Fault identification method, system and equipment for hydrogen production and charging all-in-one machine and storage medium

Abstract

The invention provides a hydrogen production and charging all-in-one machine fault identification method, system and device and a storage medium, and relates to the technical field of fault identification, and the method comprises the steps: reading a target calculation power total value, monitoring the first behavior data of a first hydrogen production and charging all-in-one machine based on a predetermined behavior characteristic index, generating a target priority sequence, and then monitoring to obtain first priority use data, calling the data to train a fault identification model, then analyzing to obtain first fault indication information, finally generating a first fault maintenance decision, and carrying out maintenance processing on the first priority hydrogen production and charging all-in-one machine. The method mainly solves the problems that due to the lack of real-time monitoring capacity, faults cannot be found in time, the automation level is insufficient, abnormal modes and potential faults cannot be automatically recognized, and the recognition accuracy and efficiency are low. Through the fault identification method for the hydrogen production and charging all-in-one machine, potential safety hazards in equipment operation can be found in time, and safe use and storage of hydrogen energy are guaranteed.

Description

Technical Field

[0001] This application relates to the technical field of fault identification, and particularly to a method, system, device and storage medium for identifying faults in a hydrogen production and filling integrated machine. Background Art

[0002] With the enhancement of environmental awareness and the continuous development of new energy technologies, hydrogen energy technology, as a clean and efficient energy form, has attracted increasing attention. As an important application of hydrogen energy technology, the research and development of a fault identification method for a hydrogen production and filling integrated machine is of great significance for ensuring the normal operation of the equipment and improving the user experience. With the continuous development of big data and artificial intelligence technologies, user behavior analysis has become an important means for enterprises to optimize products and improve service quality. By deeply analyzing the user behavior of a hydrogen production and filling integrated machine, abnormal patterns and potential faults in the equipment operation can be discovered, providing strong support for fault identification. Traditional fault identification methods often rely on the experience of professionals and regular maintenance, making it difficult to achieve real-time monitoring and automatic identification. A fault identification method for a hydrogen production and filling integrated machine uses advanced data analysis and machine learning technologies to automatically identify abnormal situations in the equipment operation, improving the accuracy and efficiency of fault identification.

[0003] However, in the process of implementing the inventive technical solution in the embodiments of this application, it is found that the above technologies have at least the following technical problems:

[0004] The lack of real-time monitoring ability results in faults not being discovered in time, the automation level is insufficient, abnormal patterns and potential faults cannot be automatically identified, and the accuracy and efficiency of identification are relatively low. Summary of the Invention

[0005] This application mainly solves the problems that the lack of real-time monitoring ability leads to faults not being discovered in time, the automation level is insufficient, abnormal patterns and potential faults cannot be automatically identified, and the accuracy and efficiency of identification are relatively low.

[0006] In view of the above problems, the present application provides a method, system, device and storage medium for fault identification of a hydrogen production and filling integrated machine. In a first aspect, the present application provides a method for fault identification of a hydrogen production and filling integrated machine, the method including: reading a target computing power total value, where the target computing power total value refers to the total computing power value for fault identification and analysis of N hydrogen production and filling integrated machines in a target area, and N is an integer greater than 1; monitoring first behavior data of a first hydrogen production and filling integrated machine based on a predetermined behavior characteristic index, where the first hydrogen production and filling integrated machine is any one of the N hydrogen production and filling integrated machines; generating a target priority sequence in descending order for the N hydrogen production and filling integrated machines according to a first behavior level obtained by analyzing the first behavior data through a user behavior model; monitoring the first-priority hydrogen production and filling integrated machine in the target priority sequence based on a predetermined usage characteristic index to obtain first-priority usage data; retrieving data in a hydrogen production and filling integrated machine fault database to train a fault identification model; analyzing the first-priority usage data through the fault identification model to obtain a first fault indication message; and generating a first fault repair decision based on the first fault indication message, where the first fault repair decision is used for repair processing of the first-priority hydrogen production and filling integrated machine.

[0007] In a second aspect, the present application provides a system for fault identification of a hydrogen production and filling integrated machine, the system including: a total computing power value acquisition module for reading a target computing power total value, where the target computing power total value refers to the total computing power value for fault identification and analysis of N hydrogen production and filling integrated machines in a target area, and N is an integer greater than 1; a first behavior data monitoring module for monitoring first behavior data of a first hydrogen production and filling integrated machine based on a predetermined behavior characteristic index, where the first hydrogen production and filling integrated machine is any one of the N hydrogen production and filling integrated machines; a target priority sequence generation module for generating a target priority sequence in descending order for the N hydrogen production and filling integrated machines according to a first behavior level obtained by analyzing the first behavior data through a user behavior model; a first-priority usage data acquisition module for monitoring the first-priority hydrogen production and filling integrated machine in the target priority sequence based on a predetermined usage characteristic index to obtain first-priority usage data; a fault identification model training module for retrieving data in a hydrogen production and filling integrated machine fault database to train a fault identification model; a first fault indication message acquisition module for analyzing the first-priority usage data through the fault identification model to obtain a first fault indication message; and a repair processing module for generating a first fault repair decision based on the first fault indication message, where the first fault repair decision is used for repair processing of the first-priority hydrogen production and filling integrated machine.

