Fault analysis, power equipment fault analysis and fault analysis model training method

Abstract

The embodiments of this specification provide fault analysis, power equipment fault analysis, and fault analysis model training methods, wherein the fault analysis method includes: obtaining fault description information of the target device; constructing multiple tasks to be processed based on the fault description information, and determining the processing strategy of the tasks to be processed, wherein the multiple tasks to be processed include a fault analysis task and at least one data generation task, and the processing strategy includes a fault analysis strategy of the fault analysis task and a data generation strategy of at least one data generation task; processing at least one data generation task according to the data generation strategy to obtain fault analysis data; using the fault analysis model, processing the fault analysis data and the fault analysis strategy to obtain the fault analysis result of the target device. Constructing multiple tasks to be processed based on the fault description information improves the flexibility of the fault analysis process, and using the fault analysis model to perform automated fault analysis can generate fault analysis results efficiently and accurately.

Description

Technical Field

[0001] The embodiments of this specification relate to the field of artificial intelligence technology, and in particular to fault analysis, power equipment fault analysis, and fault analysis model training methods. Background Art

[0002] With the rapid development of the power industry and the increasing requirements for energy management efficiency, it is particularly important to ensure the stability and accuracy of the power system. In the power system, power metering equipment plays a vital role. They are not only used to record the user's power consumption to support electricity bill settlement, but also one of the key tools for monitoring the operation status of the power grid.

[0003] However, in actual applications, various factors such as equipment aging, natural disasters, and human error may cause power metering equipment to fail. At present, power metering equipment is usually manually analyzed periodically. However, the manual analysis process requires a lot of labor costs, resulting in low efficiency of fault analysis. Summary of the invention

[0004] In view of this, an embodiment of the present specification provides a fault analysis method. One or more embodiments of the present specification also relate to a method for analyzing a fault of an electric power device, a method for training a fault analysis model, an information processing method based on a fault analysis model, a task platform, a fault analysis device, a device for analyzing a fault of an electric power device, a device for training a fault analysis model, an information processing device based on a fault analysis model, a computing device, an electronic device, a computer-readable storage medium, and a computer program product to solve the technical defects existing in the prior art.

[0005] According to the first aspect of the embodiments of this specification, a fault analysis method is provided, including: obtaining fault description information of a target device; based on the fault description information, constructing multiple tasks to be processed, and determining processing strategies for the tasks to be processed, wherein the multiple tasks to be processed include a fault analysis task and at least one data generation task, and the processing strategy includes a fault analysis strategy for the fault analysis task and a data generation strategy for at least one data generation task; processing at least one data generation task according to the data generation strategy to obtain fault analysis data; and using a fault analysis model to process the fault analysis data and the fault analysis strategy to obtain a fault analysis result of the target device.

[0006] The fault analysis method provided in an embodiment of the present specification realizes the disassembly of a fault description information processing task into multiple tasks to be processed by constructing multiple tasks to be processed based on fault description information, so that various complex fault analysis situations can be flexibly responded to, and the flexibility and efficiency of the fault analysis process are improved. By processing each data generation task, the fault analysis data required for the fault analysis process is fully generated, the comprehensiveness and accuracy of the fault analysis are ensured, and the fault analysis data and fault analysis strategy are automatically processed using the fault analysis model, so that the fault analysis results can be generated efficiently and accurately. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] Figure 1 is a flow chart of a fault analysis method provided by an embodiment of this specification; Figure 2 is an architecture diagram of a fault analysis system provided by an embodiment of this specification; Figure 3 It is a flow chart of a method for analyzing a fault of an electric power device provided by an embodiment of this specification; Figure 4 It is a process flow chart of a method for analyzing voltage loss of electric equipment provided by an embodiment of this specification; Figure 5 is a hierarchical structure diagram of a pressure loss determination condition provided by an embodiment of this specification; Figure 6 is a flow chart of a fault analysis model training method provided by an embodiment of this specification; Figure 7 is a flow chart of an information processing method based on a fault analysis model provided by an embodiment of this specification; Figure 8 It is a structural diagram of a task platform provided by an embodiment of this specification; Fig. 9 It is a structural schematic diagram of a fault analysis device provided by an embodiment of this specification; Fig.10 It is a structural schematic diagram of a power equipment fault analysis device provided by an embodiment of this specification; Fig.11 It is a structural schematic diagram of a fault analysis model training device provided by an embodiment of this specification; Fig.12 It is a structural schematic diagram of an information processing device based on a fault analysis model provided by an embodiment of this specification; Fig.13 is a structural block diagram of a computing device provided by one embodiment of this specification; Fig.14It is a structural block diagram of an electronic device provided by an embodiment of this specification. DETAILED DESCRIPTION

[0008] Many specific details are described in the following description to facilitate a full understanding of this specification. However, this specification can be implemented in many other ways than those described herein, and those skilled in the art can make similar generalizations without violating the connotation of this specification, so this specification is not limited to the specific implementation disclosed below.

[0009] The terms used in one or more embodiments of this specification are only for the purpose of describing specific embodiments, and are not intended to limit one or more embodiments of this specification. The singular forms of "a", "said" and "the" used in one or more embodiments of this specification and the appended claims are also intended to include plural forms, unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used in one or more embodiments of this specification refers to and includes any or all possible combinations of one or more associated listed items.

[0010] It should be understood that although the terms first, second, etc. may be used to describe various information in one or more embodiments of this specification, this information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of one or more embodiments of this specification, the first may also be referred to as the second, and similarly, the second may also be referred to as the first. Depending on the context, the word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".

[0011] In addition, it should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in one or more embodiments of this specification are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of the relevant countries and regions, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0012] In one or more embodiments of this specification, a large model refers to a deep learning model with large-scale model parameters, which usually contains hundreds of millions, tens of billions, hundreds of billions, trillions, or even more than 10 trillion model parameters. A large model can also be called a foundation model / foundation model. The large model is pre-trained with large-scale unlabeled corpus to produce a pre-trained model with more than 100 million parameters. This model can adapt to a wide range of downstream tasks, and the model has good generalization ability, such as a large-scale language model (LLM, Large Language Model), a multi-modal pre-training model, etc.

[0013] The development of large models has benefited from the significant improvement in computing power and progress in algorithm optimization in recent years. For example, the introduction of deep self-attention architecture has greatly promoted this trend. However, large models also face problems such as high training costs and high resource consumption. At the same time, the adaptability to small samples or specific fields still needs further research. Despite this, large models have become one of the key directions in the field of artificial intelligence to explore more efficient intelligent systems. In practical applications, large models only need a small number of samples to fine-tune the pre-trained model and can be applied to different tasks. Large models can be widely used in natural language processing (NLP), computer vision and other fields. Specifically, they can be applied to computer vision tasks such as visual question answering (VQA), image description (IC, Image Caption), image generation, as well as natural language processing tasks such as text-based sentiment classification, text summary generation, machine translation, etc. The main application scenarios of large models are digital assistants, intelligent robots, search, online education, office software, e-commerce, intelligent design, etc.

[0014] First, the terms involved in one or more embodiments of this specification are explained.

[0015] Deep Self-Attention (Transformer) Architecture: It is a network structure based on a multi-head self-attention mechanism module, mainly used to process sequence data. The Transformer architecture consists of a number of encoders and decoders that can be repeatedly stacked. This design allows the Transformer to efficiently learn long-term dependencies and is suitable for a variety of natural language processing tasks including machine translation, text summarization, and question-answering systems.

[0016] Electricity metering equipment: refers to equipment mainly used to measure and record the amount of electricity consumed by electricity users. These devices are the basis for electricity bill settlement and are also important tools for power companies to carry out load management and power market operations.

[0017] Metering equipment voltage loss diagnosis: refers to a fault detection technology for electric energy metering equipment in the power system, which aims to identify and locate the problem of inaccurate metering or malfunction caused by abnormal voltage drop (i.e. voltage loss). Metering equipment voltage loss can be referred to as metering voltage loss. In the power industry, according to the different line configurations, it can be divided into two common forms: three-phase three-wire system and three-phase four-wire system. Among them, "primary / secondary voltage loss" refers to the voltage loss on the primary side (high voltage side) or the secondary side (low voltage side); and phase A, phase B and phase C represent each phase of the three-phase alternating current. For example, in a three-phase four-wire low-voltage system, if the primary side of a phase loses voltage, it means that the phase fails to obtain sufficient voltage input from the power supply, which will directly affect all the electrical equipment connected to it and the accuracy of electric energy metering. Through effective voltage loss diagnosis methods, maintenance personnel can quickly locate the problem and take corresponding measures to restore the normal operation of the system.

[0018] Three-phase three-wire system: is a power transmission system that uses three live wires to transmit electrical energy without providing a neutral wire. This configuration is mainly used in industrial environments, especially when the load is mainly three-phase motors or other three-phase equipment, and it is assumed that these loads are balanced and symmetrical. Since there is no neutral wire, the three-phase three-wire system is not suitable for directly supplying single-phase loads.

