A method and system for distributing load power of a hydrogen fuel cell emergency power supply vehicle

Abstract

The present application relates to the field of energy technology, and discloses a load power distribution method and system for a hydrogen fuel cell emergency power supply vehicle, wherein the method comprises: obtaining performance parameters of each battery stack in the hydrogen fuel cell emergency power supply vehicle, generating output characteristic curves of each battery stack according to the performance parameters, and extracting battery stack performance characteristic vectors; based on the battery stack performance characteristic vectors, generating similarities between each battery stack and obtaining battery stack performance grouping results; based on the battery stack performance grouping results, constructing a battery stack performance prediction model, predicting the degree of damage to the battery stack under overload conditions, obtaining an overload safety threshold, and evaluating the health status of the battery stack; generating a power distribution plan according to the load priority, battery stack performance grouping results, overload safety threshold, and battery stack health status, thereby solving the technical problem of optimal power distribution of parallel battery stacks and achieving the technical effect of stable and reliable operation under various working conditions.

Description

Technical Field

[0001] The present application relates to the field of energy technology, and in particular to a load power distribution method and system for a hydrogen fuel cell emergency power supply vehicle. Background Art

[0002] In the actual application of hydrogen fuel cell emergency power supply vehicles, power distribution when the stacks are connected in parallel is a key technical issue. Emergency power supply vehicles need to work under various complex working conditions, and the performance and status of the stacks may be affected, resulting in inconsistent output power of each stack. If an average distribution strategy is simply adopted, individual stacks may be overloaded, affecting the stability and reliability of the system. Although the priority distribution strategy can ensure the power supply of critical loads, it may cause insufficient power supply to other loads. In addition, in extreme cases such as emergency repairs, although overload distribution can provide greater power in a short period of time, it may accelerate the aging and damage of the stack. How to dynamically adjust the power distribution strategy according to different emergency scenarios and stack status to balance the performance, stability and reliability of the system is a technical problem that needs to be solved urgently. Summary of the invention

[0003] The present application provides a load power distribution method and system for a hydrogen fuel cell emergency power supply vehicle, which solves the technical problem of optimal power distribution of parallel battery stacks and achieves the technical effect of ensuring that the hydrogen fuel cell emergency power supply vehicle can operate stably and reliably under various working conditions.

[0004] In order to achieve the above objectives, the main technical solutions adopted in this application include:

[0005] In a first aspect, an embodiment of the present application provides a load power distribution method for a hydrogen fuel cell emergency power supply vehicle, the method comprising: when the fuel cells are operated in parallel, obtaining performance parameters of each fuel cell in the hydrogen fuel cell emergency power supply vehicle, generating an output characteristic curve of each fuel cell according to the performance parameters and extracting a fuel cell performance characteristic vector; based on the fuel cell performance characteristic vector, generating similarities between the fuel cells and obtaining fuel cell performance grouping results; based on the fuel cell performance grouping results, constructing a fuel cell performance prediction model to predict the degree of damage to the fuel cell under overload conditions, obtain an overload safety threshold, and evaluate the health status of the fuel cell; and generating a power distribution plan according to the load priority, the fuel cell performance grouping results, the overload safety threshold and the health status of the fuel cell.

[0006] A method for distributing load power to a hydrogen fuel cell emergency power supply vehicle is proposed in an embodiment of the present application. When the stacks are running in parallel, the performance parameters of each stack in the hydrogen fuel cell emergency power supply vehicle are obtained, and based on the performance parameters, an output characteristic curve of each stack is generated and a stack performance characteristic vector is extracted; based on the stack performance characteristic vector, the similarity between the stacks is generated and a stack performance grouping result is obtained; based on the stack performance grouping result, a stack performance prediction model is constructed to predict the degree of damage to the stack under overload conditions, obtain an overload safety threshold, and evaluate the stack health status; a power distribution plan is generated based on the load priority, the stack performance grouping result, the overload safety threshold, and the stack health status, which solves the technical problem of optimal power distribution of parallel stacks and achieves the technical effect of ensuring that the hydrogen fuel cell emergency power supply vehicle can operate stably and reliably under various working conditions.

[0007] Optionally, when the fuel cells are operated in parallel, the performance parameters of each fuel cell stack in the hydrogen fuel cell emergency power supply vehicle are obtained, and based on the performance parameters, the output characteristic curve of each fuel cell stack is generated and the fuel cell stack performance characteristic vector is extracted, which specifically includes: collecting the performance parameters of each fuel cell stack in the hydrogen fuel cell emergency power supply vehicle, including temperature, humidity and catalyst activity; inputting the performance parameters into the fuel cell performance model to generate the output characteristic curve of each fuel cell stack; performing feature extraction on the output characteristic curve of each fuel cell stack to obtain the fuel cell stack performance characteristic vector; wherein the fuel cell stack performance characteristic vector includes key parameters that characterize the fuel cell performance.

[0008] Optionally, based on the battery stack performance characteristic vector, generating the similarity between each battery stack and obtaining the battery stack performance grouping result, specifically includes: based on the battery stack performance characteristic vector, calculating the cosine similarity between each battery stack to obtain a similarity matrix; using the shortest distance method to perform hierarchical clustering on the similarity matrix to obtain the battery stack performance grouping result.

[0009] Optionally, the health status indicators of each battery stack are monitored in real time, and the health level of each battery stack is determined according to a preset health status threshold, wherein the health status indicators include voltage drop and power density.

