Fuel cell hydrogen supply system fault detection method, device and medium

Abstract

The invention discloses a fuel cell hydrogen supply system fault detection method and device and a medium, and relates to the technical field of hydrogen fuel cell fault detection. According to the method, far-field sensor data and near-field sensor data of a fuel cell hydrogen supply system are collected, vehicle operation conditions are identified, a neural network model is adopted to output a far-field sensor threshold value and a near-field sensor threshold value under the current working condition, and according to the far-field sensor data, the near-field sensor data and the corresponding threshold values, the fuel cell hydrogen supply system is determined according to the far-field sensor threshold value and the near-field sensor threshold value. Whether hydrogen supply abnormity, water management abnormity and heat management abnormity exist in the fuel cell hydrogen supply system or not is judged in sequence, and finally the fault detection result of the fuel cell hydrogen supply system is determined according to the hydrogen supply detection result, the water management detection result and the heat management detection result. According to the fuel cell hydrogen supply system fault detection method and device and the medium provided by the invention, the fault of the fuel cell hydrogen supply system can be comprehensively and accurately detected.

Description

Technical Field

[0001] The present application relates to the technical field of hydrogen fuel cell fault detection, and particularly to a method, device and medium for detecting faults in a hydrogen supply system of a fuel cell. Background Art

[0002] As a clean energy conversion device, hydrogen fuel cells have advantages such as high efficiency and zero emissions, and have received extensive attention in recent years. The hydrogen supply system is one of the core components of a hydrogen fuel cell, and its performance directly affects the overall efficiency and safety of the fuel cell. However, various faults may occur in the hydrogen supply system during actual operation, such as hydrogen leakage, insufficient flow rate, abnormal pressure, etc. If these faults cannot be detected and processed in time, it may lead to a decline in fuel cell performance, potential safety hazards and even shutdown.

[0003] Existing methods for detecting faults in a hydrogen supply system of a fuel cell mainly rely on a single sensor or a single monitoring parameter. For example, only rely on a hydrogen flow sensor or a pressure sensor for monitoring. This method has the following deficiencies: 1. Low accuracy. A single sensor is easily affected by factors such as environmental interference and its own faults, resulting in inaccurate detection results. 2. Limited scope. A single monitoring parameter cannot comprehensively reflect the state of the hydrogen supply system, and it is easy to miss some faults. Summary of the Invention

[0004] The purpose of the present application is to provide a method, device and medium for detecting faults in a hydrogen supply system of a fuel cell, which can achieve comprehensive and accurate detection of faults in the hydrogen supply system of the fuel cell.

[0005] To achieve the above purpose, the present application provides the following solutions.

[0006] In the first aspect, the present application provides a method for detecting faults in a hydrogen supply system of a fuel cell, including:

[0007] Collecting far-field sensor data and near-field sensor data of the hydrogen supply system of the fuel cell; the far-field sensor data includes: far-field hydrogen concentration data and far-field infrared thermal imaging data; the near-field sensor data includes: near-field pressure data, near-field flow rate data, near-field humidity data, near-field liquid level data and near-field temperature data;

[0008] Identifying the vehicle operating conditions, and using a neural network model to output the far-field sensor thresholds and near-field sensor thresholds under the current operating conditions; the far-field sensor thresholds include: far-field hydrogen concentration thresholds; the near-field sensor thresholds include: near-field pressure drop thresholds, near-field flow rate drop thresholds, near-field humidity thresholds, near-field liquid level thresholds and near-field temperature thresholds;

[0009] Judge whether there is abnormal hydrogen supply in the fuel cell hydrogen supply system according to the far-field hydrogen concentration data, near-field pressure data, near-field flow data and the corresponding thresholds, and obtain the hydrogen supply detection result;

[0010] If the hydrogen supply is normal, judge whether there is abnormal water management in the fuel cell hydrogen supply system according to the near-field humidity data, near-field liquid level data and the corresponding thresholds, and obtain the water management detection result;

[0011] If the water management is normal, judge whether there is abnormal thermal management in the fuel cell hydrogen supply system according to the far-field infrared thermal imaging data, near-field temperature data and the corresponding thresholds, and obtain the thermal management detection result;

[0012] Determine the fault detection result of the fuel cell hydrogen supply system according to the hydrogen supply detection result, water management detection result and thermal management detection result.

[0013] Optionally, the method for determining the neural network model includes:

[0014] Collect historical data of the vehicle under different working condition categories to construct a historical data set; the working condition categories include: start-up condition, acceleration condition, constant speed condition, idle condition and braking condition; the historical data includes: working condition data, sensor data and sensor thresholds; the working condition data includes: brake pedal opening, accelerator pedal opening and vehicle speed;

[0015] Construct a multi-layer perceptron neural network with the working condition category, working condition data and sensor data as inputs and the sensor threshold as the output, and train the multi-layer perceptron neural network using an optimization algorithm according to the historical data set to obtain a trained neural network model.

[0016] Optionally, judging whether there is abnormal hydrogen supply in the fuel cell hydrogen supply system according to the far-field hydrogen concentration data, near-field pressure data and near-field flow data and the corresponding thresholds, and obtaining the hydrogen supply detection result, includes:

[0017] Judge whether the far-field hydrogen concentration is too high according to the far-field hydrogen concentration data and the far-field hydrogen concentration threshold, and obtain the first judgment result;

[0018] If the first judgment result is yes, judge whether the pressure drop exceeds the threshold according to the near-field pressure data and the near-field pressure drop threshold, and obtain the second judgment result;

[0019] If the second judgment result is yes, obtain the detection result that there is potential hydrogen leakage in the fuel cell hydrogen supply system, and calculate the hydrogen leakage probability using the first fusion function;

[0020] If the second judgment result is no, judge whether the flow rate is too low according to the near-field flow data and the near-field flow rate drop threshold, and obtain the third judgment result;

[0021] If the third judgment result is yes, a detection result of potential hydrogen leakage or pressure sensor failure in the fuel cell hydrogen supply system is obtained, and the second fusion function is used to calculate the probability of hydrogen leakage or pressure sensor failure;

[0022] If the third judgment result is no, a detection result of potential hydrogen concentration sensor failure in the fuel cell hydrogen supply system is obtained, and the third fusion function is used to calculate the probability of hydrogen concentration sensor failure;

[0023] If the first judgment result is no, it is judged whether the flow measurement value is too low according to the near-field flow data and the near-field flow decrease threshold, and a fourth judgment result is obtained;

[0024] If the fourth judgment result is yes, it is judged whether the pressure measurement value is too high according to the near-field pressure data, and a fifth judgment result is obtained;

[0025] If the fifth judgment result is yes, a detection result of potential pipeline blockage in the fuel cell hydrogen supply system is obtained, and the fourth fusion function is used to calculate the pipeline blockage probability;

[0026] If the fifth judgment result is no, a detection result of potential poor opening of the hydrogen storage bottle or hydrogen concentration sensor failure in the fuel cell hydrogen supply system is obtained, and the fifth fusion function is used to calculate the probability of poor opening of the hydrogen storage bottle or hydrogen concentration sensor failure;

[0027] If the fourth judgment result is no, a detection result that the hydrogen supply in the fuel cell hydrogen supply system is normal is obtained.