[0019] Three-phase four-wire system: is a power transmission system that adds a neutral wire to the three-phase three-wire system, so the three-phase four-wire system has a total of four wires. The additional neutral wire makes it possible to obtain single-phase power from the three-phase system, thereby supporting single-phase loads as well as three-phase loads. The four-wire system can better handle asymmetric loads, maintain voltage stability, and is necessary for some electrical appliances that require a ground reference point. This configuration is common in commercial buildings and residential areas because it provides greater flexibility and compatibility.

[0020] Phase A, Phase B, and Phase C: are the identifiers of three independent but interrelated AC voltage or current waveforms in a three-phase power system. They are the basic elements that make up three-phase electricity. Each phase represents a separate sine wave. These three sine waves have the following characteristics: The frequency is the same. The frequency of all three phases (Phase A, Phase B, and Phase C) is the same. The amplitude is the same. Ideally, the maximum or peak value (i.e., amplitude) of each phase is the same. The phase difference is 120 degrees. There is a 120-degree electrical angle difference between each two adjacent phases. This means that when one phase reaches its peak, the other phase will be at a different position in its cycle, specifically 120 degrees after the peak of the previous phase. Phase A: Often referred to as the first phase, it is the first waveform in a three-phase system. It can be used as a reference point to describe the position of the other two phases relative to it. Phase B: The second phase, 120 degrees behind Phase A. Phase C: The third phase, 120 degrees behind Phase B, or in other words, it lags 240 degrees behind Phase A (or 120 degrees ahead).

[0021] Phase Angle: For alternating current, the phase angle is the time offset of one current or voltage waveform relative to another. It is usually measured in degrees or radians (rad).

[0022] Voltage phase angle: refers to the phase angle of the voltage waveform relative to a reference point or another voltage waveform. In the power system, the voltage phase angle not only affects the stability of power transmission, but is also closely related to factors such as power factor and reactive power.

[0023] Low-Rank Adaptation (LoRA): is a technique for efficiently fine-tuning large pre-trained models. Traditionally, when fine-tuning a large pre-trained model, all parameters are updated, which requires a lot of computing resources and time. LoRA proposes to update only some parameters, especially those that can be represented by low-rank decomposition. Specifically, for each fully connected layer or convolutional layer, LoRA adds a small-scale low-rank matrix as a learnable parameter, while keeping the original large matrix unchanged. This approach greatly reduces the number of parameters that need to be trained and reduces the cost of fine-tuning while maintaining good performance.

[0024] Supervised Fine-Tuning (SFT): is a further training method based on a pre-trained model. In this method, the model is trained on a dataset containing input and expected output pairs so that the model can learn how to generate answers closer to human level. Supervised fine-tuning is usually used to adapt the model to data in a specific task or domain, thereby improving the model's performance on these tasks.

[0025] Direct Preference Optimization (DPO): is a model optimization method based on user preferences, which is used to improve the quality and conformity of model generation results.

[0026] In this specification, a fault analysis method is provided. This specification also involves an electric power equipment fault analysis method, a fault analysis model training method, an information processing method based on a fault analysis model, a task platform, a fault analysis device, an electric power equipment fault analysis device, a fault analysis model training device, an information processing device based on a fault analysis model, a computing device, an electronic device, a computer-readable storage medium and a computer program product, which are described in detail one by one in the following embodiments.

[0027] See also Figure 1 , Figure 1 A flowchart of a fault analysis method provided by an embodiment of the present specification is shown, which specifically includes the following steps: Step 102: Obtain fault description information of the target device.

[0028] It should be noted that the target device refers to the machine, system or component that is of concern during the fault analysis process. The target device can be a physical device, such as industrial machinery, power equipment, vehicles, etc., or a software device, such as an online store, social media platform, word processing software, image editing tools, etc. During the operation of the target device, some faults may occur, causing the target device to fail to perform the expected functions. For example, the target device can be a power device that loses voltage due to equipment aging, natural disasters, human operation errors, etc., or a smart home device that loses control due to interruption of the communication link. Fault description information refers to information that describes the fault condition of the target device. Fault description information includes but is not limited to the device identification of the target device, the time, location, phenomenon, affected functions, etc. of the fault. Among them, the device identification is used to uniquely identify the target device, and the phenomenon refers to the actual manifestation of the fault, such as noise, vibration, abnormal voltage drop, display of error code, etc. For example, the fault description information can be "abnormal voltage drop occurs in power equipment 001".

[0029] In practical applications, there are many ways to obtain the fault description information of the target device, which can be selected according to the actual situation. The embodiments of this specification do not impose any restrictions on this. In the first possible implementation of this specification, the fault description information of the target device can be read from other data acquisition devices or databases. In the second possible implementation of this specification, the fault description information of the target device sent by the user through the client can be received. In the third possible implementation of this specification, the target device can be monitored, and the fault description information of the target device can be automatically obtained when the target device fails. For example, the power equipment is automatically monitored for voltage abnormalities to achieve intelligent abnormality warning.

[0030] Step 104: Based on the fault description information, construct multiple tasks to be processed and determine the processing strategies for the tasks to be processed, wherein the multiple tasks to be processed include a fault analysis task and at least one data generation task, and the processing strategies include a fault analysis strategy for the fault analysis task and a data generation strategy for at least one data generation task.

[0031] In traditional fault analysis schemes, fault description information is usually directly judged based on rules to directly obtain fault analysis results, or the fault description information is directly input into a large model, and the large model directly analyzes and outputs the fault analysis results. However, fault description information is usually more complex. The above two schemes directly generate fault analysis results through one-time processing, which cannot cope with complex fault analysis scenarios and will also miss key information in the fault analysis process, resulting in poor accuracy of fault analysis results. Therefore, in the embodiments of this specification, the complete fault analysis process is split into multiple tasks to be processed, and the processing strategies of each task to be processed are used to perform task processing in a targeted manner to avoid missing key information in the fault analysis process. At the same time, it can also flexibly cope with various complex fault analysis situations, thereby improving the flexibility and efficiency of the fault analysis process.

[0032] It should be noted that the pending tasks are a series of specific tasks that are performed to perform fault analysis on the target device based on the fault description information after the fault description information is obtained. The processing strategy refers to the specific processing method or steps taken for each pending task. The data generation task refers to the data collection or generation activity performed to support the fault analysis task. The data generation task may include tasks for collecting fault analysis data from sources such as device logs and sensor readings, and may also include tasks for generating fault analysis data through simulation, calculation, modeling, etc. The fault analysis task refers to the task of investigating, diagnosing and determining the cause of the fault of the target device based on the fault analysis data. The purpose of the fault analysis task is to locate the fault mode of the target device or to provide repair guidance and suggestions for the fault mode. The data generation strategy refers to the method specified for the data generation task, and the data generation strategy is used to indicate how to effectively collect, organize and generate fault analysis data for subsequent fault analysis tasks. The data generation strategy may include selecting a specific data source to read fault analysis data, using a specific data generation function, or calling a specific task processing engine to generate fault analysis data, etc. The fault analysis strategy refers to a specific plan or method used to guide the fault analysis task. The failure analysis strategy may include the tools used, the data analyzed, the expected types of results, specific analysis rules, etc.

[0033] Exemplarily, taking the description information of the fault description information as a communication fault as an example, the data generation tasks included in the tasks to be processed can be specifically divided into signal-to-noise ratio calculation tasks, networking success rate calculation tasks, and daily collection completeness rate calculation tasks; the fault analysis tasks are fault analysis tasks based on signal-to-noise ratio, networking success rate, and daily collection completeness rate. Taking the description information of the terminal meter fault as an example, the data generation tasks included in the tasks to be processed can be specifically divided into phase angle calculation tasks, clock anomaly calculation tasks, and time period power calculation tasks; the fault analysis tasks are fault analysis tasks based on phase angle, clock anomaly information, and time period power. Taking the description information of the fault description information as the depressurization of the power equipment as an example, the data generation tasks included in the tasks to be processed can be specifically phase angle generation tasks; the fault analysis tasks are fault analysis tasks based on phase angles.

[0034] Furthermore, taking the fault description information as the description information of the power equipment losing voltage, and the data generation task as the phase angle generation task as an example, the data generation strategy can be "extracting the current and voltage values ​​of phase A, phase B, and phase C from the data collected at the metering point. Once the current and voltage values ​​of each phase are obtained, the phase angle calculation function is used to calculate the phase angle to obtain the phase angle". Taking the fault analysis task as a fault analysis task based on the phase angle as an example, the fault analysis strategy is "determining the fault analysis result based on the extracted current and voltage values ​​and the calculated phase angle".

[0035] In practical applications, there are many ways to construct multiple tasks to be processed based on fault description information and determine the processing strategies for the tasks to be processed. The specific selection is based on the actual situation, and the embodiments of this specification do not impose any restrictions on this. In one possible implementation of this specification, a task division rule corresponding to the fault description information can be constructed based on prior knowledge, and the complete process of fault analysis based on the fault description information can be divided using the task division rule to obtain multiple tasks to be processed, and the processing strategies corresponding to the tasks to be processed are read from a pre-configured strategy library, wherein the strategy library includes multiple candidate processing tasks and the processing strategies corresponding to each candidate processing task. In another possible implementation of this specification, a task planning model can be used to construct multiple tasks to be processed based on the fault description information and determine the processing strategies for the tasks to be processed.

