Method and device for determining fault solution, and program product

Abstract

The invention discloses a method and equipment for determining a fault solution and a program product, and relates to the technical field of electronic digital data processing. According to the method, when the equipment breaks down, the user operation process is determined based on the equipment operation data, the user operation process is compared with the standard operation process, if the comparison result of the user operation process and the standard operation process is inconsistent, the key field is determined according to the user operation process, and then the user operation process is determined according to the key field. Similar fault cases are determined in a fault knowledge base, corresponding fault solutions are determined based on the similar fault cases, and the fault knowledge base is a database containing historical equipment fault cases and can store and manage fault information in a centralized mode; therefore, the existing solution can be quickly searched and applied when similar faults occur in the equipment, the efficiency and accuracy of fault removal are improved, the fault diagnosis time is shortened, and the equipment maintenance time is shortened.

Description

Technical Field

[0001] The present application relates to the technical field of electronic digital data processing, and in particular to a method, device and program product for determining a fault solution. Background Art

[0002] As the complexity of equipment and systems increases, equipment failures and errors are becoming common problems. Traditional fault diagnosis methods are highly dependent on manual experience, which requires a lot of human resources and diagnostic time. For customers who use equipment, the current automated solutions mainly focus on problem identification and early warning, and cannot provide complete solutions or automatically push some solutions to customers. As a result, when customers face equipment failures, they still need to wait for the service and support of professional technicians, and the equipment maintenance time is relatively long.

[0003] The above contents are only used to assist in understanding the technical solution of the present application and do not constitute an admission that the above contents are prior art. Summary of the invention

[0004] The main purpose of this application is to provide a method, device and program product for determining a fault solution, aiming to solve the technical problem of how to automatically generate a solution to shorten the equipment maintenance time.

[0005] To achieve the above objectives, the present application proposes a method for determining a fault solution, the method comprising: When a device fails, determine the user operation process based on the device operation data, and compare the user operation process with the standard operation process; If the comparison result between the user operation process and the standard operation process is inconsistent, determining the key fields according to the user operation process; According to the key fields, similar fault cases are determined in a fault knowledge base, and corresponding fault solutions are determined based on the similar fault cases, wherein the fault knowledge base is formed by historical equipment fault cases.

[0006] In one embodiment, before the step of determining the user operation process based on the device operation data and comparing the user operation process with the standard operation process, the step further includes: Obtaining step records in historical normal work order data, and determining at least one operation step in the step records; determining a first frequency at which each of the operation steps occurs in the step record; If the first frequency is greater than a first frequency threshold, determining a sequential sequence of the operating steps; Based on the sequential sequence of each of the operating steps, the standard operating procedure is determined.

[0007] In one embodiment, if the first frequency is greater than a first frequency threshold, the step of determining the sequential sequence of the operation steps comprises: If the first frequency is greater than a first frequency threshold, generating a step rule based on the sequence of every two of the operation steps; determining a second frequency at which the step rule occurs in the step record; If the second frequency is greater than a second frequency threshold, a sequential sequence of the operation steps is determined based on the step rule.

[0008] In one embodiment, the step of determining the user operation process based on the device operation data and comparing the user operation process with the standard operation process includes: Based on the device operation data, obtain the timestamp of each operation step in the user operation process; Based on the timestamp, determining the sequence of the operation steps in the user operation process; If the sequence of the operation steps in the user operation process and the sequence of the operation steps in the standard operation process are inconsistent, it is determined that the comparison result is inconsistent.

[0009] In one embodiment, the step of determining the user operation process based on the device operation data and comparing the user operation process with the standard operation process further includes: Obtain the standard equipment operation data range corresponding to the standard operating procedure; If the device operation data corresponding to the user operation process is not within the range of the standard device operation data, it is determined that the comparison result is inconsistent.

[0010] In one embodiment, the step of determining similar fault cases in a fault knowledge base according to the key fields, and determining corresponding fault solutions based on the similar fault cases includes: Determining the weight of the key field; Determine a matching score between the user operation process and the historical device failure case based on the key fields and the weights corresponding to the key fields; If the matching score is greater than a preset score threshold, determining that the historical equipment failure case is a similar failure case; Determine corresponding fault solutions based on the similar fault cases.

[0011] In one embodiment, the step of determining a corresponding fault solution based on the similar fault case includes: Inputting the key fields and the similar fault cases into a preset random forest model, wherein the random forest model is used to predict the success rate of the reference solution according to the key fields and the reference solutions corresponding to the similar fault cases; The success rate of the random forest model output is obtained, and the fault solution is determined in the reference solution based on the success rate.

[0012] In one embodiment, after the step of determining similar fault cases in a fault knowledge base according to the key fields and determining corresponding fault solutions based on the similar fault cases, the method further includes: Based on the key fields, obtain a work order template; Generate a standard fault work order based on the work order template and the fault solution; Sending the standard fault work order to the user end; If no feedback information is received from the user end within the preset time, the standard fault work order is sent to the technician end.

[0013] In addition, to achieve the above-mentioned purpose, the present application also proposes a device for determining a fault solution, the device comprising: a memory, a processor, and a computer program stored on the memory and executable on the processor, the computer program being configured to implement the steps of the method for determining a fault solution as described above.

[0014] In addition, to achieve the above-mentioned purpose, the present application also provides a computer program product, which includes a computer program, and when the computer program is executed by a processor, the steps of the method for determining a fault solution as described above are implemented.

[0015] The present application provides a method for determining a fault solution. When a device fails, the user operation process is determined based on the device operation data, and the user operation process is compared with the standard operation process. If the comparison result between the user operation process and the standard operation process is inconsistent, the key fields are determined according to the user operation process, and then the fault solution is determined in the fault knowledge base according to the key fields. When a device fails, the above method quickly identifies deviations or anomalies in the operation by comparing the user operation process with the standard operation process of the device, and then determines the key fields related to the current device failure according to the user operation process, matches them in the fault knowledge base according to the key fields, determines similar fault cases, and determines the corresponding fault solutions based on the similar fault cases. Among them, the fault knowledge base is a database containing historical equipment failure cases, which can centrally store and manage fault information, so that when a similar fault occurs in the device, it can quickly find and apply existing solutions, improve the efficiency and accuracy of troubleshooting, and thus reduce the time of fault diagnosis and shorten the equipment maintenance time. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0018] Figure 1 A flowchart of a method for determining a fault solution according to an embodiment of the present invention is provided; Figure 2 A flowchart of a second embodiment of a method for determining a fault solution of the present application is provided; Figure 3 A flowchart of a method for determining a fault solution according to Embodiment 3 of the present application is provided; Figure 4 A schematic diagram of a method flow chart provided for Embodiment 3 of the method for determining a fault solution of the present application; Figure 5 A flowchart of a fourth embodiment of a method for determining a fault solution of the present application is provided; Figure 6 A flowchart of a method for determining a fault solution according to Embodiment 5 of the present application is provided; Figure 7 A schematic diagram of the process of generating and dispatching a standard fault work order provided in Example 5 of the present application; Figure 8 A schematic diagram of the architecture of an automatic fault resolution system involved in a method for determining a fault resolution method in an embodiment of the present application; Fig. 9 A schematic diagram of the device structure of the hardware operating environment involved in the method for determining a fault solution in an embodiment of the present application.

