Autonomous planning intelligent troubleshooting device, method, equipment and medium

Abstract

The invention discloses an autonomous planning intelligent troubleshooting device and method, equipment and a medium. The input module quickly receives a troubleshooting request of a user and starts a troubleshooting process. The troubleshooting task autonomous planning module utilizes a pre-trained equipment troubleshooting large model, combines working data and troubleshooting path text prompt prompt, and automatically reasones a troubleshooting path and diagnosis steps, so that time is saved, blind troubleshooting is avoided, and troubleshooting efficiency is improved. The troubleshooting task execution module carries out troubleshooting according to planning, the process is not interfered by manual quality, and the troubleshooting accuracy is effectively improved. Meanwhile, the module allocates maintenance resources and time windows according to fault risk levels by means of a large model, and reasonable and efficient utilization of the resources is achieved. In addition, knowledge can be obtained from an equipment operation and maintenance troubleshooting database, maintenance suggestions are determined, and powerful data support is provided for maintenance work.

Description

Technical Field

[0001] The present application relates to the fields of artificial intelligence and urban traffic operation and maintenance management, and in particular to an autonomously planned intelligent troubleshooting device, method, equipment and medium. Background Art

[0002] In the field of modern industry and transportation, the stable operation of equipment is of vital importance. Taking rail transportation as an example, the normal operation of key equipment such as trains is directly related to the safety and efficiency of public travel. At present, the commonly used equipment monitoring and maintenance technologies mainly include the following:

[0003] The first is the fixed rule engine, which monitors the operation status of the equipment in real time based on preset rules, and notifies relevant personnel in a timely manner once an alarm condition is triggered. This method can ensure the safety of equipment operation to a certain extent, and can issue an alarm in a timely manner when the equipment has obvious abnormal conditions.

[0004] The second is the expert system, which consists of a series of decision trees or rule sets written by domain experts and is mainly used to guide troubleshooting. Through the experience and knowledge accumulation of experts, it provides maintenance personnel with ideas and methods for troubleshooting.

[0005] In addition, manual recording and reporting is also a common method. Maintenance personnel need to regularly check the equipment and manually record the status information of the equipment. When encountering complex problems, they also need to rely on external support to solve them.

[0006] However, these traditional technologies have exposed many problems in practical applications. First, the response speed is slow. It takes a long time from problem discovery to problem solving, especially in the face of sudden failures, which may cause train delays or even suspension of operation, seriously affecting the normal operation order. Second, excessive reliance on manual experience and differences in professional knowledge and technical levels of different maintenance personnel may lead to unstable processing results and fail to ensure that every failure can be solved in a timely and effective manner. Third, lack of predictive ability, traditional systems are difficult to warn of potential failures in advance and cannot achieve preventive maintenance, which not only increases the cost of later repairs, but also increases the difficulty of repairs. Fourth, insufficient data utilization and failure to effectively integrate historical fault data and real-time monitoring data limit the value mining of data analysis and fail to give full play to the role of data in equipment monitoring and maintenance. Summary of the invention

[0007] The present application provides an autonomously planned intelligent troubleshooting device, method, equipment and medium, which are used to solve the problems of low troubleshooting efficiency and poor troubleshooting quality in existing troubleshooting systems.

[0008] In a first aspect, the present application provides an autonomous planning intelligent troubleshooting device, the device comprising:

[0009] An input module, used for receiving a troubleshooting request for a target device input by a user;

[0010] The troubleshooting task autonomous planning module is used to obtain the working data of the target device in response to the troubleshooting request; through the pre-trained equipment troubleshooting model, based on the working data and the troubleshooting path text prompt, the working data is inferred to determine the troubleshooting path of the target device and the diagnostic steps corresponding to each fault node in the troubleshooting path;

[0011] A troubleshooting task execution module is used to troubleshoot the target device according to the execution order between each fault node in the troubleshooting path and the diagnostic steps of the current fault node to determine the target fault currently existing in the target device; obtain the risk level corresponding to each target fault; through the equipment troubleshooting big model, according to the risk level corresponding to each target fault, allocate corresponding resources and time windows for the maintenance of each target fault; through the equipment troubleshooting big model, according to the knowledge information in the pre-built equipment operation and maintenance troubleshooting database, determine the maintenance suggestions for each target fault; wherein the knowledge information includes: the structural principle of the equipment, the maintenance manual, and the common fault solutions;

[0012] The output module is used to output the resources and time windows corresponding to the maintenance allocation of each of the target faults, as well as the maintenance suggestions for each of the target faults.