[0010] Optionally, the evaluating the health status of the battery stack further includes constructing a battery stack health status evaluation model by utilizing the laws of battery stack performance degradation and failure to predict the remaining life of the battery stack.

[0011] Optionally, generating a power allocation plan based on the load priority, the stack performance grouping result, the overload safety threshold and the stack health status specifically includes: establishing a fuzzy control rule base, the fuzzy control rule base including power allocation plans for parallel stacks under various working conditions, the power allocation plan including the power allocation ratio of each stack; using the load priority, the stack performance grouping result and the stack health status as inputs to the fuzzy control rule base to generate a power allocation plan.

[0012] Optionally, the method further includes: generating a power adjustment instruction according to the power allocation plan; and the control unit of each fuel cell stack adjusts the output power of each fuel cell stack according to the power adjustment instruction.

[0013] In a second aspect, an embodiment of the present application provides a load power distribution system for a hydrogen fuel cell emergency power supply vehicle, the system comprising: a feature extraction module, for obtaining the performance parameters of each battery stack in the hydrogen fuel cell emergency power supply vehicle when the battery stacks are operated in parallel; for generating an output characteristic curve of each battery stack based on the performance parameters and extracting a battery stack performance characteristic vector; a grouping module, for generating similarities between each battery stack based on the battery stack performance characteristic vector and obtaining a battery stack performance grouping result; a prediction module, for constructing a battery stack performance prediction model based on the battery stack performance grouping result, predicting the degree of damage to the battery stack under overload conditions, obtaining an overload safety threshold, and evaluating the health status of the battery stack; a power distribution module, for generating a power distribution plan based on the load priority, the battery stack performance grouping result, the overload safety threshold and the health status of the battery stack.

[0014] Optionally, the feature extraction module includes: a collection unit for collecting performance parameters of each fuel cell stack in the hydrogen fuel cell emergency power supply vehicle, including temperature, humidity and catalyst activity; a generation unit for inputting the performance parameters into a fuel cell stack performance model to generate an output characteristic curve of each fuel cell stack; an extraction unit for performing feature extraction on the output characteristic curve of each fuel cell stack to obtain a fuel cell stack performance characteristic vector; wherein the fuel cell stack performance characteristic vector includes key parameters that characterize the fuel cell stack performance.

[0015] In a third aspect, an embodiment of the present application provides a computer device, comprising: a memory and a processor, wherein the memory and the processor are communicatively connected to each other, computer instructions are stored in the memory, and the processor executes the above method by executing the computer instructions.

[0016] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, on which computer instructions are stored, and the computer instructions are used to enable a computer to execute the above method.

[0017] In a fifth aspect, an embodiment of the present application provides a computer program product, including computer instructions, which are used to enable a computer to execute the above method. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the specific implementation methods of the present application or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0019] Figure 1 A flow chart of a method for distributing load power to a hydrogen fuel cell emergency power supply vehicle provided in an embodiment of the present application;

[0020] Figure 2 A schematic diagram of a load power distribution system for a hydrogen fuel cell emergency power supply vehicle provided in an embodiment of the present application;

[0021] Figure 3 A schematic diagram of the structure of a computer device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0022] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present application.

[0023] In the practical application of hydrogen fuel cell emergency power supply vehicles, power distribution when the stacks are connected in parallel is a key technical issue. Since the emergency power supply vehicle needs to work in various complex environments, the performance and status of the stacks may be affected, resulting in inconsistent output power of each stack. If the average distribution strategy is simply adopted, individual stacks may be overloaded, affecting the stability and reliability of the system. Although the priority distribution strategy can ensure the power supply of key loads, it may cause insufficient power supply for other loads. In addition, in extreme cases such as emergency repairs, although overload distribution can provide greater power in a short time, it may accelerate the aging and damage of the stack. Therefore, how to dynamically adjust the power distribution strategy according to different emergency scenarios and stack states to balance the performance, stability and reliability of the system is a technical problem that needs to be solved urgently. This requires comprehensive consideration of multiple parameters such as the temperature, humidity, and gas flow of the stack, the establishment of an accurate stack performance model, and the combination of real-time monitoring data and intelligent control algorithms to achieve the optimal distribution of stack power, ensuring that the hydrogen fuel cell emergency power supply vehicle can operate stably and reliably under various working conditions.

[0024] According to an embodiment of the present application, an embodiment of a method for distributing load power to a hydrogen fuel cell emergency power supply vehicle is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in an order different from that shown here.

[0025] Figure 1 : is a flow chart of a method for distributing load power of a hydrogen fuel cell emergency power supply vehicle provided in an embodiment of the present application, such as Figure 1 As shown, the process includes the following steps:

[0026] Step S1, when the stacks are running in parallel, obtain the performance parameters of each stack in the hydrogen fuel cell emergency power supply vehicle, generate the output characteristic curve of each stack according to the performance parameters and extract the stack performance characteristic vector, specifically including:

[0027] The performance parameters of each fuel cell stack in the hydrogen fuel cell emergency power supply vehicle are collected, including temperature, humidity and catalyst activity; the performance parameters are input into the fuel cell stack performance model to generate the output characteristic curve of each fuel cell stack; the output characteristic curve of each fuel cell stack is subjected to feature extraction to obtain the fuel cell stack performance characteristic vector; wherein the fuel cell stack performance characteristic vector includes key parameters characterizing the fuel cell stack performance.