[0019] The purpose, features and advantages of this application will be further described in conjunction with the embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0020] It should be understood that the specific embodiments described herein are only used to explain the technical solutions of the present application and are not used to limit the present application.

[0021] In order to better understand the technical solution of the present application, a detailed description will be given below in conjunction with the accompanying drawings and specific implementation methods.

[0022] As the complexity of equipment and systems increases, failures and errors are becoming common problems. Traditional fault diagnosis methods are highly dependent on manual experience, which requires a lot of human resources and diagnostic time. For customers who use equipment, the current automated solutions mainly focus on problem identification and early warning, and cannot provide complete solutions or automatically push some solutions to customers. As a result, when customers face equipment failures, they still need to wait for the service and support of professional technicians, and the equipment maintenance time is relatively long.

[0023] In view of the above problems, the present application proposes a method for determining a fault solution. When a device fails, the user operation process is determined based on the device operation data, and the user operation process is compared with the standard operation process. If the comparison result between the user operation process and the standard operation process is inconsistent, the key fields are determined according to the user operation process, and then similar fault cases are determined in the fault knowledge base based on the key fields, and the corresponding fault solutions are determined based on the similar fault cases. Among them, the fault knowledge base is a database containing historical equipment fault cases, which can centrally store and manage fault information, so that when a similar fault occurs in the equipment, the existing solutions can be quickly found and applied, thereby improving the efficiency and accuracy of troubleshooting, thereby reducing the time for fault diagnosis and shortening the equipment maintenance time.

[0024] It should be noted that the execution subject of this embodiment can be a computing service device with data processing, network communication and program running functions, such as a tablet computer, a personal computer, etc., or an electronic device capable of realizing the above functions, etc. The following takes the automatic fault resolution system as an example to illustrate this embodiment and the following embodiments.

[0025] Based on this, the first embodiment proposed in this application provides a method for determining a fault solution, referring to Figure 1 In this embodiment, the method for determining a fault solution includes steps S10 to S30: Step S10, when a device fails, a user operation process is determined based on the device operation data, and the user operation process is compared with a standard operation process.

[0026] It should be noted that equipment operation data refers to the equipment operation status information collected through the equipment’s own sensors, monitoring systems or remote monitoring platforms, including but not limited to the equipment’s working parameters, operating time, operation process, etc.

[0027] Optionally, a large amount of historical equipment operation data is trained using a machine learning algorithm so that it can automatically identify and extract user operation processes. The collected equipment operation data is converted into recognizable operation step information through the trained machine learning algorithm.

[0028] Exemplarily, collect historical equipment operation data, including sensor readings such as temperature, pressure, and equipment status. Annotate the historical equipment operation data to mark the start and end of each operation step. And define labels for operation steps, such as "start equipment", "execute production process", "close equipment", etc. Select a machine learning model suitable for sequence data, and divide the preprocessed data into a training set and a validation set. Use the training set to train the model so that the model learns the mapping relationship between sensor readings and operation steps. Use the validation set to evaluate model performance and adjust model parameters or structure to optimize performance. Then, according to the trained model, convert the collected equipment operation data into recognizable operation step information.

[0029] Specifically, taking the automated assembly line as an example, historical equipment operation data, including temperature, humidity, vibration, current consumption, equipment status, etc., are first collected from the sensors and control systems of the assembly line. The collected data is preprocessed, such as normalizing the data and scaling the temperature and humidity data to between 0 and 1. According to the timestamps of the operation steps, the data is divided into segments of fixed time windows. Feature data is extracted based on the data segments, such as calculating the average temperature, humidity, vibration frequency, etc. in each time window. Operation steps are defined such as assembly line, assembly of component A, assembly of component B, quality inspection, packaging, shutdown, etc. After that, each data segment is labeled, and the operation steps corresponding to each segment are marked. A multi-layer perceptron (MLP) model is selected, and the labeled data is divided into a training set and a test set. The training set is used to train the MLP model, with the input as feature data and the output as operation step labels. The performance of the MLP model is evaluated using the test set, and indicators such as accuracy and recall are recorded. The MLP model parameters are adjusted according to the test results, such as increasing the number of hidden layer nodes and changing the learning rate. Deploy the trained MLP model to the automated fault resolution system to ensure that the MLP model can receive real-time equipment operation data and output prediction results. Then collect sensor data in real time, convert the sensor data into feature data through preprocessing and feature engineering steps, and input the feature data into the deployed MLP model to obtain the user operation process.

[0030] After obtaining the user operation process, compare the user operation process with the standard operation process to determine the differences or anomalies between each operation step of the user and the operation steps in the standard operation process.

[0031] Optionally, the user operation process is compared with the standard operation process in chronological order to find out the difference in the order of steps. For each operation step, the corresponding equipment operation data such as temperature, pressure, speed, etc. are compared with the corresponding standard equipment operation data range in the standard operation process to see if they are consistent.

[0032] Optionally, the standard operation process is represented as a graph structure, and then the user operation process is mapped to the graph, and the differences are found through a graph matching algorithm.

[0033] In the graph structure, each node represents an operation step, and each node can contain multiple attributes, such as step name, execution condition, expected result, etc. Edges are used to represent dependencies between operation steps. Dependencies include sequential dependencies, i.e., one step must be executed after another step, conditional dependencies, i.e., the execution of one step depends on the result of another step, etc. The direction of the edge represents the execution order between steps.

[0034] Exemplarily, first, the standard operation process is decomposed into a series of operation steps, a node is created for each operation step, and edges are created according to the dependency relationship between the steps. Then, the data of the user operation process is preprocessed to match it with the node attributes in the graph structure. The user operation process is matched with the nodes in the graph structure by comparing the attributes such as the step name and the execution condition. According to the matching result of the user operation process, a path is constructed in the graph structure to represent the execution order of the user operation process. In the graph structure, a subgraph corresponding to the user operation process path is defined, and the subgraph composed of the user operation process path is matched with the standard operation process graph structure using the subgraph isomorphism algorithm to find the difference between the two, including checking whether some steps in the standard operation process are missing in the user operation process path; checking whether the execution order of the user operation process is consistent with the order in the standard operation process; checking whether the attributes of the user operation process such as the execution condition and the expected result are consistent with the corresponding attributes in the standard operation process, etc. According to the results of the difference analysis, a difference report is generated, which lists the differences between the user operation process and the standard operation process in detail.

