Safety monitoring system and equipment for water electrolysis hydrogen production equipment and monitoring and analysis method

Through the safety monitoring and analysis method of electrolytic hydrogen production equipment, the parameter logic structure and BP neural network model are used, combined with Monte Carlo simulation, the equipment abnormal probability is evaluated, and the accuracy of the reliability analysis of electrolytic hydrogen production equipment in the existing technology is solved. The problem is to achieve accurate evaluation of equipment performance indicators and fault prediction.

CN118422274BActive Publication Date: 2024-11-26CHINA THREE GORGES TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410526033.3
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-04-29
Publication Date
2024-11-26
Estimated Expiration
2044-04-29

AI Technical Summary

Technical Problem

In the reliability analysis of electrolytic water hydrogen production equipment, the prior art mainly relies on historical abnormal information, the evaluation accuracy is not high, and the equipment reliability cannot be effectively evaluated dynamically in real time, which may cause serious consequences when equipment failures.

Method used

A safety monitoring and analysis method for electrolytic hydrogen production equipment is adopted. By obtaining operating parameters and environmental parameters, the parameter logical structure is established, the abnormal probability is calculated, the performance change curve is constructed using the BP neural network model, and combined with Monte Carlo simulation, the abnormal probability is calculated 2. Finally, the performance indicators are calculated by weighted average to evaluate the current status of the device.

Benefits of technology

The accurate evaluation of the performance indicators of electrolytic hydrogen production equipment is achieved, and the core function abnormality can be predicted regularly, and inspection or replacement is carried out in a timely manner to avoid serious consequences caused by undetected diaphragm problems.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118422274B_ABST
    Figure CN118422274B_ABST
Patent Text Reader

Abstract

The present invention provides a safety monitoring system and equipment for water electrolysis hydrogen production equipment and a monitoring and analysis method. The method comprises: step 1 obtaining the operating parameters and environmental parameters of each water electrolysis hydrogen production equipment during operation, and analyzing to obtain the parameter logic structure and abnormal data of each water electrolysis hydrogen production equipment. Step 2 calculating the abnormal probability one of each water electrolysis hydrogen production equipment through the parameter logic structure. Step 3 analyzing the abnormal data to determine the performance change curve of the water electrolysis hydrogen production equipment. Step 4 generating the abnormal probability two of the water electrolysis hydrogen production equipment based on the performance change curve of the water electrolysis hydrogen production equipment. Step 5 analyzing the performance index of the current state of the water electrolysis hydrogen production equipment based on the abnormal probability one and abnormal probability two of each water electrolysis hydrogen production equipment. The present invention can accurately evaluate the performance index of the water electrolysis hydrogen production equipment after the water electrolysis hydrogen production equipment is installed in the water electrolysis hydrogen production workshop.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of water electrolysis hydrogen production, and in particular to a safety monitoring system and equipment and a monitoring and analysis method for water electrolysis hydrogen production equipment. Background Art

[0002] Water electrolysis hydrogen production equipment is a general term for equipment used to produce hydrogen by water electrolysis. With the continuous advancement of intelligent means, water electrolysis hydrogen production equipment faces more and more complex risks and failure points. It is necessary to conduct real-time dynamic evaluation and prediction of the reliability of water electrolysis hydrogen production equipment.

[0003] If the availability of components such as sensors, wireless communications, data accuracy, alarm and warning functions, and balancing functions contained in the water electrolysis hydrogen production equipment is not effectively evaluated, more serious consequences will occur once problems occur in the hydrogen production equipment.

[0004] At present, the existing technology for the reliability analysis of water electrolysis hydrogen production equipment is more focused on discovering and solving abnormal points. The reliability evaluation of water electrolysis hydrogen production equipment is mainly based on the statistics and summary of historical abnormal information of the equipment, and the accuracy of the evaluation results is not high. Summary of the invention

[0005] In order to solve the deficiencies in the prior art, one of the purposes of the present invention is to provide a safety monitoring and analysis method for a water electrolysis hydrogen production device, comprising:

[0006] Step 1: Analyze the water electrolysis hydrogen production equipment, obtain the operating parameters and environmental parameters of each water electrolysis hydrogen production equipment during operation, and obtain the parameter logic structure and abnormal data of each water electrolysis hydrogen production equipment based on the operating parameters and environmental parameters.

[0007] Step 2 Calculate the abnormal probability of each water electrolysis hydrogen production equipment through the parameter logic structure.

[0008] Step 3: Analyze the abnormal data of each water electrolysis hydrogen production device to determine the performance change curve of each water electrolysis hydrogen production device.

[0009] Step 4: Generate the second abnormal probability of each water electrolysis hydrogen production device based on the performance change curve of each water electrolysis hydrogen production device.

[0010] Step 5: Based on the abnormal probability 1 and abnormal probability 2 of each water electrolysis hydrogen production equipment, the performance index of the current state of the water electrolysis hydrogen production equipment is obtained, and the current state of the water electrolysis hydrogen production equipment is evaluated through the performance index.

[0011] Furthermore, the operating parameters described in step 1 include the rated power, capacity, voltage and hydrogen production of the water electrolysis hydrogen production equipment.

[0012] The environmental parameters in step 1 include temperature, humidity and operating volume.

[0013] Furthermore, the method for obtaining the parameter logic structure based on the required parameter analysis in step 1 is: creating the parameter logic structure according to the relationship between various parameters of the water electrolysis hydrogen production equipment.

[0014] The parameter logic structure includes different connection families, each connection family includes several node parameters, and each node parameter corresponds to one or several association relationships.

[0015] The correlation between each set of parameters indicates that there is a certain probability that one parameter will cause another parameter to change.

[0016] Furthermore, the method for obtaining abnormal data of each water electrolysis hydrogen production equipment based on the required parameter analysis described in step 1 is: classifying the collected multiple parameters one by one, so as to obtain abnormal data among the multiple parameters.

[0017] The classification one by one is specifically as follows: each parameter is judged by the threshold range or experience value range of each parameter, and the parameters within the threshold range or experience value range are determined to be normal data, and the parameters not within the threshold range or experience value range are determined to be abnormal data.

[0018] Furthermore, step 2 specifically includes:

[0019] Step 2.1 Calculate the abnormal rate corresponding to each abnormal mode at the bottom of the fault tree according to the ratio of the abnormal mode occurrence time to the component operation time.