[0028] In some embodiments, the performance parameters such as temperature, humidity and catalyst activity of each stack in the hydrogen fuel cell emergency power supply vehicle are collected in real time by sensors, and the collected data is transmitted to the central control unit. The central control unit receives the performance parameter data from each stack, and according to the preset stack performance model, comprehensively analyzes the influence of temperature, humidity and catalyst activity on the stack output characteristics, and obtains the output characteristic curve equation of each stack under the current working condition. Feature extraction is performed on the output characteristic curve equation of each stack, and the key parameters that can characterize the performance of the stack are selected as feature quantities to construct the stack performance feature vector. The stack performance feature vector of each stack is input into the pre-trained support vector machine classification model to distinguish the performance state of the stack and identify the stack with abnormal or declining performance. If the discrimination result shows that the performance of a stack is abnormal, the performance optimization strategy is generated by using the decision tree algorithm according to the temperature, humidity and catalyst activity parameters of the stack, and the working conditions of the stack are adjusted. According to the optimized stack working conditions, the output characteristic curve equation of the stack is updated, its feature vector is recalculated, and it is judged whether the stack performance has returned to normal. If the performance of all the fuel cells is normal, the total output characteristics of the fuel cell parallel system are determined using the weighted average method based on the output characteristic curves of each fuel cell, and the performance evaluation results of the fuel cell parallel system are output.

[0029] In some embodiments, a plurality of temperature sensors, humidity sensors and catalyst activity sensors are installed in the hydrogen fuel cell emergency power supply vehicle to collect the performance parameters of each stack in real time. For example, the sampling frequency of the temperature sensor is 1Hz, the sampling frequency of the humidity sensor is 5Hz, and the sampling frequency of the catalyst activity sensor is 2Hz. The data collected by the sensor is transmitted to the central control unit via the CAN bus at a rate of 500kbps. Based on the stack performance model, the central control unit adopts a multivariate linear regression algorithm, with temperature, humidity and catalyst activity as independent variables and stack output voltage as the dependent variable, to fit the output characteristic curve equation of each stack. Among them, for every 1°C increase in temperature, the stack output voltage decreases by 2mV; for every 1% increase in humidity, the stack output voltage increases by 1mV; for every 1% decrease in catalyst activity, the stack output voltage decreases by 5mV. Feature extraction is performed on the output characteristic curve equation, and the slope, intercept and curvature of the curve are selected as feature quantities to construct a three-dimensional stack performance feature vector. The stack performance feature vector is input into the support vector machine classification model, and the radial basis kernel function is used. The classification decision boundary is set to 8 to identify the stack with abnormal performance. If the performance of the battery stack is abnormal, the decision tree algorithm is used to generate an optimization strategy. The root node is temperature, the middle node is humidity, and the leaf node is catalyst activity. The battery stack is restored to normal by adjusting the working conditions. For example, when the temperature exceeds 60°C, increase the coolant flow rate; when the humidity is lower than 50%, increase the humidifier power; when the catalyst activity is lower than 80%, increase the hydrogen flow rate. After adjustment, the battery stack performance characteristic vector is recalculated. If the battery stack performance characteristic vector falls within the classification decision boundary, it is considered that the battery stack performance has returned to normal. The weighted average method is used to determine the total output characteristics of the parallel battery stack system. The weight coefficient is the ratio of the rated power of each battery stack to the total rated power. The performance evaluation results of the battery stack parallel system are output to provide a basis for the safe and efficient operation of hydrogen fuel cell emergency power supply vehicles.

[0030] Step S3, based on the battery stack performance feature vector, generates similarities between the battery stacks and obtains the battery stack performance grouping results, specifically including:

[0031] Based on the characteristic vector of battery stack performance, the cosine similarity between each battery stack is calculated to obtain the similarity matrix; the shortest distance method is used to perform hierarchical clustering on the similarity matrix to obtain the battery stack performance grouping results.

[0032] In some embodiments, the battery stack performance characteristic vector includes various performance indicator data of the battery stack. According to a preset similarity measurement method, the cosine similarity between the battery stack performance characteristic vectors is calculated to obtain a similarity matrix. According to the calculated similarity matrix, a hierarchical clustering algorithm is used to group the battery stacks. During the clustering process, similar battery stacks are continuously merged according to preset grouping criteria, such as the principle of minimizing intra-group differences, until a preset number of groups is reached. For each generated battery stack group, the center point of the battery stack characteristic vector within the group is calculated as the representative performance indicator of the group. According to the group to which each battery stack belongs, the battery stack performance grouping result is obtained. The grouping results are evaluated. If the intra-group difference is too large or the inter-group difference is too small, the parameters of the clustering algorithm are adjusted and the battery stacks are regrouped until a satisfactory grouping result is obtained.