[0035] Step S20: If the comparison result between the user operation process and the standard operation process is inconsistent, determine the key fields according to the user operation process.

[0036] Step S30, determining similar fault cases in a fault knowledge base according to the key fields, and determining corresponding fault solutions based on the similar fault cases, wherein the fault knowledge base is formed by historical equipment fault cases.

[0037] Optionally, if the comparison result between the user operation process and the standard operation process is inconsistent, the user operation process, the equipment operation data corresponding to the user operation process, the log and other information are analyzed to extract key fields related to troubleshooting, including equipment type, fault type, fault cause, operation steps, fault description, etc.

[0038] It should be noted that the fault knowledge base is a database used to store information such as fault types, fault causes, and fault solutions.

[0039] Optionally, in order to ensure the flexibility and scalability of the fault knowledge base, NoSQL (Not Only SQL, non-relational database) can be used to store historical equipment fault cases in order to process unstructured and semi-structured data. NoSQL databases support large-scale data storage and can quickly respond to query requirements, which is suitable for complex and diverse fault data storage requirements. In order to achieve efficient query, the fault knowledge base includes a multi-level index structure, including hash index, inverted index, etc., which can quickly retrieve historical equipment fault cases based on key fields such as fault type, equipment model, fault description, and similar fault cases that are the same or similar to the current equipment fault situation, and determine the corresponding fault solution based on the similar fault case.

[0040] Optionally, if the difference between the user operation process and the standard operation process matches the known fault mode in the fault knowledge base, a fault prompt can be automatically triggered to provide maintenance personnel with fault location and repair suggestions, and the difference analysis results between the user operation process and the standard operation process can be synchronized to the fault knowledge base.

[0041] In this embodiment, based on the equipment operation data and machine learning algorithm, the user operation process is acquired and identified in real time, and the user operation process is compared with the standard operation process to determine the comparison result. If the comparison result is inconsistent, the fault knowledge base is matched according to the user operation process to find the corresponding fault solution. This method not only improves the efficiency and accuracy of fault diagnosis and reduces equipment downtime, but also enhances the safety and stability of the equipment through standardized operation procedures. At the same time, the establishment and update of the fault knowledge base also provides a platform for the accumulation and inheritance of fault handling knowledge. When the equipment user encounters the same or similar problems, he can directly find solutions without waiting for technicians to come to the door for repair. At the same time, it reduces the repeated fault diagnosis work of technicians, thereby improving overall work efficiency and reducing equipment maintenance time.

[0042] Based on the first embodiment of the present application, in the second embodiment of the present application, the same or similar contents as those in the above-mentioned embodiment 1 can refer to the above introduction, and will not be repeated later. Figure 2 Before step S10, the method for determining the fault solution includes steps S40 to S70: Step S40, obtaining step records in historical normal work order data, and determining at least one operation step in the step records.

[0043] Step S50, determining a first frequency of each of the operation steps in the step record.

[0044] It should be noted that historical normal work order data refers to the work order data recorded during the normal operation of the equipment in the past period of time. In the historical normal work order, a series of operation steps required to complete a task or handle a situation are recorded in detail.

[0045] Optionally, collect all work order data generated during normal operation of the equipment. This includes daily maintenance records, inspection records, and regular maintenance records of the equipment. Clean the collected data to remove duplicate, invalid, or abnormal data records, and extract step records related to equipment operation and maintenance from the cleaned data, including specific operations such as equipment startup, operation, shutdown, inspection, and maintenance. Standardize the extracted steps to form a unified operating process and specification. This includes unifying the naming, sequence, and description of the operating steps to ensure the accuracy and consistency of subsequent analysis.

[0046] Extract all step records from historical normal work order data, split the step records in each work order into individual independent operation steps, count the number of times each operation step appears in the step record, divide the number of times each operation step appears by the total number of all operation steps in the step record, and get the first frequency of the operation step. By calculating the first frequency, it is possible to filter out those common and important operation steps in the step records in historical normal work order data.

[0047] Step S60: if the first frequency is greater than a first frequency threshold, determining a sequential order of the operation steps.

[0048] It should be noted that the operation steps whose first frequency is greater than the first frequency threshold are screened out, and the sequence of each operation step is determined according to the sequence information or logical relationship in the step record.

[0049] Optionally, step S60 includes steps S61 to S63: Step S61: if the first frequency is greater than a first frequency threshold, a step rule is generated based on the sequence of every two operation steps.

[0050] Optionally, any two operation steps whose first frequency is greater than the first frequency threshold form a step pair to generate a step rule. The step rule is used to indicate the sequence between two operation steps. It can be expressed as "step A->step B", which means that step A must be executed before step B.

[0051] Step S62, determining a second frequency of the step rule appearing in the step record.

[0052] It should be noted that the second frequency of the step rule appearing in the step record is determined to quantify the importance or prevalence of each step rule in the actual operation step record.

[0053] Exemplarily, each step record is parsed to extract the step sequence of the operation steps in the step record, and the parsed step sequence is matched with the previously generated step rule to check whether it contains the step pair in the step rule, and determine whether the sequence of the step pair is consistent with the step rule. For each step rule, the number of times it appears in the step record is counted. Then, the number of occurrences of each step rule is divided by the total number of step records to calculate the second frequency of the step rule.

[0054] For example, assume that there are operation step combination 1 "start device - check parameters - adjust parameters", operation step combination 2 "start device - check parameters", operation step combination 3 "start device - check parameters - adjust parameters - record data", and operation step combination 4 "check parameters - start device" in the step record. Assume that step rule 1 "start device->check parameters" and step rule 2 "check parameters->adjust parameters" are generated. Then step rule 1 appears in operation step combinations 1, 2, and 3 for a total of 3 times, and the second frequency is 3 / 4=75%. Step rule 2 appears in operation step combinations 1 and 3 for a total of 2 times, and the second frequency is 2 / 4=50%.

[0055] Step S63: if the second frequency is greater than a second frequency threshold, determining a sequential order of the operation steps based on the step rule.

[0056] It should be noted that the second frequency threshold is used to determine whether the second frequency of the step rule is high enough, thereby determining whether the step rule should be used to construct a sequential sequence of operation steps.

[0057] Based on the calculated second frequency, step rules that frequently appear in actual operations and have high importance are screened out, and then the step pairs in the screened step rules are connected in sequence to form a complete operation step chain.

[0058] Step S70, determining the standard operating procedure based on the sequential sequence of each of the operating steps.

[0059] After screening out the operation steps with a high repetition rate and screening out the sequential sequence of the operation steps with a high repetition rate, the above operation steps are connected according to the determined sequential sequence, and the parallel and repeated operation steps are removed to obtain a standard operation process.