[0013] In a second aspect, the present application also provides a method for troubleshooting equipment operation and maintenance failures, the method comprising:

[0014] Receiving a troubleshooting request for a target device input by a user;

[0015] Acquire working data of the target device;

[0016] Through the pre-trained equipment troubleshooting model, based on the working data and the troubleshooting path text prompt, the working data is inferred to determine the troubleshooting path of the target equipment and the diagnostic steps corresponding to each fault node in the troubleshooting path;

[0017] According to the execution order between each fault node in the troubleshooting path and according to the diagnosis steps of the current fault node, troubleshoot the target device to determine the target fault currently existing in the target device;

[0018] Obtaining the risk level corresponding to each of the target faults;

[0019] By using the equipment troubleshooting model, corresponding resources and time windows are allocated for the maintenance of each target fault according to the risk level corresponding to each target fault;

[0020] Through the equipment troubleshooting model, the maintenance suggestions for each target fault are determined according to the knowledge information in the pre-built equipment operation and maintenance troubleshooting database; wherein the knowledge information includes: the structural principle of the equipment, maintenance manual, and common fault solutions;

[0021] Output the resources and time window corresponding to the maintenance allocation of each of the target faults, as well as the maintenance suggestions for each of the target faults.

[0022] In a third aspect, the present application provides a computer device, which includes a processor, and the processor is used to implement the steps of the autonomous planning and intelligent troubleshooting method as described above when executing a computer program stored in a memory.

[0023] In a fourth aspect, the present application provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the autonomous planning intelligent troubleshooting method as described above.

[0024] The beneficial effects of this application are as follows:

[0025] 1. The input module can quickly receive the user's troubleshooting request and quickly start the troubleshooting process. The troubleshooting task autonomous planning module uses the pre-trained equipment troubleshooting model, combined with work data and troubleshooting path text prompts, to automatically infer the troubleshooting path and the corresponding diagnostic steps for each fault node. Compared with manual experience to explore the troubleshooting sequence, this automated planning method greatly saves time, avoids blind troubleshooting, and makes troubleshooting more organized and targeted, so that the fault point can be quickly located, significantly improving the overall troubleshooting efficiency.

[0026] 2. When reasoning about work data, the pre-trained equipment troubleshooting model can comprehensively consider various factors and accurately determine the troubleshooting path and diagnostic steps. The troubleshooting task execution module performs troubleshooting according to the planned path and steps, and can systematically find the target fault, reducing the misjudgment and omission of faults caused by human negligence or lack of experience, and improving the accuracy of troubleshooting.

[0027] 3. The troubleshooting task execution module uses the equipment troubleshooting model to allocate corresponding resources and time windows for maintenance according to the risk level of each target fault. This resource allocation method based on risk level can make maintenance resources more reasonable and efficient, and optimize the resource allocation of the enterprise.

[0028] 4. The troubleshooting task execution module uses the equipment troubleshooting model to obtain knowledge information from the pre-built equipment operation and maintenance troubleshooting database, and then determines the maintenance suggestions for each target fault. The database contains a variety of knowledge such as the structural principles of the equipment, maintenance manuals, and common fault solutions, making the maintenance suggestions more comprehensive, professional, and scientific. Maintenance personnel can quickly formulate maintenance plans based on these suggestions to improve the quality and efficiency of maintenance work. At the same time, it also provides maintenance personnel with learning and reference materials to help improve their maintenance skills. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0030] Figure 1 A schematic diagram of the structure of an autonomous planning intelligent troubleshooting device provided in an embodiment of the present application;

[0031] Figure 2 A schematic diagram of a process of autonomous planning and intelligent troubleshooting provided in an embodiment of the present application;

[0032] Figure 3 It is a structural schematic diagram of a computer device provided in an optional embodiment of the present application. DETAILED DESCRIPTION

[0033] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. Obviously, the described embodiments are only 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 ordinary technicians in this field without making creative work are within the scope of protection of the present application.

[0034] In order to improve the efficiency and quality of troubleshooting, the present application provides an autonomously planned intelligent troubleshooting device, method, equipment and medium.

[0035] Embodiment 1:

[0036] This application provides an autonomous planning intelligent troubleshooting device. Figure 1 A schematic diagram of the structure of an autonomous planning intelligent troubleshooting device provided in an embodiment of the present application, the device comprising:

[0037] An input module 11 is used to receive a troubleshooting request for a target device input by a user;

[0038] The troubleshooting task autonomous planning module 12 is used to obtain the working data of the target device in response to the troubleshooting request; through the pre-trained equipment troubleshooting model, based on the working data and the troubleshooting path text prompt, the working data is inferred to determine the troubleshooting path of the target device and the diagnostic steps corresponding to each fault node in the troubleshooting path;

[0039] The troubleshooting task execution module 13 is used to troubleshoot the target device according to the execution order between each fault node in the troubleshooting path and the diagnostic steps of the current fault node to determine the target fault currently existing in the target device; obtain the risk level corresponding to each target fault; through the equipment troubleshooting big model, according to the risk level corresponding to each target fault, allocate corresponding resources and time windows for the maintenance of each target fault; through the equipment troubleshooting big model, according to the knowledge information in the pre-built equipment operation and maintenance troubleshooting database, determine the maintenance suggestions for each target fault; wherein the knowledge information includes: the structural principle of the equipment, the maintenance manual, and the common fault solutions;

[0040] The output module 14 is used to output the resources and time windows corresponding to the maintenance allocation of each of the target faults, as well as the maintenance suggestions for each of the target faults.