[0036] In an optional embodiment of the present specification, the above-mentioned constructing multiple tasks to be processed based on the fault description information and determining the processing strategy of the tasks to be processed may include the following steps: Acquire task generation prompt information, wherein the task generation prompt information is used to guide the task planning model to determine a plurality of tasks to be processed and processing strategies based on the fault description information; The task generation prompt information and fault description information are input into the task planning model to obtain multiple tasks to be processed and processing strategies.

[0037] It should be noted that the task planning model refers to a model that can receive fault description information and task generation prompt information as input, and output a series of specific tasks to be processed and a processing strategy for each task to be processed. The task planning model can be a large model with task planning capabilities, or it can be a deep learning model trained based on sample fault description information, multiple sample processing tasks of the sample fault description information, and the processing strategy corresponding to each sample processing task. The task planning model can perceive and model the environment of fault analysis, understand the fault analysis intention through the fault description information, and perform more fine-grained goal setting and decomposition on the overall goal of the fault analysis, thereby subdividing the complete fault analysis process into a series of tasks to be processed. Task generation prompt information refers to a set of guiding instructions for the task planning model, which is selected according to the actual situation, and the embodiments of this specification do not make any restrictions on this. The task generation prompt information can guide the task planning model to determine the multiple tasks to be processed and the corresponding processing strategies for each task to be processed based on the fault description information.

[0038] In actual applications, multiple pending tasks can be completed using different task processing engines, and the task planning model can collect data fed back by the task processing engine or implement the overall execution of the fault analysis according to the arranged fault analysis process. For example, the task generation prompt information is as follows: You are a fault diagnosis assistant. Please answer the following questions as much as possible. Before drawing conclusions, you should refer to some specific data. Please do not create data yourself. You need to split the input fault description information into multiple tasks to be executed, and think about querying the input fault description information step by step. You can selectively use some of the following task processing engines, or call calculation tools to calculate the results through mathematical analysis, or rewrite and split the text.

[0039] {{Engine description information of the task processing engine}} For each task to be processed, you must provide a processing strategy consisting of three parts, namely "Thought", "Action" and "Observation", in the following format: Question: Enter the fault description information Thought: You should always think about what to do Action: The action to be taken. It can be one of the above task processing engines or mathematical analysis calculation. If you use the above task processing engines, please directly enter the task processing engine name engine Action Input: Operation input Observation: The result of the operation ... (this thought / action / action-input / observation can be repeated zero or more times) Thought: Now I know the end result Final Answer: Final output start! Question:{{Fault description}} By applying the solution of the embodiment of this specification, the task planning model is guided to determine multiple pending tasks and processing strategies for each pending task through task generation prompt information, which provides a clear direction for the task planning model and improves the efficiency of determining pending tasks and processing strategies.

[0040] Step 106: Process at least one data generation task according to the data generation strategy to obtain fault analysis data.

[0041] It should be noted that fault analysis data is a data set obtained by executing data generation tasks. Fault analysis data refers to the data required for fault analysis of the target device. Taking the fault description information of power equipment voltage loss as an example, the fault analysis data includes the voltage, active power, reactive power, phase angle, voltage phase angle, etc. of phase A, phase B, and phase C.

[0042] In practical applications, at least one data generation task is processed according to the data generation strategy, and there are multiple ways to obtain fault analysis data, which are selected according to the actual situation. The embodiments of this specification do not impose any restrictions on this. In one possible implementation of this specification, when the data generation strategy is to select a specific data source to read the fault analysis data, when processing the data generation task, the fault analysis data can be read directly from the specific data source. In another possible implementation of this specification, when the data generation strategy is to use a specific data generation function or call a specific task processing engine to generate fault analysis data, the data generation function or task processing engine can be called to process at least one data generation task to obtain fault analysis data.

[0043] In an optional embodiment of the present specification, the data generation strategy is used to indicate a task processing engine and a reference data type of a data generation task; the above-mentioned processing of at least one data generation task according to the data generation strategy to obtain fault analysis data may include the following steps: Based on the reference data type, obtain task reference data; The task processing engine is called to process at least one data generation task according to the task reference data to obtain fault analysis data.

[0044] It should be noted that the task processing engine refers to a specific tool or algorithm for performing data generation tasks. Data generation tasks can be divided based on the capabilities of the task processing engine. For example, if the task processing engine includes a current and voltage calculation engine and a phase angle calculation engine, the data generation task may include a current and voltage generation task and a phase angle generation task. If the task processing engine only includes a phase angle calculation engine, the data generation task may include a phase angle generation task. The current and voltage can be obtained by reading. The task processing engine can be an engine designed specifically for a specific data generation task, such as a phase angle calculation engine, or it can be a more general data processing platform. The task processing engine can generate the required fault analysis data based on the given task reference data and the task parameters of the data generation task. The reference data type refers to the data type of the data required by the task processing engine when performing the data generation task. The reference data type can ensure that the task processing engine can accurately perform the data generation task. Task reference data refers to the specific data obtained according to the specified reference data type. Task reference data is the basis for the task processing engine to generate fault analysis data.

[0045] In actual applications, there are multiple ways to obtain task reference data based on the reference data type, which can be selected based on actual conditions, and the embodiments of this specification do not impose any restrictions on this. In one possible implementation of this specification, the task reference data corresponding to the reference data type can be read from other databases or data acquisition devices. In another possible implementation of this specification, the reference data type can be sent to the client, and the task reference data sent by the user through the client is received.

[0046] By applying the solution of the embodiments of this specification, by clearly specifying the reference data type, it is ensured that the acquired task reference data is directly related to the specific fault condition, reducing the impact of irrelevant or redundant data, so that the task processing engine can accurately generate fault analysis data, and by calling the task processing engine to process specific task reference data, high-quality fault analysis data can be quickly acquired.

[0047] In an optional embodiment of the present specification, the calling of the task processing engine to process at least one data generation task according to the task reference data and before obtaining the fault analysis data may further include the following steps: A task processing engine is selected from the plurality of candidate processing engines according to engine description information and data generation strategies of the plurality of candidate processing engines.

[0048] It should be noted that the engine description information of the candidate processing engine is used to describe the key attributes and functions of the candidate processing engine. Since the data generation strategy can describe the way of reading or generating fault analysis data, the task processing engine can be screened out from multiple candidate processing engines according to the engine description information and data generation strategy of multiple candidate processing engines.

[0049] In actual applications, there are many ways to filter out a task processing engine from multiple candidate processing engines based on the engine description information and data generation strategies of multiple candidate processing engines. The specific selection is based on the actual situation, and the embodiments of this specification do not impose any restrictions on this. In one possible implementation of this specification, the engine description information and data generation strategies of multiple candidate processing engines can be input into a task planning model to obtain the target engine description information, and the candidate processing engine corresponding to the target engine description information is determined as the task processing engine. In another possible implementation of this specification, the similarity between the data generation strategy and each engine description information can be calculated, and the task processing engine can be filtered out from multiple candidate processing engines based on the similarity.

[0050] It is worth noting that since the data generation strategy can also describe the data source of the fault analysis data, it is also possible to screen out the target data source from multiple candidate processing engines based on the data source description information and data generation strategy of multiple candidate data sources, and read the fault analysis data from the target data source.

[0051] By applying the solution of the embodiment of this specification, a task processing engine is determined from multiple candidate processing engines according to the data generation strategy and the engine description information of each candidate processing engine, so that the task processing engine can be determined efficiently and accurately.

[0052] In an optional embodiment of the present specification, taking the target device including power equipment, the data generation task including a phase angle generation task, the task processing engine including a phase angle calculation engine, and the task reference data including a reference electrical parameter as an example, the above-mentioned calling of the task processing engine, processing at least one data generation task according to the task reference data, and obtaining the fault analysis data may include the following steps: The phase angle calculation engine is called to process the phase angle generation task according to the reference electrical parameters to obtain the phase angle of the power equipment.

[0053] It should be noted that power equipment refers to various equipment used for power generation, transmission, distribution and consumption. Power equipment plays a key role in all aspects of the power system, ensuring the efficient transmission and use of electric energy from production to end users. Power equipment includes but is not limited to electric energy metering equipment, power generation equipment, and power distribution equipment, which are selected according to actual conditions, and the embodiments of this specification do not impose any restrictions on this. Faults that occur in power equipment include but are not limited to loss of voltage, overvoltage, and short circuit. The phase angle generation task refers to a task that focuses on calculating and generating the phase angles between various components (such as voltage and current) in the power system. The phase angle calculation engine is used to calculate the accurate phase angle based on the input reference electrical parameters. The phase angle calculation engine can be constructed based on inverse trigonometric functions. Reference electrical parameters refer to standard or known electrical parameters used when calculating the phase angle. They can come from values ​​measured by the measurement platform, data provided by the manufacturer, or records from previous similar situations. The reference electrical parameters ensure the accuracy of the phase angle calculation.