[0008] In a third aspect, the present application provides an electronic device, including: a processor; and a memory communicatively connected to the at least one processor; wherein, the memory stores a computer program executable by the at least one processor, and when the computer program is executed by the at least one processor, the at least one processor is enabled to execute the steps of the method according to any one of the above first aspects.

[0009] In a fourth aspect, the present application provides a computer-readable storage medium storing computer instructions for causing a processor to implement the steps of the method according to any one of the above first aspects when executed.

[0010] One or more technical solutions provided in the present application have at least the following technical effects or advantages:

[0011] The present application provides a method, system, device and storage medium for fault identification of a hydrogen production and filling integrated machine, relating to the technical field of fault identification. The method includes: reading a target total computing power value, monitoring first behavior data of a first hydrogen production and filling integrated machine based on predetermined behavior characteristic indicators, generating a target priority sequence, then monitoring to obtain first priority usage data, invoking data to train a fault identification model, then analyzing to obtain first fault indication information, and finally generating a first fault repair decision to perform repair processing on the first priority hydrogen production and filling integrated machine.

[0012] The present application mainly solves the problems that the lack of real-time monitoring ability leads to failures not being discovered in time, the automation level is insufficient, abnormal patterns and potential failures cannot be automatically identified, and the accuracy and efficiency of identification are relatively low. Through a method for fault identification of a hydrogen production and filling integrated machine, potential safety hazards in the operation of the device can be discovered in time, ensuring the safe use and storage of hydrogen energy.

[0013] The above description is only an overview of the technical solutions of the present application. In order to be able to understand the technical means of the present application more clearly, it can be implemented according to the content of the description. And in order to make the above and other purposes, features and advantages of the present application more obvious and understandable, the following specifically gives the specific implementation manners of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the technical solutions in the present application or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only exemplary, and for those of ordinary skill in the art, other drawings can be obtained according to the provided drawings without creative efforts.

[0015] Figure 1 FIG. is a schematic flow chart of a method for fault identification of a hydrogen production and filling integrated machine provided by an embodiment of the present application;

[0016] Figure 2 This is a schematic flowchart of the method for obtaining the first behavior level in a hydrogen production and filling integrated machine fault identification method provided by an embodiment of the present application;

[0017] Figure 3 This is a schematic flowchart of the method for obtaining a fault identification model in a hydrogen production and filling integrated machine fault identification method provided by an embodiment of the present application;

[0018] Figure 4 This is a schematic structural diagram of a hydrogen production and filling integrated machine fault identification system provided by an embodiment of the present application;

[0019] Figure 5 This is a schematic structural diagram of an exemplary electronic device of the present application.

[0020] Explanation of reference numerals: The total computing power value acquisition module 10, the first behavior data monitoring module 20, the target priority sequence generation module 30, the first priority usage data acquisition module 40, the fault identification model training module 50, the first fault indication information acquisition module 60, the maintenance processing module 70, the electronic device 300, the memory 301, the processor 302, the communication interface 303, the bus architecture 304. Detailed implementation manners

[0021] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying 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 shall fall within the protection scope of the present application.

[0022] The present application mainly solves the problems that the lack of real-time monitoring ability leads to the failure not being discovered in time, the automation level is insufficient, the abnormal patterns and potential faults cannot be automatically identified, and the accuracy and efficiency of identification are relatively low. Through a hydrogen production and filling integrated machine fault identification method, potential safety hazards in the operation of the equipment can be discovered in time, ensuring the safe use and storage of hydrogen energy.

[0023] To better understand the above technical solutions, the above solutions will be introduced in detail below in conjunction with the accompanying drawings of the specification and specific implementation manners:

[0024] Embodiment 1:

[0025] As Figure 1 shown, a hydrogen production and filling integrated machine fault identification method, the method includes:

[0026] Read the total target computing power value, where the total target computing power value refers to the total computing power value for fault identification and analysis of N hydrogen production and filling integrated machines in the target area, and N is an integer greater than 1;