[0020] Step 2.2 Calculate the abnormal probability of each node parameter corresponding to each family through the abnormal data corresponding to each parameter of the parameter logic structure, and calculate the abnormal probability of the family according to the order of node parameters from small to large associated parameters according to the weighted sum algorithm, that is, the abnormal probability of each family is one.

[0021] Further, the method of analyzing the abnormal data of each water electrolysis hydrogen production device to determine the performance change curve of each water electrolysis hydrogen production device in step 3 includes:

[0022] Step 3.1: Using the abnormal data of each water electrolysis hydrogen production device as model building data, and combining it with the BP neural network model to build an abnormality recognition model.

[0023] Step 3.2: Train the anomaly recognition model until the model converges and meets the accuracy requirements.

[0024] Step 3.3 Set the monitoring period, obtain the operating parameters and environmental parameters of each water electrolysis hydrogen production equipment within the monitoring period, and construct the performance change curve A of each parameter of each water electrolysis hydrogen production equipment: operating parameter / environmental parameter-time.

[0025] Step 3.4: Use the anomaly identification model to identify the degree of abnormality of the operating parameters and environmental parameters of each water electrolysis hydrogen production equipment relative to the abnormal threshold during the monitoring period.

[0026] Step 3.5: Capture the performance change curve portion of the performance change curve A of each water electrolysis hydrogen production device where the abnormal state change degree is greater than the abnormal threshold, and obtain the performance change curve B. Obtain the working time of the performance change curve B corresponding to the time node and the real-time parameter value within the working time.

[0027] Step 3.6 Based on the parameter logic structure, the working hours at the time nodes corresponding to the performance change curve B and the real-time parameter values ​​within the working hours are weighted to obtain the water electrolysis hydrogen production equipment state value of each water electrolysis hydrogen production equipment at each time node.

[0028] Step 3.7 Based on the water electrolysis hydrogen production equipment state value of each water electrolysis hydrogen production equipment at each time node, a performance change curve of each water electrolysis hydrogen production equipment corresponding to the water electrolysis hydrogen production equipment in the monitoring period is constructed.

[0029] Furthermore, the model training method in step 3.2 includes:

[0030] The abnormal data is randomly divided into a training data set, a validation data set and a test data set, and the abnormality recognition model is trained, verified and tested using the training data set, the validation data set and the test data set until the model converges and meets the accuracy requirement.

[0031] Furthermore, the model training in step 3.2 also includes: performing an accuracy test on the trained anomaly recognition model, specifically including:

[0032] Step 3.2.1 Call sample association and substitute the parameters corresponding to the abnormal data sample into the abnormal recognition model for testing.

[0033] Step 3.2.2 Observe whether the anomaly recognition model can recognize the abnormal data parameters, and use this as the training test result.

[0034] Step 3.2.3: Generate feedback constraints for the anomaly recognition model based on the training test results, and adjust and optimize the anomaly recognition model through the feedback constraints.

[0035] Furthermore, if the anomaly recognition model identifies negative samples with low accuracy, the number of anomaly data samples is increased in the training data set and model training is continued until the model converges and meets the accuracy requirements, and then the action of adding anomaly data samples to the training data set is stopped.

[0036] Furthermore, step 4 specifically includes:

[0037] Step 4.1 Based on the performance change curves of each water electrolysis hydrogen production device, the time node information of the abnormality of each water electrolysis hydrogen production device is obtained.

[0038] Step 4.2: Input the time node information of abnormal occurrence of each water electrolysis hydrogen production equipment into the simulation system, perform a preset number of Monte Carlo simulations, and obtain the number of abnormal occurrences of the simulation system within the preset number of times.

[0039] Step 4.3 Calculate the abnormal probability 2 of each water electrolysis hydrogen production equipment by the formula Pf = Nf / N. Where Pf represents the abnormal probability 2 of the water electrolysis hydrogen production equipment at the specified abnormal time, N represents the total simulation times of the water electrolysis hydrogen production equipment at the specified abnormal time, and Nf represents the total abnormal times in the total simulation times.

[0040] Furthermore, the step 4.1 of obtaining the time node information of abnormality of each water electrolysis hydrogen production equipment includes:

[0041] Step 4.1.1 Obtain the abnormal data of each water electrolysis hydrogen production device for analysis, so as to determine the performance change curve C of each water electrolysis hydrogen production device for representing the change rule of the abnormal data over time.

[0042] Step 4.1.2 Determine the time node corresponding to the abnormal data position of each water electrolysis hydrogen production equipment from the performance change curve C.

[0043] Step 4.1.3 The time nodes corresponding to the abnormal data positions from high to low are used as output to obtain the time node information of the abnormal occurrence of each water electrolysis hydrogen production device.

[0044] Furthermore, the method for obtaining the performance change curve C described in step 4.1.1 is: fitting the abnormal data of the same water electrolysis hydrogen production equipment among the abnormal data obtained for each water electrolysis hydrogen production equipment, modeling to obtain a distribution function that satisfies the abnormal data, and using the distribution function as the performance change curve C of the water electrolysis hydrogen production equipment.

[0045] Furthermore, in step 4.1, the time node information of the abnormality of each water electrolysis hydrogen production equipment is obtained and the average is taken to obtain the average abnormal time of the entire water electrolysis hydrogen production equipment, and the average abnormal time is used as the output time node information of the abnormality of each water electrolysis hydrogen production equipment.

[0046] Furthermore, the method described in step 5 for calculating the performance index of the current state of the water electrolysis hydrogen production equipment based on the abnormal probability one and the abnormal probability two of each water electrolysis hydrogen production equipment is as follows: according to the abnormal situation that may occur in each water electrolysis hydrogen production equipment, the corresponding abnormal probability one and the abnormal probability two are weighted averaged by the importance coefficients of the abnormal probability one and the abnormal probability two to obtain the performance index.

[0047] Furthermore, the performance index (P) = a*abnormal probability 1 (P1) + b*abnormal probability 2 (P2). Wherein, a and b are importance coefficients, both have a value range of 0-1 and a+b=1. The values ​​of a and b are preset.

[0048] Furthermore, the values ​​of a and b are set in such a way that a=b=0.5 is the limit if:

[0049] When the relationship between the abnormal data of the current water electrolysis hydrogen production equipment is relatively complex or close, the value of a is set to be higher than the value of b.

[0050] When the abnormal data relationship of the current water electrolysis hydrogen production equipment is simple and mainly changes independently, the value of b is set to be higher than the value of a.