[0033] For example, first, the similarity between the battery stack performance feature vectors is calculated based on the cosine similarity to obtain a similarity matrix. Then, the shortest distance method is used to perform hierarchical clustering on the similarity matrix, and the battery stacks with similarities greater than 8 are gradually merged until the number of groups reaches the preset 5. For each battery stack grouping, the arithmetic mean of all battery stack feature vectors in the group is calculated as the representative performance indicator of the group. According to the distance between the similarity of each battery stack and the representative performance indicator, it is divided into corresponding groups to obtain the battery stack performance grouping results. Finally, the standard deviation of the battery stack feature vectors in each group is calculated. If the maximum standard deviation exceeds 1, the clustering similarity threshold is increased by 05, and clustering is performed again until the standard deviation within the group is less than 1 and the distance between the center points of the groups is greater than 5, so as to obtain a satisfactory grouping result. For the clustering grouping of 100 battery stacks, by calculating the cosine similarity matrix and using the shortest distance method for clustering, when the similarity threshold is 85, the battery stacks can be divided into 5 groups. The standard deviation within the group is less than 08, and the distance between the center points of the groups is greater than 6, which meets the grouping criteria. A more reasonable battery stack performance grouping result is obtained, which provides data support for the subsequent battery stack performance analysis and management.

[0034] Step S5, based on the results of the battery stack performance grouping, construct a battery stack performance prediction model, predict the damage degree of the battery stack under overload conditions, obtain the overload safety threshold, and evaluate the health status of the battery stack.

[0035] The input of the stack performance prediction model is the stack overload degree, and the output is the stack damage degree. The power allocation scheme aims to meet the load power demand and minimize stack damage.

[0036] In some embodiments, according to each battery stack performance grouping, a corresponding battery stack performance prediction model is established to predict the degree of damage of the battery stack under different overload conditions and obtain an overload safety threshold. The health status indicators of each battery stack are monitored in real time, and the health level of each battery stack is determined according to the preset health status threshold. When emergency power supply is required to provide high power, the battery stack is allocated from the battery stack group with good health status first, and the tolerable overload degree is determined according to the battery stack performance prediction model to minimize the overload of the decaying or faulty battery stack. If the battery stack group with good health status cannot meet the high power demand, the battery stack is allocated from the sub-healthy battery stack group, and the overload degree is reasonably controlled according to the battery stack performance prediction model to slow down the damage rate of the battery stack. During the emergency power supply process, the real-time performance and health status changes of each battery stack are continuously monitored, and the battery stack allocation strategy is dynamically adjusted to ensure that the damage to the battery stack is minimized while meeting the high power demand. After the emergency power supply is completed, the health status of each battery stack is evaluated, and the battery stack grouping is updated according to the evaluation results to prepare for the next emergency power supply. At the same time, the overload-damaged battery stack is maintained or repaired to restore its performance and health status.

[0037] For example, for each battery stack performance group, 80% of its historical performance data is selected as the training set and 20% as the test set, and the support vector machine regression (SVR) algorithm is used to establish a battery stack performance prediction model. The model input is the battery stack overload degree (such as 120%, 150% of the rated power, etc.), and the output is the corresponding battery stack damage degree (such as 5%, 10% reduction in remaining life, etc.). Real-time monitoring of the health status indicators of each battery stack, such as voltage drop, power density, etc., set 3 levels of health status thresholds (such as voltage drop exceeding 10% for mild decline, exceeding 20% ​​for severe decline, and exceeding 30% for failure), and judge the health level of each battery stack. When emergency power supply is required, the battery stack is allocated from the high-performance battery stack group first, and the overload safety threshold is determined according to the prediction model, that is, the tolerable overload degree (such as not exceeding 130% of the rated power). If the high-performance battery stack group cannot meet the power demand, the battery stack is allocated from the medium-performance battery stack group, and the overload degree is controlled within the damage threshold given by the prediction model (such as not exceeding 120% of the rated power). During the power supply process, the health status of each battery stack is updated every 5 minutes. If the health level of the battery stack is found to be declining, the battery stack allocation strategy is dynamically adjusted to reduce the overload on the degraded battery stack and increase the load sharing of the healthy battery stack. After the power supply is completed, the health status of each battery stack is re-evaluated, the battery stack grouping is updated, and the damaged battery stack is maintained or replaced to ensure that the power supply vehicle is always in the best working condition and ready for the next emergency power supply.

[0038] In some embodiments, during the parallel operation of the battery stacks, the health status of the battery stack can also be evaluated by using the temperature and humidity of the battery stack, and the changes in parameters such as the temperature and humidity of each battery stack are continuously monitored to determine whether the health status of the battery stack is abnormal. If an abnormality is found, the possible cause of the fault is located through a fault diagnosis algorithm, and the power allocation strategy is adjusted in time to reduce or cut off the load of the faulty battery stack.

[0039] In some embodiments, the real-time temperature and humidity and other operating parameters of each stack are obtained, and the parameter data is transmitted to the central control unit for summary analysis. According to the preset normal parameter range, it is determined whether the parameters of each stack fluctuate abnormally. If the parameters exceed the normal range, the abnormal health status alarm of the stack is triggered. When a stack parameter is detected to be abnormal, a fault diagnosis algorithm based on a decision tree is used to comprehensively analyze the abnormal parameter characteristics and locate the potential cause of the fault. According to the diagnosis results, the power adjustment strategy of the faulty stack is determined. If the fault is serious, the stack is removed and its operation is stopped; if the fault is minor, its load power is reduced. Through the load balancing algorithm, the power distribution ratio of other normal stacks is dynamically adjusted to compensate for the power gap of the removed or reduced load stack to maintain the stability of the total output power. The changes in the parameters of each stack are continuously monitored to determine whether the fault is eliminated. If the fault is eliminated, the normal power distribution strategy is restored; if the fault persists, the adjusted operation strategy is maintained. The stack fault diagnosis results and disposal conditions are recorded and archived as data support for subsequent stack maintenance and optimization, and the fault diagnosis and fault tolerance capabilities of the system are continuously improved.