[0027] Specifically, determine the number of hydrogen production and filling integrated machines in the target area: First, clarify the number N of hydrogen production and filling integrated machines that need to be fault-identified in the target area. Relevant information can be obtained through on-site investigations or system monitoring. For each hydrogen production and filling integrated machine, it is necessary to understand its data scale (such as the amount of data generated per second, the amount of data per day, etc.) and processing requirements (such as data processing speed, real-time requirements, etc.). This information can be obtained through equipment manuals, technical documents, or actual tests. Based on the data scale and processing requirements of each hydrogen production and filling integrated machine, evaluate the computing power value required for each device. This requires considering factors such as the performance of the processor, the size of the memory, and the storage capacity. Relevant tools or software can be used for evaluation. Add up the computing power values required for all hydrogen production and filling integrated machines to obtain the total target computing power value. According to actual needs, the configuration of computing power resources can be appropriately adjusted and optimized to ensure the accuracy and real-time nature of fault identification and analysis. The total target computing power value can be read relatively accurately, providing strong support for a hydrogen production and filling integrated machine fault identification method. The total target computing power value is a dynamic value, and it will change accordingly as the number of hydrogen production and filling integrated machines, the data scale, and the processing requirements change.

[0028] Monitor the first behavior data of the first hydrogen production and filling integrated machine based on a predetermined behavior characteristic index, where the first hydrogen production and filling integrated machine is any one of the N hydrogen production and filling integrated machines;

[0029] Specifically, monitoring the first behavior data of the first hydrogen production and filling integrated machine based on a predetermined behavior characteristic index can ensure accurate identification and analysis of the operating state and potential faults of the device. First, select the predetermined behavior characteristic index. These indexes should be able to reflect the normal operating state and potential fault modes of the hydrogen production and filling integrated machine. For example, they can include characteristic indexes such as temperature, pressure, flow rate, and sound. Collect the first behavior data of the first hydrogen production and filling integrated machine in real time through sensors or a monitoring system. This data should include various parameters and state information during the operation of the device, such as current, voltage, air pressure, and temperature. Then, compare and analyze these first behavior data with the predetermined behavior characteristic index. By observing the change trend and abnormal mode of the data, it is possible to preliminarily judge whether the operating state of the hydrogen production and filling integrated machine is normal. Monitoring the first behavior data of the first hydrogen production and filling integrated machine based on a predetermined behavior characteristic index helps to detect device abnormalities and potential faults in a timely manner, providing strong support for fault identification and prediction.

[0030] Generate a target priority sequence in descending order for the N hydrogen production and filling integrated machines according to the first behavior level obtained by analyzing the first behavior data through a user behavior model;

[0031] Specifically, by analyzing the first behavior data through a user behavior model, the behavior levels of the hydrogen production and filling integrated machines can be identified, such as normal, warning, fault, etc. Based on these behavior levels, the N hydrogen production and filling integrated machines can be sorted in descending order to generate a target priority sequence. First, obtain the first behavior level by analyzing the first behavior data through a user behavior model. This level can be a relative assessment, such as a level from normal to warning to fault. Then, sort the N hydrogen production and filling integrated machines according to the first behavior level. Normal devices can be ranked first, warning devices in the middle, and faulty devices at the end. In this way, a target priority sequence can be formed. By generating the target priority sequence in this descending order, the hydrogen production and filling integrated machines can be better managed, and those more urgent or important devices can be processed first. This can improve the efficiency and response speed of fault handling, and ensure the normal operation of the hydrogen production and filling integrated machines and the safe use of hydrogen energy.

[0032] Monitor the first-priority hydrogen production and filling integrated machine in the target priority sequence based on predetermined usage characteristic indicators to obtain first-priority usage data;

[0033] Specifically, first, select the predetermined usage characteristic indicators. These indicators should be able to reflect the usage status and performance characteristics of the hydrogen production and filling integrated machine. For example, they can include characteristic indicators such as usage frequency, hydrogen filling speed, and energy consumption. Then, collect the first-priority usage data of the first-priority hydrogen production and filling integrated machine in real time through sensors or a monitoring system. These data should include various parameters and status information during the actual operation of the device, such as the actual hydrogen filling volume, energy consumption, and operation time. Next, compare and analyze these first-priority usage data with the predetermined usage characteristic indicators. By observing the change trends and abnormal patterns of the data, the performance and usage of the device can be understood. Comparing these first-priority usage data with the data of other hydrogen production and filling integrated machines can better understand the performance differences and usage situations between different devices. This helps to optimize the configuration and usage of the devices, and improve the overall efficiency and reliability. Monitoring the first-priority hydrogen production and filling integrated machine in the target priority sequence based on the predetermined usage characteristic indicators can provide more detailed and accurate usage data, providing strong support for the optimization and usage of the devices.