[0054] In practical applications, taking the fault description information of power equipment depressurization as an example, the reference electrical parameters may include the current and voltage values ​​of phase A, phase B, and phase C. These current and voltage values ​​can be extracted and calculated from the data collected by sensors or other monitoring equipment in the power system, such as those calculated based on the basic principles of electrical engineering (Ohm's law). Since the phase relationship between different phases directly affects the stability of the power system and the energy transmission efficiency, once the current and voltage values ​​of each phase are obtained, the phase angle calculation engine can be called to process the phase angle generation task according to the current and voltage values ​​to obtain the phase angle between different phases of the power equipment.

[0055] By applying the solution of the embodiment of this specification, the phase angle calculation engine is called, and the phase angle generation task is processed according to the reference electrical parameters to obtain the phase angle of the power equipment, thereby improving the generation efficiency of the phase angle.

[0056] Step 108: Using the fault analysis model, the fault analysis data and the fault analysis strategy are processed to obtain the fault analysis result of the target device.

[0057] It should be noted that a fault analysis model refers to a model that can receive fault analysis data and fault analysis strategies as input and output fault analysis results. The fault analysis model can be a large model with rule reasoning, fault understanding and interpretation capabilities, or it can be a deep learning model trained based on sample fault analysis data, sample fault analysis strategies and sample fault analysis results. The fault analysis engine can also be understood as a task processing engine corresponding to the fault analysis task. The fault analysis result is the final output obtained through the fault analysis model. The fault analysis result can provide detailed insights into the nature, cause and possible solutions of the fault, thereby helping technicians quickly locate the problem and take effective corrective measures.

[0058] In practical applications, there are many ways to obtain the fault analysis results of the target device by processing the fault analysis data and the fault analysis strategy using the fault analysis model. The specific selection is based on the actual situation, and the embodiments of this specification do not impose any restrictions on this. In one possible implementation of this specification, the fault analysis data and the fault analysis strategy can be directly input into the fault analysis model to obtain the fault analysis results of the target device. In another possible implementation of this specification, since the fault analysis strategy may only include the tools used by the fault analysis model, that is, the fault analysis strategy only indicates that the fault analysis task is performed by the fault analysis model, the actual processing process of the fault analysis model only refers to the fault analysis data, resulting in poor accuracy of the fault analysis results. Therefore, according to the fault analysis strategy, fault analysis prompt information can be constructed, and the fault analysis prompt information and the fault analysis data can be input into the fault analysis model to obtain the fault analysis results of the target device.

[0059] By applying the scheme of the embodiments of this specification, a plurality of tasks to be processed are constructed based on the fault description information, so that a fault description information processing task is decomposed into a plurality of tasks to be processed, so that various complex fault analysis situations can be flexibly responded to, and the flexibility and efficiency of the fault analysis process are improved. By processing each data generation task, the fault analysis data required for the fault analysis process is fully generated, the comprehensiveness and accuracy of the fault analysis are ensured, and the fault analysis data and fault analysis strategy are automatically processed using the fault analysis model, so that the fault analysis results can be generated efficiently and accurately.

[0060] In an optional embodiment of the present specification, the above-mentioned processing of the fault analysis data and the fault analysis strategy using the fault analysis model to obtain the fault analysis result of the target device may include the following steps: According to the fault analysis strategy, construct fault analysis prompt information, wherein the fault analysis prompt information is used to guide the fault analysis model to analyze the fault analysis data to obtain the fault analysis result; The fault analysis prompt information and the fault analysis data are input into the fault analysis model to obtain the fault analysis result of the target device.

[0061] It should be noted that the fault analysis prompt information is a set of model guidance instructions generated based on the fault analysis strategy, which is intended to guide the fault analysis model to effectively process the fault analysis data and obtain accurate fault analysis results. The fault analysis prompt information includes but is not limited to fault analysis samples (including analysis data samples and analysis result samples corresponding to the analysis data samples), fault determination conditions, fault analysis focus, fault analysis priority, etc., which are selected according to actual conditions, and the embodiments of this specification do not impose any restrictions on this.

[0062] In practical applications, there are many ways to construct fault analysis prompt information according to the fault analysis strategy, which are selected according to the actual situation, and the embodiments of this specification do not limit this. In one possible implementation of this specification, a fault analysis sample corresponding to the fault analysis strategy can be obtained, and fault analysis prompt information can be constructed according to the fault analysis sample. Through the fault analysis sample, the fault analysis model can be helped to learn a wider range of fault modes and characteristics, so that the fault analysis model can perform fault analysis more accurately and provide more accurate fault analysis results. In another possible implementation of this specification, the fault determination condition corresponding to the fault analysis strategy can be obtained, and the fault analysis prompt information can be constructed according to the fault determination condition. Among them, the fault determination condition refers to a set of standards, parameters or rules used to clearly determine whether the target device has indeed failed. These conditions are usually based on specific thresholds, pattern recognition, logical judgment, etc., and are intended to distinguish between normal operating states and fault states. When the target device meets the fault determination condition, it can be determined that the target device is in a fault state or fault mode corresponding to the fault determination condition.

[0063] By applying the solution of the embodiment of this specification, the fault analysis model is guided to analyze the fault analysis data through the fault analysis prompt information, thereby improving the accuracy of the fault analysis result.

[0064] In an optional embodiment of the present specification, the above-mentioned construction of fault analysis prompt information according to the fault analysis strategy may include the following steps: Obtain the fault determination condition corresponding to the fault analysis strategy, and construct fault analysis prompt information according to the fault determination condition.

[0065] It should be noted that the fault determination conditions corresponding to different types of target devices or different modules in the same target device may be different. Therefore, the fault determination conditions corresponding to the fault analysis strategy can be obtained. Here, the fault determination conditions corresponding to the fault analysis strategy are also the fault determination conditions corresponding to the fault description information, and then the fault analysis prompt information is constructed according to the fault determination conditions. For example, when the fault analysis strategy is a programmable logic controller fault analysis strategy, the acquired fault determination conditions include input unit fault determination conditions and output unit fault determination conditions.

[0066] In practical applications, there are many ways to obtain the fault determination conditions corresponding to the fault analysis strategy, which are selected according to the actual situation, and the embodiments of this specification do not limit this. In one possible implementation of this specification, the fault determination conditions corresponding to the fault analysis strategy sent by the user through the client can be received. In another possible implementation of this specification, the fault determination conditions corresponding to the fault analysis strategy can be read from other data acquisition devices or databases.

[0067] By applying the solution of the embodiments of this specification, the accuracy of the fault analysis prompt information is improved by obtaining the fault determination condition corresponding to the fault analysis strategy and constructing the fault analysis prompt information according to the fault determination condition.

[0068] In an optional embodiment of the present specification, taking the fault determination condition including the primary side fault determination condition and the secondary side fault determination condition as an example, the above-mentioned acquisition of the fault determination condition corresponding to the fault analysis strategy and the construction of the fault analysis prompt information according to the fault determination condition may include the following steps: When the fault analysis strategy is a power equipment voltage loss analysis strategy, the primary side fault determination condition and the secondary side fault determination condition are obtained, and fault analysis prompt information is constructed according to the primary side fault determination condition and the secondary side fault determination condition.

[0069] It should be noted that voltage loss refers to the situation where the voltage of one or several phases in the power system is significantly lower than the normal working range. According to the different locations where voltage loss occurs, voltage loss can be divided into primary side voltage loss and secondary side voltage loss. For users, voltage loss may cause household appliances to fail to start normally or reduce working efficiency, seriously affecting the quality of daily life; for power supply companies, voltage loss will not only affect the accuracy of energy metering and cause economic losses, but also may be difficult to accurately assess the operation of the power grid due to large data deviations, thus burying safety hazards; in extreme cases, long-term severe voltage loss may even cause power outages in the entire area, causing huge inconvenience and losses to society. The power equipment voltage loss analysis strategy refers to an analysis method for the voltage loss problem in power equipment. If the fault analysis strategy is the power equipment voltage loss analysis strategy, it means that the fault analysis task is the power equipment voltage loss analysis task. The primary side usually refers to the high-voltage side in the power system, that is, the high-voltage end input to the transformer or other power conversion equipment. The primary side is responsible for receiving electrical energy from the power station or the upper power grid and transmitting it to the next level of equipment for processing or distribution. The secondary side refers to the low-voltage side of the power system, that is, the low-voltage end output from the transformer or other power conversion equipment. The secondary side is responsible for providing power to the end user or the next-level distribution network to ensure safe and reliable power supply. The primary side fault determination condition refers to the specific standard or parameter used to determine whether there is a fault on the primary side. The secondary side fault determination condition refers to the specific standard or parameter used to determine whether there is a fault on the secondary side.

[0070] In actual application, there are many ways to obtain the primary side fault determination condition and the secondary side fault determination condition, which are selected according to the actual situation, and the embodiments of this specification do not limit this. In one possible implementation of this specification, the primary side fault determination condition and the secondary side fault determination condition sent by the user through the client can be received. In another possible implementation of this specification, the primary side fault determination condition and the secondary side fault determination condition can be inferred based on the basic principles of electrical engineering.

[0071] By applying the solution of the embodiments of this specification, additional fault determination conditions are added to the input of the fault analysis model through prompt information engineering, so that the fault analysis model can independently perform fault analysis and realize the fault type judgment of the power equipment. In addition, the fault analysis model can also distinguish between abnormal conditions occurring on the primary side and the secondary side, thereby providing maintenance personnel with more accurate fault analysis results.