[0040] For example, in a hydrogen fuel cell system, the operating parameters of each stack, such as temperature and humidity, are collected in real time by sensors and transmitted to the central control unit at a frequency of 10 times per second. The control unit uses the Kalman filter algorithm to preprocess the data, filter out high-frequency noise, and extract stable parameter trends. According to the operation experience of the stack, the preset temperature range is 60℃-80℃, and the humidity range is 50%-70%. If the parameters exceed this range for more than 5 seconds, an abnormal alarm is triggered. When it is detected that the temperature of a stack drops to 50℃ and the humidity rises to 80%, the control unit uses a decision tree model based on the C5 algorithm to comprehensively analyze six key parameters such as temperature, humidity, and voltage, and locates the cause of the fault as "hydrogen leakage" and the fault severity level is "high". Based on this, the control unit issues an order to remove the stack and stop its operation. At the same time, the power distribution ratio of the remaining 9 stacks is adjusted from 10% to 11%, and the total output power is maintained at 100kW through load balancing. After the faulty stack is removed, the system continuously monitors its parameter changes. If the temperature returns to 70°C, the humidity drops to 60%, and remains stable within the normal range for more than 30 minutes, the fault is determined to have been eliminated and the equal power distribution strategy is restored. The diagnostic results and handling of the entire process are automatically recorded in the system log database as sample data for subsequent optimization of fault diagnosis rules and fault-tolerant strategies. The decision tree model is continuously iterated and updated through machine learning algorithms to improve the intelligence and robustness of the system.

[0041] In some embodiments, based on the historical data of parallel operation of battery stacks, big data analysis technology is used to explore the patterns of battery stack performance degradation and failure, establish an assessment model for the health status of the battery stack, and predict the remaining life of the battery stack, providing a basis for formulating a battery stack maintenance and replacement plan, thereby ensuring the long-term stable power supply emergency of the hydrogen fuel cell emergency power supply vehicle.

[0042] In some embodiments, a stack operation status data set is constructed based on the performance parameters of the stack in parallel operation, including parameters such as voltage, current, and temperature. The stack operation status data is preprocessed to remove outliers and noise data, and key characteristic parameters such as voltage mean, current mean, and temperature mean are extracted. A time series analysis algorithm, such as an ARIMA model, is used to model the stack operation status data, and the stack performance degradation law is mined to obtain a stack performance degradation trend curve. According to the stack failure history data, fault characteristic parameters, such as voltage mutation, current mutation, etc., are extracted to construct a stack failure characteristic data set. An association rule mining algorithm, such as an Apriori algorithm, is used to mine the stack failure characteristic data to obtain the stack failure mode and failure law. Based on the stack performance degradation trend and failure law, a stack health status assessment model is constructed, and the stack health status level is judged according to the current stack operation status parameters. Based on the stack health status assessment model, the remaining life prediction algorithm, such as the support vector machine regression algorithm, is used to combine the historical operation data of the stack and the current health status to predict the remaining life of the stack, providing a decision-making basis for formulating the stack maintenance and replacement plan, and ensuring the long-term and stable power supply of the hydrogen fuel cell emergency power supply vehicle.

[0043] For example, a battery stack performance data set of parameters such as voltage, current, and temperature can be obtained from the historical data of parallel operation of the battery stack. For example, the voltage data of each battery stack at different time points, such as 6V, 7V, 8V, etc., the current data such as 20A, 25A, 30A, etc., and the temperature data such as 60℃, 70℃, 80℃, etc. can be collected. Then the battery stack operation status data is preprocessed, the 3σ principle is used to remove outliers, the Kalman filter is used to remove noise data, and key characteristic parameters such as voltage mean, current mean, and temperature mean are extracted. Then the ARIMA time series analysis model is used to model the battery stack operation status data. By analyzing the changing trends of parameters such as voltage, current, and temperature over time, the performance degradation law of the battery stack is excavated, and the performance degradation trend curve of the battery stack is obtained. According to the historical data of battery stack failures, fault characteristic parameters are extracted, such as the voltage suddenly changes from 8V to 2V, the current suddenly changes from 30A to 5A, etc., to construct a battery stack failure characteristic data set. The Apriori association rule mining algorithm is used to mine the fault feature data of the battery stack, and it is found that voltage mutation and current mutation often occur at the same time, based on which the failure mode and failure law of the battery stack are obtained. The battery stack health status assessment model is constructed by integrating the performance degradation trend and failure law of the battery stack. When the battery stack voltage is lower than 6V, the current is less than 20A, and the temperature is higher than 80℃, the health status of the battery stack is judged to be poor. On this basis, the support vector machine regression algorithm is used, combined with the historical operation data and current health status of the battery stack, and the remaining life of the battery stack is predicted by the trained model. For example, the current health status of the battery stack is medium, and historical data show that the average remaining life of the battery stack under similar conditions is 5000 hours, then the current remaining life of the battery stack is predicted to be about 5000 hours. Based on this, a maintenance and replacement plan for the battery stack is formulated, such as maintenance when the remaining life of the battery stack is less than 1000 hours, and replacement when it is less than 500 hours, so as to ensure the long-term and stable power supply of the hydrogen fuel cell emergency power supply vehicle.