[0034] Retrieve the data in the hydrogen production and filling integrated machine fault database to train a fault recognition model;

[0035] Specifically, data preparation: Retrieve historical fault data from the fault database of the hydrogen production and filling integrated machine. Preprocess the data, including data cleaning (removing duplicate, incorrect, or irrelevant data), standardization (bringing the data to the same dimension), and feature extraction (selecting features related to faults). Model selection: Select a suitable machine learning model, such as Support Vector Machine (SVM), Random Forest, deep learning models, etc. Model training: Divide the preprocessed data into a training set and a test set. Use the training set to train the selected model and adjust the model parameters to achieve the best performance. Use methods such as cross-validation to evaluate the generalization ability of the model. Model evaluation and optimization: Use the test set to evaluate the performance of the model. Commonly used evaluation metrics include accuracy, recall, F1 score, etc. According to the evaluation results, optimize the model, such as adjusting the model parameters, trying different feature combinations, or adopting different model structures. By training a fault identification model with the data in the fault database of the hydrogen production and filling integrated machine, automatic identification and prediction of equipment faults can be achieved, improving the efficiency and accuracy of fault handling, thereby ensuring the normal operation of the hydrogen production and filling integrated machine and the safe use of hydrogen energy.

[0036] Analyze the first-priority usage data through the fault identification model to obtain first fault indication information;

[0037] Specifically, first, input the first-priority usage data into the fault identification model. This model has been trained with historical fault data and has the ability to identify fault features. Then, deeply analyze the usage data through the model and compare it with the normal behavior pattern. The model will judge whether the state of the hydrogen production and filling integrated machine is normal or whether there are potential faults based on the characteristics and pattern changes of the data. Finally, output the first fault indication information. This information can be the specific fault type, fault degree, or fault location, etc., providing guidance for subsequent fault handling and repair. By analyzing the first-priority usage data through the fault identification model, more accurate and timely fault indication information can be obtained, helping to improve the efficiency and accuracy of fault handling. At the same time, this method can also achieve automation and real-time monitoring, reducing the burden of manual analysis.

[0038] Generate a first fault repair decision based on the first fault indication information, and the first fault repair decision is used to perform repair processing on the first-priority hydrogen production and filling integrated machine.

[0039] Specifically, first, identify the specific fault type, location, and degree based on the first fault indication information. This information can provide detailed information about the equipment failure to the maintenance personnel. Then, formulate corresponding maintenance decisions according to the fault indication information. This decision should include specific maintenance steps, required tools and spare parts, safety precautions, etc. Next, implement the first fault maintenance decision. The maintenance personnel perform maintenance on the first-priority hydrogen production and filling integrated machine according to the decision, repair the fault or replace the damaged components. Finally, record and evaluate the maintenance process and results. This helps to summarize experience and lessons, optimize the maintenance process, and improve the efficiency of equipment maintenance. By generating the first fault maintenance decision based on the first fault indication information, rapid and accurate maintenance can be carried out on the hydrogen production and filling integrated machine, and the normal operation state of the equipment can be restored in a timely manner. At the same time, this method can also improve the pertinence and efficiency of maintenance work, reduce unnecessary maintenance costs and equipment downtime.

[0040] Furthermore, as Figure 2 shown, in the method of the present application, the predetermined behavior characteristic indicators include usage frequency, hydrogen filling times, and usage duration. Generating a target priority sequence in descending order for the N hydrogen production and filling integrated machines according to the first behavior level obtained by analyzing the first behavior data through the user behavior model includes:

[0041] Read the predetermined weight coefficient allocation information, where the predetermined weight coefficient allocation information refers to the weight allocation information for the usage frequency, the hydrogen filling times, and the usage duration;

[0042] Perform standardization processing on the first behavior data to obtain first standardized behavior data, where the first behavior data includes the first usage frequency, the first hydrogen filling times, and the first usage duration;

[0043] The user behavior model calculates by combining the predetermined weight coefficient allocation information and the first standardized behavior data to obtain the first behavior level.

[0044] Specifically, first, read the predefined weight coefficient allocation information, which defines the weight allocation for features such as usage frequency, number of hydrogen filling times, and usage duration. For example, if the weight of the usage frequency is higher, then the influence of this feature in the model will be greater. Then, perform normalization processing on the first row of data. Normalization is a commonly used method for data preprocessing, which is used to eliminate the influence of different feature dimensions on the analysis. Specifically, if the first row of data includes the first usage frequency, the first number of hydrogen filling times, and the first usage duration, then these data should be converted into a unified standard for easy comparison and analysis. Next, the user behavior model calculates by combining the predefined weight coefficient allocation information and the first normalized behavior data. This includes the establishment and application of a mathematical model. Based on the weight information and the normalized data, the model can obtain a more accurate judgment. Finally, through calculation, the user behavior model obtains the first behavior level. This level can be a relative evaluation result, such as normal, warning, or fault, etc. This level is based on the results of weight analysis and normalization processing, and can more accurately reflect the behavior state of the hydrogen production and filling integrated machine. By reading the predefined weight coefficient allocation information and normalizing the first row of data, the user behavior model can more accurately identify the behavior level of the hydrogen production and filling integrated machine. This method helps to improve the accuracy and reliability of fault identification, and provides strong support for the maintenance and management of the equipment.