[0028] Optionally, it is judged whether there is abnormal water management in the fuel cell hydrogen supply system according to the near-field humidity data, the near-field liquid level data and the corresponding thresholds, and a water management detection result is obtained, including:

[0029] It is judged whether the humidity measurement value is abnormal according to the near-field humidity data and the near-field humidity threshold, and a sixth judgment result is obtained; among them, the abnormal humidity measurement value includes too high humidity and too low humidity;

[0030] If the sixth judgment result is yes and the humidity is too high, it is judged whether the liquid level is too high according to the near-field liquid level data and the near-field liquid level threshold, and a seventh judgment result is obtained;

[0031] If the seventh judgment result is yes, a detection result of potential drain valve failure in the fuel cell hydrogen supply system is obtained, and the sixth fusion function is used to calculate the drain valve failure probability;

[0032] If the seventh judgment result is no, a detection result of potential humidifier failure in the fuel cell hydrogen supply system is obtained, and the seventh fusion function is used to calculate the humidifier failure probability;

[0033] If the result of the sixth determination is yes and the humidity is too low, a detection result that there is a potential humidifier failure or purge valve failure in the fuel cell hydrogen supply system is obtained, and the eighth fusion function is used to calculate the probability of humidifier failure or purge valve failure;

[0034] If the result of the sixth determination is no, a detection result that the water management in the fuel cell hydrogen supply system is normal is obtained.

[0035] Optionally, it is determined whether there is abnormal thermal management in the fuel cell hydrogen supply system according to the far-field infrared thermal imaging data and the near-field temperature data and the corresponding thresholds, and a thermal management detection result is obtained, including:

[0036] It is determined whether the temperature measurement value is too high according to the near-field temperature data and the near-field temperature threshold, and an eighth determination result is obtained;

[0037] If the result of the eighth determination is yes, it is determined whether there is local high temperature according to the far-field infrared thermal imaging data, and a ninth determination result is obtained;

[0038] If the result of the ninth determination is yes, a detection result that there is a potential coolant blockage in the fuel cell hydrogen supply system is obtained, and the ninth fusion function is used to calculate the probability of coolant blockage;

[0039] If the result of the ninth determination is no, a detection result that there is a potential cooling circulation pump failure in the fuel cell hydrogen supply system is obtained, and the tenth fusion function is used to calculate the probability of cooling circulation pump failure;

[0040] If the result of the eighth determination is no, a detection result that the thermal management in the fuel cell hydrogen supply system is normal is obtained.

[0041] Optionally, the hydrogen supply detection result includes: hydrogen leakage, pressure sensor failure, hydrogen concentration sensor failure, pipeline blockage, poor opening of the hydrogen storage bottle, and normal hydrogen supply;

[0042] The expression for calculating the hydrogen leakage probability using the first fusion function is:

[0043]

[0044] where, P leak is the hydrogen leakage probability, C is the actual far-field hydrogen concentration, C th is the far-field hydrogen concentration threshold, ΔP is the actual near-field pressure drop value, ΔP th is the near-field pressure drop threshold, t is the time, and α, β, γ, and δ are the first set of weight coefficients;

[0045] The expression for calculating the hydrogen leakage or pressure sensor failure probability using the second fusion function is:

[0046]

[0047] Among them, P potential is the probability of hydrogen leakage or pressure sensor failure, F is the actual near-field flow value, F th is the near-field flow decline threshold, and w1, w2, and w3 are the second set of weight coefficients;

[0048] The expression for calculating the probability of hydrogen concentration sensor failure using the third fusion function is:

[0049]

[0050] Among them, P sensor is the probability of hydrogen concentration sensor failure, and a and b are the third set of weight coefficients;

[0051] The expression for calculating the pipeline blockage probability using the fourth fusion function is:

[0052]

[0053] Among them, P block is the pipeline blockage probability, P is the actual near-field pressure, P high is the near-field pressure threshold, and c and d are the fourth set of weight coefficients;

[0054] The expression for calculating the probability of poor opening of the hydrogen storage bottle or hydrogen concentration sensor failure using the fifth fusion function is:

[0055]

[0056] Among them, P open is the probability of poor opening of the hydrogen storage bottle or hydrogen concentration sensor failure, P normal is the normal value of the near-field pressure, and e and f are the fifth set of weight coefficients.

[0057] Optionally, the water management detection results include: drain valve failure, humidifier failure, purge valve failure, and normal water management;

[0058] The expression for calculating the drain valve failure probability using the sixth fusion function is:

[0059]

[0060] Among them, P drain is the drain valve failure probability, H is the actual near-field humidity, H th is the near-field humidity threshold, L is the actual near-field liquid level, L th is the near-field liquid level threshold, and k1 and k2 are the sixth set of weight coefficients;

[0061] The expression for calculating the humidifier failure probability using the seventh fusion function is:

[0062]

[0063] Among them, P humidifier is the failure probability of the humidifier, L low is a specific value of the liquid level lower than the threshold, and m1 and m2 are the weight coefficients of the seventh group;

[0064] The expression for calculating the failure probability of the humidifier or the purge valve using the eighth fusion function is:

[0065]

[0066] Among them, P humidifier-purge is the failure probability of the humidifier or the purge valve, r1, s1 and t1 are the weight coefficients of the eighth group, and u is an adjustment parameter.

[0067] Optionally, the thermal management detection result includes: coolant blockage, cooling circulation pump failure, and normal thermal management;

[0068] The expression for calculating the coolant blockage probability using the ninth fusion function is:

[0069]

[0070] Among them, P coolant-block is the coolant blockage probability, T is the actual near-field temperature, T th is the near-field temperature threshold, I local is the local high-temperature characteristic parameter of the far-field infrared thermal imaging, and q1 and q2 are the weight coefficients of the ninth group;

[0071] The expression for calculating the cooling circulation pump failure probability using the tenth fusion function is:

[0072]

[0073] Among them, P pump is the cooling circulation pump failure probability, I global is the overall high-temperature characteristic parameter of the far-field infrared thermal imaging, and q3 and q4 are the weight coefficients of the tenth group.

[0074] In a second aspect, the present application provides a computer device, including: a memory, a processor, and a computer program stored on the memory and executable on the processor, and the processor executes the computer program to implement the above-mentioned fuel cell hydrogen supply system fault detection method.