[0044] Step S7, generating a power allocation plan according to the load priority, the stack performance grouping result, the overload safety threshold and the stack health status, specifically including:

[0045] In some embodiments, when a hydrogen fuel cell emergency power supply vehicle needs to provide high power in a short period of time in an emergency, the degree of damage to the battery stack under overload conditions can be predicted based on the battery stack performance grouping and the battery stack performance model, and priority can be given to battery stacks in good condition based on the health status of the battery stack to minimize overload on degraded or faulty battery stacks.

[0046] In some embodiments, a fuzzy control rule base is established, wherein the fuzzy control rule base includes power allocation schemes under various operating conditions; the load priority, fuel cell performance grouping results, overload safety threshold and fuel cell health status are used as inputs to the fuzzy control rule base to generate a power allocation scheme, including the power allocation ratio of each fuel cell.

[0047] In some embodiments, a fuzzy control rule base is established, and the load priority and the stack performance grouping results are comprehensively considered to formulate the stack power allocation strategy under various working conditions. The load demand of the power system and the stack operating status data are obtained in real time as the input of the fuzzy control algorithm. The fuzzy control algorithm is run to dynamically calculate the power allocation ratio of each stack according to the current load demand and stack status, and generate power adjustment instructions for each stack. The power adjustment instruction is sent to the control unit of each stack to adjust the output power of each stack and implement the power allocation plan. The stack operating parameters and load power supply status are continuously monitored. If there are abnormal conditions such as insufficient power supply to the critical load or overload operation of the stack, the power allocation plan is adjusted in time to ensure safe and stable operation of the system.

[0048] In some embodiments, the priority of the load is determined by the characteristics of the emergency scenario. According to the characteristics of the scenario, the load is divided into critical load, secondary load and non-critical load according to the priority from high to low. Among them, the critical load has the highest weight, followed by the secondary load, and the non-critical load has the lowest weight.

[0049] For example, when establishing a fuzzy control rule base, the load priority and stack performance grouping are comprehensively considered to formulate the stack power allocation strategy under various working conditions. When the critical load demand is high, the high-performance stack group is scheduled first, and its power allocation ratio is set to 60%. The load demand of the power system, the load priority, the stack performance grouping results, the overload safety threshold and the stack health status data are obtained as the input of the fuzzy control algorithm. The fuzzy control algorithm is run to dynamically generate the power allocation ratio of each stack. For example, the Mamdani reasoning method is used to calculate the membership function to obtain the power adjustment instructions of each stack, such as adjusting the output power of stack A to 80% of the rated power. The power adjustment instructions are sent to the control unit of each stack, and the output power of each stack is adjusted through the PID control algorithm to achieve accurate execution of the power allocation plan. The stack operating parameters and load power supply status are continuously monitored. If there are abnormal conditions such as insufficient power supply to the critical load or overload operation of the stack, the power allocation plan is adjusted in time. For example, when the voltage of critical loads is lower than 90% of the rated value, the emergency dispatch mechanism is activated to prioritize the power supply to critical loads, and through load reduction strategies, the power supply ratio of non-critical loads is reduced to ensure safe and stable operation of the system.

[0050] In some embodiments, the health status data of each battery stack is monitored in real time, including parameters such as voltage, current, and temperature, and a machine learning algorithm such as a support vector machine is used to determine whether the battery stack has a risk of performance degradation or failure. If the battery stack is in good health, its rated power is allocated according to the load priority weight; if it is determined that the battery stack has a risk of performance degradation or failure, its priority weight is dynamically reduced. According to the load priority weight and the health status of the battery stack, a heuristic optimization algorithm such as a genetic algorithm is used to obtain the power allocation plan for each battery stack, giving priority to power supply to critical loads. The optimized battery stack power allocation plan is sent to the energy management system of the hydrogen fuel cell power supply vehicle to control the output power of each battery stack. The emergency power supply of the power supply vehicle is continuously tracked. When the emergency scenario or load demand changes, the data allocation weight is re-acquired, and the battery stack power allocation is dynamically optimized to ensure the reliability and stability of the emergency power supply.

[0051] In some embodiments, after obtaining the emergency scenario information of the hydrogen fuel cell emergency power supply vehicle, the critical load, secondary load and non-critical load are determined according to the characteristics of the scenario. For example, in the hospital emergency power supply scenario, operating room lighting, ventilator and other equipment are critical loads, ward lighting is a secondary load, and office area lighting is a non-critical load. For each load type, the corresponding priority weight is set, the critical load weight is set to 6, the secondary load weight is set to 3, and the non-critical load weight is set to 1. The health status data of each battery stack is monitored in real time, including parameters such as voltage, current, and temperature, and the support vector machine algorithm is used to determine whether the battery stack has a risk of performance degradation or failure. If the battery stack is in good health, its rated power is allocated according to the load priority weight; if it is judged that the battery stack has a risk of performance degradation or failure, its priority weight is dynamically reduced, such as by 20%. According to the load priority weight and the health status of the battery stack, a genetic algorithm is used to optimize the power allocation scheme of each battery stack. The fitness function of the genetic algorithm is designed to minimize the power gap of the critical load, while taking into account the health status of the battery stack and the load priority. Through the selection, crossover, mutation and other operations of the genetic algorithm, the optimal battery stack power allocation scheme is searched to give priority to the power supply of critical loads. The optimized stack power distribution plan is sent to the energy management system of the hydrogen fuel cell power supply vehicle to control the output power of each stack. The emergency power supply of the power supply vehicle is continuously tracked. When the emergency scenario or load demand changes, such as the addition of an operating room load, the above steps are repeated to dynamically optimize the stack power distribution and ensure the reliability and stability of the emergency power supply. Through this method, the hydrogen fuel cell emergency power supply vehicle can intelligently allocate the stack power according to the load priority and the health status of the stack, improve the pertinence and effectiveness of the emergency power supply, and provide reliable emergency power supply guarantee for critical loads.