[0045] Furthermore, in the method of the present application, before the user behavior model calculates by combining the predefined weight coefficient allocation information and the first normalized behavior data to obtain the first behavior level, it further includes:

[0046] Call the predefined weight coefficient verification function to obtain the first verified weight coefficient allocation information;

[0047] Verify the predefined weight coefficient allocation information according to the first verified weight coefficient allocation information to obtain the weight coefficient verification adjustment result;

[0048] Use the weight coefficient verification adjustment result to replace the predefined weight coefficient allocation information for calculation and analysis.

[0049] Specifically, call a predetermined weight coefficient verification function: Call a pre-defined weight coefficient verification function. This function can evaluate the weight coefficients. Obtain the first verified weight coefficient allocation information: Based on the output of the verification function, obtain the first verified weight coefficient allocation information. This information may be an adjustment or correction to the original predetermined weight coefficient allocation information. Verify according to the first verified weight coefficient allocation information: Use the first verified weight coefficient allocation information to verify the predetermined weight coefficient allocation information. This includes comparing the differences between the two sets of weight coefficients and checking their logical relationships, etc. Obtain the weight coefficient verification adjustment result: Based on the verification result, obtain a weight coefficient adjustment plan or a verification adjustment result. This includes adjusting certain weights, increasing or decreasing them. Use the weight coefficient verification adjustment result to replace the predetermined weight coefficient allocation information for calculation and analysis: Use the weight coefficient verification adjustment result to replace the original predetermined weight coefficient allocation information. Then, continue with the calculation and analysis process. Through such verification and adjustment, the accuracy and rationality of the weight coefficients can be ensured, thereby improving the accuracy and reliability of fault identification.

[0050] Furthermore, for the method of the present application, the expression of the predetermined weight coefficient verification function is as follows:

[0051]

[0052]

[0053] Wherein, V i represents the coefficient of variation of the i-th behavior index in the first row of data, σ i represents the standard deviation of the i-th behavior index in the first row of data, represents the average value of the i-th behavior index in the first row of data, W i represents the i-th weight of the i-th behavior index and constitutes the first verified weight coefficient allocation information.

[0054] Specifically, the coefficient of variation: is used to measure the degree of dispersion of the i-th behavior index, that is, the magnitude of data fluctuation. The larger the coefficient of variation, the greater the volatility of the index. The standard deviation: represents the overall "average" fluctuation amplitude of the i-th behavior index. The standard deviation is used to measure the degree of dispersion of data. The average value: the average level of the i-th behavior index, reflecting the average state of the index over a period of time. By comparing the average values, the weight: represents the importance of the i-th behavior index in fault identification. The magnitude of the weight is determined according to its contribution degree to the fault indication information, and the allocation of the weight directly affects the final fault identification result.

[0055] Furthermore, for the method of the present application, the predetermined usage characteristic indicators include temperature, pressure, flow rate, current, voltage, and hydrogen purity.

[0056] Specifically, during the use of the hydrogen production and filling integrated machine, parameters such as temperature, pressure, flow rate, current, voltage, and hydrogen purity are important usage characteristic indicators. These indicators can reflect the operating status and performance of the equipment and are crucial for the fault identification and warning system. Temperature indicator: During the operation of the hydrogen production and filling integrated machine, the change in temperature can reflect the working state of the equipment. Excessive temperature may mean that the equipment is overheating or there is a problem with the cooling system. Continuous high-temperature operation may cause irreversible damage to the equipment. Pressure indicator: Pressure is another key parameter during the operation of the hydrogen production and filling integrated machine. Too high or too low pressure may affect the normal operation of the equipment. The pressure sensor can monitor the pressure status of the equipment in real time and issue an alarm in a timely manner when the pressure exceeds the safe range. Flow rate indicator: The flow rate refers to the amount of hydrogen filled by the hydrogen production and filling integrated machine within a certain period of time. The change in flow rate can reflect the hydrogen filling efficiency of the equipment. If the flow rate suddenly drops, it may mean that there is a blockage or leakage problem with the equipment. Current and voltage indicators: Current and voltage are the basis for the normal operation of the hydrogen production and filling integrated machine. Abnormal current and voltage may mean that there is a circuit fault or power supply problem with the equipment. Continuous high-current or high-voltage operation may burn out the electronic components inside the equipment. Hydrogen purity indicator: Hydrogen purity is an important standard for measuring the performance of the hydrogen production and filling integrated machine. Too low purity may mean that there is a leakage or contamination problem during hydrogen production or storage. Substandard purity will not only affect the performance of the equipment but may also cause safety problems. By real-time monitoring and analysis of these usage characteristic indicators, the operating status of the hydrogen production and filling integrated machine can be comprehensively understood, potential faults and problems can be discovered in a timely manner, and strong support can be provided for the maintenance and upkeep of the equipment.