[0075] In a third aspect, the present application provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the above-mentioned fuel cell hydrogen supply system fault detection method is implemented.

[0076] According to the specific embodiments provided in the present application, the following technical effects are disclosed in the present application:

[0077] The present application provides a method, device and medium for fault detection of a fuel cell hydrogen supply system. By collecting far-field sensor data and near-field sensor data of the fuel cell hydrogen supply system and identifying the vehicle operating conditions, a neural network model is used to output the far-field sensor threshold and near-field sensor threshold under the current conditions. According to the far-field sensor data, near-field sensor data and the corresponding thresholds, it is successively determined whether there are abnormalities in hydrogen supply, water management and thermal management in the fuel cell hydrogen supply system. Finally, the fault detection result of the fuel cell hydrogen supply system is determined according to the hydrogen supply detection result, water management detection result and thermal management detection result. The present application can automatically adjust the sensor thresholds according to different conditions of the fuel cell, adapt to various operating conditions, improve the accuracy and reliability of fault detection. Among them, the far-field sensor monitors the overall state, and the near-field sensor monitors local details, enabling comprehensive monitoring. The fusion of far-field sensor data and near-field sensor data can more accurately evaluate the fault probability. Description of the Drawings

[0078] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the following will briefly introduce the drawings required in the embodiments. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0079] Figure 1 It is a flowchart of the method for fault detection of the fuel cell hydrogen supply system provided by the present application.

[0080] Figure 2 It is a schematic diagram of the fault detection process of the fuel cell hydrogen supply system provided by the present application. Detailed Embodiments

[0081] The following will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present application.

[0082] In the monitoring of fuel cell systems, near-field sensors and far-field sensors each have their own characteristics and advantages. Near-field sensors are deployed in a small area near the device or system to be monitored, focusing on obtaining fine parameters of key parts of the device, providing a direct basis for the local state assessment of the system. Far-field sensors are arranged at a relatively far position from the object to be monitored, used to obtain information related to the overall state of the system or the surrounding environment, and help to grasp the operating state of the system in a large range and its interaction with the surrounding environment.

[0083] The linkage of near-field and far-field sensors is a technical strategy in monitoring and control systems that comprehensively uses near-field sensors and far-field sensors, enabling them to cooperate with each other and interact with information, so as to achieve more comprehensive and accurate monitoring and control. Near-field and far-field sensors achieve data interaction and collaborative work through a fusion function. The fusion function takes the data of near-field and far-field sensors as inputs for fusion, and processes these inputs through specific algorithms to calculate the fault probability.

[0084] The purpose of this application is to provide a fuel cell hydrogen supply system fault detection method, device, and medium. Through the data fusion of near-field and far-field sensors, comprehensive and accurate detection of faults in the fuel cell hydrogen supply system can be achieved.

[0085] To make the above objects, features, and advantages of this application more obvious and understandable, the following further details this application in conjunction with the drawings and specific embodiments.

[0086] In an exemplary embodiment, this application provides a fuel cell hydrogen supply system fault detection method. In the embodiment of this application, as Figure 1 shown, the method includes the following steps 1 to step 6.

[0087] Step 1: Collect far-field sensor data and near-field sensor data of the fuel cell hydrogen supply system. The far-field sensor data includes: far-field hydrogen concentration data and far-field infrared thermal imaging data. The near-field sensor data includes: near-field pressure data, near-field flow data, near-field humidity data, near-field liquid level data, and near-field temperature data.

[0088] Among them, the far-field hydrogen concentration sensor is used to obtain the hydrogen concentration data in the environment of the fuel cell hydrogen supply system, the far-field infrared thermal imaging sensor is used to obtain the overall thermal distribution data of the fuel cell hydrogen supply system, the near-field pressure sensor is used to obtain the internal hydrogen pressure value of the hydrogen supply system, the near-field flow sensor is used to obtain the internal hydrogen flow value of the hydrogen supply system, the near-field humidity sensor is used to obtain the hydrogen humidity value entering the fuel cell anode, the near-field liquid level sensor is used to obtain the liquid level height value in the fuel cell steam-water separator, and the near-field temperature sensor is used to obtain the temperature value of the hydrogen supply system.

[0089] Step 2: Identify the vehicle operating conditions and use a neural network model to output the far-field sensor threshold and near-field sensor threshold under the current conditions. The far-field sensor threshold includes: the far-field hydrogen concentration threshold. The near-field sensor threshold includes: the near-field pressure drop threshold, the near-field flow rate drop threshold, the near-field humidity threshold, the near-field liquid level threshold, and the near-field temperature threshold.

[0090] Among them, the categories of vehicle operating conditions include: start-up condition, acceleration condition, constant-speed condition, idle condition, and braking condition. The vehicle operating conditions are jointly determined by the brake pedal opening, the accelerator pedal opening, and the vehicle speed. Specifically: Monitor the vehicle speed (V) in real time and determine whether the vehicle is in a stationary state, that is, whether the speed (V) is 0 km / h. If the vehicle is in a stationary state, further detect the accelerator pedal opening (APO). If the APO is greater than 0%, it is determined that the current condition is "start-up". If the APO is 0%, it is determined that the current condition is "idle". If the vehicle speed (V) is not 0 km / h, further detect the brake pedal opening (BPO). If the BPO is greater than 0%, it is determined that the current condition is "braking". If the BPO is 0% and the APO remains stable, further monitor the change in the vehicle speed (V). If the vehicle speed remains relatively stable, it is determined that the current condition is "constant-speed". If the vehicle speed (V) is increasing, it is determined that the current condition is "acceleration".

[0091] Preferably, the method for determining the neural network model includes: collecting historical data of the vehicle under different condition categories to construct a historical data set; constructing a multi-layer perceptron neural network with the condition category, condition data, and sensor data as inputs and the sensor threshold as the output, and training the multi-layer perceptron neural network using an optimization algorithm according to the historical data set to obtain a trained neural network model. The historical data includes: condition data, sensor data, and sensor thresholds. The condition data includes: brake pedal opening, accelerator pedal opening, and vehicle speed.

[0092] Specifically, vehicle operation data is collected, including the brake pedal opening, accelerator pedal opening, and vehicle speed. Based on the vehicle operation data, the vehicle operation conditions are judged, and for the five conditions of starting, accelerating, cruising, idling, and braking, one-hot encoding is performed. For example, the starting condition is encoded as [1, 0, 0, 0, 0], the accelerating condition is [0, 1, 0, 0, 0], the cruising condition is [0, 0, 1, 0, 0], the idling condition is [0, 0, 0, 1, 0], and the braking condition is [0, 0, 0, 0, 1]. For each condition, relevant sensor data is collected, and the collected data is preprocessed, including removing outliers and data normalization. A Multilayer Perceptron (MLP) neural network is constructed, which consists of an input layer, a hidden layer, and an output layer. The input layer inputs the data training sample set, data validation sample set, and data test sample set. The hidden layer learns and trains according to the data training sample set to obtain a trained neural network model. The current condition is identified based on the real-time monitored brake pedal opening, accelerator pedal opening, and vehicle speed, and the real-time monitored condition data and the condition category encoding (in one-hot encoding form) are input into the trained neural network model, and the model will output the thresholds of each sensor under the current condition.