[0060] Optionally, by using association rule learning techniques such as the Apriori algorithm, common operation step combinations in historical normal work order data are analyzed, and frequently occurring operation step combinations are extracted as standard operation procedures.

[0061] Among them, association rule learning is used to discover the association relationships between data items in the data set, which are usually presented in the form of "if...then...". The Apriori algorithm can discover frequent item sets and association rules from a large amount of transaction data. In the generation scenario of standard operating procedures, each operation step in the step record in the historical normal work order data is regarded as an item, and a series of operation steps performed in sequence is regarded as a transaction. Through the Apriori algorithm, frequently occurring operation step combinations, namely frequent item sets, can be mined from the historical normal work order data, and association rules representing the order of operation steps can be further generated, and standard operating procedures can be constructed based on these association rules.

[0062] Exemplarily, relevant operation data is collected from historical normal work order data to ensure the integrity and accuracy of the operation data. A minimum support threshold is set to indicate the minimum frequency of an item set appearing in the data set, which is used to filter frequent item sets. A minimum confidence threshold is set to indicate the minimum probability that a consequent item set will also appear given the appearance of a preceding item set, which is used to filter strong association rules. Then, all transactions are traversed, the support of each operation step, i.e., a single item set, is calculated, and the initial frequent item set is filtered out. A new candidate item set is constructed by combining each item in the initial frequent item set. All transactions are traversed again, and the support of each candidate item set is calculated. The candidate item sets whose support is less than the minimum support threshold are deleted to obtain the final frequent item set. All possible association rules, i.e., the mapping of the preceding item set to the consequent item set, are extracted from the final frequent item set. For each rule, calculate its confidence, that is, the probability that the consequent item set will appear when the antecedent item set appears. After deleting the association rules whose confidence is less than the minimum confidence threshold, sort the remaining association rules and merge the rules with the same antecedent or consequent items to form a coherent standard operating procedure.

[0063] In this embodiment, by analyzing the historical normal work order data, the key operation steps during the normal operation of the equipment can be automatically identified and the execution order of these steps can be determined. By using algorithms to mine the association rules between the operation steps, the standard operation process can be constructed more accurately, reducing the time and cost of manual analysis.

[0064] Based on the above embodiments of the present application, in the third embodiment of the present application, the same or similar contents as those in the above embodiments can be referred to the above introduction, and will not be described in detail later. Figure 3 , step S10 includes steps S11 to S13: Step S11, based on the device operation data, obtaining the timestamp of each operation step in the user operation process.

[0065] Step S12: determining the sequence of the operation steps in the user operation process based on the timestamp.

[0066] Step S13: If the sequence of the operation steps in the user operation process and the corresponding operation steps in the standard operation process are inconsistent, then the comparison result is determined to be inconsistent.

[0067] Parse the device operation data, extract specific timestamp fields or convert the string representing the date in the device operation data into a date and time object to mark the data point at the exact time when a certain operation occurred. Sort the extracted timestamps, and after the timestamps are sorted, identify the operation steps corresponding to each timestamp. Determine the order of the operation steps in the user operation process based on the order of the timestamps. Compare the operation steps corresponding to the user operation process and the standard operation process one by one. If the order of the operation steps is inconsistent, determine that the comparison result is inconsistent.

[0068] For example, suppose there is an automated production equipment that has a series of preset standard operating procedures, including the correct order of each step: 1 start the equipment, 2 load raw materials, 3 adjust equipment parameters, 4 start production, 5 end production, 6 unload products.

[0069] When a device fails, timestamps are extracted from the device operation data, and the sequence of operation steps in the user operation process is determined based on the sequence of timestamps: Step 1, start the device (2023-10-01, 08:00); Step 2, loading raw materials (2023-10-01, 08:05); Step 3: Start production (2023-10-01, 08:10); Step 4, end production (2023-10-01, 08:25); Step 5, uninstall the product (2023-10-01, 08:30).

[0070] When comparing with the standard operating procedure, it was found that the "loading raw materials" step was followed by the "start production" step, and the "adjusting equipment parameters" step was omitted, which determined that the comparison result was inconsistent.

[0071] Optionally, in a feasible implementation manner, referring to Figure 4 , step S10 also includes steps S14-S15: Step S14, obtaining the standard equipment operation data range corresponding to the standard operation process.

[0072] Step S15: If the device operation data corresponding to the user operation process is not within the range of the standard device operation data, it is determined that the comparison result is inconsistent.

[0073] It should be noted that for each operating step in the standard operating procedure, the equipment operating data should reach or remain within a specific range after executing the standard operating steps to ensure the normal operation of the equipment. This specific range is the standard equipment operating data range.

[0074] Optionally, the device operation data is collected in real time through the device's sensor, instrument or monitoring system, and the standard device operation data range corresponding to the current operation step is obtained from the device manual, operation guide or fault knowledge base. The collected device operation data is compared with the standard device operation data range, and if the device operation data exceeds the standard device operation data range, the comparison result is determined to be inconsistent.

[0075] Optionally, a normal fluctuation range of the equipment operation data is set according to historical normal work order data. If the equipment operation data of the current faulty equipment exceeds the normal fluctuation range and the duration exceeds a certain threshold, the comparison result is determined to be inconsistent.

[0076] Optionally, it is first determined whether the order of the operation steps in the user operation process and the corresponding operation steps in the standard operation process are consistent. If they are inconsistent, the equipment operation data corresponding to the user operation process is compared with the standard equipment operation data range. By first checking the order of the operation steps, possible logical errors or omissions in the user operation process can be quickly identified. If the order of the operation steps is inconsistent, it can be determined that the comparison results are inconsistent, and there is no need to perform a detailed comparison of the equipment operation data, which can reduce the workload of data processing and improve the speed of fault diagnosis.

[0077] Optionally, analyze the differences between the user operation process and the standard operation process to determine whether the equipment operation data corresponding to the user operation process exceeds the normal operation range or design limit of the equipment. Compare the changes in equipment operation data before and after the failure, identify abnormal values ​​or value change trends, determine the association between abnormal operation data and the failure phenomenon, and combine the results of operation step analysis and equipment operation data analysis to determine the cause of the failure.

[0078] For example, suppose the standard operating procedure of a production equipment requires the operator to preheat the equipment before starting it, the preheating time is 5 minutes, and the preheating temperature is set to 200°C. However, in the user operation process, the operator directly started the equipment without preheating. At this time, the difference is analyzed as the cause of the fault, that is, "no preheating".

[0079] Exemplarily, the normal range and design limit of the equipment operation data are obtained from the specification sheet or technical document provided by the equipment manufacturer, and the equipment operation data corresponding to the user operation process is evaluated to see whether it exceeds the normal range or design limit of the equipment. Continuing with the above example, assume that the preheating temperature in the normal range of the equipment operation data is 180°C~220°C, and although preheating is performed in the current operation step, the preheating temperature is set to 250°C, which exceeds the normal range. Therefore, it can be determined that the cause of the fault is beyond the normal range of the equipment operation data.