[0057] Furthermore, as Figure 3 shown, the method of this application retrieves data from the fault database of the hydrogen production and filling integrated machine to train the fault identification model, including:

[0058] Randomly extract the first data record from the fault database of the hydrogen production and filling integrated machine, and the first data record has the identifier of the first fault record;

[0059] Traverse the first data record based on the predetermined usage characteristic indicators to obtain the first set of usage characteristic parameter records;

[0060] Based on the first set of usage characteristic parameter records and the first fault record, form the first training data set;

[0061] Based on the support vector machine principle, conduct supervised learning, training, and testing on the first training data set to obtain the fault identification model.

[0062] Specifically, a first data record is randomly extracted: one or more data records are randomly selected from the hydrogen production and filling device's fault database. These records should contain detailed information about the device failure, such as the time of failure, symptoms, and the device's operating status. A first set of usage characteristic parameter records is obtained based on predetermined usage characteristic indicators: corresponding characteristic parameters are extracted from the selected data records based on previously determined predetermined usage characteristic indicators (such as temperature, pressure, flow rate, etc.). These characteristic parameters will be used to describe the device's operating status and fault conditions. A first training dataset is constructed based on the first set of usage characteristic parameter records and the first fault record: the extracted characteristic parameters are combined with the fault information in the corresponding data records to form a training dataset. This dataset will be used to train and optimize the fault identification model. Supervised learning, training, and validation based on support vector machine principles are performed: a support vector machine (SVM) is selected as the machine learning model. SVM is a commonly used classification and regression model suitable for solving various problems, including fault identification. The SVM model is trained using the first training dataset. During this process, the model identifies faults by learning patterns and relationships in the data. The trained model is validated and tested to ensure its performance and accuracy. This involves using a portion of the data as a test set to evaluate the model's performance on unseen data. This process also involves obtaining a fault recognition model that has learned how to identify different failure modes based on the equipment's operating status. The core of this process is to leverage historical data to train and optimize the model, enabling it to predict and diagnose equipment failures.

[0063] Furthermore, the method of the present application, analyzing the first priority usage data using the fault identification model to obtain first fault indication information, includes:

[0064] Eliminating the first usage characteristic parameter record set from the first data record to obtain a first remaining data record;

[0065] performing a correlation analysis on the first remaining data record and the first fault record to obtain a first correlation analysis result;

[0066] The first fault indication information is verified and analyzed according to the first correlation analysis result.

[0067] Specifically, eliminate the first set of used feature parameter records: Eliminate the "first set of used feature parameter records" that have been used for training the fault identification model from the first data record. This part of the data has been used in model training and is therefore not suitable for use in correlation analysis again. Conduct correlation analysis on the first remaining data record and the first fault record: Conduct correlation analysis on the data remaining after eliminating the feature parameter record set (i.e., the first remaining data record) and the first fault record. Correlation analysis can reveal the relationships between various variables in the data, such as whether there are causal relationships or strong correlations. Obtain the first correlation analysis result: Based on the results of the correlation analysis, the degree of association between each feature and the fault can be obtained. This helps to understand which features are more sensitive or predictive when a fault occurs. Conduct verification analysis on the first fault indication information based on the first correlation analysis result: Based on the first correlation analysis result, the first fault indication information can be verified. This helps to verify whether the preliminary fault identification is accurate and which features play a key role in fault indication.

[0068] Embodiment 2

[0069] Based on the same inventive concept as the hydrogen production and filling integrated machine fault identification method in the foregoing Embodiment 1, as Figure 4 shown, the present application provides a hydrogen production and filling integrated machine fault identification system, and the system includes:

[0070] The total computing power value acquisition module 10 is configured to read the target total computing power value, where the target total computing power value refers to the total computing power value for fault identification analysis of N hydrogen production and filling integrated machines in the target area, and N is an integer greater than 1;

[0071] The first row data monitoring module 20 is configured to monitor the first row data of the first hydrogen production and filling integrated machine based on a predetermined behavior feature index, and the first hydrogen production and filling integrated machine is any one of the N hydrogen production and filling integrated machines;

[0072] The target priority sequence generation module 30 is configured to generate a target priority sequence in descending order for the N hydrogen production and filling integrated machines according to the first behavior level obtained by analyzing the first row data through a user behavior model;

[0073] The first priority usage data acquisition module 40 is configured to monitor the first priority hydrogen production and filling integrated machine in the target priority sequence based on a predetermined usage feature index to obtain first priority usage data;

[0074] A fault identification model training module 50, which is used to retrieve data from the hydrogen production and filling integrated machine fault database to train the fault identification model;

[0075] A first fault indication information acquisition module 60, which is used to analyze the first-priority usage data through the fault identification model to obtain first fault indication information;

[0076] An overhaul processing module 70, which generates a first fault overhaul decision based on the first fault indication information, and the first fault overhaul decision is used to perform overhaul processing on the first-priority hydrogen production and filling integrated machine.