[0093] For each condition, relevant sensor data is collected, including far-field hydrogen concentration data, far-field infrared thermal imaging data, near-field pressure data, near-field flow data, near-field humidity data, near-field liquid level data, etc. The collected data is preprocessed, including removing outliers (for example, using the 3-sigma method, if a data point exceeds the range of the mean ± 3 standard deviations, it is regarded as an outlier for correction or deletion) and data normalization, mapping the data to the [0, 1] interval, and the normalization formula is where x is the original data, x min and x max are the minimum and maximum values of this feature, respectively.

[0094] The multi-layer perceptron (MLP) neural network consists of an input layer, hidden layers, and an output layer. The number of nodes in the input layer is determined by the input data, including the operating condition category (using one-hot encoding, the starting condition is [1,0,0,0,0], the acceleration condition is [0,1,0,0,0], the constant speed condition is [0,0,1,0,0], the idle condition is [0,0,0,1,0], the braking condition is [0,0,0,0,1], a total of 5 nodes), key operating condition data (brake pedal opening, accelerator pedal opening, and vehicle speed, a total of 3 nodes), and key sensor data (far-field hydrogen concentration data, far-field infrared thermal imaging data, near-field pressure data, near-field flow data, near-field humidity data, near-field liquid level data, near-field temperature data, a total of 7 nodes). So the number of nodes in the input layer is 15. The hidden layers are set to 2 layers, with 16 nodes in the first hidden layer and 8 nodes in the second hidden layer. The number of nodes in the output layer is the same as the number of sensors whose thresholds need to be adjusted. The activation function for the hidden layers selects the ReLU (Rectified Linear Unit) function ReLU(x1) = max(0,x1), and the activation function for the output layer selects the Softplus function Softplus(x2) = ln(1 + e x2 ).

[0095] Collect a large amount of historical data under different operating conditions, including the pedal opening, vehicle speed in the above-mentioned operating condition judgment basis, and the corresponding sensor data and ideal thresholds. Divide the data according to the ratio of 70% for the training set, 20% for the validation set, and 10% for the test set.

[0096] Use the stochastic gradient descent (SGD) algorithm and its variants (such as the Adam optimization algorithm) to optimize the weights and biases of the neural network. The loss function selects the Mean Square Error (MSE) function where n is the number of training samples, y i is the true sensor threshold of the i-th sample, is the sensor threshold predicted by the neural network model for the i-th sample. Set appropriate learning rates (such as 0.001) and the number of training iterations (such as 2000 times). During training, use the Early Stopping method to prevent overfitting, that is, when the loss function value on the validation set no longer decreases or the decrease amplitude is less than a certain threshold (such as 0.0001), stop training. Adopt the L2 regularization (weight decay) method and add a regularization term to the loss function (where λ1 is the regularization coefficient, w i is the weight), to limit the model complexity and improve the generalization ability of the model.

[0097] Step 3: Determine whether there is an abnormal hydrogen supply in the fuel cell hydrogen supply system based on the far-field hydrogen concentration data, near-field pressure data, and near-field flow rate data and their corresponding thresholds, and obtain the hydrogen supply detection result.

[0098] Step 3 specifically includes: judging whether the far-field hydrogen concentration is too high according to the far-field hydrogen concentration data and the far-field hydrogen concentration threshold to obtain a first judgment result; if the first judgment result is yes, then judging whether the pressure drop exceeds the threshold according to the near-field pressure data and the near-field pressure drop threshold to obtain a second judgment result; if the second judgment result is yes, then obtain the detection result that there is potential hydrogen leakage in the fuel cell hydrogen supply system, and calculate the hydrogen leakage probability using the first fusion function; if the second judgment result is no, then judge whether the flow rate is too low according to the near-field flow rate data and the near-field flow rate drop threshold to obtain a third judgment result; if the third judgment result is yes, then obtain the detection result that there is potential hydrogen leakage or pressure sensor failure in the fuel cell hydrogen supply system, and calculate the hydrogen leakage or pressure sensor failure probability using the second fusion function; if the third judgment result is no, then obtain the detection result that there is potential hydrogen concentration sensor failure in the fuel cell hydrogen supply system, and calculate the hydrogen concentration sensor failure probability using the third fusion function; if the first judgment result is no, then judge whether the flow rate measurement value is too low according to the near-field flow rate data and the near-field flow rate drop threshold to obtain a fourth judgment result; if the fourth judgment result is yes, then judge whether the pressure measurement value is too high according to the near-field pressure data to obtain a fifth judgment result; if the fifth judgment result is yes, then obtain the detection result that there is potential pipeline blockage in the fuel cell hydrogen supply system, and calculate the pipeline blockage probability using the fourth fusion function; if the fifth judgment result is no, then obtain the detection result that there is potential poor opening of the hydrogen storage bottle or hydrogen concentration sensor failure in the fuel cell hydrogen supply system, and calculate the poor opening of the hydrogen storage bottle or hydrogen concentration sensor failure probability using the fifth fusion function; if the fourth judgment result is no, then obtain the detection result that the hydrogen supply in the fuel cell hydrogen supply system is normal.

[0099] First, set the far-field hydrogen concentration threshold C th , the near-field pressure drop threshold ΔP th and the near-field flow rate drop threshold F th . Compare the far-field hydrogen concentration data with the near-field pressure and flow rate data. If the far-field hydrogen concentration data is higher than C th , then further judge the near-field pressure data. If the near-field pressure data drops by more than ΔP th , at this time, the far-field hydrogen concentration data and the near-field pressure data confirm each other, indicating that there is a high possibility of hydrogen leakage in the fuel cell hydrogen supply system. The data fusion model calculates the hydrogen leakage probability P leak = f(C - C th , ΔP - ΔPth ), where C is the actual far - field hydrogen concentration, ΔP is the actual near - field pressure drop value, and f is the first fusion function determined based on experience or a machine - learning model.

[0100] Preferably, the following is an example of a fusion function to calculate the hydrogen leakage probability:

[0101]

[0102] where, P leak is the hydrogen leakage probability, C is the actual far - field hydrogen concentration, C th is the far - field hydrogen concentration threshold, ΔP is the actual near - field pressure drop value, ΔP th is the near - field pressure drop threshold. and respectively represent the deviation degrees of the far - field hydrogen concentration and the near - field pressure drop value relative to the threshold. and respectively represent the change trends (derivative with respect to time t) of the far - field hydrogen concentration and the near - field pressure drop value. α, β, γ, δ are the first set of weight coefficients, which can be adjusted according to the actual situation and experience to reflect the importance of the deviation degree and the change trend in the calculation of the fault probability.