[0051] When the correlation and individuality of the abnormal data of the current water electrolysis hydrogen production equipment are not much different, the value of a is set to be the same as or close to the value of b.

[0052] Furthermore, steps 1 to 5 are performed periodically to re-evaluate the reliability of the water electrolysis hydrogen production equipment.

[0053] Furthermore, the safety monitoring and analysis method for the water electrolysis hydrogen production equipment also includes:

[0054] Step 6: According to the different performance indicators of each water electrolysis hydrogen production equipment, adjust the parameters of the water electrolysis hydrogen production equipment.

[0055] Furthermore, step 6 specifically includes:

[0056] Step 6.1 When simulating the abnormal conditions corresponding to the multiple water electrolysis hydrogen production equipment at each abnormal time, adjust the input parameters of the simulated water electrolysis hydrogen production equipment.

[0057] Step 6.2 Count the changes in performance indicators of the water electrolysis hydrogen production equipment after the input parameters are adjusted.

[0058] Step 6.3 Determine the parameters that are prone to abnormalities in the simulated water electrolysis hydrogen production equipment based on the magnitude of the change in performance indicators.

[0059] The second object of the present invention is to provide a safety monitoring system for water electrolysis hydrogen production equipment, comprising:

[0060] The abnormal data acquisition module is used to execute the method of step 1 in the above-mentioned safety monitoring and analysis method for water electrolysis hydrogen production equipment.

[0061] The curve fitting module is used to execute the method of step 3 in the above-mentioned safety monitoring and analysis method for water electrolysis hydrogen production equipment.

[0062] The abnormal probability calculation module executes the methods of step 2 and step 4 in the above-mentioned safety monitoring and analysis method for water electrolysis hydrogen production equipment.

[0063] The performance indicator analysis module executes the method of step 5 in the above-mentioned safety monitoring and analysis method for water electrolysis hydrogen production equipment.

[0064] The parameter adjustment module executes the method of step 6 in the above-mentioned safety monitoring and analysis method for water electrolysis hydrogen production equipment.

[0065] The third object of the present invention is to provide a safety monitoring device for a water electrolysis hydrogen production device, including a processor and a memory, wherein:

[0066] The processor executes various functional applications and data processing of the processor by running non-transitory software programs, instructions and modules stored in the memory to implement the method steps in the above-mentioned safety monitoring and analysis method for water electrolysis hydrogen production equipment.

[0067] The memory includes a program storage area and a data storage area. The program storage area is used to store an operating system and an application required by at least one function. The data storage area is used to store data created by the processor.

[0068] The present invention has at least one of the following beneficial effects:

[0069] 1. The present invention can accurately evaluate the performance indicators of the water electrolysis hydrogen production equipment after the water electrolysis hydrogen production equipment is installed in the water electrolysis hydrogen production workshop through specific method steps.

[0070] 2. The present invention can periodically execute the evaluation step, and can regularly predict the time when the core functions of the water electrolysis hydrogen production equipment will become abnormal, and promptly conduct on-site inspection or replacement of the water electrolysis hydrogen production equipment when the water electrolysis hydrogen production equipment is about to become abnormal, thereby avoiding serious consequences such as problems with the diaphragm not being accurately detected. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] Figure 1 The present invention is a flow chart of the steps of the safety monitoring and analysis method for the water electrolysis hydrogen production equipment. DETAILED DESCRIPTION

[0072] The present invention will now be described in further detail with reference to the accompanying drawings. These drawings are simplified schematic diagrams, which only illustrate the basic structure of the present invention in a schematic manner, and therefore only show the components related to the present invention.

[0073] Please note that the terms "above", "below", "left", "right", "top", "top", "bottom", "bottom", etc. used in the present invention to describe the positional relationship do not represent the absolute positional relationship between the modules / components / assemblies / components / parts, but the relative positional relationship between the modules / components / assemblies / components / parts. Example 1

[0074] A safety monitoring and analysis method for water electrolysis hydrogen production equipment, such as Figure 1 As shown, including:

[0075] Step 1: Analyze the water electrolysis hydrogen production equipment, obtain the operating parameters and environmental parameters of each water electrolysis hydrogen production equipment during operation, and obtain the parameter logic structure and abnormal data of each water electrolysis hydrogen production equipment based on the operating parameters and environmental parameters.

[0076] Before calculating multiple parameters of each water electrolysis hydrogen production equipment, it is first necessary to collect the operating parameters of the water electrolysis hydrogen production equipment and determine the abnormal data of each component to prepare for the subsequent analysis of the operating safety level of the water electrolysis hydrogen production equipment.

[0077] The operating parameters described in step 1 include the rated power, capacity, voltage and hydrogen production of the water electrolysis hydrogen production equipment.

[0078] The environmental parameters in step 1 include temperature, humidity and operating volume.

[0079] The method for obtaining the parameter logic structure based on the required parameter analysis described in step 1 is: creating the parameter logic structure according to the relationship between various parameters of the water electrolysis hydrogen production equipment.

[0080] The parameter logic structure includes different connection families, each connection family includes several node parameters, and each node parameter corresponds to one or several association relationships.

[0081] The correlation between each set of parameters indicates that there is a certain probability that one parameter will cause another parameter to change.

[0082] The method for obtaining abnormal data of each water electrolysis hydrogen production equipment based on the required parameter analysis described in step 1 is: classifying the collected multiple parameters one by one, so as to obtain abnormal data among the multiple parameters.

[0083] The classification one by one is specifically as follows: each parameter is judged by the threshold range or experience value range of each parameter, and the parameters within the threshold range or experience value range are determined to be normal data, and the parameters not within the threshold range or experience value range are determined to be abnormal data.

[0084] In actual applications, the same device usually contains multiple parameter relationships, among which some operating parameters are affected by other environmental parameters, a certain environmental parameter may be affected by one or more operating parameters of the current device, and some operating parameters or environmental parameters may be associated with the same type of operating parameters or environmental parameters. Therefore, there is usually a certain connection between the various parameters.

[0085] Step 2 Calculate the abnormal probability of each water electrolysis hydrogen production equipment through the parameter logic structure.

[0086] Considering that simply using the abnormal data of the equipment to calculate the abnormal probability of each device ignores the relationship between the various parameters, which leads to certain errors in the calculation of the abnormal probability of each device. Based on this, the present invention proposes a method for calculating the abnormal probability of each water electrolysis hydrogen production device through a parameter logical structure.