[0080] For example, suppose that the vibration parameter of a certain production equipment is stable below 0.2mm / s before the failure, and the vibration parameter suddenly increases to 0.5mm / s after the failure. At this time, when the change in the vibration parameter value is identified, the abnormal increase in the vibration parameter may be related to the damage or looseness of the bearing inside the equipment, and the damage or looseness of the bearing is one of the fault phenomena. Therefore, it can be determined that there is a correlation between the abnormal increase in the vibration parameter and the fault phenomenon, and the cause of the fault is determined to be the damage or looseness of the bearing inside the equipment.

[0081] In this embodiment, by combining the timestamp analysis of the user operation process and the comparison of the equipment operation data, the fault can be diagnosed quickly and accurately. First, by extracting the timestamps in the equipment operation data, the order of user operations can be determined. If this order is inconsistent with the preset standard operation process, the operation errors or omissions can be quickly identified, thereby speeding up the fault diagnosis and reducing the workload of data processing. In addition, by analyzing the differences between the user operation process and the standard operation process, the specific cause of the fault can be determined more accurately, so that the fault solution can be matched more accurately based on the compared differences and fault causes.

[0082] Based on the above embodiments of the present application, in the fourth embodiment of the present application, the same or similar contents as those in the above embodiments can be referred to the above introduction, and will not be described in detail later. Figure 5 , step S30 includes steps S31 to S34: Step S31, determining the weight of the key field.

[0083] Optionally, the user operation process, the equipment operation data corresponding to the user operation process, the log and other information are parsed to form a fault description. After the key fields related to the fault troubleshooting are extracted from the fault description, a weight value is calculated for each key field based on the statistical characteristics, location information or semantic analysis of the key fields in the fault description according to preset rules or algorithms.

[0084] For example, the frequency of occurrence of each key field in the fault description is counted, and the weight is adjusted in combination with the inverse document frequency to reflect the importance of the key field; or a weight value is assigned to each position according to the position of the key field in the fault case description, such as the title, the beginning of the text, the middle of the text, the end of the text, etc. For each key field, the corresponding position weight is found according to its position. If the key field appears in multiple positions, the average or maximum value of the position weights can be taken as the weight of the key field.

[0085] Step S32: determining a matching score between the user operation process and the historical device failure case based on the key fields and the weights corresponding to the key fields.

[0086] Step S33: If the matching score is greater than a preset score threshold, it is determined that the historical equipment failure case is a similar failure case.

[0087] It should be noted that the matching score is used to measure the similarity or matching degree between the current equipment failure and the historical equipment failure cases, and can be calculated based on the key fields and the weights of the key fields.

[0088] Optionally, key fields are extracted from the user operation process description and the historical equipment failure case description, respectively, to find the intersection of the key fields in the user operation process description and the historical equipment failure case description. A weighted calculation is performed based on the number of key fields in the intersection and the weight of the key fields to obtain a matching score.

[0089] Optionally, use a pre-trained word vector model such as Word2Vec (Word Embeddings), BERT (Bidirectional Encoder Representations from Transformers), etc. to convert each key field into a word vector representation. For each key field in the user operation process description, calculate its word vector similarity value with the key field in the historical equipment failure case description, such as the cosine similarity value, as the matching score. The larger the cosine similarity value, the more semantically similar the two descriptions are.

[0090] The calculated matching score is compared with the preset score threshold. If the matching score is greater than the preset score threshold, the historical equipment failure case is considered similar to the current equipment failure, otherwise it is considered not similar. By setting the score threshold, historical equipment failure cases that are sufficiently similar to the current equipment failure can be screened out, providing strong support for the subsequent determination of the fault solution.

[0091] For example, when retrieving similar fault cases through Elasticsearch, weights can be set for different fields, where Elasticsearch is an open source distributed search and analysis engine suitable for fast search, analysis and storage of large-scale data. For example, the weight of the fault type field may be higher than that of the device type field, which means that when the fault type has a higher matching degree, the automatic fault resolution system will give priority to this solution as a similar device fault case to improve the accuracy of the matching results. The query results of Elasticsearch are sorted by matching scores, and the fault cases with higher scores are more consistent with the characteristics of the current device fault. And based on these matching scores, similar fault cases are selected.

[0092] It should be noted that in the process of equipment fault diagnosis and fault solution determination, the fault knowledge base stores a large number of historical fault cases, including key information such as fault cause, fault type, fault description, and solution. By using this key information, cases similar to the current equipment fault can be quickly located, thereby providing effective solutions.

[0093] Optionally, according to the fault cause and / or fault type, a condition for querying in the fault knowledge base is constructed. The query is executed according to the query condition, and historical equipment fault cases with the same fault cause or the same fault type are screened out from the fault knowledge base.

[0094] Optionally, use the specific fault cause code and fault type name to form an accurate structure query statement and search in the fault knowledge base. Alternatively, use natural language processing or machine learning algorithms to calculate the similarity between the query conditions and historical equipment failure cases in the fault knowledge base, and use cases with similarity greater than a certain threshold as similar failure cases.

[0095] Exemplarily, Elasticsearch is used to store historical equipment failure cases, and fast retrieval is achieved through its inverted index mechanism. Each historical equipment failure case is indexed and contains multiple key fields, such as equipment type, failure type, operation steps, and historical success rate. These key fields enable the fault automation solution system to quickly locate the relevant historical equipment failure cases and the solutions corresponding to the historical equipment failure cases according to different query conditions. In Elasticsearch, each solution is stored as a document. These documents contain multiple fields, such as fault description, solution steps, related resources, and processing time. Elasticsearch shards the solution database, that is, distributes the indexed data on different nodes, realizes distributed storage and parallel query of data, and supports the storage and efficient retrieval of massive solutions. At the same time, Elasticsearch can improve the ability of fault recovery and high data availability by setting multiple shard copies. Even if a node fails, Elasticsearch can still continue to provide services and ensure the integrity of the data.

[0096] Optionally, collect user operation process, status information, error log and other information, parse this information, and extract key fields related to troubleshooting, including the name of the operation step, error code, exception information, etc. Match the extracted key fields with the fault knowledge base. The matching algorithm can determine the accuracy of the match based on factors such as the similarity between the key fields of the current equipment fault and the historical equipment fault cases recorded in the fault knowledge base, the frequency of occurrence of the key fields in the historical equipment fault cases, etc. According to the matching results, sort the historical equipment fault cases and select the similar fault cases that are most likely to be related to the current equipment fault. According to the description of the fault phenomenon in the similar fault case, compare and verify it with the fault phenomenon of the current device to ensure the consistency of the two. If the fault phenomenon in the target case is consistent with the fault phenomenon of the current device, the solution provided by it can be directly used for troubleshooting and repair.