[0041] This autonomous planning intelligent troubleshooting intelligent agent device is designed to realize autonomous troubleshooting and maintenance planning of target equipment, and improve the efficiency and accuracy of equipment fault handling. The device is mainly composed of core parts such as input module 11, autonomous troubleshooting task planning module 12, troubleshooting task execution module 13, and output module 14. Optionally, the device also includes auxiliary modules such as equipment operation and maintenance fault diagnosis API protocol call module and equipment operation and maintenance fault diagnosis memory storage module. The specific implementation method of each module will be described in detail below.

[0042] (I) Input module 11

[0043] The input module 11 is an interface for the device to interact with the user, and its main function is to receive a troubleshooting request for the target device input by the user. The user can send a troubleshooting request to the input module 11 in a variety of ways, such as inputting request information on the device's operation interface, or sending a request in the form of a data packet in a specific format through a network interface.

[0044] In order to ensure the accuracy and completeness of the troubleshooting request, the input module 11 can perform preliminary verification and processing on the information input by the user. For example, it can check whether the request contains the necessary identification information of the target device, such as the device name, model, serial number, etc. If it is found that the input information is incomplete or erroneous, the input module 11 can prompt the user to re-enter or supplement it.

[0045] (II) Troubleshooting Task Autonomous Planning Module 12

[0046] After receiving the troubleshooting request, the troubleshooting task autonomous planning module 12 has the primary task of obtaining the working data of the target device. Among them, the working data includes one or more of the equipment operation data, historical fault data and environmental monitoring data. The equipment operation data can be collected in real time through various sensors installed on the equipment, such as temperature sensors, pressure sensors, current sensors, etc. These sensors convert the operating parameters of the equipment into electrical signals and transmit them to the device through the data acquisition system. The historical fault data can be stored in the equipment operation and maintenance troubleshooting database, and the troubleshooting task autonomous planning module 12 can directly query the historical fault records of the target equipment from the database, including information such as the time of the fault, the fault phenomenon, and the fault handling results. Environmental monitoring data can be obtained through environmental monitoring sensors, such as temperature and humidity sensors, air quality sensors, etc. These data can help analyze whether the equipment failure is related to environmental factors.

[0047] In a possible implementation, the device may also include an equipment operation and maintenance fault diagnosis API protocol calling module, through which the device can use a standardized API interface to call external tools and services to obtain the working data in the above embodiments. For example, by calling the API interface of the sensor data acquisition system, the equipment operation data can be obtained in real time; by calling the API interface of the historical fault database, the historical fault data of the target device can be queried. This method enables the device to be flexibly integrated with various external systems, improving the efficiency and accuracy of data acquisition.

[0048] These API interfaces provide a unified access method, making it easy for devices to integrate various services and technical resources, and also allow devices to receive status updates and feedback information from other systems in real time. This mechanism helps expand the information sources of intelligent agents, enhance their ability to perceive the global situation, and promote information sharing and collaboration.

[0049] In this application, before reasoning, it is necessary to pre-train a large model for equipment troubleshooting. The model can use deep learning algorithms, such as neural networks, Transformers, etc. The training data can come from the historical working data, maintenance records, expert experience, etc. of the equipment. For example, a large amount of working data and troubleshooting information related to the target equipment can be collected, including data when the equipment is operating normally, data when various faults occur, and corresponding troubleshooting paths and diagnostic steps. The collected data is pre-processed by cleaning, labeling, and feature extraction, and converted into a format suitable for model training. The model is trained using the pre-processed data, and the parameters of the model are continuously adjusted so that the model can accurately infer the troubleshooting path and diagnostic steps based on the working data and the troubleshooting path text prompt.

[0050] In one example, the equipment troubleshooting model can also be obtained by fine-tuning the general text understanding model based on the equipment operation and maintenance industry knowledge samples. These equipment operation and maintenance industry knowledge samples include but are not limited to the following types: equipment failure case reports, equipment technical documents, industry standards and specifications, expert experience, equipment operation logs, and maintenance records. The equipment failure case report records in detail the specific circumstances of failures of different equipment under various working conditions, including failure phenomena, troubleshooting process, final determination of the cause of the failure, and maintenance measures taken. By learning a large number of failure case reports, the equipment troubleshooting model can master the characteristics and troubleshooting methods of common faults. Equipment technical documents, such as equipment design drawings, technical specifications, operating manuals, etc., contain key information such as the structure, working principle, and performance parameters of the equipment, which helps the equipment troubleshooting model understand the internal operating mechanism of the equipment, so as to more accurately reason about the fault. Industry standards and specifications cover the standards and requirements of each link of equipment installation, commissioning, operation, and maintenance. After learning these standards and specifications, the equipment troubleshooting model can judge whether the operating status of the equipment meets the industry requirements, providing an important basis for fault diagnosis. Expert experience collects the valuable experience accumulated by experts in the field of equipment operation and maintenance in long-term practice, including fault diagnosis skills, methods for quickly locating faults, and strategies for handling special faults. These expert experiences can help the equipment troubleshooting model to handle some complex and rare fault conditions. The equipment operation log records the real-time operating status, parameter changes, start and stop time, and other information of the equipment. By analyzing a large number of operation logs, the equipment troubleshooting model can understand the normal operating mode and abnormal characteristics of the equipment and improve the accuracy of fault diagnosis. The maintenance record contains the details of the equipment's previous maintenance, such as maintenance time, maintenance content, replaced parts, etc. These maintenance records help the equipment troubleshooting model analyze the recurrence of faults and the maintenance effect, and provide a reference for formulating a more reasonable troubleshooting path.