[0103] Again, if the far - field hydrogen concentration data is higher than C th but the near - field pressure data does not drop by more than ΔP th , then the near - field flow data is further analyzed. If the near - field flow data drops by more than F th , the data fusion model comprehensively calculates the potential fault probability, that is, the hydrogen leakage or pressure sensor fault probability P potential , considering that the pressure does not drop significantly, at this time the flow data has a relatively high weight in the calculation, P potential = g(C - C th , ΔP, F - F th ), where F is the actual near - field flow rate and g is the second fusion function.

[0104] Preferably, the following is an example of a fusion function to calculate the hydrogen leakage or pressure sensor fault probability:

[0105]

[0106] where, P potential is the hydrogen leakage or pressure sensor fault probability, C is the far - field hydrogen concentration value, C th is the far - field hydrogen concentration threshold, ΔP is the actual near - field pressure drop value, ΔP th is the near - field pressure drop threshold, F is the actual near - field flow rate value, F this the near-field flow rate drop threshold. w1, w2, and w3 are the second set of weight coefficients, representing the weights of the deviation degree of the far-field hydrogen concentration, the deviation degree of the near-field pressure drop value, and the deviation degree of the near-field flow rate drop value, respectively. Since the weight of the flow rate data is relatively high when the pressure does not drop significantly, it can be set that w3 > w1 and w3 > w2. The specific values of the weight coefficients can be adjusted according to the actual situation and experience.

[0107] If the near-field flow rate data does not drop by more than F th , then the data fusion model combines the far-field hydrogen concentration data and the near-field flow rate data to determine whether there is a hydrogen concentration sensor failure, and the hydrogen concentration sensor failure probability P sensor = h(C - C th ), where h is the third fusion function.

[0108] Preferably, the following is an example of a fusion function to calculate the hydrogen concentration sensor failure probability:

[0109]

[0110] Among them, P sensor is the hydrogen concentration sensor failure probability, C is the actual far-field hydrogen concentration, C th is the far-field hydrogen concentration threshold, F is the actual near-field flow rate, and F th is the near-field flow rate drop threshold. a and b are the third set of weight coefficients, representing the weights of the deviation degree of the far-field hydrogen concentration and the deviation degree of the near-field flow rate drop value, respectively.

[0111] If the far-field hydrogen concentration data does not show abnormalities, the judgment is mainly based on the near-field flow rate and pressure data.

[0112] When the near-field flow rate data is lower than F th , if the near-field pressure data is higher than the set pressure threshold P high , the data fusion model calculates the pipeline blockage probability P block = k(F - F th , P - P high ), where P is the actual near-field pressure and k is the fourth fusion function.

[0113] Preferably, the following is an example of a fusion function to calculate the pipeline blockage probability:

[0114]

[0115] Among them, P block is the pipeline blockage probability, F is the actual near-field flow rate, F th is the near-field flow rate drop threshold, P is the actual near-field pressure, and P highis a set pressure threshold, i.e., the near-field pressure threshold, and c and d are the fourth set of weight coefficients, respectively representing the weights of the deviation degree of the near-field flow rate decrease value and the degree of the near-field pressure being higher than the threshold.

[0116] If the near-field pressure data is not higher than P high , the data fusion model comprehensively judges whether there is a problem with the opening of the hydrogen storage bottle or a hydrogen concentration sensor failure based on the near-field flow rate and pressure data, and calculates the failure probability P open = m(F - F th , P), where m is the fifth fusion function.

[0117] Preferably, the following is an example of a fusion function to calculate the failure probability of the opening of the hydrogen storage bottle or the hydrogen concentration sensor:

[0118]

[0119] where P open is the failure probability of the opening of the hydrogen storage bottle or the hydrogen concentration sensor, F is the actual near-field flow rate, F th is the near-field flow rate decrease threshold, P is the actual near-field pressure, and P normal is the near-field pressure value under normal conditions, i.e., the near-field pressure normal value. e and f are the fifth set of weight coefficients, respectively representing the weights of the deviation degree of the near-field flow rate decrease value and the deviation degree of the near-field pressure from the normal value.

[0120] Step 4: If the hydrogen supply is normal, judge whether there is abnormal water management in the fuel cell hydrogen supply system according to the near-field humidity data and near-field liquid level data and the corresponding thresholds, and obtain the water management detection result.

[0121] Step 4 specifically includes: judging whether the humidity measurement value is abnormal according to the near-field humidity data and the near-field humidity threshold to obtain the sixth judgment result; among them, the abnormal humidity measurement value includes too high humidity and too low humidity; if the sixth judgment result is yes and the humidity is too high, judge whether the liquid level is too high according to the near-field liquid level data and the near-field liquid level threshold to obtain the seventh judgment result; if the seventh judgment result is yes, obtain the detection result that there is a potential drain valve failure in the fuel cell hydrogen supply system, and calculate the drain valve failure probability using the sixth fusion function; if the seventh judgment result is no, obtain the detection result that there is a potential humidifier failure in the fuel cell hydrogen supply system, and calculate the humidifier failure probability using the seventh fusion function; if the sixth judgment result is yes and the humidity is too low, obtain the detection result that there is a potential humidifier failure or purge valve failure in the fuel cell hydrogen supply system, and calculate the humidifier failure or purge valve failure probability using the eighth fusion function; if the sixth judgment result is no, obtain the detection result that the water management in the fuel cell hydrogen supply system is normal.

[0122] First, set the near-field humidity threshold H based on the above neural network modelth and the near - field liquid level threshold L th .

[0123] If the near - field humidity measurement value exceeds H th , when the near - field liquid level sensor measurement value is higher than L th , the data fusion model combines the humidity and liquid level data to calculate the drainage valve failure probability P drain = n(H - H th , L - L th ), where H is the actual near - field humidity, L is the actual near - field liquid level, and n is the sixth fusion function, indicating a relatively high probability of drainage valve failure.

[0124] Preferably, the following is an example of a fusion function to calculate the drainage valve failure:

[0125]

[0126] Among them, H is the actual near - field humidity, H th is the near - field humidity threshold, L is the actual near - field liquid level, and L th is the near - field liquid level threshold. k1 and k2 are the weight coefficients of the sixth group, respectively characterizing the weights of the humidity deviation degree and the liquid level deviation degree.

[0127] If the near - field liquid level sensor measurement value is lower than L t h, then calculate the humidifier failure probability P humidifier = p(H - H th , L), where p is the seventh fusion function, indicating a relatively high probability of humidifier failure.