[0087] A parameter logic structure is created according to the relationship between various parameters of the water electrolysis hydrogen production equipment. The parameter logic structure includes different connection families. Each family contains multiple node parameters. Each node parameter corresponds to one or more association relationships. The connection relationship between parameters indicates that there is a certain probability that the parameter will cause another parameter to change.

[0088] At this time, step 2 specifically includes:

[0089] Step 2.1 Calculate the abnormal rate corresponding to each abnormal mode at the bottom of the fault tree according to the ratio of the abnormal mode occurrence time to the component operation time.

[0090] Step 2.2 Calculate the abnormal probability of each node parameter corresponding to each family through the abnormal data corresponding to each parameter of the parameter logic structure, and calculate the abnormal probability of the family according to the order of node parameters from small to large associated parameters according to the weighted sum algorithm, that is, the abnormal probability of each family is one.

[0091] Step 3: Analyze the abnormal data of each water electrolysis hydrogen production equipment to determine the performance change curve of each water electrolysis hydrogen production equipment, including:

[0092] Step 3.1: Using the abnormal data of each water electrolysis hydrogen production device as model building data, and combining it with the BP neural network model to build an abnormality recognition model.

[0093] The BP neural network is a multi-layer feedforward neural network trained according to the error back propagation algorithm. It does not need to determine the mathematical equation of the mapping relationship between input and output in advance. It only learns certain rules through its own training to obtain the result closest to the expected output value when the input value is given.

[0094] Step 3.2: Train the anomaly recognition model until the model converges and meets the accuracy requirements.

[0095] Step 3.3 Set the monitoring period, obtain the operating parameters and environmental parameters of each water electrolysis hydrogen production equipment within the monitoring period, and construct the performance change curve A of each parameter of each water electrolysis hydrogen production equipment: operating parameter / environmental parameter-time.

[0096] The performance change curve A is composed of various parameters. Taking the hydrogen production in the operating parameters as an example, the performance change curve A divides the monitoring period into several sub-time points, obtains the hydrogen production at each sub-time point, and uses time as the horizontal coordinate and the hydrogen production as the vertical coordinate to establish a two-dimensional coordinate system for hydrogen production. The hydrogen production at each sub-time point obtained during the monitoring period is plotted on the two-dimensional coordinate system of hydrogen production by drawing points, thereby obtaining a hydrogen production curve.

[0097] Step 3.4: Use the anomaly identification model to identify the degree of abnormality of the operating parameters and environmental parameters of each water electrolysis hydrogen production equipment relative to the abnormal threshold during the monitoring period.

[0098] Step 3.5: Capture the performance change curve portion of the performance change curve A of each water electrolysis hydrogen production device where the abnormal state change degree is greater than the abnormal threshold, and obtain the performance change curve B. Obtain the working time of the performance change curve B corresponding to the time node and the real-time parameter value within the working time.

[0099] Step 3.6 Based on the parameter logic structure, the working hours at the time nodes corresponding to the performance change curve B and the real-time parameter values ​​within the working hours are weighted to obtain the water electrolysis hydrogen production equipment state value of each water electrolysis hydrogen production equipment at each time node.

[0100] Step 3.7 Based on the water electrolysis hydrogen production equipment state value of each water electrolysis hydrogen production equipment at each time node, a performance change curve of each water electrolysis hydrogen production equipment corresponding to the water electrolysis hydrogen production equipment in the monitoring period is constructed.

[0101] Step 4: Generate the second abnormal probability of each water electrolysis hydrogen production device based on the performance change curve of each water electrolysis hydrogen production device.

[0102] Step 4 specifically includes:

[0103] Step 4.1 Based on the performance change curves of each water electrolysis hydrogen production device, the time node information of the abnormality of each water electrolysis hydrogen production device is obtained.

[0104] Step 4.2: Input the time node information of abnormal occurrence of each water electrolysis hydrogen production equipment into the simulation system, perform a preset number of Monte Carlo simulations, and obtain the number of abnormal occurrences of the simulation system within the preset number of times.

[0105] Monte Carlo simulation, also known as random sampling or statistical test method, is a traditional empirical method that cannot approximate the real physical process and is difficult to obtain satisfactory results. However, the Monte Carlo method can truly simulate the actual physical process, so the solution to the problem is very consistent with the actual situation and can obtain very satisfactory results. Based on this, in this embodiment, for each abnormal moment, the embodiment of the present invention adopts the Monte Carlo simulation method, inputs the abnormal moment into the simulation safety monitoring system, performs a preset number of simulations, and confirms whether the electrolysis water hydrogen production equipment really has an abnormality within the abnormal moment. At the same time, the number of abnormalities that occur within the preset number of simulations is counted, so that the reliability indicators such as the abnormal probability can be calculated according to the number of abnormalities that occur within the preset number of simulations, which is used to simulate a large number of abnormal situations of real electrolysis water hydrogen production equipment in actual applications, so as to facilitate the current state evaluation of the real electrolysis water hydrogen production equipment according to the performance indicators of the simulation output.

[0106] For example, during the abnormal moment of each device, determine whether each device has an abnormality, the number of times the abnormality occurs, the time point when the abnormality occurs, etc., and then calculate the reliability indicators that can evaluate the reliability of the water electrolysis hydrogen production equipment based on each abnormal situation, such as mean, variance, frequency, probability, etc., and finally evaluate the reliability of the water electrolysis hydrogen production equipment through reliability indicators.

[0107] For example: if a threshold is set, such as 75%, the abnormal situation of the water electrolysis hydrogen production equipment is judged as normal and abnormal (the abnormal probability is greater than 75%); if two or more thresholds are set, such as 20%, 70%, etc., the abnormal situation of the water electrolysis hydrogen production equipment is judged as normal, warning and abnormal or more states.

[0108] Step 4.3 Calculate the abnormal probability 2 of each water electrolysis hydrogen production equipment by the formula Pf = Nf / N. Where Pf represents the abnormal probability 2 of the water electrolysis hydrogen production equipment at the specified abnormal time, N represents the total simulation times of the water electrolysis hydrogen production equipment at the specified abnormal time, and Nf represents the total abnormal times in the total simulation times.

[0109] The reliability of the water electrolysis hydrogen production equipment is calculated by the formula R=1-Pf, where R represents the reliability of the water electrolysis hydrogen production equipment, which is used to characterize the probability that the water electrolysis hydrogen production equipment will not have an abnormality within a specified abnormal time.