[0072] In an optional embodiment of the present specification, since the faults of the target device may be various, it may take a long time for the fault analysis model to perform fault analysis based only on the fault analysis prompt information and the fault analysis data. Therefore, the device association information of the target device may be additionally added to the input of the fault analysis model, so that the fault analysis model performs fault analysis more specifically. That is, before the fault analysis prompt information and the fault analysis data are input into the fault analysis model and the fault analysis result of the target device is obtained, the following steps may be further included: Get device association information of the target device; Inputting the fault analysis prompt information and the fault analysis data into the fault analysis model to obtain the fault analysis result of the target device may include the following steps: The fault analysis prompt information, device association information and fault analysis data are input into the fault analysis model to obtain the fault analysis result of the target device.

[0073] It should be noted that device-related information refers to information related to the target device, such as the region where the target device is located, the system structure, and so on. Taking the target device as an electric power device as an example, the device-related information of the target device may be the electric power system structure information, such as a three-phase three-wire system or a three-phase four-wire system. If the device-related information "three-phase three-wire system" is input into the fault analysis model, the fault analysis model can locate the real fault of the target device only among the possible faults of the three-phase three-wire system, without considering the possible faults of the three-phase four-wire system. Alternatively, the fault analysis model can also perform fault analysis only using the fault determination conditions corresponding to the three-phase three-wire system, without considering the fault determination conditions corresponding to the three-phase four-wire system.

[0074] By applying the solution of the embodiments of this specification, fault analysis prompt information, device association information and fault analysis data are input into the fault analysis model to obtain the fault analysis result of the target device, thereby reducing the fault consideration scope of the fault analysis model and improving the fault analysis efficiency.

[0075] In an optional embodiment of the present specification, after the above-mentioned use of the fault analysis model to process the fault analysis data and the fault analysis strategy to obtain the fault analysis result of the target device, the following steps may also be included: Generate a fault diagnosis report for the target device based on the fault description information, the processing strategies of multiple pending tasks, the fault analysis data, and the fault analysis results.

[0076] It should be noted that a fault diagnosis report is a detailed document that usually contains comprehensive content from fault description information to the final fault analysis results, such as all data involved in the fault analysis and diagnosis process, the determined fault mode and fault solution. The fault diagnosis report can provide clear guidance and support for technicians, maintenance personnel and management. The fault diagnosis report can be a plain text document or a diversified graphic report.

[0077] In practical applications, there are many ways to generate a fault diagnosis report for a target device based on fault description information, processing strategies for multiple tasks to be processed, fault analysis data, and fault analysis results. The specific selection is based on the actual situation, and the embodiments of this specification do not impose any restrictions on this. In one possible implementation of this specification, a pre-set report template can be obtained, and the fault description information, processing strategies for multiple tasks to be processed, fault analysis data, and fault analysis results can be filled into the report template to obtain a fault diagnosis report for the target device. In another possible implementation of this specification, the fault description information, processing strategies for multiple tasks to be processed, fault analysis data, and fault analysis results can be input into a report generation model, and the fault description information, processing strategies for multiple tasks to be processed, fault analysis data, and fault analysis results can be summarized using the powerful natural language processing capabilities of the report generation model to generate a detailed and clearly organized fault diagnosis report. Optionally, report generation prompt information can also be input into the report generation model to guide the fault analysis report generation process of the report generation model, wherein the report generation prompt information is as follows: You are a summary assistant. Please read the reference materials below and answer the user's questions in a detailed, professional and focused manner based on your professional knowledge. Note: (1) User questions may require judgment, summary, and reasoning based on data. Please be careful to identify and then make corresponding reasoning answers. (2) The answer should be comprehensive and do not miss key points. At the same time, the answer should be relevant to the question and do not answer other irrelevant information. (3) The answer should pay attention to summary, point division, line breaks and other formats to ensure good readability. (4) If you cannot answer based on the reference materials, please give an explanation and reply "Sorry, there is not enough information". Do not try to make up content that is not in the reference materials.

[0078] [Original data] Current and voltage values [Phase angle calculation data] Phase angle calculation results [Fault diagnosis result] Three-phase three-wire metering voltage loss-A phase voltage loss-one wire break [Troubleshooting suggestions] Maintenance of related equipment It is worth noting that the report generation model can be a large model, and its core capabilities are reflected in the following aspects: Multi-source information fusion: First, the report generation model can receive and process various forms of input data, including but not limited to sensor data, current and voltage data, etc., and can efficiently parse these heterogeneous data to lay the foundation for subsequent steps. Accurate diagnosis generation: The report generation model can use its text generation skills to present complex fault analysis processes and results to users in an easy-to-understand way. This not only covers an accurate description of the nature of the fault problem, but also includes recommended solutions, preventive measures and other suggestions. At the same time, the output form of the report generation model is flexible and diverse, and can be customized to generate reports, dialogue responses, etc. according to the needs of the application scenario.

[0079] By applying the solution of the embodiments of this specification, a detailed and clearly structured fault diagnosis report is generated, which improves the transparency of the fault analysis process and further improves user satisfaction.

[0080] In an optional embodiment of the present specification, the fault analysis model can be iteratively optimized as more fault analysis tasks are processed, thereby improving fault detection accuracy and response speed. That is, after the fault analysis data and the fault analysis strategy are processed using the fault analysis model to obtain the fault analysis result of the target device, the following steps may also be included: Receiving result feedback information sent by the client, wherein the result feedback information is information provided by the client on the fault analysis result based on the analysis requirements; Based on the feedback information from the results, build a model to optimize the data; The model optimization data is used to adjust the parameters of the fault analysis model to obtain the adjusted fault analysis model.

[0081] It should be noted that object feedback information refers to the client's response and evaluation of the fault analysis results, such as ratings, comments, likes and other interactive information. The quality and effect of the fault analysis results can be determined through the result feedback information. The adjusted fault analysis model refers to a fault analysis model with better performance after parameter adjustment. The adjusted fault analysis model can better perform the fault analysis task, thereby providing more accurate fault analysis results.

[0082] In practical applications, there are many ways to construct model optimization data based on the result feedback information, and the selection is made according to the actual situation. The embodiments of this specification do not impose any restrictions on this. In one possible implementation of this specification, an updated fault analysis result can be generated based on the result feedback information, and the model optimization data is determined based on the updated fault analysis result, the fault analysis data and the fault analysis strategy. In another possible implementation of this specification, the model optimization data can be read from other databases or data acquisition devices based on the result feedback information. Furthermore, the method of adjusting the parameters of the fault analysis model using the model optimization data includes but is not limited to SFT, DPO, LoRA and other methods, which are selected according to the actual situation. The embodiments of this specification do not impose any restrictions on this.

[0083] It is worth noting that in the embodiments of this specification, not only can the parameters of the fault analysis model be adjusted, but also model optimization data for optimizing other models (such as task planning models, report generation models, etc.) involved in the embodiments of this specification can be constructed based on the result feedback information, and the model parameters of the corresponding models can be adjusted using these model optimization data to obtain the adjusted model. For example, based on the result feedback information, model optimization data for optimizing the task planning model is constructed, and the model optimization data of the task planning model is used to adjust the parameters of the task planning model to obtain the adjusted fault analysis model. The parameter adjustment methods of other models involved in the embodiments of this specification include but are not limited to SFT, DPO, LoRA and other methods.

[0084] By applying the solution of the embodiments of this specification, the fault analysis model is continuously learned and optimized according to the result feedback information sent by the client, so that the fault analysis model can generate more accurate fault analysis results. At the same time, the fault analysis model can quickly adapt to changes when encountering new types of faults, showing good generalization performance.

[0085] Considering that the model parameters of the fault analysis model are relatively large and the computing resources of the client are limited, the fault analysis method proposed in the embodiment of this specification can be applied to the following: Figure 2 The fault analysis system shown is, but not limited to, Figure 2 , Figure 2 An architecture diagram of a fault analysis system provided by an embodiment of the present specification is shown, and the fault analysis system may include a client 202 and a server 204; The client 202 is used to send the fault description information of the target device to the server 204; The server 204 is used to construct multiple tasks to be processed based on the fault description information, and determine the processing strategies of the tasks to be processed, wherein the multiple tasks to be processed include a fault analysis task and at least one data generation task, and the processing strategies include a fault analysis strategy of the fault analysis task and a data generation strategy of at least one data generation task; process at least one data generation task according to the data generation strategy to obtain fault analysis data; use the fault analysis model to process the fault analysis data and the fault analysis strategy to obtain a fault analysis result of the target device; and send the fault analysis result of the target device to the client 202; The client 202 is also used to receive the fault analysis result of the target device sent by the server 204.