[0052] A method for distributing load power to a hydrogen fuel cell emergency power supply vehicle is proposed in an embodiment of the present application. When the stacks are running in parallel, the performance parameters of each stack in the hydrogen fuel cell emergency power supply vehicle are obtained, and based on the performance parameters, an output characteristic curve of each stack is generated and a stack performance characteristic vector is extracted; based on the stack performance characteristic vector, the similarity between the stacks is generated and a stack performance grouping result is obtained; based on the stack performance grouping result, a stack performance prediction model is constructed to predict the degree of damage to the stack under overload conditions, obtain an overload safety threshold, and evaluate the stack health status; a power distribution plan is generated based on the load priority, the stack performance grouping result, the overload safety threshold, and the stack health status, which solves the technical problem of optimal power distribution of parallel stacks and achieves the technical effect of ensuring that the hydrogen fuel cell emergency power supply vehicle can operate stably and reliably under various working conditions.

[0053] Please refer to Figure 2 , Figure 2 A schematic diagram of a load power distribution system for a hydrogen fuel cell emergency power supply vehicle provided in an embodiment of the present application, the power distribution system comprising: a feature extraction module, for obtaining the performance parameters of each battery stack in the hydrogen fuel cell emergency power supply vehicle when the battery stacks are operated in parallel; for generating an output characteristic curve of each battery stack according to the performance parameters and extracting a battery stack performance characteristic vector; a grouping module, for generating similarities between each battery stack based on the battery stack performance characteristic vector and obtaining a battery stack performance grouping result; a prediction module, for constructing a battery stack performance prediction model based on the battery stack performance grouping result, predicting the degree of damage to the battery stack under overload conditions, obtaining an overload safety threshold, and evaluating the health status of the battery stack; a power distribution module, for generating a power distribution plan according to the load priority, the battery stack performance grouping result, the overload safety threshold and the health status of the battery stack.

[0054] Among them, the feature extraction module includes: an acquisition unit, which is used to collect the performance parameters of each battery stack in the hydrogen fuel cell emergency power supply vehicle, including temperature, humidity and catalyst activity; a generation unit, which is used to input the performance parameters into the battery stack performance model to generate the output characteristic curve of each battery stack; an extraction unit, which is used to extract features from the output characteristic curve of each battery stack to obtain a battery stack performance characteristic vector; wherein the battery stack performance characteristic vector includes key parameters for characterizing the battery stack performance.

[0055] The further functional description of each of the above modules and units is the same as that of the above corresponding embodiments and will not be repeated here.

[0056] The system in this embodiment is presented in the form of functional units, where the units refer to ASIC (Application Specific Integrated Circuit) circuits, processors and memories that execute one or more software or fixed programs, and / or other devices that can provide the above functions.

[0057] See also Figure 3 , Figure 3 is a schematic diagram of the structure of a computer device provided in an embodiment of the present application, such as Figure 3 As shown, the computer device includes: one or more processors 10, a memory 20, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. Various components are connected to each other using different buses for communication, and can be installed on a common mainboard or installed in other ways as needed. The processor can process the instructions executed in the computer device, including instructions stored in or on the memory to display the graphical information of the GUI on an external input / output device (such as, a display device coupled to the interface). In some optional embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Similarly, multiple computer devices can be connected, and each device provides some necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system). Figure 3 A processor 10 is taken as an example.

[0058] The processor 10 may be a central processing unit, a network processor or a combination thereof. The processor 10 may further include a hardware chip. The hardware chip may be a dedicated integrated circuit, a programmable logic device or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable gate array, a general purpose array logic or any combination thereof.

[0059] The memory 20 stores instructions executable by at least one processor 10, so that at least one processor 10 executes the method shown in the above embodiment.

[0060] The memory 20 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created according to the use of the computer device, etc. In addition, the memory 20 may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some optional embodiments, the memory 20 may optionally include a memory remotely arranged relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0061] The memory 20 may include a volatile memory, such as a random access memory; the memory may also include a non-volatile memory, such as a flash memory, a hard disk or a solid state drive; the memory 20 may also include a combination of the above types of memory.

[0062] The computer device further comprises a communication interface 30 for the computer device to communicate with other devices or a communication network.

[0063] The embodiment of the present application also provides a computer-readable storage medium. The above method according to the embodiment of the present application can be implemented in hardware, firmware, or can be implemented as a computer code that can be recorded in a storage medium, or can be implemented as a computer code that is originally stored in a remote storage medium or a non-temporary machine-readable storage medium and will be stored in a local storage medium through a network download, so that the method described herein can be stored in such software processing on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only storage memory, a random access memory, a flash memory, a hard disk or a solid-state hard disk, etc.; further, the storage medium can also include a combination of the above types of memories. It can be understood that a computer, a processor, a microprocessor controller, or programmable hardware includes a storage component that can store or receive software or computer code. When the software or computer code is accessed and executed by a computer, a processor, or hardware, the method shown in the above embodiment is implemented.