[0110] Specifically, through the Monte Carlo simulation, the number of abnormalities and simulation times of each device under various abnormal time conditions are counted. Under any set abnormal time conditions, according to the simulation situation of each device, the total number of abnormalities and the total number of simulations of all components at the specified abnormal time are calculated, and then the ratio of the total number of abnormalities to the total number of simulations is calculated, and then the abnormal probability that can characterize the water electrolysis hydrogen production equipment at the specified abnormal time is obtained, which is used to evaluate the possibility of abnormality of the water electrolysis hydrogen production equipment at the specified abnormal time.

[0111] For example, when the user specifies an abnormal time of 1 hour, under all abnormal time conditions within 1 hour (for example, the abnormal time input when simulating the water electrolysis hydrogen production equipment includes 1 hour, 40 minutes, and 30 minutes), the total number of simulations performed by each device and the total number of abnormalities of the simulated water electrolysis hydrogen production equipment are counted, thereby obtaining the total number of abnormalities and the total number of simulations. Similarly, the reliability of the water electrolysis hydrogen production equipment can be calculated by R=1-Pf, which is used to characterize the probability that the water electrolysis hydrogen production equipment does not have an abnormality within the specified abnormal time.

[0112] Step 5: Based on the abnormal probability 1 and abnormal probability 2 of each water electrolysis hydrogen production equipment, the performance index of the current state of the water electrolysis hydrogen production equipment is obtained, and the current state of the water electrolysis hydrogen production equipment is evaluated through the performance index.

[0113] The method described in step 5 for calculating the performance index of the current state of the water electrolysis hydrogen production equipment based on the abnormal probability one and abnormal probability two of each water electrolysis hydrogen production equipment is as follows: according to the abnormal situation that may occur in each water electrolysis hydrogen production equipment, the corresponding abnormal probability one and abnormal probability two are weighted averaged by the importance coefficients of the abnormal probability one and abnormal probability two to obtain the performance index.

[0114] The performance index (P) = a*abnormal probability 1 (P1) + b*abnormal probability 2 (P2). Wherein, a and b are importance coefficients, both in the range of 0-1 and a+b=1. The values ​​of a and b are preset.

[0115] The values ​​of a and b are set in such a way that a=b=0.5 is the limit if:

[0116] When the relationship between the abnormal data of the current water electrolysis hydrogen production equipment is relatively complex or close, the value of a is set to be higher than the value of b.

[0117] When the abnormal data relationship of the current water electrolysis hydrogen production equipment is simple and mainly changes independently, the value of b is set to be higher than the value of a.

[0118] When the correlation and individuality of the abnormal data of the current water electrolysis hydrogen production equipment are not much different, the value of a is set to be the same as or close to the value of b.

[0119] For example, if the relationship between the current equipment abnormal data is relatively complex or close, a is set higher than b, such as P=0.7P1+0.3P2; if the relationship between the current equipment abnormal data is simple and basically changes independently, b is set higher than a, such as P=0.2P1+0.8P2; if the correlation and independence of the current equipment abnormal data are not much different, a is set to be the same as or close to b, such as P=0.5P1+0.5P2.

[0120] The present invention fully considers the relationship between various abnormal types, thereby improving the accuracy of abnormal data judgment of each device.

[0121] Through the above method and steps, the reliability of the water electrolysis hydrogen production equipment can be accurately evaluated after the safety monitoring system is installed in the water electrolysis hydrogen production workshop. Example 2

[0122] Based on the safety monitoring and analysis method for water electrolysis hydrogen production equipment in Example 1, the model training method described in step 3.2 includes:

[0123] The abnormal data is randomly divided into a training data set, a validation data set and a test data set, and the abnormality recognition model is trained, verified and tested using the training data set, the validation data set and the test data set until the model converges and meets the accuracy requirement. Example 3

[0124] Based on the safety monitoring and analysis method for water electrolysis hydrogen production equipment in Example 2, the model training described in step 3.2 further includes: performing an accuracy test on the trained abnormality recognition model, specifically including:

[0125] Step 3.2.1 Call sample association and substitute the parameters corresponding to the abnormal data sample into the abnormal recognition model for testing.

[0126] Step 3.2.2 Observe whether the anomaly recognition model can recognize the abnormal data parameters, and use this as the training test result.

[0127] Step 3.2.3: Generate feedback constraints for the anomaly recognition model based on the training test results, and adjust and optimize the anomaly recognition model through the feedback constraints.

[0128] At this time, if the accuracy of the anomaly recognition model in identifying negative samples is low, the number of abnormal data samples is increased in the training data set and the model training is continued until the model converges and meets the accuracy requirements, and the action of adding abnormal data samples to the training data set is stopped.

[0129] The anomaly recognition model trained using the training method can identify more accurate abnormal data. Example 4

[0130] Based on the safety monitoring and analysis method for water electrolysis hydrogen production equipment in Example 1, the step 4.1 of obtaining the time node information of abnormality of each water electrolysis hydrogen production equipment includes:

[0131] Step 4.1.1 Obtain the abnormal data of each water electrolysis hydrogen production device for analysis, so as to determine the performance change curve C of each water electrolysis hydrogen production device for representing the change rule of the abnormal data over time.

[0132] Step 4.1.2 Determine the time node corresponding to the abnormal data position of each water electrolysis hydrogen production equipment from the performance change curve C.

[0133] Step 4.1.3 The time nodes corresponding to the abnormal data positions from high to low are used as output to obtain the time node information of the abnormal occurrence of each water electrolysis hydrogen production device.

[0134] The method for obtaining the performance change curve C described in step 4.1.1 is: fitting the abnormal data of the same water electrolysis hydrogen production equipment among the abnormal data obtained for each water electrolysis hydrogen production equipment, modeling to obtain a distribution function that satisfies the abnormal data, and using the distribution function as the performance change curve C of the water electrolysis hydrogen production equipment.

[0135] The present invention can fit the abnormal data of the current component through the Weibull distribution model, model the distribution function that satisfies the abnormal data, and use the distribution function as the performance change curve of the current component to characterize the change law of the abnormal probability of the current component over time. The Weibull distribution model is currently widely used in abnormal analysis scenarios such as radar, electromechanical, and weather.

[0136] Based on the data characteristics of abnormal probability 2, the present invention introduces the Weibull distribution model, which can accurately obtain the changing law of abnormal probability 2 over time according to the abnormal data analysis of each device. Afterwards, the abnormal moments when the abnormalities are most likely to occur are determined according to the different abnormalities of each device, and multiple groups of abnormal moments are generated. For example: Assuming that the threshold of 80% is the abnormal probability 2 threshold of a certain device, as long as the abnormal probability 2 exceeds 75%, it is considered that the safety monitoring system is abnormal due to the current device, and the time corresponding to when the abnormal probability 2 of a certain device reaches 75% is determined by the performance change curve is the abnormal moment.