[0097] Step S34: determining a corresponding fault solution based on the similar fault cases.

[0098] Optionally, step S34 includes steps S341-S342: Step S341, inputting the key fields and the similar fault cases into a preset random forest model, wherein the random forest model is used to predict the success rate of the reference solution according to the key fields and the reference solutions corresponding to the similar fault cases.

[0099] Step S342, obtaining the success rate output by the random forest model, and determining the fault solution in the reference solution based on the success rate.

[0100] It should be noted that the random forest model can improve the prediction accuracy by constructing multiple decision trees and synthesizing their output results. In this embodiment, the random forest model is used to predict the success rate of the reference solution corresponding to the similar fault case based on the key fields and similar fault cases.

[0101] First, use historical equipment failure cases as training data to train a random forest model. The goal of the random forest model is to predict the success rate of the reference solution based on the input key fields and the reference solutions corresponding to similar failure cases. Input the key fields of the current equipment failure into the trained random forest model and output the predicted success rate of each reference solution. Use the success rate output by the random forest model as an evaluation indicator and select the reference solution with the highest success rate as the fault solution for the current equipment failure.

[0102] Optionally, use a decision tree model to predict the effect of the reference solution corresponding to each similar fault case based on the input key fields and similar fault cases. Select a suitable decision tree construction algorithm, such as ID3 (Iterative Dichotomiser 3), CART (Classification and Regression Tree), etc. Use historical data to train the decision tree model. During the training process, the decision tree model learns how to split the data to predict the effect of the solution based on the input fault features. Use indicators such as cross-validation, accuracy, and recall to evaluate the performance of the model. Adjust the decision tree model parameters, such as the depth of the tree and the number of leaf nodes, based on the evaluation results to improve the prediction accuracy of the decision tree model. Input the new key field into the trained decision tree model, and the decision tree model will traverse the decision tree from the root node according to the input key field until it reaches the leaf node. During the traversal process, the decision tree model will select the next child node according to the splitting criterion of each node until it finds the leaf node that best matches the input key field. The leaf node will give the predicted reference solution effect, which can be a specific value such as success rate, or a category label such as "valid", "invalid", etc.

[0103] Optionally, in fault similarity matching, a KNN (K-Nearest Neighbors) algorithm is used to recommend the closest solution based on the similarity between the fault characteristics and historical equipment failure cases.

[0104] Exemplarily, historical device failure case data containing failure features and their corresponding solutions is collected, and key failure features are extracted from the historical device failure case data as input variables for the KNN algorithm. The failure features can be the extracted keyword fields or other device operating parameters. The failure features are standardized to eliminate the dimensional differences between different failure features and improve the accuracy of the algorithm. An appropriate K value is selected, and the KNN model is trained using the historical device failure case data, that is, the failure features and their corresponding solutions are stored. When a new failure occurs, its failure features are extracted and input into the KNN model. The KNN model will search for the K nearest neighbors (i.e., failures) most similar to the input failure features in the historical device failure case data, and measure the similarity between the input failure features and the historical device failure cases by calculating the distances such as Euclidean distance and Manhattan distance between the input failure features and each historical device failure case. According to the failure features and their corresponding solutions of the K nearest neighbors, the majority voting method or the average value method is used to recommend the historical device failure case closest to the input, and the solution corresponding to the historical device failure case is used as the failure solution.

[0105] Optionally, natural language input by the user is received through the user operation interface, and through NLP (Natural Language Processing) algorithms, the text is converted into retrievable keyword fields. Then, these keyword fields are indexed and matched through fuzzy matching algorithms such as edit distance and Jaccard coefficient, and the most matching similar failure cases are returned.

[0106] Exemplarily, the user operation interface can be a simple text box where the user can input failure information described in natural language. For example, there may be a search box on the interface labeled "Please enter failure description". The failure automatic resolution system receives the text input by the user, such as: "Device A cannot be powered on, the screen is always black and the power light is flashing." The NLP algorithm is used to split the text into individual words or phrases, identify the part of speech and specific entities of each word, and remove the words meaningless for retrieval, such as "of", "already", etc. The keyword fields meaningful for retrieval are extracted from the processed text, such as: "Device A", "cannot be powered on", "screen black", "power light flashing". Then, the case descriptions in the historical device failure cases are also processed by NLP, and keyword fields are extracted. The similarity between the two sets of keyword fields is calculated by the Jaccard coefficient, and the historical device failure cases are sorted according to the similarity, and the case most matching the user input is returned.

[0107] Based on the above embodiments of the present application, in the fifth embodiment of the present application, the same or similar content as the above embodiments can be referred to the above introduction and will not be repeated hereinafter. On this basis, please refer to Figure 6After step S30, the method for determining a fault solution further includes steps S80 to S110: Step S80, obtaining a work order template based on the key fields.

[0108] Step S90: Generate a standard fault work order based on the work order template and the fault solution.

[0109] Exemplarily, the automated fault resolution system has multiple built-in work order templates, which are classified according to key fields such as fault cause and fault type. According to the key fields, the most suitable work order template is selected from the preset template library. For example, for a hardware failure caused by a damaged hard disk, a hardware repair template containing hard disk replacement steps is selected. After selecting the work order template and the fault solution, the fault solution is filled into the work order template to generate a standard fault work order. The content of the standard fault work order includes detailed information such as fault description, solution steps, required resources, and estimated completion time.

[0110] Step S100: sending the standard fault work order to the user end.

[0111] Step S110: If no feedback information from the user terminal is received within a preset time, the standard fault work order is sent to the technician terminal.

[0112] Exemplarily, the automated fault resolution system also supports automatic scheduling, which first pushes the standard fault work order to the user end, such as sending the work order to the user end such as a personal computer, mobile device, etc. through email, SMS or internal message, so that the user can handle the equipment failure by himself according to the fault solution and reduce waiting time. If the user solves the equipment failure by himself within the preset time, feedback information will be provided to the automated fault resolution system. If the user cannot solve the equipment failure by himself according to the fault solution within the preset time, the automated fault resolution system will reassign the order to the technician or team, and the technician or team will assist in solving the failure to shorten the equipment failure processing time.

[0113] Optionally, for simple faults, the automated fault resolution system will directly push fault resolution solutions to users via SMS, email, or a self-service portal. Users can solve the problem by themselves through the detailed steps provided by the automated fault resolution system, reducing manual intervention. Figure 7 , Figure 7The flowchart for the generation and dispatch of standard fault work orders in this embodiment is as follows. Specifically: the user operation process is obtained from the equipment monitoring system, and the user operation process is compared and matched with the fault cases in the fault knowledge base to find similar fault cases similar to the current equipment fault, and then the fault solution is determined based on the similar fault case. After the fault solution is generated, it can be determined whether technical support is needed based on the fault solution. If technical support is not required, the fault solution corresponding to the similar fault case is directly pushed to the user, and the user solves the fault independently according to the fault solution. If technical support is required, a standard fault work order is generated and sent to the technician, who will provide on-site service to assist in troubleshooting. Among them, the newly generated standard fault work order information is synchronized to the fault knowledge base.