[0051] Exemplarily, a general text understanding large model can be selected as the base model of the equipment troubleshooting large model, and then the base model can be fine-tuned based on the collected equipment operation and maintenance industry knowledge samples, so that the fine-tuned large model is more suitable for the field of equipment troubleshooting data analysis. Specifically, a deep learning framework is used to pre-train on large-scale general text data, so that the trained general text understanding large model has basic language understanding and generation capabilities. Then, based on the collected equipment operation and maintenance industry knowledge samples, the pre-trained general text understanding large model is fine-tuned. During the fine-tuning process, the equipment operation and maintenance industry knowledge samples and the corresponding instructions are input into the model, so that the general text understanding large model learns how to extract key information from the equipment operation and maintenance industry knowledge samples according to the instructions, perform logical reasoning, and generate outputs that meet the equipment troubleshooting analysis needs. In this way, the general text understanding large model can better adapt to the equipment troubleshooting task and can more accurately understand and process information related to equipment troubleshooting.

[0052] In a possible implementation, the collected training samples can be cleaned to remove duplicate and erroneous data, unify the data format, and ensure the quality of the training samples. High-quality training samples can improve the training effect and accuracy of the model and avoid deviations caused by erroneous or inconsistent data.

[0053] After obtaining the pre-trained equipment troubleshooting model based on the above embodiment, the pre-trained equipment troubleshooting model can be directly deployed in the server cluster where the system is located, or the pre-trained equipment troubleshooting model can be deployed on an independent server or cloud platform by using the system call engine, and called through the network interface. In this way, the model deployment method can be flexibly selected according to actual business needs and system architecture to meet the equipment troubleshooting analysis needs in different scenarios.

[0054] Based on the above embodiment, after acquiring the working data, the troubleshooting task autonomous planning module 12 can use a pre-trained equipment troubleshooting model to infer the working data based on the acquired working data and the troubleshooting path text prompt, thereby determining the troubleshooting path of the target equipment and the diagnostic steps corresponding to each fault node in the troubleshooting path.

[0055] The troubleshooting path text prompt is a text message that guides the model to perform reasoning, and it can contain some rules, strategies, or prior knowledge about troubleshooting. For example, the prompt can be "prioritize troubleshooting of fault nodes related to equipment alarm information."

[0056] The equipment troubleshooting model will analyze and reason based on the working data and the troubleshooting path prompt to determine the troubleshooting path of the target device and the diagnostic steps corresponding to each fault node in the troubleshooting path. Among them, the troubleshooting path is an ordered sequence of fault nodes, indicating the order of troubleshooting; the diagnostic steps are the specific operations for fault diagnosis for each fault node, such as checking the output value of a sensor, testing the connectivity of a circuit, etc.

[0057] (III) Troubleshooting task execution module 13

[0058] The troubleshooting task execution module 13 performs troubleshooting on the target device according to the troubleshooting path and diagnostic steps determined by the troubleshooting task autonomous planning module 12, and completes subsequent maintenance resource allocation and maintenance suggestion determination. Specifically, the troubleshooting task execution module 13 performs troubleshooting on the target device according to the execution order between each fault node in the troubleshooting path determined by the troubleshooting task autonomous planning module 12 and according to the diagnostic steps of the current fault node. For example, if the first fault node in the troubleshooting path is "sensor failure", the corresponding diagnostic step is "check whether the power supply of the sensor is normal", and the troubleshooting task execution module 13 will check the power supply of the sensor by controlling the corresponding detection equipment or tools. If the power supply is found to be abnormal, further check whether the power supply line is broken and whether the power supply module is damaged. During the troubleshooting process, the troubleshooting task execution module 13 can control the relevant detection equipment to detect the target device according to the requirements of the diagnostic steps, such as using a multimeter to detect the voltage and current of the circuit, and using an oscilloscope to detect the waveform of the signal.

[0059] In a possible implementation, the troubleshooting task execution module 13 can also monitor the changes in the working data during the troubleshooting process in real time. If new abnormal data is found, the new abnormal data is fed back to the troubleshooting task autonomous planning module 12. The troubleshooting task autonomous planning module 12 re-reasons through the equipment troubleshooting model and updates the troubleshooting path and diagnostic steps. This real-time feedback and dynamic adjustment mechanism enables the device to respond to complex and changeable equipment failures more flexibly.

[0060] Considering that different faults have different effects on target devices, some faults may only cause minor performance degradation, such as sensor reading offset or software error, which can be solved by simple calibration or restart; while others may cause serious safety accidents, such as fire caused by motor overheating, or equipment damage, such as circuit board burnout. These faults require immediate measures to prevent catastrophic consequences. Therefore, the degree of harm varies significantly, and accurate classification and risk assessment of faults are crucial.