[0137] This embodiment only takes the example of determining the performance change curve through the Weibull distribution model for explanation. Other methods of generating the performance change curve corresponding to the component by fitting the abnormal data are also possible, such as exhaustive drawing by computer, etc. The embodiment of the present invention is not particularly limited to this. Example 5

[0138] Based on the safety monitoring and analysis method for water electrolysis hydrogen production equipment in Example 1, step 4.1 obtains the time node information of abnormal occurrence of each water electrolysis hydrogen production equipment and takes the average to obtain the average abnormal time of the entire water electrolysis hydrogen production equipment, and uses the average abnormal time as the output time node information of abnormal occurrence of each water electrolysis hydrogen production equipment.

[0139] This embodiment takes the average of the expected value of the abnormal time calculated for each water electrolysis hydrogen production equipment to obtain the average abnormal time of the entire water electrolysis hydrogen production equipment, which can effectively evaluate the average life of the wireless water electrolysis hydrogen production equipment. Finally, through the performance indicators composed of abnormal probability, reliability and average abnormal time, a quantitative evaluation of the water electrolysis hydrogen production equipment is achieved, the service life of the water electrolysis hydrogen production equipment and the abnormal probability under various abnormal time conditions are predicted, and an accurate evaluation of the reliability of the water electrolysis hydrogen production equipment is achieved after the water electrolysis hydrogen production equipment is installed in the water electrolysis hydrogen production workshop. Example 6

[0140] Based on the safety monitoring and analysis method for the water electrolysis hydrogen production equipment of Example 1, steps 1 to 5 are periodically performed to re-evaluate the reliability of the water electrolysis hydrogen production equipment.

[0141] By executing the above-mentioned evaluation steps at a preset period, it is possible to regularly predict the time when the core functions of the water electrolysis hydrogen production equipment will become abnormal, and to promptly conduct on-site inspections or replace the water electrolysis hydrogen production equipment when the water electrolysis hydrogen production equipment is about to become abnormal, thereby avoiding serious consequences such as problems with the diaphragm not being accurately detected. Example 7

[0142] Based on the safety monitoring and analysis method for the water electrolysis hydrogen production equipment in Example 1, Figure 1 As shown, the safety monitoring and analysis method for the water electrolysis hydrogen production equipment also includes:

[0143] Step 6: According to the different performance indicators of each water electrolysis hydrogen production equipment, adjust the parameters of the water electrolysis hydrogen production equipment.

[0144] Step 6 specifically includes:

[0145] Step 6.1 When simulating the abnormal conditions corresponding to the multiple water electrolysis hydrogen production equipment at each abnormal time, adjust the input parameters of the simulated water electrolysis hydrogen production equipment.

[0146] Step 6.2 Count the changes in performance indicators of the water electrolysis hydrogen production equipment after the input parameters are adjusted.

[0147] Step 6.3 Determine the parameters that are prone to abnormalities in the simulated water electrolysis hydrogen production equipment based on the magnitude of the change in performance indicators.

[0148] This embodiment also analyzes the input parameters of the simulated water electrolysis hydrogen production equipment during the evaluation process for parameters that are prone to abnormalities, so as to understand the degree of influence of these parameters on the model output.

[0149] Specifically, it includes: first determine the key parameters in the input parameters, then adjust each parameter, record the changes in the model output, and count the influence of each parameter on the model output. In the analysis of parameters prone to abnormalities, commonly used indicators include parameter influence PI, which refers to the influence of a certain parameter on the model output, which can be calculated using the following formula:

[0150] PI=(Y max -Y min ) / (X max -X min )

[0151] Among them, "Y max ” and “Y min " respectively represent the maximum and minimum output values ​​of the simulated water electrolysis hydrogen production equipment, "X max ” and “X min " are the maximum and minimum values ​​of the input parameters respectively.

[0152] By sorting the influence of each input parameter from large to small, the sensitive parameters with greater influence in the simulated water electrolysis hydrogen production equipment can be determined. This is used to understand the influence of each parameter on the overall reliability evaluation results, which is convenient for researchers to determine the parameter adjustment step size in subsequent parameter adjustments and improve the accuracy of the evaluation results.

[0153] For example, if the input parameter is a certain performance change curve, once it is adjusted, it will have a significant impact on the overall reliability index, which means that the curve is very sensitive and is currently a more important parameter.

[0154] For example, if a certain abnormal time interval changes, causing the reliability index to change greatly, it means that there are some problems in this time difference interval, causing the system to be more sensitive. It is necessary to check the factors that affect the reliability index in this time period, such as manual error operation, communication failure, etc. Example 8

[0155] A safety monitoring system for water electrolysis hydrogen production equipment, comprising:

[0156] The abnormal data acquisition module is used to execute the method of step 1 in the safety monitoring and analysis method for water electrolysis hydrogen production equipment in the above embodiment.

[0157] The curve fitting module is used to execute the method of step 3 in the safety monitoring and analysis method for water electrolysis hydrogen production equipment in the above embodiment.

[0158] The abnormal probability calculation module executes the methods of step 2 and step 4 in the above-mentioned safety monitoring and analysis method for water electrolysis hydrogen production equipment.

[0159] The performance indicator analysis module executes the method of step 5 in the safety monitoring and analysis method for the water electrolysis hydrogen production equipment in the above embodiment.

[0160] The parameter adjustment module executes the method of step 6 in the safety monitoring and analysis method for the water electrolysis hydrogen production equipment in the above embodiment. Example 9

[0161] A safety monitoring device for a water electrolysis hydrogen production device comprises a processor and a memory, wherein:

[0162] The processor executes various functional applications and data processing of the processor by running non-transitory software programs, instructions and modules stored in the memory to implement the method steps in the above-mentioned safety monitoring and analysis method for water electrolysis hydrogen production equipment.

[0163] The memory includes a program storage area and a data storage area. The program storage area is used to store an operating system and an application required by at least one function. The data storage area is used to store data created by the processor.

[0164] The processor may be a central processing unit (CPU). The processor 901 may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or a combination of the above chips.

[0165] As a non-transitory computer-readable storage medium, the memory can be used to store non-transitory software programs, non-transitory computer executable programs and modules, such as program instructions / modules corresponding to the methods in the above method embodiments.