[0114] Exemplarily, a self-service portal is provided for users, through which users can query fault information, obtain operation guides and solutions, or handle faults by themselves.

[0115] Among them, the homepage of the self-service portal adopts a card-style layout, placing common operation guides and common problems in a prominent position for customers to quickly access. At the same time, the homepage contains quick entrances to fault types and equipment types. Users can directly select the corresponding key fields such as fault types to enter the relevant solution page. The navigation menu at the top or side of the homepage of the self-service portal allows users to quickly switch between modules such as equipment classification, fault type, and solution documents. The self-service portal also includes a global search bar that supports key field search and advanced search. Advanced search allows users to perform precise queries based on conditions such as equipment type and fault code. In order to improve the search experience, users can select the equipment-related information fields defined in the fault automation solution system or user-defined fields for fuzzy or precise matching searches. The self-service portal adopts a responsive design to ensure a good user experience on different devices such as computers and tablets. The homepage page layout will automatically adjust according to the screen size of the device to ensure the readability of the content and the convenience of operation.

[0116] When the user enters a fault description or selects a device type or fault type in the self-service portal, the automated fault resolution system automatically recommends possible solutions through built-in NLP algorithms and search algorithms. Users can view detailed steps by clicking on the solution card. For each solution, the self-service portal provides a detailed operation guide, including text instructions, diagrams, and video tutorials to ensure that users can understand and execute the solution steps. If the user encounters any problems during the operation, he can send a message to the technical support team or get further guidance through the real-time feedback function built into the self-service portal. At the same time, the automated fault resolution system will automatically adjust the recommended solution or generate a new work order based on the user's feedback.

[0117] When users solve problems on their own according to the solutions provided by the self-service portal, the automated fault resolution system will monitor the user's operation progress in real time to determine whether the problem has been resolved. If the problem is successfully resolved, the automated fault resolution system will ask the user to confirm and close the relevant work order or task. If the user cannot solve the problem by self-service, after the user provides feedback, the self-service portal will automatically generate a new work order and notify the relevant technical support team. Users can also choose to communicate with technicians online for further help.

[0118] In order to ensure information security and controllability of the automated fault resolution system, the self-service portal has designed a strict permission management automated fault resolution system to ensure that only authorized users can access specific information or perform certain operations. This includes setting different roles and permissions for different users, such as ordinary users, advanced users, administrators, etc. Ordinary users can view common fault resolution solutions and operation guides; advanced users such as technical support personnel can access more detailed technical documents and diagnostic tools; administrators have management permissions for the entire automated fault resolution system, including user management, content updates, and other operations. The scope of permissions corresponding to each role is finely defined. For example, some advanced solutions or sensitive information may only be open to specific user groups, and ordinary users cannot access them. The automated fault resolution system dynamically determines the user's permission level through the access control list and performs corresponding access control.

[0119] When users access the self-service portal, they can log in through multi-factor authentication such as password + SMS verification code to ensure that the access is by authenticated users. For high-authority operations, the automated fault resolution system will verify permissions again to ensure that only users with corresponding permissions can perform operations. All user data is encrypted during transmission and storage to prevent information leakage. At the same time, the automated fault resolution system also records all operation logs, so that administrators can audit user behavior at any time and perform risk control when abnormal operations are found.

[0120] Optionally, the fault automation solution system will regularly extract valuable information from newly generated standard fault work orders and fault data. In particular, when a new type of fault occurs and is effectively solved, the fault automation solution system will automatically capture the key points of the fault solution and add it as a new knowledge item to the fault knowledge base. The fault automation solution system can also regularly analyze all new standard fault work order data according to the set time interval, and filter out the knowledge items that need to be added or updated for regular updates; at the same time, when a certain type of fault occurs frequently in a short period of time, or a new fault solution is verified to be effective many times, the fault automation solution system will immediately trigger the update of the fault knowledge base. In order to ensure the accuracy of new information, the fault automation solution system can design a multi-layer verification mechanism. Whenever a new fault solution is added to the fault knowledge base, the fault automation solution system will first perform internal verification, that is, match it with the existing fault mode and solution to check the rationality of the new solution. Secondly, the fault automation solution system will perform external verification based on the feedback of real-time applications, such as evaluating the actual effectiveness of the new information through user feedback or the solution results of the re-occurrence of the fault. For more complex fault types or new fault solutions, the fault automation solution system can be set to require expert review and confirmation. In addition, the automated fault resolution system also has an automatic optimization function, which regularly reorganizes and simplifies the information in the fault knowledge base based on data mining technology. The automated fault resolution system analyzes the application frequency and effect of various fault resolution solutions, prioritizes the most effective and most commonly used solutions, and eliminates duplicate or redundant information. This dynamic optimization mechanism ensures that the fault knowledge base can remain efficient and accurate over time.

[0121] For example, please refer to Figure 8 , Figure 8The architecture diagram of the fault automation solution system is provided. Specifically, the B-end is the top layer of the entire architecture and is an interface for multiple tenants, including tenant A, tenant B, and tenant C. The B-end contains multiple subsystems to meet the different needs of tenants. These subsystems include the warehousing system, logistics system, work order system, and process system. Among them, the warehousing system is used to manage and track inventory to ensure efficient storage and distribution of items; the logistics system is responsible for the transportation and distribution of goods to ensure that the items can arrive at the designated location on time. The work order system is used to handle repair and maintenance requests submitted by customers or internal personnel, and track the progress and completion of tasks. The business middle platform mainly includes functions such as equipment technical support, equipment maintenance management, and work order time management. Among them, equipment technical support includes functions such as fault diagnosis, remote support, and technical document provision to help users solve problems encountered during the use of equipment; equipment maintenance management is used to formulate and implement equipment maintenance plans to ensure the normal operation of equipment and extend its service life; work order time management is used to monitor the creation, allocation, processing, and completion status of work orders to ensure that tasks can be completed efficiently within the specified time. These functions work together to ensure efficient operation of equipment and timely processing of work orders. The functional background includes knowledge base integration, fault monitoring and early warning, and fault analysis and push services to support the stable operation and fault handling of the entire system. The enterprise knowledge base contains equipment operation manuals, troubleshooting guides, FAQs, etc., providing users with rich information support.

[0122] It should be noted that the above examples are only used to understand the present application and do not constitute a limitation on the method for determining the fault solution of the present application. More simple transformations based on this technical concept are all within the scope of protection of the present application.