[0061] In the present application, in order to efficiently and accurately manage these fault problems, after completing the troubleshooting, the troubleshooting task execution module 13 will determine the target faults currently existing in the target device, and then determine the risk level corresponding to each target fault. Among them, the risk level can be determined by pre-configuring the risk levels corresponding to different fault types, such as in scenarios with a small amount of data and relatively fixed fault types, or it can be determined by a pre-trained risk assessment model, such as in scenarios with rich data and requiring more accurate assessments. For example, in the case of pre-configured risk levels, equipment short-circuit faults may be pre-set to a high-risk level, because once a short-circuit fault occurs, it is very likely to cause a fire, which will not only cause serious damage to the equipment, but may also endanger the life safety of the operator, causing huge economic losses and social impacts; while minor surface wear faults of the equipment may be set to a low-risk level. Such faults will not have a substantial impact on the normal operation of the equipment in a short period of time, and only need to be paid attention to in subsequent maintenance.

[0062] In an example, the troubleshooting task execution module 13 is specifically used to determine the risk level of each target fault based on the working data and the target fault through a pre-trained risk assessment model.

[0063] Considering that the same fault type has different effects on the equipment under different degrees of loss. Based on this, in this application, a pre-trained risk assessment model can also be used to determine the risk level, and the risk assessment model will take the degree of loss of the fault as an important assessment dimension. For example, in the scenario of automobile engine fault assessment, the same engine oil leakage fault, if it is only a slight oil leakage, will not have a significant impact on the performance of the engine in a short time, nor will it endanger driving safety. At this time, the risk assessment model evaluates the risk level of the fault based on the collected oil leakage degree data, combined with the working state of the engine, vehicle mileage and other working data, and assesses that the risk level of the fault is low; when the engine has a serious oil leakage, the oil pressure drops sharply, and the engine may be damaged at any time due to insufficient lubrication, which may cause the vehicle to break down or even cause a traffic accident. At this time, the risk assessment model will determine the risk level of the fault as high based on this information through complex calculation logic. By incorporating the degree of fault loss into the risk assessment model, it is possible to more carefully grade the risks of faults in different situations, so that the risk assessment results are more in line with the actual situation, provide more targeted guidance for equipment maintenance and fault handling, and effectively improve the efficiency and accuracy of equipment fault management. For example, after completing the troubleshooting, the troubleshooting task execution module 13 will determine the target fault currently existing in the target device, and can determine the risk level of each target fault based on the working data of the target device and the target fault through a pre-trained risk assessment model. The working data can reflect the degree of loss of the target fault.

[0064] In one example, the risk assessment model is trained as follows:

[0065] Obtain any working data sample in the target device sample set and its corresponding fault types; wherein each fault type of the working data sample corresponds to an actual risk level;

[0066] For each of the fault types, determining the expected risk level corresponding to the fault type based on the working data sample by using the original risk assessment model;

[0067] Based on the actual risk levels and expected risk levels corresponding to the fault types, the original risk assessment model is trained to obtain a trained risk assessment model.

[0068] In order to obtain a risk assessment model that can accurately assess risks, in this application, a target equipment sample set can be collected in advance to train the original risk assessment model based on the working data samples in the target equipment sample set and the various fault types corresponding to the working data samples. Among them, each fault type of any working data sample corresponds to an actual risk level. The target equipment sample set can be obtained from the equipment operation and maintenance troubleshooting database, which stores a large amount of fault data and working data accumulated by the equipment during daily operation and maintenance, covering the operating status and fault conditions of the equipment under different time periods and different working conditions, and providing rich historical data support for model training. At the same time, it can also be obtained through actual equipment operation monitoring and fault record collection. In the actual monitoring process, various sensors are used to collect the key parameters of the equipment in real time, such as temperature, pressure, vibration frequency, etc. Once the equipment fails, the working data and fault type when the failure occurs are recorded in time to ensure that the sample data obtained is true, reliable and timely.

[0069] It should be noted that the situations that the target device may encounter in the application scenario should be considered and the working data samples should be collected so that the working data samples under different fault types are as many as possible to improve the robustness of the risk assessment model.

[0070] Exemplarily, any working data sample and each fault type corresponding to the working data sample are obtained from the target device sample set. For each fault type, the expected risk level corresponding to the fault type is determined based on the working data sample through the original risk assessment model. Among them, the original risk assessment model can be constructed using a machine learning algorithm, such as a neural network, a decision tree, etc. Based on the actual risk level corresponding to each fault type and the expected risk level corresponding to each fault type, the parameters in the original risk assessment model are adjusted to obtain a trained risk assessment model.