[0166] In addition, the memory may include a high-speed random access memory, and may also include a non-transitory memory, such as at least one disk storage device, a flash memory device, or other non-transitory solid-state storage device. In some embodiments, the memory may optionally include a memory remotely arranged relative to the processor, and these remote memories may be connected to the processor 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.

[0167] In addition, those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the implemented program can be stored in a computer-readable storage medium, and when the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, the storage medium can be a disk, an optical disk, a read-only memory (ROM), a random access memory (RAM), a flash memory, a hard disk drive (HDD) or a solid-state drive (SSD), etc.; the storage medium can also include a combination of the above-mentioned types of memory.

[0168] Based on the above ideal embodiments of the present invention, the relevant staff can make various changes and modifications without departing from the technical concept of the present invention through the above description. The technical scope of the present invention is not limited to the contents of the specification, and its technical scope must be determined according to the scope of the claims.

Claims

1. A safety monitoring and analysis method for water electrolysis hydrogen production equipment, characterized in that: include: Step 1: Analyze the water electrolysis hydrogen production equipment, obtain the operating parameters and environmental parameters of each water electrolysis hydrogen production equipment during operation, and obtain the parameter logic structure and abnormal data of each water electrolysis hydrogen production equipment based on the operating parameters and environmental parameters; The method for obtaining the parameter logic structure based on the analysis of the operating parameters and the environmental parameters is as follows: creating the parameter logic structure according to the relationship between the various parameters of the water electrolysis hydrogen production equipment; The parameter logic structure includes different connection families, each connection family includes several node parameters, and each node parameter corresponds to one or several association relationships; The correlation between each set of parameters indicates that there is a certain probability that the parameter will cause another parameter to change; The method for obtaining abnormal data of each water electrolysis hydrogen production device based on operating parameter and environmental parameter analysis is: classifying the collected multiple parameters one by one, thereby obtaining abnormal data in the multiple parameters; The classification one by one specifically includes: judging each parameter by the threshold range or the empirical value range of each parameter, determining that the parameter within the threshold range or the empirical value range is normal data, and determining that the parameter not within the threshold range or the empirical value range is abnormal data; Step 2: Calculate the abnormal probability of each water electrolysis hydrogen production equipment through the parameter logic structure; Create a parameter logic structure according to the relationship between various parameters of the water electrolysis hydrogen production equipment. The parameter logic structure includes different connection families, each family contains multiple node parameters, each node parameter corresponds to one or more association relationships, and the connection relationship between parameters indicates that the parameter has a certain probability of causing another parameter to change; At this time, step 2 specifically includes: Step 2.1 calculates the abnormal rate corresponding to each abnormal mode at the bottom of the fault tree according to the ratio of the abnormal mode occurrence time to the component operation time; Step 2.2 calculates the abnormal probability of each node parameter corresponding to each family through the abnormal data corresponding to each parameter of the parameter logic structure, and calculates the abnormal probability of the family according to the node parameter order from less to more of the associated parameters according to the weighted summation algorithm, that is, the abnormal probability of each family is one; Step 3 analyzes the abnormal data of each water electrolysis hydrogen production device to determine the performance change curve of each water electrolysis hydrogen production device; specifically includes: Step 3.1 uses the abnormal data of each water electrolysis hydrogen production device as model building data, and builds an abnormality recognition model in combination with the BP neural network model; Step 3.2: train the anomaly recognition model until the model converges and meets the accuracy requirements; Step 3.3 sets a monitoring period, obtains the operating parameters and environmental parameters of each water electrolysis hydrogen production equipment within the monitoring period, and constructs: operating parameter / environmental parameter-time performance change curve A of each parameter of each water electrolysis hydrogen production equipment; Step 3.4 uses an abnormality recognition model to identify the abnormal degree of change of operating parameters and environmental parameters of each water electrolysis hydrogen production equipment relative to the abnormal threshold during the monitoring period; Step 3.5 captures the performance change curve portion of the performance change curve A of each water electrolysis hydrogen production device where the abnormal degree of state change is greater than the abnormal threshold, and obtains the performance change curve B; obtains the working time of the corresponding time node of the performance change curve B and the real-time parameter value within the working time; Step 3.6: Based on the parameter logic structure, weight the working time of the corresponding time node of the performance change curve B and the real-time parameter value within the working time to obtain the water electrolysis hydrogen production equipment state value of each water electrolysis hydrogen production equipment at each time node; Step 3.7 constructs a performance change curve of each water electrolysis hydrogen production device within the monitoring period corresponding to the water electrolysis hydrogen production device based on the water electrolysis hydrogen production device state value of each water electrolysis hydrogen production device at each time node; Step 4 generates the second abnormal probability of each water electrolysis hydrogen production device based on the performance change curve of each water electrolysis hydrogen production device; Step 4 specifically includes: Step 4.1: based on the performance change curve of each water electrolysis hydrogen production device, obtain the time node information of the abnormality of each water electrolysis hydrogen production device; Step 4.2 inputs the time node information of abnormal occurrence of each water electrolysis hydrogen production equipment into the simulation system, performs a preset number of Monte Carlo simulations, and obtains the number of abnormal occurrences of the simulation system within the preset number of times; Step 4.3 calculates the abnormal probability 2 of each water electrolysis hydrogen production equipment by the formula Pf=Nf / N; wherein Pf represents the abnormal probability 2 of the water electrolysis hydrogen production equipment at the specified abnormal time, N represents the total simulation times of the water electrolysis hydrogen production equipment at the specified abnormal time, and Nf represents the total abnormal times within the total simulation times; Step 5: analyzing the abnormal probability 1 and abnormal probability 2 of each water electrolysis hydrogen production equipment to obtain the performance index of the current state of the water electrolysis hydrogen production equipment, and evaluating the current state of the water electrolysis hydrogen production equipment through the performance index; The method described in step 5 for calculating the performance index of the current state of the water electrolysis hydrogen production equipment based on the abnormal probability one and abnormal probability two of each water electrolysis hydrogen production equipment is as follows: according to the abnormal situation that may occur in each water electrolysis hydrogen production equipment, the corresponding abnormal probability one and abnormal probability two are weighted averaged by the importance coefficients of the abnormal probability one and abnormal probability two to obtain the performance index.