[0077] Furthermore, the system further includes:

[0078] A first behavior level acquisition module, which is used to read the predetermined weight coefficient allocation information, where the predetermined weight coefficient allocation information refers to the weight allocation information for the usage frequency, the hydrogen filling times, and the usage duration; perform normalization processing on the first behavior data to obtain first normalized behavior data, where the first behavior data includes a first usage frequency, a first hydrogen filling times, and a first usage duration; the user behavior model calculates based on the predetermined weight coefficient allocation information and the first normalized behavior data to obtain the first behavior level.

[0079] Furthermore, the system further includes:

[0080] A calculation and analysis module, which is used to call a predetermined weight coefficient verification function to obtain first verification weight coefficient allocation information; verify the predetermined weight coefficient allocation information according to the first verification weight coefficient allocation information to obtain a weight coefficient verification adjustment result; use the weight coefficient verification adjustment result to replace the predetermined weight coefficient allocation information for calculation and analysis.

[0081] Furthermore, the system further includes:

[0082] A fault identification model acquisition module, which is used to randomly extract a first data record from the hydrogen production and filling integrated machine fault database, and the first data record has the identifier of the first fault record; traverse in the first data record based on the predetermined usage feature index to obtain a first usage feature parameter record set; form a first training data set based on the first usage feature parameter record set and the first fault record; perform supervised learning, training, and testing on the first training data set based on the support vector machine principle to obtain the fault identification model.

[0083] Furthermore, the system further includes:

[0084] A verification analysis module is configured to eliminate the first usage feature parameter record set in the first data record to obtain a first remaining data record; perform a correlation analysis on the first remaining data record and the first fault record to obtain a first correlation analysis result; and perform a verification analysis on the first fault indication information according to the first correlation analysis result.

[0085] Exemplary electronic device

[0086] The electronic device 300 includes: a processor 302, a communication interface 303, and a memory 301. Optionally, the electronic device 300 may further include a bus architecture 304. Among them, the communication interface 303, the processor 302, and the memory 301 may be interconnected through the bus architecture 304; the bus architecture 304 may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The bus architecture 304 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 5 only a thick line is used to represent it in the figure, but it does not mean that there is only one bus or one type of bus.

[0087] The processor 302 may be a CPU, a microprocessor, an ASIC, or one or more integrated circuits for controlling the execution of the program of the present application solution.

[0088] The communication interface 303 uses any device such as a transceiver to communicate with other devices or communication networks, such as Ethernet, radio access network (RAN), wireless local area networks (WLAN), wired access networks, etc.

[0089] The memory 301 can be a ROM or other types of static storage devices that can store static information and instructions, a RAM or other types of dynamic storage devices that can store information and instructions, or can also be an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM), or other optical disc storage, optical disc storage (including compact discs, laser discs, optical discs, digital versatile discs, Blu-ray discs, etc.), magnetic disk storage media, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto. The memory can exist independently and be connected to the processor through the bus architecture 304. The memory can also be integrated with the processor.

[0090] Among them, the memory 301 is used to store the computer execution instructions for implementing the solution of this application, and is controlled by the processor 302 to execute. The processor 302 is used to execute the computer execution instructions stored in the memory 301, so as to implement a method for identifying faults of a hydrogen production and filling integrated machine provided in the above embodiments of this application.

[0091] Through the detailed description of the foregoing method for identifying faults of a hydrogen production and filling integrated machine, those skilled in the art can clearly know a system for identifying faults of a hydrogen production and filling integrated machine in this embodiment. For the system disclosed in the embodiment, since it corresponds to the disclosed device in the embodiment, the description is relatively simple, and the relevant parts can be referred to the description in the method part.

[0092] The above description of the disclosed embodiments enables those skilled in the art to implement or use this application. Various modifications to these embodiments will be obvious to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application will not be limited to these embodiments shown herein, but will be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for identifying faults of a hydrogen production and filling integrated machine, characterized in that, Including: Read the total target computing power, where the total target computing power refers to the total computing power for fault identification and analysis of N hydrogen production and filling integrated machines in the target area, and N is an integer greater than 1; Monitor the first behavior data of the first hydrogen production and filling integrated machine based on predetermined behavior characteristic indicators, where the first hydrogen production and filling integrated machine is any one of the N hydrogen production and filling integrated machines; Generate a target priority sequence in descending order for the N hydrogen production and filling integrated machines according to the first behavior level obtained by analyzing the first behavior data through a user behavior model; Monitor the first-priority hydrogen production and filling integrated machine in the target priority sequence based on predetermined usage characteristic indicators to obtain first-priority usage data; Retrieve data from the hydrogen production and filling integrated machine fault database to train a fault identification model; Analyze the first-priority usage data through the fault identification model to obtain first fault indication information; Generate a first fault repair decision based on the first fault indication information, where the first fault repair decision is used to repair and process the first-priority hydrogen production and filling integrated machine.