[0071] Based on the above embodiment, after determining the risk level corresponding to each target fault, the troubleshooting task execution module 13 can also allocate corresponding resources and time windows for the maintenance of each target fault through the equipment troubleshooting model according to the risk level corresponding to each target fault. Among them, resources include human, material and financial resources, such as maintenance personnel, maintenance tools, parts and components. The time window refers to the reasonable time range for completing the maintenance task. Considering the use requirements of the equipment and the urgency of maintenance, the time windows corresponding to faults of different risk levels are also different. Specifically, the higher the risk level of the target fault, the greater its potential impact and harm, so more resources need to be allocated, and it is ensured that the maintenance task can be completed within a more timely time window; on the contrary, the target fault with a lower risk level can allocate fewer resources and appropriately postpone the maintenance time window. For example, when a high-risk fault is detected in the core component of a key device, the system will immediately trigger the emergency maintenance process, mobilize the best maintenance team, the most advanced maintenance tools and spare parts, and ensure that the normal operation of the equipment is restored in the shortest time; and for some low-risk, non-critical component failures, the system may arrange them in the daily maintenance plan and repair them step by step.

[0072] After the troubleshooting task execution module 13 based on the above embodiment determines the target fault, the equipment troubleshooting big model can be used to determine the maintenance suggestions for each target fault based on the knowledge information in the pre-built equipment operation and maintenance troubleshooting database. Among them, the knowledge information includes the structural principles, maintenance manuals, common fault solutions, etc. of the equipment. The equipment troubleshooting big model can analyze and reason about these knowledge information, and give targeted maintenance suggestions based on the specific circumstances of the target fault. Among them, the maintenance suggestions not only cover the analysis of the cause of the fault and the guidance of the maintenance steps, but also include practical information such as the required tools and material lists to help maintenance personnel quickly understand and execute the correct repair measures.

[0073] (IV) Output module 14

[0074] The main function of the output module 14 is to output the corresponding resources and time windows for the maintenance allocation of each target fault, as well as the maintenance suggestions for each target fault. There are many ways to output, such as displaying in the form of text or charts on the display screen of the device, or sending the information to a designated terminal device such as a mobile phone or computer through a network interface, so that users can obtain and view it in time. It is presented to maintenance personnel through a human-computer interaction interface to help them quickly understand and execute the correct repair measures.

[0075] In a possible implementation, when outputting the resources and time windows corresponding to the maintenance allocation of each target fault and the maintenance suggestions of each target fault, the output module 14 can generate and output a detailed troubleshooting report based on the resources and time windows corresponding to the maintenance allocation of each target fault and the maintenance suggestions of each target fault according to a pre-configured troubleshooting report template. The troubleshooting report includes the basic information of the target equipment, the troubleshooting process, the diagnosis result of the target fault, the maintenance allocation information and the maintenance suggestions.

[0076] It should be noted that the troubleshooting report can be saved in PDF, Word and other formats to facilitate users to archive and share.

[0077] In one example, the device may also include an equipment operation and maintenance fault diagnosis memory storage module, which is used to record key information involved in the entire fault diagnosis process of the target device. Among them, the key information includes fault type, occurrence time, diagnosis result, maintenance measures, maintenance personnel, maintenance time, etc. This information not only provides a valuable reference basis for future fault diagnosis, but also facilitates traceability and auditing. The equipment operation and maintenance fault diagnosis memory storage module can update this key information as knowledge information into the equipment operation and maintenance troubleshooting database to support the task of continuous learning and optimization of the device.

[0078] By continuously recording and updating these key information, the knowledge information in the equipment operation and maintenance troubleshooting database will become increasingly rich and accurate, providing a more reliable basis for subsequent fault diagnosis and maintenance. At the same time, these data can also be used to further train and optimize the equipment troubleshooting model and risk assessment model to improve the performance and adaptability of the device.

[0079] The beneficial effects of this application are as follows:

[0080] 1. The input module 11 can quickly receive the user's troubleshooting request and quickly start the troubleshooting process. The troubleshooting task autonomous planning module 12 uses the pre-trained equipment troubleshooting model, combined with the work data and the troubleshooting path text prompt, to automatically infer the troubleshooting path and the corresponding diagnostic steps for each fault node. Compared with manual experience to explore the troubleshooting sequence, this automated planning method greatly saves time, avoids blind troubleshooting, and makes the troubleshooting work more organized and targeted, so that the fault point can be quickly located, significantly improving the overall troubleshooting efficiency.

[0081] 2. When reasoning about the working data, the pre-trained equipment troubleshooting model can comprehensively consider various factors and accurately determine the troubleshooting path and diagnostic steps. The troubleshooting task execution module 13 performs troubleshooting according to the planned path and steps, and can systematically find the target fault, reducing the misjudgment and omission of faults caused by human negligence or lack of experience, and improving the accuracy of troubleshooting.

[0082] 3. The troubleshooting task execution module 13 allocates corresponding resources and time windows for the maintenance of each target fault according to its risk level through the equipment troubleshooting model. This resource allocation method based on risk level can make maintenance resources more reasonably and efficiently used, and optimize the resource allocation of the enterprise.

[0083] 4. The troubleshooting task execution module 13 uses the equipment troubleshooting model to obtain knowledge information from the pre-built equipment operation and maintenance troubleshooting database, and then determines the maintenance suggestions for each target fault. The database contains a variety of knowledge such as the structural principles of the equipment, maintenance manuals, and common fault solutions, making the maintenance suggestions more comprehensive, professional, and scientific. Maintenance personnel can quickly formulate maintenance plans based on these suggestions to improve the quality and efficiency of maintenance work. At the same time, it also provides maintenance personnel with learning and reference materials to help improve their maintenance skills.