[0128] Preferably, the following is an example of a fusion function to calculate the humidifier failure probability:

[0129]

[0130] Among them, P humidifier is the humidifier failure probability, H is the actual near - field humidity, H th is the humidity threshold, and L low is a specific value when the liquid level is lower than the threshold. m1 and m2 are the weight coefficients of the seventh group, respectively characterizing the weights of the humidity deviation degree and the liquid level deviation degree (in the direction below the threshold).

[0131] If the near - field humidity measurement value is lower than H th , the data fusion model judges the humidifier failure or purge valve failure probability P humidifier-purge = q(H) only according to the humidity data, because at this time the liquid level data has relatively little contribution to the judgment of these two failures.

[0132] Preferably, the following is an example of a fusion function for calculating the probability of a humidifier failure or a purge valve failure:

[0133]

[0134] Where P humidifier-purge is the probability of a humidifier failure or a purge valve failure, H is the actual near-field humidity, and H th is the humidity threshold. r1, s1, and t1 are the weight coefficients of the eighth group, and u is an adjustment parameter.

[0135] Step 5: If the water management is normal, determine whether there is a thermal management abnormality in the fuel cell hydrogen supply system based on the far-field infrared thermal imaging data, the near-field temperature data, and the corresponding thresholds, and obtain the thermal management detection result.

[0136] Step 5 specifically includes: determining whether the temperature measurement value is too high based on the near-field temperature data and the near-field temperature threshold to obtain the eighth judgment result; if the eighth judgment result is yes, determining whether there is a local high temperature based on the far-field infrared thermal imaging data to obtain the ninth judgment result; if the ninth judgment result is yes, obtaining the detection result that there is a potential coolant blockage in the fuel cell hydrogen supply system, and calculating the coolant blockage probability using the ninth fusion function; if the ninth judgment result is no, obtaining the detection result that there is a potential cooling circulation pump failure in the fuel cell hydrogen supply system, and calculating the cooling circulation pump failure probability using the tenth fusion function; if the eighth judgment result is no, obtaining the detection result that the thermal management in the fuel cell hydrogen supply system is normal.

[0137] First, set the near-field temperature threshold T th .

[0138] If the near-field temperature sensor detects that the temperature exceeds T th , further assist in the judgment based on the far-field infrared thermal imaging data.

[0139] If the infrared thermal imaging data shows that there is a local high temperature in the fuel cell hydrogen supply system, the data fusion model calculates the coolant blockage failure probability P coolant-block = r(T - T th , I local ), where T is the actual near-field temperature, I local is the local high temperature characteristic parameter of the far-field infrared thermal imaging, and r is the ninth fusion function.

[0140] Preferably, the following is an example of a fusion function for calculating the coolant blockage failure probability:

[0141]

[0142] Where P coolant-blockLet \(P\) be the probability of coolant blockage, \(T\) be the actual near-field temperature, and \(T\) th be the near-field temperature threshold, and \(I\) local be the local high-temperature characteristic parameter of the far-field infrared thermal imaging. \(q1\) and \(q2\) are the weight coefficients of the ninth group, respectively representing the weights of the temperature deviation degree and the local high-temperature characteristic parameter.

[0143] If the infrared thermal imaging data shows that there is an overall high temperature in the fuel cell hydrogen supply system, calculate the failure probability \(P\) of the cooling circulation pump pump = s(T - T th , I global ), where \(I\) global is the overall high-temperature characteristic parameter of the far-field infrared thermal imaging, and \(s\) is the tenth fusion function.

[0144] Preferably, the following is an example of a fusion function to calculate the failure probability of the cooling circulation pump:

[0145]

[0146] where \(P\) pump is the failure probability of the cooling circulation pump, \(T\) is the actual near-field temperature, \(T\) th is the near-field temperature threshold, and \(I\) global is the overall high-temperature characteristic parameter of the far-field infrared thermal imaging. \(q3\) and \(q4\) are the weight coefficients of the tenth group, respectively representing the weights of the temperature deviation degree and the overall high-temperature characteristic parameter.

[0147] Step 6: Determine the failure detection result of the fuel cell hydrogen supply system according to the hydrogen supply detection result, water management detection result, and thermal management detection result.

[0148] For the convenience of understanding the solution of this application, the following is illustrated by a specific example. As Figure 2 shown, the fuel cell hydrogen supply system failure detection process provided by this application includes the following steps:

[0149] S10: Collect the data of the far-field hydrogen concentration sensor, far-field infrared thermal imaging sensor, near-field pressure sensor, flow sensor, humidity sensor, liquid level sensor, and temperature sensor respectively.

[0150] S11: Judge whether the hydrogen concentration in the environment of the fuel cell hydrogen supply system is too high according to the far-field hydrogen concentration data. If so, enter S12; if not, enter S17.

[0151] S12: Judge whether the hydrogen pressure in the fuel cell hydrogen supply system is too high according to the near-field pressure sensor data. If so, enter S13; if not, enter S14.

[0152] S13: Calculate the probability of hydrogen leakage in the fuel cell hydrogen supply system.

[0153] S14: Determine whether the hydrogen flow rate in the fuel cell hydrogen supply system is too low based on the near-field flow sensor data. If so, proceed to S15; if not, proceed to S16.

[0154] S15: Calculate the probabilities of hydrogen leakage and near-field pressure sensor failure in the fuel cell hydrogen supply system.

[0155] S16: Calculate the probability of hydrogen concentration sensor failure in the fuel cell hydrogen supply system.

[0156] S17: Determine whether the hydrogen flow rate in the fuel cell hydrogen supply system is too low based on the near-field flow sensor data. If so, proceed to S18; if not, proceed to S21.

[0157] S18: Determine whether the hydrogen pressure in the fuel cell hydrogen supply system is too high based on the near-field pressure sensor data. If so, proceed to S19; if not, proceed to S20.

[0158] S19: Calculate the probability of hydrogen supply pipeline blockage in the fuel cell hydrogen supply system.

[0159] S20: Calculate the probabilities of poor opening of the hydrogen storage bottle or hydrogen concentration sensor failure in the fuel cell hydrogen supply system.

[0160] S21: Determine whether the humidity of the hydrogen entering the anode of the fuel cell is too high or too low based on the near-field humidity sensor data. If it is too high, proceed to S22; if it is too low, proceed to S23.

[0161] S22: Obtain the judgment result that the humidity of the hydrogen entering the anode of the fuel cell is too high.

[0162] S23: Obtain the judgment result that the humidity of the hydrogen entering the anode of the fuel cell is too low.

[0163] S24: Determine whether the liquid level of the steam-water separator in the fuel cell hydrogen supply system is too high based on the near-field liquid level sensor data. If so, proceed to S26; if not, proceed to S27.

[0164] S25: Calculate the probabilities of humidifier failure or purge valve failure in the fuel cell hydrogen supply system.