2. The method according to claim 1, wherein The predetermined behavior characteristic indicators include usage frequency, hydrogen filling times, and usage duration. Generating a target priority sequence in descending order for the N hydrogen production and filling integrated machines according to the first behavior level obtained by analyzing the first behavior data through a user behavior model includes: Read the predetermined weight coefficient allocation information, where the predetermined weight coefficient allocation information refers to the weight allocation information for the usage frequency, the hydrogen filling times, and the usage duration; Perform normalization processing on the first behavior data to obtain first-normalized behavior data, where the first behavior data includes a first usage frequency, a first hydrogen filling times, and a first usage duration; The user behavior model calculates in combination with the predetermined weight coefficient allocation information and the first-normalized behavior data to obtain the first behavior level.

3. The method according to claim 2, wherein Before the user behavior model calculates in combination with the predetermined weight coefficient allocation information and the first-normalized behavior data to obtain the first behavior level, it further includes: Call a predetermined weight coefficient verification function to obtain first-verified weight coefficient allocation information; Verify the predetermined weight coefficient allocation information according to the first-verified weight coefficient allocation information to obtain a weight coefficient verification adjustment result; Use the weight coefficient verification adjustment result to replace the predetermined weight coefficient allocation information for calculation and analysis.

4. The method according to claim 3, wherein The expression of the predetermined weight coefficient verification function is as follows: Among them, V i represents the coefficient of variation of the i-th behavioral index in the first behavioral data, σ i represents the standard deviation of the i-th behavioral index in the first behavioral data, represents the average of the i-th behavioral index in the first behavioral data, W i represents the i-th weight of the i-th behavioral index and forms the first verification weight coefficient allocation information.

5. The method according to claim 1, wherein The predetermined usage characteristic indicators include temperature, pressure, flow rate, current, voltage, and hydrogen purity.

6. The method according to claim 5, characterized in that, Retrieving data from the hydrogen production and filling integrated machine fault database to train a fault identification model includes: Randomly extract a first data record from the hydrogen production and filling integrated machine fault database, where the first data record has the identifier of the first fault record; Traverse in the first data record based on the predetermined usage characteristic indicators to obtain a first usage characteristic parameter record set; Construct a first training data set based on the first usage characteristic parameter record set and the first fault record; Perform supervised learning, training, and testing on the first training data set based on the support vector machine principle to obtain the fault identification model.

7. The method according to claim 6, wherein Analyzing the first-priority usage data through the fault identification model to obtain first fault indication information, including: Eliminating the first usage feature parameter record set in the first data record to obtain a first remaining data record; Performing a correlation analysis on the first remaining data record and the first fault record to obtain a first correlation analysis result; Performing a verification analysis on the first fault indication information according to the first correlation analysis result.

8. A fault recognition system for a hydrogen production and filling integrated machine, characterized in that, Including: A total computing power value acquisition module, which is used to read the total target computing power value. The total target computing power value refers to the total computing power value for fault identification and analysis of N hydrogen production and filling integrated machines in a target area, and N is an integer greater than 1; A first behavior data monitoring module, which monitors the first behavior data of the first hydrogen production and filling integrated machine based on a predetermined behavior feature index. The first hydrogen production and filling integrated machine is any one of the N hydrogen production and filling integrated machines; A target priority sequence generation module, which is used to generate a target priority sequence in descending order for the N hydrogen production and filling integrated machines according to the first behavior level obtained by analyzing the first behavior data through a user behavior model; A first-priority usage data acquisition module, which monitors the first-priority hydrogen production and filling integrated machine in the target priority sequence based on a predetermined usage feature index to obtain first-priority usage data; A fault identification model training module, which is used to retrieve data in a hydrogen production and filling integrated machine fault database to train a fault identification model; A first fault indication information acquisition module, which is used to analyze the first-priority usage data through the fault identification model to obtain first fault indication information; An overhaul processing module, which generates a first fault overhaul decision based on the first fault indication information. The first fault overhaul decision is used to perform overhaul processing on the first-priority hydrogen production and filling integrated machine.

9. An electronic device, characterized in that, The electronic device includes: At least one processor; and a memory communicatively connected to the at least one processor; wherein, the memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the method according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable the processor to implement the method according to any one of claims 1-7 when executed.

Images