient of each charging device. The specific calculation formula is as follows: In the formula, LSC i is the starting load stability factor of the i-th charging device.

5. The EPS emergency power supply control method based on rail transit according to claim 4 is characterized in that: The logic for obtaining the standard deviation of the comprehensive instantaneous power variation coefficient and the standard deviation of the comprehensive starting load stability coefficient is as follows: Extract the instantaneous power variation coefficient IPVC of each charging device i and starting load stability factor LSC i , i=1, 2, 3, …, k, k is a positive integer; Calculate the instantaneous power variation coefficient IPVC of all charging devices i The average value of the starting load stability factor LSC of all charging equipment i The specific calculation formula is as follows: In the formula, IPVC - The instantaneous power variation coefficient IPVC of all charging devices i The average value, LSC - The starting load stability factor LSC for all charging equipment i The average value of Calculate the standard deviation of the comprehensive instantaneous power variation coefficient and the standard deviation of the comprehensive starting load stability coefficient. The specific calculation formula is as follows: In the formula, σ IPVC is the standard deviation of the comprehensive instantaneous power variation coefficient, σ LSC is the standard deviation of the comprehensive starting load stability coefficient.

6. The EPS emergency power supply control method based on rail transit according to claim 5 is characterized in that: A load fluctuation assessment model is constructed based on the generated comprehensive instantaneous power variation coefficient standard deviation and the comprehensive starting load stability coefficient standard deviation to generate a load fluctuation assessment coefficient, which specifically includes the following steps: Collect several comprehensive instantaneous power variation coefficient standard deviations, comprehensive starting load stability coefficient standard deviations and corresponding load fluctuation assessment coefficients generated in the past period of time, and calibrate them as and LFEC x , x represents the number of the integrated instantaneous power variation coefficient standard deviation, the integrated starting load stability coefficient standard deviation and the corresponding load fluctuation assessment coefficient generated in the past period of time, x = 3, 4, 5, ..., d, d is a positive integer, and the data collected in the past period of time is formed into a historical data set; The multivariate regression model is selected as the load fluctuation assessment model, and is trained through historical data sets to determine the value of the regression coefficient according to the formula: Where β0, β1 and β2 are regression coefficients; By minimizing the error between the predicted value and the actual value, the regression coefficient is optimized, and finally the values ​​of the regression coefficients β0, β1 and β2 are determined; Using the final determined regression coefficient, the constructed load fluctuation assessment model is input with the real-time generated comprehensive instantaneous power variation coefficient standard deviation σ IPVC and the standard deviation of the comprehensive starting load stability factor σ LSC , generate load fluctuation assessment coefficient LFEC in real time.

7. The EPS emergency power supply control method based on rail transit according to claim 6 is characterized in that: The generated load fluctuation evaluation coefficient LFEC is compared with the preset load fluctuation evaluation coefficient threshold LFEC yuzhi Compare and evaluate whether the load fluctuation on the power grid caused by the startup of all charging equipment in the depot is a normal charging load change according to the comparison results, and divide it into normal charging load fluctuation and abnormal load fluctuation according to the evaluation results. The specific comparison analysis and classification are as follows: If LFEC≤LFEC yuzhi , the load fluctuation on the power grid caused by the startup of all charging equipment in the depot belongs to normal charging load changes, so this situation is classified as normal charging load fluctuation; If LFEC>LFEC yuzhi If the load fluctuation on the power grid caused by the startup of all charging equipment in the depot does not belong to the normal charging load change, then the situation is classified as abnormal load fluctuation.

8. The EPS emergency power supply control method based on rail transit according to claim 7 is characterized in that: Based on the division results, the power supply switching strategy of the EPS system is dynamically adjusted according to different load fluctuation conditions, specifically: For load fluctuations that are normal charging load fluctuations, the power supply switching strategy for dynamically adjusting the EPS system is: maintain the current grid power supply, do not start EPS switching, and continue monitoring; For abnormal load fluctuations, the power supply switching strategy for the EPS system is to dynamically adjust the power supply of the EPS to give priority to key equipment, adjust the power of the charging equipment and suspend some charging at the same time, and then gradually resume the charging plan after the load returns to normal.

9. An EPS emergency power supply control system based on rail transit, used to implement an EPS emergency power supply control method based on rail transit as described in any one of claims 1 to 8, characterized in that: It includes real-time monitoring and data acquisition module, load fluctuation data analysis module, comprehensive fluctuation analysis module, load fluctuation assessment modeling module, dynamic power supply adjustment module and adaptive optimization and feedback module; The real-time monitoring and data acquisition module monitors the startup time and instantaneous charging current of each charging device in real time through the sensor network when charging is carried out after the subway depot stops operating at night. It also establishes a real-time monitoring framework for charging load fluctuation assessment and EPS response based on the load fluctuation when the charging device is started. The load fluctuation data analysis module obtains the load fluctuation information on the power grid caused by the startup of each charging device in the depot in real time, analyzes it after acquisition, and generates the instantaneous power variation coefficient and startup load stability coefficient of each charging device respectively; The comprehensive fluctuation analysis module comprehensively analyzes the instantaneous power variation coefficient and the starting load stability coefficient of each charging device, and generates the standard deviation of the comprehensive instantaneous power variation coefficient and the standard deviation of the comprehensive starting load stability coefficient respectively; The load fluctuation assessment modeling module constructs a load fluctuation assessment model for the generated standard deviation of the comprehensive instantaneous power variation coefficient and the standard deviation of the comprehensive starting load stability coefficient, generates a load fluctuation assessment coefficient, and analyzes it after generation to assess whether the load fluctuation caused by the startup of all charging equipment in the depot to the power grid is a normal charging load change, and divides it into normal charging load fluctuation and abnormal load fluctuation according to the assessment results; Dynamic power supply adjustment module, based on the division results, dynamically adjusts the power supply switching strategy of the EPS system according to different load fluctuation conditions; The adaptive optimization and feedback module continuously monitors the load fluctuation of the charging equipment to the power grid and the response effect of the EPS system, adjusts the load assessment model and switching strategy in real time, and optimizes the EPS control during the startup of the charging equipment through feedback data.

Images