[0084] Embodiment 2:

[0085] Based on the same inventive concept, the present application also provides an autonomous planning intelligent troubleshooting method. Figure 2 A schematic diagram of a process of autonomous planning and intelligent troubleshooting provided in an embodiment of the present application, the process includes:

[0086] S201: receiving a troubleshooting request for a target device input by a user.

[0087] S202: Acquire the working data of the target device.

[0088] S203: Using a pre-trained equipment troubleshooting model, based on the working data and the troubleshooting path text prompt, the working data is inferred to determine the troubleshooting path of the target equipment and the diagnostic steps corresponding to each fault node in the troubleshooting path.

[0089] S204: According to the execution order between each fault node in the troubleshooting path and according to the diagnosis steps of the current fault node, troubleshoot the target device to determine the target fault currently existing in the target device.

[0090] S205: Obtain the risk level corresponding to each of the target faults.

[0091] S206: Allocate corresponding resources and time windows for repairing each of the target faults according to the risk level corresponding to each of the target faults through the equipment troubleshooting model.

[0092] S207: Determine maintenance suggestions for each of the target faults through the equipment troubleshooting model and based on knowledge information in a pre-built equipment operation and maintenance troubleshooting database; wherein the knowledge information includes: structural principles of equipment, maintenance manuals, and common fault solutions.

[0093] S208: Outputting resources and time windows corresponding to maintenance allocations of each of the target faults, and maintenance suggestions for each of the target faults.

[0094] Since the principle of solving the problem by the above method is similar to that of the autonomous planning intelligent troubleshooting device, the implementation of the above method can refer to the embodiment of the device, and the repeated parts will not be repeated.

[0095] Embodiment 3:

[0096] See also Figure 3 , Figure 3 is a schematic diagram of the structure of a computer device provided by an optional embodiment of the present application, such as Figure 3 As shown, the computer device includes: one or more processors 10, a memory 20, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. Various components are connected to each other using different buses for communication, and can be installed on a common mainboard or installed in other ways as needed. The processor can process the instructions executed in the computer device, including instructions stored in or on the memory to display the graphical information of the GUI on an external input / output device (such as, a display device coupled to the interface). In some optional embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Similarly, multiple computer devices can be connected, and each device provides some necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system). Figure 3 A processor 10 is taken as an example.

[0097] The processor 10 may be a central processing unit, a network processor or a combination thereof. The processor 10 may further include a hardware chip. The hardware chip may be a dedicated integrated circuit, a programmable logic device or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable gate array, a general purpose array logic or any combination thereof.

[0098] The memory 20 stores instructions executable by at least one processor 10, so that at least one processor 10 executes the method shown in the above embodiment.

[0099] The memory 20 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created by the use of a computer device based on the presentation of a small program landing page, etc. In addition, the memory 20 may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some optional embodiments, the memory 20 may optionally include a memory remotely arranged relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0100] The memory 20 may include a volatile memory, such as a random access memory; the memory may also include a non-volatile memory, such as a flash memory, a hard disk or a solid state drive; the memory 20 may also include a combination of the above types of memory.

[0101] The computer device also includes an input device 30 and an output device 40. The processor 10, the memory 20, the input device 30 and the output device 40 may be connected via a bus or other means. Figure 3 The example of connecting through bus is taken in the following.

[0102] The input device 30 can receive input digital or character information, and generate key signal input related to the user settings and function control of the computer device, such as a touch screen, a keypad, a mouse, a track pad, a touch pad, an indicator bar, one or more mouse buttons, a trackball, a joystick, etc. The output device 40 may include a display device, an auxiliary lighting device (e.g., an LED) and a tactile feedback device (e.g., a vibration motor), etc. The above-mentioned display device includes but is not limited to a liquid crystal display, a light emitting diode, a display and a plasma display. In some optional embodiments, the display device can be a touch screen.

[0103] Embodiment 4:

[0104] On the basis of the above embodiments, an embodiment of the present application further provides a computer-readable storage medium, in which a computer program executable by a processor is stored. When the program runs on the processor, the processor implements the following steps when executing:

[0105] Receiving a troubleshooting request for a target device input by a user;

[0106] Acquire working data of the target device;

[0107] Through the pre-trained equipment troubleshooting model, based on the working data and the troubleshooting path text prompt, the working data is inferred to determine the troubleshooting path of the target equipment and the diagnostic steps corresponding to each fault node in the troubleshooting path;

[0108] According to the execution order between each fault node in the troubleshooting path and according to the diagnosis steps of the current fault node, troubleshoot the target device to determine the target fault currently existing in the target device;

[0109] Obtaining the risk level corresponding to each of the target faults;

[0110] By using the equipment troubleshooting model, corresponding resources and time windows are allocated for the maintenance of each target fault according to the risk level corresponding to each target fault;

[0111] Through the equipment troubleshooting model, the maintenance suggestions for each target fault are determined according to the knowledge information in the pre-built equipment operation and maintenance troubleshooting database; wherein the knowledge information includes: the structural principle of the equipment, maintenance manual, and common fault solutions;

[0112] Output the resources and time window corresponding to the maintenance allocation of each of the target faults, as well as the maintenance suggestions for each of the target faults.