[0165] S26: Calculate the probability of drain valve failure in the fuel cell hydrogen supply system.

[0166] S27: Calculate the probability of humidifier failure in the fuel cell hydrogen supply system.

[0167] S28: Determine whether the hydrogen temperature in the fuel cell hydrogen supply system is too high based on the near-field temperature sensor data. If so, proceed to S29; if not, end the fault judgment process.

[0168] S29: Determine whether there is overall high temperature or local high temperature in the hydrogen supply pipeline of the fuel cell based on the temperature distribution image obtained by the infrared thermal imaging sensor. If it is local high temperature, go to S30; if it is overall high temperature, go to 31.

[0169] S30: Obtain the detection result that there is local high temperature in the fuel cell hydrogen supply system.

[0170] S31: Obtain the detection result that there is overall high temperature in the fuel cell hydrogen supply system.

[0171] S32: Calculate the probability of coolant blockage in the fuel cell.

[0172] S33: Calculate the probability of cooling system pump failure in the fuel cell.

[0173] In summary, the present application provides a method for detecting faults in a fuel cell hydrogen supply system. Based on the linkage of near-field and far-field sensors with adaptive thresholds, the overall state of the system is monitored by the far-field sensor, and local details are monitored by the near-field sensor to achieve comprehensive monitoring. By combining the data of the far-field sensor and the near-field sensor, the fault area can be more accurately located, and the reliability of the detection result can be improved.

[0174] In an exemplary embodiment, the present application further provides a computer device, including a memory and a processor. A computer program is stored in the memory, and when the processor executes the computer program, the steps in the above method embodiments are implemented.

[0175] In an exemplary embodiment, the present application further provides a computer-readable storage medium storing a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0176] In an exemplary embodiment, the present application further provides a computer program product including a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0177] In the present application, all actions of obtaining signals, information, or data are carried out on the premise of complying with the corresponding data protection regulations and policies of the country where the location is located and obtaining authorization from the owner of the corresponding device. It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in the present application are all information and data that have been authorized by the user or fully authorized by all parties, and the collection, use, and processing of relevant data need to comply with relevant laws and regulations.

[0178] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, database, or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memories. Non-volatile memories can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memories can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.

[0179] The databases involved in the embodiments provided in the present application can include at least one of relational databases and non-relational databases. Non-relational databases can include distributed databases based on blockchain, etc., without limitation. The processors involved in the embodiments provided in the present application can be general-purpose processors, central processors, graphics processors, digital signal processors, programmable logics, data processing logics based on quantum computing, etc., without limitation.

[0180] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered to be within the scope described in this specification.

[0181] In this text, specific examples are used to illustrate the principle and implementation manner of the present application. The description of the above embodiments is only used to help understand the method and its core idea of the present application; at the same time, for those of ordinary skill in the art, according to the idea of the present application, there will be changes in the specific implementation manner and application scope. In summary, the content of this specification should not be construed as a limitation to the present application.

Claims

1. A method for detecting a fault in a fuel cell hydrogen supply system, characterized in that: include: Collect far-field sensor data and near-field sensor data of the fuel cell hydrogen supply system; The far-field sensor data includes: far-field hydrogen concentration data and far-field infrared thermal imaging data; the near-field sensor data includes: near-field pressure data, near-field flow data, near-field humidity data, near-field liquid level data and near-field temperature data; Identify the vehicle operating conditions, and use a neural network model to output the far-field sensor threshold and near-field sensor threshold under the current operating conditions; the far-field sensor threshold includes: a far-field hydrogen concentration threshold; the near-field sensor threshold includes: a near-field pressure drop threshold, a near-field flow drop threshold, a near-field humidity threshold, a near-field liquid level threshold, and a near-field temperature threshold; According to the far-field hydrogen concentration data, the near-field pressure data, the near-field flow data and the corresponding thresholds, it is determined whether there is a hydrogen supply abnormality in the fuel cell hydrogen supply system, and a hydrogen supply detection result is obtained; If the hydrogen supply is normal, determine whether there is water management abnormality in the fuel cell hydrogen supply system based on the near-field humidity data, the near-field liquid level data and the corresponding thresholds, and obtain the water management detection result; If the water management is normal, the fuel cell hydrogen supply system is judged to have thermal management abnormality based on the far-field infrared thermal imaging data, the near-field temperature data and the corresponding thresholds, and the thermal management detection result is obtained; The fault detection result of the fuel cell hydrogen supply system is determined based on the hydrogen supply detection result, the water management detection result and the thermal management detection result.

2. The method for detecting a fault in a fuel cell hydrogen supply system according to claim 1, characterized in that: The method for determining the neural network model includes: Collect historical data of the vehicle under different working condition categories to construct a historical data set; the working condition categories include: starting working condition, acceleration working condition, constant speed working condition, idling working condition and braking working condition; the historical data include: working condition data, sensor data and sensor threshold; the working condition data include: brake pedal opening, accelerator pedal opening and vehicle speed; Taking the working condition category, working condition data and sensor data as input and the sensor threshold as output, a multi-layer perceptron neural network is constructed. Based on the historical data set, the multi-layer perceptron neural network is trained using an optimization algorithm to obtain a trained neural network model.

3. The method for detecting a fault in a fuel cell hydrogen supply system according to claim 1, characterized in that: According to the far-field hydrogen concentration data, the near-field pressure data, the near-field flow data and the corresponding thresholds, it is judged whether there is a hydrogen supply abnormality in the fuel cell hydrogen supply system, and the hydrogen supply detection result is obtained, including: Determine whether the far-field hydrogen concentration is too high according to the far-field hydrogen concentration data and the far-field hydrogen concentration threshold, and obtain a first determination result; If the first judgment result is yes, judging whether the pressure drop exceeds the threshold value according to the near-field pressure data and the near-field pressure drop threshold value, and obtaining a second judgment result; If the second judgment result is yes, a detection result indicating that there is a potential hydrogen leakage in the fuel cell hydrogen supply system is obtained, and the first fusion function is used to calculate the hydrogen leakage probability; If the second judgment result is no, judging whether the flow is too low according to the near-field flow data and the near-field flow decrease threshold, and obtaining a third judgment result; If the third judgment result is yes, a detection result of a potential hydrogen leakage or pressure sensor failure in the fuel cell hydrogen supply system is obtained, and the probability of hydrogen leakage or pressure sensor failure is calculated using the second fusion function; If the third judgment result is no, a detection result of a potential hydrogen concentration sensor failure in the fuel cell hydrogen supply system is obtained, and the third fusion function is used to calculate the hydrogen concentration sensor failure probability; If the first judgment result is no, judging whether the flow measurement value is too low according to the near-field flow data and the near-field flow decrease threshold, and obtaining a fourth judgment result; If the fourth judgment result is yes, judging whether the pressure measurement value is too high according to the near-field pressure data, and obtaining a fifth judgment result; If the fifth judgment result is yes, a detection result indicating that there is potential pipeline blockage in the fuel cell hydrogen supply system is obtained, and the pipeline blockage probability is calculated using the fourth fusion function; If the fifth judgment result is no, a detection result of a potential hydrogen storage bottle opening failure or a hydrogen concentration sensor failure in the fuel cell hydrogen supply system is obtained, and the fifth fusion function is used to calculate the probability of the hydrogen storage bottle opening failure or the hydrogen concentration sensor failure; If the fourth judgment result is no, a detection result indicating that the hydrogen supply in the fuel cell hydrogen supply system is normal is obtained.