[0064] The embodiment of the present application provides a computer program product, which includes computer instructions, which are stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes the method of any embodiment of the present application.

[0065] Although the embodiments of the present application are described in conjunction with the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present application, and such modifications and variations are all within the scope defined by the appended claims.

[0066] The systems, modules or units described in the above embodiments may be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, the computer may be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

[0067] For the convenience of description, the above device is described in various units according to their functions. Of course, when implementing the present application, the functions of each unit can be implemented in the same or multiple software and / or hardware.

[0068] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0069] The present application is described with reference to the flowcharts and / or block diagrams of the methods, systems, and computer program products according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of the processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0070] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0071] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0072] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0073] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0074] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.

[0075] Although the embodiments of the present application have been described in conjunction with the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present application, and such modifications and variations are all within the scope defined by the appended claims.

Claims

1. A method for distributing load power of a hydrogen fuel cell emergency power supply vehicle, characterized in that: The method comprises: When the stacks are operated in parallel, the performance parameters of each stack in the hydrogen fuel cell emergency power supply vehicle are obtained, and according to the performance parameters, the output characteristic curve of each stack is generated and the stack performance characteristic vector is extracted, which specifically includes: collecting the performance parameters of each stack in the hydrogen fuel cell emergency power supply vehicle, including temperature, humidity and catalyst activity; inputting the performance parameters into the stack performance model to generate the output characteristic curve of each stack; extracting features from the output characteristic curve of each stack to obtain the stack performance characteristic vector; wherein the stack performance characteristic vector includes key parameters characterizing the stack performance; Based on the battery stack performance characteristic vector, the similarity between each battery stack is generated and the battery stack performance grouping result is obtained, specifically including: based on the battery stack performance characteristic vector, the cosine similarity between each battery stack is calculated to obtain a similarity matrix; hierarchical clustering is performed on the similarity matrix using the shortest distance method to obtain the battery stack performance grouping result; Based on the battery stack performance grouping results, a battery stack performance prediction model is constructed to predict the damage degree of the battery stack under overload conditions, obtain an overload safety threshold, and evaluate the health status of the battery stack, wherein the input of the battery stack performance prediction model is the battery stack overload degree, and the output is the battery stack damage degree; A power allocation plan is generated according to the load priority, the battery stack performance grouping result, the overload safety threshold and the battery stack health status, specifically including: establishing a fuzzy control rule base, the fuzzy control rule base including power allocation plans for parallel battery stacks under various working conditions, the power allocation plan including the power allocation ratio of each battery stack; using the load priority, the battery stack performance grouping result, the overload safety threshold and the battery stack health status as inputs to the fuzzy control rule base to generate a power allocation plan; wherein the power allocation plan aims to meet the load power demand and minimize battery stack damage.

2. The method according to claim 1, characterized in that The evaluation of the health status of the battery stack specifically includes: Monitor the health status indicators of each battery stack in real time, and determine the health level of each battery stack based on a preset health status threshold, wherein the health status indicators include voltage drop and power density.

3. The method according to claim 1, characterized in that The evaluating the health status of the battery stack also includes constructing a battery stack health status evaluation model by utilizing the laws of battery stack performance degradation and failure to predict the remaining life of the battery stack.

4. The method according to claim 1, characterized in that: The method further comprises: Generate power adjustment instructions according to the power allocation plan; The control unit of each fuel cell stack adjusts the output power of each fuel cell stack according to the power adjustment instruction.

5. A load power distribution system for a hydrogen fuel cell emergency power supply vehicle, used to implement the load power distribution method for a hydrogen fuel cell emergency power supply vehicle as described in any one of claims 1 to 4, characterized in that: The system comprises: A feature extraction module is used to obtain the performance parameters of each battery stack in the hydrogen fuel cell emergency power supply vehicle when the battery stacks are running in parallel; and is used to generate an output characteristic curve of each battery stack based on the performance parameters and extract a battery stack performance feature vector; A grouping module, used to generate similarities between the battery stacks and obtain battery stack performance grouping results based on the battery stack performance feature vector; A prediction module is used to construct a stack performance prediction model based on the stack performance grouping results, predict the damage degree of the stack under overload conditions, obtain an overload safety threshold, and evaluate the health status of the stack, wherein the input of the stack performance prediction model is the stack overload degree, and the output is the stack damage degree; The power allocation module is used to generate a power allocation plan based on the load priority, the stack performance grouping result, the overload safety threshold and the stack health status, wherein the power allocation plan aims to meet the load power demand and minimize stack damage.

6. The system according to claim 5, characterized in that The feature extraction module comprises: The collection unit is used to collect the performance parameters of each battery stack in the hydrogen fuel cell emergency power supply vehicle, including temperature, humidity and catalyst activity; A generating unit, used for inputting the performance parameters into a fuel cell performance model to generate an output characteristic curve of each fuel cell; The extraction unit is used to extract features from the output characteristic curves of each battery stack to obtain a battery stack performance feature vector; wherein the battery stack performance feature vector includes key parameters that characterize the battery stack performance.

7. A computer device, characterized in that: include: A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the method according to any one of claims 1 to 4 by executing the computer instructions.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute the method according to any one of claims 1 to 4.

9. A computer program product, characterized in that The method comprises computer instructions for causing a computer to execute the method according to any one of claims 1 to 4.

Images