[0086] In practical applications, the fault analysis model can be deployed in the server 204, and the server 204 can be connected to one or more clients 202 through a local area network connection, a wide area network connection, an Internet connection, or other types of data networks. The data transmitted by the client 202 may need to be encoded, transcoded, compressed, etc. before being published to the server 204. The client 202 can be a browser, an application (APP, Application), or a web application such as Hypertext Markup Language Version 5 (H5, HyperText Markup Language5) application, or a light application (also known as a mini-program, a lightweight application) or a cloud application, etc. The client 202 can be based on the software development kit (SDK, Software Development Kit) of the corresponding service provided by the server 204, such as based on the real-time communication (RTC, Real Time Communication) SDK development and acquisition. The client 202 can be deployed in an electronic device, and needs to rely on the device to run or some APPs in the device to run, etc. The electronic device can, for example, have a display screen and support information browsing, etc., such as a personal mobile terminal such as a mobile phone, a tablet computer, a personal computer (PC, Personal Computer), etc. Various other types of applications can also be configured in electronic devices, such as human-computer dialogue applications, model training applications, text processing applications, web browser applications, shopping applications, search applications, instant messaging tools, email clients, social platform software, etc. The client 202 can also interact with the user through a graphical user interface to call the fault analysis model, thereby implementing the fault analysis method provided in the embodiments of this specification.

[0087] The server 204 may include servers that provide various services, such as servers that provide communication services to multiple clients, servers for background training that provide support for models used on the client, and servers that process data sent by the client. It should be noted that the server 204 can be implemented as a distributed server cluster consisting of multiple servers, or as a single server. The server can also be a server of a distributed system, or a server combined with a blockchain. The server can also be a cloud server for basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content distribution networks (CDN, Content Delivery Network), and big data and artificial intelligence platforms, or an intelligent cloud computing server or intelligent cloud host with artificial intelligence technology.

[0088] It is worth noting that the fault analysis method provided in the embodiments of this specification is generally executed by the server, but in other embodiments of this specification, when the operating resources of the client can meet the deployment and operating conditions of the fault analysis model, the client can also have similar functions as the server, thereby executing the fault analysis method provided in the embodiments of this specification. In other embodiments, the fault analysis method provided in the embodiments of this specification can also be jointly executed by the client and the server.

[0089] The following combination Figure 3 , taking the application of the fault analysis method provided in this specification in the power equipment fault analysis scenario as an example, the fault analysis method is further explained. Figure 3 A flowchart of a method for analyzing a fault of an electric power device provided by an embodiment of the present specification is shown, which specifically includes the following steps: Step 302: Obtain fault description information of the power equipment.

[0090] Step 304: Based on the fault description information, construct multiple tasks to be processed and determine the processing strategies for the tasks to be processed, wherein the multiple tasks to be processed include a fault analysis task and at least one data generation task, and the processing strategies include a fault analysis strategy for the fault analysis task and a data generation strategy for at least one data generation task.

[0091] Step 306: Process at least one data generation task according to the data generation strategy to obtain fault analysis data.

[0092] Step 308: Using the fault analysis model, process the fault analysis data and the fault analysis strategy to obtain the fault analysis result of the power equipment.

[0093] It should be noted that the implementation of steps 302 to 308 may refer to the implementation of steps 102 to 108 described above, and will not be described in detail in this embodiment of the specification.

[0094] By applying the scheme of the embodiment of this specification, by constructing multiple tasks to be processed based on fault description information, it is possible to decompose a fault description information processing task into multiple tasks to be processed, so that various complex power equipment fault analysis situations can be flexibly responded to, and the flexibility and efficiency of the fault analysis process are improved. By processing each data generation task, the fault analysis data required for the fault analysis process is fully generated, ensuring the comprehensiveness and accuracy of the power equipment fault analysis, and the fault analysis model is used to automatically process the fault analysis data and the fault analysis strategy, so that the fault analysis results of the power equipment can be generated efficiently and accurately.

[0095] The following combination Figure 4 Taking the application of the power equipment fault analysis method provided in this specification in the scenario of voltage loss diagnosis of metering equipment as an example, the power equipment fault analysis method is further explained. Figure 4The present specification shows a processing flow chart of a method for analyzing power equipment undervoltage provided by an embodiment of the present specification. The method for analyzing power equipment undervoltage is applied to a metering equipment undervoltage diagnosis intelligent agent system. The metering equipment undervoltage diagnosis intelligent agent system includes a task planning model, a phase angle calculation engine, a metering data platform, a fault analysis model, and a report generation model. Among them, the metering data platform can collect electrical parameters at regular intervals, perform real-time deduction of metering users, and generate real-time status portraits; the phase angle calculation engine can use the inverse trigonometric function calculation method to derive the precise phase angle from a given current and voltage waveform; the report generation model can summarize and output the diagnosis results. Specifically, receiving fault description information of the power equipment input by the user; inputting the task generation prompt information and the fault description information into the task planning model, obtaining multiple tasks to be processed and the processing strategies of each task to be processed, wherein the multiple tasks to be processed include a phase angle generation task and a fault analysis task, and the processing strategies include a fault analysis strategy of the fault analysis task and a phase angle generation strategy of the phase angle generation task, the phase angle generation strategy is used to indicate that the phase angle generation task processing is performed by the phase angle calculation engine, and the fault analysis strategy is used to indicate that the fault analysis task is performed by the fault analysis model; calling the phase angle calculation engine to perform the phase angle generation task based on the electrical parameters collected by the metering data platform, and obtaining the phase angles of different phases of the power equipment; constructing the fault analysis prompt information according to the primary side fault determination conditions and the secondary side fault determination conditions; inputting the fault analysis prompt information, the power system structure information (three-phase three-wire system or three-phase four-wire system) and the phase angle into the fault analysis model to perform the fault analysis task processing, and obtaining the fault analysis results of the target equipment; using the report generation model to summarize the above-mentioned metering equipment pressure loss diagnosis process, and generate a metering equipment pressure loss diagnosis report.

[0096] By applying the scheme of the embodiment of this specification, the user intention is understood through the task planning model, and the goal is set and decomposed according to the capabilities of the external engine, so that the system can flexibly respond to various complex fault situations and improve overall efficiency. Through the method of prompt word engineering, the fault analysis model is used to realize the judgment of the fault type of power equipment, so as to better adapt to new fault types and abnormal situations based on dynamic analysis and reasoning methods, and improve the flexibility and generalization ability of the system. By using the advanced natural language processing capabilities of the report generation model, data from different sources are integrated to generate easy-to-understand diagnostic reports, providing detailed fault descriptions and solutions, and improving user experience. In general, the power equipment fault analysis method proposed in the embodiment of this specification improves the accuracy and reliability of fault detection through automated calculation and intelligent reasoning, while improving the flexibility and adaptability of the system and improving user experience. The metering equipment pressure loss diagnosis intelligent agent system can monitor the voltage level of each phase, issue an alarm immediately once an abnormality is detected, and provide detailed fault information. It is suitable for different types of power configuration environments, including three-phase three-wire and three-phase four-wire systems. It can greatly shorten the troubleshooting cycle and improve maintenance efficiency.

[0097] See also Figure 5 , Figure 5 A hierarchical structure diagram of a pressure loss determination condition provided by an embodiment of this specification is shown. The metering pressure loss can be divided into three-phase three-wire metering pressure loss, three-phase four-wire metering pressure loss (low-voltage metering) and three-phase four-wire metering pressure loss (high-voltage metering). Next, the pressure loss determination conditions of these three types of metering pressure loss are explained respectively.

[0098] For three-phase three-wire metering pressure loss: three-phase three-wire metering pressure loss can be divided into phase A pressure loss, phase B pressure loss and phase C pressure loss. The failure mode of phase A pressure loss can be divided into phase A primary side line break and phase A secondary side line break. The failure mode of phase B pressure loss can be divided into phase B primary side line break and phase B secondary side line break. The failure mode of phase C pressure loss can be divided into phase C primary side line break and phase C secondary side line break. If the metering equipment pressure loss diagnosis intelligent body system determines that the metering equipment meets the pressure loss determination condition 1, then the failure mode of the metering equipment is determined to be the phase A primary side line break in the three-phase three-wire metering pressure loss. If the metering equipment pressure loss diagnosis intelligent body system determines that the metering equipment meets the pressure loss determination condition 2, then the failure mode of the metering equipment is determined to be the phase A secondary side line break in the three-phase three-wire metering pressure loss. If the metering equipment pressure loss diagnosis intelligent body system determines that the metering equipment meets the pressure loss determination condition 3, then the failure mode of the metering equipment is determined to be the phase B primary side line break in the three-phase three-wire metering pressure loss. If the metering equipment pressure loss diagnosis intelligent agent system determines that the metering equipment meets the pressure loss determination condition 4, then the failure mode of the metering equipment is determined to be the secondary side line break of phase B in the three-phase three-wire metering pressure loss. If the metering equipment pressure loss diagnosis intelligent agent system determines that the metering equipment meets the pressure loss determination condition 5, then the failure mode of the metering equipment is determined to be the primary side line break of phase C in the three-phase three-wire metering pressure loss. If the metering equipment pressure loss diagnosis intelligent agent system determines that the metering equipment meets the pressure loss determination condition 6, then the failure mode of the metering equipment is determined to be the secondary side line break of phase C in the three-phase three-wire metering pressure loss.