4. The method for detecting a fault in a fuel cell hydrogen supply system according to claim 3, characterized in that: Based on the near-field humidity data, the near-field liquid level data and the corresponding thresholds, it is determined whether there is water management abnormality in the fuel cell hydrogen supply system, and the water management detection results are obtained, including: Determine whether the humidity measurement value is abnormal according to the near-field humidity data and the near-field humidity threshold, and obtain a sixth determination result; wherein the abnormal humidity measurement value includes excessively high humidity and excessively low humidity; If the sixth judgment result is yes and the humidity is too high, whether the liquid level is too high is judged according to the near-field liquid level data and the near-field liquid level threshold, to obtain a seventh judgment result; If the seventh judgment result is yes, a detection result of a potential drain valve failure in the fuel cell hydrogen supply system is obtained, and the drain valve failure probability is calculated using the sixth fusion function; If the seventh judgment result is no, a detection result indicating a potential humidifier failure in the fuel cell hydrogen supply system is obtained, and the seventh fusion function is used to calculate the humidifier failure probability; If the sixth judgment result is yes and the humidity is too low, a detection result indicating a potential humidifier failure or a purge valve failure in the fuel cell hydrogen supply system is obtained, and the eighth fusion function is used to calculate the probability of a humidifier failure or a purge valve failure; If the sixth judgment result is no, the detection result that the water management in the fuel cell hydrogen supply system is normal is obtained.

5. The method for detecting a fault in a fuel cell hydrogen supply system according to claim 4, characterized in that: Based on the far-field infrared thermal imaging data, near-field temperature data and corresponding thresholds, it is determined whether there is thermal management anomaly in the fuel cell hydrogen supply system, and the thermal management detection results are obtained, including: determining whether the temperature measurement value is too high according to the near-field temperature data and the near-field temperature threshold, to obtain an eighth determination result; If the eighth judgment result is yes, judging whether there is local high temperature according to the far-field infrared thermal imaging data, and obtaining a ninth judgment result; If the ninth judgment result is yes, a detection result indicating that there is potential coolant blockage in the fuel cell hydrogen supply system is obtained, and the coolant blockage probability is calculated using the ninth fusion function; If the ninth judgment result is no, a detection result of a potential cooling circulation pump failure in the fuel cell hydrogen supply system is obtained, and the tenth fusion function is used to calculate the cooling circulation pump failure probability; If the eighth judgment result is no, then a detection result indicating that the thermal management in the fuel cell hydrogen supply system is normal is obtained.

6. The method for detecting a fault in a fuel cell hydrogen supply system according to claim 3, characterized in that: The hydrogen supply detection results include: hydrogen leakage, pressure sensor failure, hydrogen concentration sensor failure, pipeline blockage, hydrogen storage bottle opening failure, and normal hydrogen supply; The expression for calculating the hydrogen leakage probability using the first fusion function is: Among them, P leak is the hydrogen leakage probability, C is the actual far-field hydrogen concentration, C th is the far-field hydrogen concentration threshold, ΔP is the actual near-field pressure drop value, ΔP th is the near-field pressure drop threshold, t is the time, α, β, γ and δ are the first set of weight coefficients; The expression for calculating the probability of hydrogen leakage or pressure sensor failure using the second fusion function is: Among them, P potential is the probability of hydrogen leakage or pressure sensor failure, F is the actual near-field flow value, and F th is the near-field flow rate drop threshold, w1, w2 and w3 are the second set of weight coefficients; The expression for calculating the failure probability of the hydrogen concentration sensor using the third fusion function is: Among them, P sensor is the failure probability of the hydrogen concentration sensor, a and b are the third group of weight coefficients; The expression for calculating the pipeline blockage probability using the fourth fusion function is: Among them, P block is the probability of pipeline blockage, P is the actual near-field pressure, and P high is the near-field pressure threshold, c and d are the fourth group of weight coefficients; The expression for calculating the probability of hydrogen storage bottle opening failure or hydrogen concentration sensor failure using the fifth fusion function is: Among them, P open is the probability of hydrogen storage bottle opening failure or hydrogen concentration sensor failure, P normal is the normal value of near-field pressure, e and f are the fifth group of weight coefficients.

7. The method for detecting a fault in a fuel cell hydrogen supply system according to claim 4, characterized in that: The water management test results include: drain valve failure, humidifier failure, purge valve failure, and normal water management; The expression for calculating the failure probability of the drain valve using the sixth fusion function is: Among them, P drain is the failure probability of the drain valve, H is the actual near-field humidity, H th is the near-field humidity threshold, L is the actual near-field liquid level, and L th is the near-field liquid level threshold, k1 and k2 are the sixth group of weight coefficients; The expression for calculating the humidifier failure probability using the seventh fusion function is: Among them, P humidifier is the failure probability of the humidifier, L low is the specific value of the liquid level below the threshold, m1 and m2 are the seventh group of weight coefficients; The expression for calculating the probability of humidifier failure or purge valve failure using the eighth fusion function is: Among them, P humidifier-purge is the probability of humidifier failure or purge valve failure, r1, s1 and t1 are the eighth group of weight coefficients, and u is the adjustment parameter.

8. The method for detecting a fault in a fuel cell hydrogen supply system according to claim 5, characterized in that: The thermal management detection results include: coolant blockage, cooling circulation pump failure, and normal thermal management; The expression for calculating the coolant blockage probability using the ninth fusion function is: Among them, P coolant-block is the coolant blockage probability, T is the actual near-field temperature, T th is the near-field temperature threshold, I local is the local high temperature characteristic parameter of far-field infrared thermal imaging, q1 and q2 are the ninth group of weight coefficients; The expression for calculating the failure probability of the cooling circulation pump using the tenth fusion function is: Among them, P pump is the failure probability of the cooling circulation pump, I global is the overall high temperature characteristic parameter of far-field infrared thermal imaging, and q3 and q4 are the tenth group of weight coefficients.

9. A computer device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the fuel cell hydrogen supply system fault detection method according to any one of claims 1 to 8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the fuel cell hydrogen supply system fault detection method according to any one of claims 1 to 8 is implemented.

Images