[0123] The present application provides a device for determining a fault solution, the device for determining a fault solution comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the method for determining a fault solution in the above-mentioned embodiment one.

[0124] Reference below Fig. 9 , which shows a schematic diagram of a structure of a device for determining a fault solution suitable for implementing an embodiment of the present application. The device for determining a fault solution in the embodiment of the present application may include but is not limited to a mobile terminal such as a laptop computer, a PAD (Portable Application Description: tablet computer), etc., and a fixed terminal such as a desktop computer. Fig. 9 The fault solution determination device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0125] like Fig. 9 As shown, the device for determining a fault solution may include a processing device 1001 (e.g., a central processing unit, a graphics processor, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM: Read Only Memory) 1002 or a program loaded from a storage device 1003 to a random access memory (RAM: Random Access Memory) 1004. In RAM1004, various programs and data required for the operation of the device for determining a fault solution are also stored. The processing device 1001, ROM1002, and RAM1004 are connected to each other through a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Generally, the following systems can be connected to the I / O interface 1006: an input device 1007 including, for example, a touch screen, a touch pad, a keyboard, a mouse, an image sensor, a microphone, etc.; an output device 1008 including, for example, a liquid crystal display (LCD: Liquid Crystal Display), a speaker, a vibrator, etc.; a storage device 1003 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 1009. The communication device 1009 can allow the fault solution determination device to communicate with other devices wirelessly or by wire to exchange data. Although the figure shows a fault solution determination device with various systems, it should be understood that it is not required to implement or have all the systems shown. More or fewer systems can be implemented or have alternatively.

[0126] In particular, according to the embodiments disclosed in the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present application include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device, or installed from a storage device 1003, or installed from a ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiment disclosed in the present application are executed.

[0127] The device for determining a fault solution provided by the present application adopts the method for determining a fault solution in the above embodiment, and can solve the technical problem of how to automatically generate a solution to shorten the equipment maintenance time. Compared with the prior art, the beneficial effects of the device for determining a fault solution provided by the present application are the same as the beneficial effects of the method for determining a fault solution provided by the above embodiment, and the other technical features in the device for determining a fault solution are the same as the features disclosed in the method of the previous embodiment, and are not described in detail here.

[0128] It should be understood that the various parts disclosed in this application can be implemented by hardware, software, firmware or a combination thereof. In the description of the above embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0129] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0130] The present application also provides a computer program product, including a computer program, which implements the steps of the method for determining a fault solution as described above when the computer program is executed by a processor.

[0131] The computer program product provided by the present application can solve the technical problem of how to automatically generate solutions to shorten the equipment maintenance time. Compared with the prior art, the beneficial effects of the computer program product provided by the present application are the same as the beneficial effects of the method for determining the fault solution provided by the above embodiment, which will not be repeated here.

[0132] The above descriptions are only some embodiments of the present application, and are not intended to limit the patent scope of the present application. All equivalent structural changes made using the contents of the present application specification and drawings under the technical concept of the present application, or direct / indirect applications in other related technical fields are included in the patent protection scope of the present application.

Claims

1. A method for determining a fault solution, characterized in that: The method comprises: When a device fails, determine the user operation process based on the device operation data, and compare the user operation process with the standard operation process; If the comparison result between the user operation process and the standard operation process is inconsistent, determining the key fields according to the user operation process; According to the key fields, similar fault cases are determined in a fault knowledge base, and corresponding fault solutions are determined based on the similar fault cases, wherein the fault knowledge base is formed by historical equipment fault cases.

2. The method for determining a fault solution according to claim 1, characterized in that: Before the step of determining the user operation process based on the device operation data and comparing the user operation process with the standard operation process, the method further includes: Obtaining step records in historical normal work order data, and determining at least one operation step in the step records; determining a first frequency at which each of the operation steps occurs in the step record; If the first frequency is greater than a first frequency threshold, determining a sequential sequence of the operating steps; Based on the sequential sequence of each of the operating steps, the standard operating procedure is determined.

3. The method for determining a fault solution according to claim 2, characterized in that: The step of determining the sequential sequence of the operation steps if the first frequency is greater than a first frequency threshold comprises: If the first frequency is greater than a first frequency threshold, generating a step rule based on the sequence of every two of the operation steps; determining a second frequency at which the step rule occurs in the step record; If the second frequency is greater than a second frequency threshold, a sequential sequence of the operation steps is determined based on the step rule.

4. The method for determining a fault solution according to claim 1, characterized in that: The step of determining the user operation process based on the device operation data and comparing the user operation process with the standard operation process includes: Based on the device operation data, obtain the timestamp of each operation step in the user operation process; Based on the timestamp, determining the sequence of the operation steps in the user operation process; If the sequence of the operation steps in the user operation process and the sequence of the operation steps in the standard operation process are inconsistent, it is determined that the comparison result is inconsistent.

5. The method for determining a fault solution according to claim 1, characterized in that: The step of determining the user operation process based on the device operation data and comparing the user operation process with the standard operation process also includes: Obtain the standard equipment operation data range corresponding to the standard operating procedure; If the device operation data corresponding to the user operation process is not within the range of the standard device operation data, it is determined that the comparison result is inconsistent.

6. The method for determining a fault solution according to claim 1, wherein: The step of determining similar fault cases in a fault knowledge base according to the key fields, and determining corresponding fault solutions based on the similar fault cases includes: Determining the weight of the key field; Determine a matching score between the user operation process and the historical device failure case based on the key fields and the weights corresponding to the key fields; If the matching score is greater than a preset score threshold, determining that the historical equipment failure case is a similar failure case; Determine corresponding fault solutions based on the similar fault cases.

7. The method for determining a fault solution according to claim 6, characterized in that: The step of determining a corresponding fault solution based on the similar fault case comprises: Inputting the key fields and the similar fault cases into a preset random forest model, wherein the random forest model is used to predict the success rate of the reference solution according to the key fields and the reference solutions corresponding to the similar fault cases; The success rate of the random forest model output is obtained, and the fault solution is determined in the reference solution based on the success rate.

8. The method for determining a fault solution according to claim 1, characterized in that: After the step of determining similar fault cases in a fault knowledge base according to the key fields and determining corresponding fault solutions based on the similar fault cases, the method further includes: Based on the key fields, obtain a work order template; Generate a standard fault work order based on the work order template and the fault solution; Sending the standard fault work order to the user end; If no feedback information is received from the user end within the preset time, the standard fault work order is sent to the technician end.

9. A device for determining a fault solution, characterized in that: The device comprises: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program is configured to implement the steps of the method for determining a fault solution according to any one of claims 1 to 8.

10. A computer program product, characterized in that The computer program product comprises a computer program, and when the computer program is executed by a processor, the steps of the method for determining a fault solution according to any one of claims 1 to 8 are implemented.

Images