[0113] Since the principle of solving the problem by the above-mentioned computer-readable storage medium is similar to that of the autonomous planning intelligent troubleshooting method, the implementation of the above-mentioned computer-readable storage medium can refer to the embodiment of the method, and the repeated parts will not be repeated.

[0114] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.

Claims

1. An autonomous planning intelligent troubleshooting device, characterized in that: The device comprises: An input module, used for receiving a troubleshooting request for a target device input by a user; The troubleshooting task autonomous planning module is used to obtain the working data of the target device in response to the troubleshooting request; through the pre-trained equipment troubleshooting model, based on the working data and the troubleshooting path text prompt, the working data is inferred to determine the troubleshooting path of the target device and the diagnostic steps corresponding to each fault node in the troubleshooting path; A troubleshooting task execution module is used to troubleshoot the target device according to the execution order between each fault node in the troubleshooting path and the diagnostic steps of the current fault node to determine the target fault currently existing in the target device; obtain the risk level corresponding to each target fault; through the equipment troubleshooting big model, according to the risk level corresponding to each target fault, allocate corresponding resources and time windows for the maintenance of each target fault; through the equipment troubleshooting big model, according to the knowledge information in the pre-built equipment operation and maintenance troubleshooting database, determine the maintenance suggestions for each target fault; wherein the knowledge information includes: the structural principle of the equipment, the maintenance manual, and the common fault solutions; The output module is used to output the resources and time windows corresponding to the maintenance allocation of each of the target faults, as well as the maintenance suggestions for each of the target faults.

2. The device according to claim 1, characterized in that The troubleshooting task execution module is specifically used to determine the risk level of each target fault based on the working data and the target fault through a pre-trained risk assessment model.

3. The device according to claim 1, characterized in that The risk assessment model is trained in the following way: Obtain any working data sample in the target device sample set and its corresponding fault types; wherein each fault type of the working data sample corresponds to an actual risk level; For each of the fault types, determining the expected risk level corresponding to the fault type based on the working data sample by using the original risk assessment model; Based on the actual risk levels and expected risk levels respectively corresponding to the fault types, the original risk assessment model is trained to obtain a trained risk assessment model.

4. The device according to claim 1 or 2, characterized in that The working data includes one or more of the following: equipment operation data, historical fault data and environmental monitoring data.

5. The device according to claim 4, characterized in that The device also includes: an equipment operation and maintenance fault diagnosis API protocol calling module; The equipment operation and maintenance fault diagnosis API protocol calling module is used to call external tools and services through a standardized API interface to obtain the working data.

6. The device according to claim 1, characterized in that The device also includes: an equipment operation and maintenance fault diagnosis memory storage module; The equipment operation and maintenance fault diagnosis memory storage module is used to record the key information involved in the entire fault diagnosis process of the target equipment, and update the key information as knowledge information to the equipment operation and maintenance troubleshooting database; wherein the key information includes but is not limited to one or more of the following: fault type, occurrence time, diagnosis result, maintenance measures, maintenance personnel, and maintenance time.

7. The device according to claim 1, characterized in that The equipment troubleshooting big model is obtained by fine-tuning the general text understanding big model based on the equipment operation and maintenance industry knowledge samples; wherein the types of the equipment operation and maintenance industry knowledge samples include one or more of the following: equipment failure case reports, equipment technical documents, industry standards and specifications, expert experience, equipment operation logs, and maintenance records.

8. An autonomous planning intelligent troubleshooting method, characterized in that: The method comprises: Receiving a troubleshooting request for a target device input by a user; Acquire working data of the target device; Through the pre-trained equipment troubleshooting model, based on the working data and the troubleshooting path text prompt, the working data is inferred to determine the troubleshooting path of the target equipment and the diagnostic steps corresponding to each fault node in the troubleshooting path; According to the execution order between each fault node in the troubleshooting path and according to the diagnosis steps of the current fault node, troubleshoot the target device to determine the target fault currently existing in the target device; Obtaining the risk level corresponding to each of the target faults; By using the equipment troubleshooting model, corresponding resources and time windows are allocated for the maintenance of each target fault according to the risk level corresponding to each target fault; Through the equipment troubleshooting model, the maintenance suggestions for each target fault are determined according to the knowledge information in the pre-built equipment operation and maintenance troubleshooting database; wherein the knowledge information includes: the structural principle of the equipment, maintenance manual, and common fault solutions; Output the resources and time window corresponding to the maintenance allocation of each of the target faults, as well as the maintenance suggestions for each of the target faults.

9. A computer device, characterized in that: The computer device includes a processor, and the processor is used to implement the steps of the autonomous planning intelligent troubleshooting method as described in claim 8 above when executing the computer program stored in the memory.

10. A computer-readable storage medium, characterized in that: It stores a computer program executable by a computer device. When the program is run on the computer device, the computer device executes the steps of the autonomous planning intelligent troubleshooting method as described in claim 8 above.

Images