[0099] For three-phase four-wire metering pressure loss (low-voltage metering): Three-phase four-wire metering pressure loss (low-voltage metering) can be divided into phase A pressure loss, phase B pressure loss and phase C pressure loss. The fault mode of phase A pressure loss is the primary side line break of phase A. The fault mode of phase B pressure loss is the primary side line break of phase B. The fault mode of phase C pressure loss is the primary side line break of phase C. If the metering equipment pressure loss diagnosis intelligent body system determines that the metering equipment meets the pressure loss determination condition 7, then the fault mode of the metering equipment is determined to be the phase A pressure loss in the three-phase four-wire metering pressure loss (low-voltage metering). If the metering equipment pressure loss diagnosis intelligent body system determines that the metering equipment meets the pressure loss determination condition 8, then the fault mode of the metering equipment is determined to be the phase B pressure loss in the three-phase four-wire metering pressure loss (low-voltage metering). If the metering equipment pressure loss diagnosis intelligent body system determines that the metering equipment meets the pressure loss determination condition 9, then the fault mode of the metering equipment is determined to be the phase C pressure loss in the three-phase four-wire metering pressure loss (low-voltage metering).

[0100] For three-phase four-wire metering pressure loss (high-voltage metering): Three-phase four-wire metering pressure loss (high-voltage metering) can be divided into A-phase pressure loss, B-phase pressure loss and C-phase pressure loss. The failure mode of A-phase pressure loss can be divided into A-phase primary side line break and A-phase secondary side line break. The failure mode of B-phase pressure loss can be divided into B-phase primary side line break and B-phase secondary side line break. The failure mode of C-phase pressure loss can be divided into C-phase primary side line break and C-phase secondary side line break. If the metering equipment pressure loss diagnosis intelligent body system determines that the metering equipment meets the pressure loss determination condition 10, then the failure mode of the metering equipment is determined to be the A-phase primary side line break in the three-phase four-wire metering pressure loss (high-voltage metering). If the metering equipment pressure loss diagnosis intelligent body system determines that the metering equipment meets the pressure loss determination condition 11, then the failure mode of the metering equipment is determined to be the A-phase secondary side line break in the three-phase four-wire metering pressure loss (high-voltage metering). If the metering equipment pressure loss diagnosis intelligent body system determines that the metering equipment meets the pressure loss determination condition 12, the failure mode of the metering equipment is determined to be the primary side line break of phase B in the three-phase four-wire metering pressure loss (high-voltage metering). If the metering equipment pressure loss diagnosis intelligent body system determines that the metering equipment meets the pressure loss determination condition 13, the failure mode of the metering equipment is determined to be the secondary side line break of phase B in the three-phase four-wire metering pressure loss (high-voltage metering). If the metering equipment pressure loss diagnosis intelligent body system determines that the metering equipment meets the pressure loss determination condition 14, the failure mode of the metering equipment is determined to be the primary side line break of phase C in the three-phase four-wire metering pressure loss (high-voltage metering). If the metering equipment pressure loss diagnosis intelligent body system determines that the metering equipment meets the pressure loss determination condition 15, the failure mode of the metering equipment is determined to be the secondary side line break of phase C in the three-phase four-wire metering pressure loss (high-voltage metering).

[0101] It should be noted that the pressure loss determination condition 1 is obtained based on the following formulas (1) to (4), formula (3) is deduced based on formula (1), and formula (4) is deduced based on formula (2). If the metering equipment meets the following criteria, it is determined that the metering equipment meets the pressure loss determination condition 1: Main criteria, U 12 <40 and U 32 ≈100; auxiliary criterion, when φ 1 >0,P 1 <0; when φ 1 <0, P 1 >0; when φ 1 =0,P 1 =0, where φ 1 ≈φ 2 =acosθ 2 +30°, Q 2 >0, acosθ 2 >0; when Q 2 <0, acosθ 2 <0; when Q 2 =0, acosθ 2 =0.

[0102] U 12 =U cb =U 2e * (0~30) / U 2e ≈0~30<40(1) U 32 =U cb =U 2e ≈100 (2) P 1 =U 12 I 1 cosθ 1 =U cb I a cos(90°+φ 1 ), when φ 1 >0,P 1 <0; when φ 1 <0, P 1 >0; when φ 1 =0,P 1 =0(3) P 2 =U 32 I 2 cosθ 2 =U cb I c cos(30°-φ 2 ), φ 2 =acosθ 2 +30°,Q 2 >0, acosθ 2 >0; when Q 2 <0, acosθ 2 <0; when Q 2 =0, acosθ 2 =0(4) The pressure loss determination condition 2 is obtained based on the following formulas (5) to (8). Formula (7) is deduced based on formula (5), and formula (8) is deduced based on formula (6). If the metering equipment meets the following criteria, it is determined that the metering equipment meets the pressure loss determination condition 2: Main criteria, U 12 =0 or 33<U 12 <67 and U 32 ≈100; auxiliary criterion, when φ 1 >0,P 1 <0; when φ 1 <0, P 1 >0; when φ 1 =0,P 1 =0, where φ 1 ≈φ 2 =acosθ2 +30°, Q 2 >0, acosθ 2 >0; when Q 2 <0, acosθ 2 <0; when Q 2 =0, acosθ 2 =0.

[0103] U 12 =U cb =U 2e * (0 or 33~67) / U 2e ≈0 or 33~67 (5) U 32 =U cb =U 2e ≈100 (6) P 1 =U 12 I 1 cosθ 1 =U cb I a cos(90°+φ 1 ), when φ 1 >0,P 1 <0; when φ 1 <0, P 1 >0; when φ 1 =0,P 1 =0(7) P 2 =U 32 I 2 cosθ 2 =U cb I c cos(30°-φ 2 ), φ 2 =acosθ 2 +30°, when Q 2 >0, acosθ 2 >0; when Q 2 <0, acosθ 2 <0; when Q 2 =0, acosθ 2 =0(8) The pressure loss determination condition 3 is obtained based on the following formulas (9) to (12). Formula (11) is deduced based on formula (9), and formula (12) is deduced based on formula (10). If the metering equipment meets the following criteria, it is determined that the metering equipment meets the pressure loss determination condition 3: Main criteria, U 12 =40~50andU 32 =50~60; auxiliary criterion, when φ>0, P 1 >P2 ; When φ<0, P 1 <P 2 ; When φ=0, P 1 =P 2 ; where φ 1 ≈φ 2 ≈φ.

[0104] U 12 =U ac =U 2e * (40~50) / U 2e ≈40~50 (9) U 32 =U ca =U 2e * (50~60) / U 2e ≈50~60 (10) P 1 =U 12 I 1 cosθ 1 =U cb I a cos(30°-φ 1 )(11) P 2 =U 32 I 2 cosθ 2 =U cb I c cos(30°+φ 2 )(12) The pressure loss determination condition 4 is obtained based on the following formulas (13) to (16). Formula (15) is deduced based on formula (13), and formula (16) is deduced based on formula (14). If the metering equipment meets the following criteria, it is determined that the metering equipment meets the pressure loss determination condition 4: Main criteria, U1 2 =40~50andU 32 =50~60; auxiliary criterion, when φ>0, P 1 >P 2 ; When φ<0, P 1 <P 2 ; When φ=0, P 1 =P 2 ; where φ 1 ≈φ 2 ≈φ.

[0105] U 12 =U ac =U 2e * (40~50) / U 2e ≈40~50 (13) U 32 =Uca =U 2e * (50~60) / U 2e ≈50~60 (14) P 1 =U 12 I 1 cosθ 1 =U cb I a cos(30°-φ 1 )(15) P 2 =U 32 I 2 cosθ 2 =U cb I c cos(30°+φ 2 )(16) The pressure loss determination condition 5 is obtained based on the following formulas (17) to (20). Formula (19) is deduced based on formula (17), and formula (20) is deduced based on formula (18). If the metering equipment meets the following criteria, it is determined that the metering equipment meets the pressure loss determination condition 5: Main criteria, U 32 ≈0 or U 32 ≈33<40 and U 12 ≈100; auxiliary criterion, when φ 2 >0,P 2 >0; when φ 2 <0, P 2 <0; when φ 2 =0,P 2 =0; where φ 2 ≈φ 1 =acosθ 1 -30°, Q 1 >0, acosθ 1 >0; when Q 1 <0, acosθ 1 <0;Q 1 =0, acosθ 1 =0.

[0106] U 12 =U ab =U 2e ≈100 (17) U 32 =U ab =U 2e * (0~33) / U 2e ≈0~33<40(18) P 1 =U 12 I 1 cosθ1 =U ab I a cos(30°+φ), φ=acosθ 1 -30° when Q 1 >0, acosθ 1 >0; when Q 1 <0, acosθ 1 <0;Q 1 =0, acosθ 1 =0(19) P 2 =U 32 I 2 cosθ 2 =U ab I c cos(90°-φ 2 ), when φ2>0, P2>0; φ2<0, P2<0; φ2=0, P2=0 (20) The pressure loss determination condition 6 is obtained based on the following formulas (21) to (24). Formula (23) is deduced based on formula (21), and formula (24) is deduced based on formula (22). If the metering equipment meets the following criteria, it is determined that the metering equipment meets the pressure loss determination condition 6: Main criteria, U 32 ≈0 or U 32 ≈50andU 12 ≈100; auxiliary criterion, when φ 2 >0,P 2 >0; when φ 2 <0, P 2 <0; when φ 2 =0,P 2 =0; where φ 2 ≈φ 1 =acosθ 1 -30°, Q 1 >0, acosθ 1 >0; when Q 1 <0, acosθ 1 <0;Q 1 =0, acosθ 1 =0.