2. The safety monitoring and analysis method for water electrolysis hydrogen production equipment according to claim 1, characterized in that: The operating parameters in step 1 include the rated power, capacity, voltage and hydrogen production of the water electrolysis hydrogen production equipment; The environmental parameters in step 1 include temperature, humidity and operating volume.

3. The safety monitoring and analysis method for water electrolysis hydrogen production equipment according to claim 1, characterized in that: The model training method in step 3.2 includes: The abnormal data is randomly divided into a training data set, a verification data set and a test data set, and the training data set, the verification data set and the test data set are used to train, verify and test the abnormality recognition model until the model converges and meets the accuracy requirements.

4. The safety monitoring and analysis method for water electrolysis hydrogen production equipment according to claim 1, characterized in that: The model training in step 3.2 also includes: performing an accuracy test on the trained anomaly recognition model, specifically including: Step 3.2.1 calls sample association and substitutes the parameters corresponding to the abnormal data sample into the abnormal recognition model for testing; Step 3.2.2 observes whether the anomaly recognition model can recognize the abnormal data parameters, and uses this as the training test result; Step 3.2.3 generates feedback constraints for the anomaly recognition model based on the training test results, and adjusts and optimizes the anomaly recognition model through the feedback constraints.

5. The safety monitoring and analysis method for water electrolysis hydrogen production equipment according to claim 4, characterized in that: If the anomaly recognition model identifies negative samples with low accuracy, the number of anomaly data samples is increased in the training data set and model training is continued until the model converges and meets the accuracy requirements, and the action of adding anomaly data samples to the training data set is stopped.

6. The safety monitoring and analysis method for water electrolysis hydrogen production equipment according to claim 1, characterized in that: The step 4.1 of obtaining the time node information of abnormality of each water electrolysis hydrogen production equipment includes: Step 4.1.1 obtains abnormal data of each water electrolysis hydrogen production device for analysis, thereby determining a performance change curve C of each water electrolysis hydrogen production device for representing the law of abnormal data changing over time; Step 4.1.2 determines the time node corresponding to the abnormal data position of each water electrolysis hydrogen production equipment from the performance change curve C; Step 4.1.3 uses the time nodes corresponding to the abnormal data positions from high to low as the output of the time node information of the abnormalities of each water electrolysis hydrogen production equipment.

7. The safety monitoring and analysis method for water electrolysis hydrogen production equipment according to claim 6, characterized in that: The method for obtaining the performance change curve C described in step 4.1.1 is: fitting the abnormal data of the same water electrolysis hydrogen production equipment among the abnormal data obtained for each water electrolysis hydrogen production equipment, modeling to obtain a distribution function that satisfies the abnormal data, and using the distribution function as the performance change curve C of the water electrolysis hydrogen production equipment.

8. The safety monitoring and analysis method for water electrolysis hydrogen production equipment according to claim 1, characterized in that: In step 4.1, the time node information of abnormal occurrence of each water electrolysis hydrogen production equipment is obtained and the average is taken to obtain the average abnormal time of the entire water electrolysis hydrogen production equipment, and the average abnormal time is used as the output time node information of abnormal occurrence of each water electrolysis hydrogen production equipment.

9. The safety monitoring and analysis method for water electrolysis hydrogen production equipment according to claim 1, characterized in that: The performance index (P) = a*abnormal probability one (P1) + b*abnormal probability two (P2); wherein a and b are importance coefficients, both in the range of 0-1 and a+b=1; the values ​​of a and b are preset.

10. The safety monitoring and analysis method for water electrolysis hydrogen production equipment according to claim 9, characterized in that: The values ​​of a and b are set in such a way that a=b=0.5 is the limit, if: When the relationship between the abnormal data of the current water electrolysis hydrogen production equipment is relatively complex or close, the value of a is set higher than the value of b; When the abnormal data relationship of the current water electrolysis hydrogen production equipment is simple and mainly changes independently, the value of b is set higher than the value of a; When the correlation and individuality of the abnormal data of the current water electrolysis hydrogen production equipment are not much different, the value of a is set to be the same as the value of b.

11. The safety monitoring and analysis method for water electrolysis hydrogen production equipment according to claim 1, characterized in that: Steps 1 to 5 are performed periodically to re-evaluate the reliability of the water electrolysis hydrogen production equipment.

12. The safety monitoring and analysis method for water electrolysis hydrogen production equipment according to claim 1, characterized in that: The safety monitoring and analysis method for the water electrolysis hydrogen production equipment also includes: Step 6: According to the different performance indicators of each water electrolysis hydrogen production equipment, adjust the parameters of the water electrolysis hydrogen production equipment.

13. The safety monitoring and analysis method for water electrolysis hydrogen production equipment according to claim 12, characterized in that: Step 6 specifically includes: Step 6.1, when simulating the abnormal conditions corresponding to the multiple water electrolysis hydrogen production equipment at each abnormal time, adjusting the input parameters of the simulated water electrolysis hydrogen production equipment; Step 6.2: Count the changes in performance indicators of the hydrogen production equipment after the input parameters are adjusted; Step 6.3 determines the parameters that are prone to abnormalities in the simulated water electrolysis hydrogen production equipment based on the magnitude of the change in performance indicators.

14. A safety monitoring system for water electrolysis hydrogen production equipment, characterized in that: include: An abnormal data acquisition module, used to execute the method of step 1 in the safety monitoring and analysis method for water electrolysis hydrogen production equipment according to any one of claims 1 to 11; A curve fitting module, used to execute the method of step 3 in the safety monitoring and analysis method for water electrolysis hydrogen production equipment according to any one of claims 1 to 11; An abnormal probability calculation module, executing the method of step 2 and step 4 in the safety monitoring and analysis method for water electrolysis hydrogen production equipment according to any one of claims 1-11; A performance index analysis module, executing the method of step 5 in the safety monitoring and analysis method for water electrolysis hydrogen production equipment according to any one of claims 1 to 11; The parameter adjustment module executes the method of step 6 in the safety monitoring and analysis method for water electrolysis hydrogen production equipment according to any one of claims 12-13.

15. A safety monitoring device for water electrolysis hydrogen production equipment, characterized in that: comprising a processor and a memory, wherein: The processor executes various functional applications and data processing of the processor by running non-transient software programs, instructions and modules stored in the memory, so as to implement the method steps in the safety monitoring and analysis method for water electrolysis hydrogen production equipment according to any one of claims 1 to 13; The memory includes a program storage area and a data storage area; the program storage area is used to store an operating system and an application program required by at least one function; and the data storage area is used to store data created by the processor.