[0107] U 12 =U ab =U 2e ≈100 (21) U 32 =U ab =U 2e * (0~50) / U 2e ≈0~50<60(22) P 1 =U12 I 1 cosθ 1 =U ab I a cos (30°+φ), φ=acosθ 1 -30° when Q 1 >0, acosθ 1 >0; when Q 1 <0, acosθ 1 <0;Q 1 =0, acosθ 1 =0(23) P 2 =U 32 I 2 cosθ 2 =U ab I c cos(90°-φ 2 ), when φ 2 >0,P 2 >0; when φ 2 <0, P 2 <0; when φ 2 =0,P 2 =0; (24) The pressure loss determination condition 7 is obtained based on the following formulas (25) to (26). If the metering equipment meets the following criteria, it is determined that the metering equipment meets the pressure loss determination condition 7: Main criteria, U 1 ≈0, U 2 =U 3 ≈220, auxiliary criterion, when the three phases are balanced, P 1 ≈0;P 2 =P 3 ; P≈P 2 +P 3 .

[0108] U 1 =U a ≈0 (25) U 2 =U b ≈U 3 =U c =U 2e ≈220 (26) The pressure loss determination condition 8 is obtained based on the following formulas (27) to (28). If the metering equipment meets the following criteria, it is determined that the metering equipment meets the pressure loss determination condition 8: Main criteria, U 2 ≈0, U 1 =U 3 ≈220, auxiliary criterion, when the three phases are balanced, P 2 ≈0;P 1=P 3 ; P≈P 1 +P 3 .

[0109] U 2 =U b ≈0 (27) U 1 =U a ≈U 3 =U c =U 2e ≈220 (28) The pressure loss determination condition 9 is obtained based on the following formulas (29) to (30). If the metering equipment meets the following criteria, it is determined that the metering equipment meets the pressure loss determination condition 9: Main criteria, U 3 ≈0, U 1 =U 2 ≈220, auxiliary criterion, when the three phases are balanced, P 3 ≈0;P 1 =P 2 ; P≈P 1 +P 2 .

[0110] U 3 =U c ≈0 (29) U 1 =U a ≈U 2 =U b =U 2e ≈220 (30) The pressure loss determination condition 10 is obtained based on the following formulas (31) to (32). If the metering equipment meets the following criteria, it is determined that the metering equipment meets the pressure loss determination condition 10: Main criteria, U 1 ≈0 and U 2 =U 3 ≈58, auxiliary criterion, when the three phases are balanced, P 1 ≈0;P 2 =P 3 ; P≈P 2 +P 3 .

[0111] U 1 =U a ≈0 (31) U 2 =U b ≈U 3 =U c =U 2e ≈58 (32) The pressure loss determination condition 11 is obtained based on the following formulas (33) to (34). If the metering equipment meets the following criteria, it is determined that the metering equipment meets the pressure loss determination condition 10: Main criteria, U 1 ≈0 and U 2 =U 3 ≈58, auxiliary criterion, when the three phases are balanced, P 1 ≈0;P 2 =P 3 ; P≈P 2 +P 3 .

[0112] U 1 =U a ≈0 (33) U 2 =U b ≈U 3 =U c =U 2e ≈58 (34) The pressure loss determination condition 12 is obtained based on the following formulas (35) to (36). If the metering equipment meets the following criteria, it is determined that the metering equipment meets the pressure loss determination condition 11: Main criteria, U 2 ≈0, U 1 =U 3 ≈58, auxiliary criterion, when the three phases are balanced, P 2 ≈0;P 1 =P 3 ; P≈P 1 +P 3 .

[0113] U 2 =U b ≈0 (35) U 1 =U a ≈U 3 =U c =U 2e ≈58 (36) The pressure loss determination condition 13 is obtained based on the following formulas (37) to (38). If the metering equipment meets the following criteria, it is determined that the metering equipment meets the pressure loss determination condition 11: Main criteria, U 2 ≈0, U 1 =U 3 ≈58, auxiliary criterion, when the three phases are balanced, P 2 ≈0;P 1 =P 3 ; P≈P 1 +P 3 .

[0114] U 2 =U b≈0 (37) U 1 =U a ≈U 3 =U c =U 2e ≈58 (38) The pressure loss determination condition 14 is obtained based on the following formulas (39) to (40). If the metering equipment meets the following criteria, it is determined that the metering equipment meets the pressure loss determination condition 12: Main criteria, U 3 ≈0, U 1 =U 2 ≈58, auxiliary criterion, when the three phases are balanced, P 3 ≈0;P 1 =P 2 ; P≈P 1 +P 2 .

[0115] U 3 =U c ≈0 (39) U 1 =U a ≈U 2 =U b =U 2e ≈58 (40) The pressure loss determination condition 15 is obtained based on the following formulas (41) to (42). If the metering equipment meets the following criteria, it is determined that the metering equipment meets the pressure loss determination condition 12: Main criteria, U 3 ≈0, U 1 =U 2 ≈58, auxiliary criterion, when the three phases are balanced, P 3 ≈0;P 1 =P 2 ; P≈P 1 +P 2 .

[0116] U 3 =U c ≈0 (41) U 1 =U a ≈U 2 =U b =U 2e ≈58 (42) It is worth noting that in formula (1) to formula (42), U 1 , U 12 and U a Indicates the voltage of phase A; U 2 , U b Indicates the B phase voltage; U 3 , U 32 and Uc Indicates the C phase voltage; P 1 Indicates the active power of phase A; P 2 Indicates the active power of phase C; Q 1 Indicates the reactive power of phase A; Q 2 Indicates the reactive power of phase C; φ 1 Indicates the phase angle 1; φ 2 Represents the phase angle 2; θ 1 Indicates the voltage phase angle 1; θ 2 Indicates voltage phase angle 2; U ac Indicates the line voltage between phase A and phase C; U cb Indicates the line voltage between phase C and phase B; U ca Indicates the voltage between phase C and phase A; U 2e Indicates standard voltage; I 1 Indicates the current of phase A; I 2 Indicates the C phase current; I a Indicates the current of phase A; I c Indicates the C phase current.

[0117] By applying the solution of the embodiment of this specification, automatic data processing and analysis are realized, and the relationship between the phase angle and the current and voltage is derived through the data formula to realize automatic calculation, which not only improves the accuracy of the calculation, but also reduces the need for manual intervention.

[0118] See also Figure 6 , Figure 6 A flowchart of a fault analysis model training method provided by an embodiment of this specification is shown, which specifically includes the following steps: Step 602: Obtain a sample set, wherein the sample set includes sample fault analysis data, a sample fault analysis strategy, and a sample fault analysis result, the sample fault analysis data is obtained by processing at least one sample data generation task based on a sample data generation strategy, and the sample fault analysis strategy, at least one sample data generation task, and the sample data generation strategy are obtained based on sample fault description information of a sample target device.

[0119] Step 604: Input the sample fault analysis data and the sample fault analysis strategy into the fault analysis model to obtain the predicted fault analysis result.

[0120] Step 606: Train the fault analysis model according to the sample fault analysis results and the predicted fault analysis results to obtain a trained fault analysis model.

[0121] It should be noted that the definition and generation method of "sample fault analysis data, sample fault analysis strategy and sample fault analysis result" can refer to the definition and generation method of the above-mentioned "fault analysis data, fault analysis strategy and fault analysis result", and the embodiments of this specification will not be repeated. The training method of the fault analysis model is supervised fine-tuning, that is, the sample set includes the real processing label (sample fault analysis result) of the fault analysis model, and the sample fault analysis result is the generation target of the fault analysis model, which is used to guide the training process of the fault analysis model. There are many ways to obtain the sample set, which are selected according to the actual situation, and the embodiments of this specification do not make any restrictions on this. In one possible implementation method of this specification, a sample set sent by a user through a client can be received, and in another possible implementation method of this specification, a sample set can be read from other data acquisition devices or databases. The implementation method of "inputting sample fault analysis data and sample fault analysis strategy into the fault analysis model to obtain the predicted fault analysis result" can refer to the implementation method of "using the fault analysis model to process the fault analysis data and fault analysis strategy to obtain the fault analysis result of the target device", and the embodiments of this specification will not be repeated.

[0122] In practical applications, when training a fault analysis model based on sample fault analysis results and predicted fault analysis results, the analysis loss value can be calculated based on the sample fault analysis results and predicted fault analysis results, and the model parameters of the fault analysis model can be adjusted based on the analysis loss value until the training process meets the preset stop conditions, and a trained fault analysis model is obtained. There are many functions for calculating the analysis loss value, such as the cross entropy loss function, the L1 norm loss function, the maximum loss function, the mean square error loss function, the logarithmic loss function, etc., which are selected according to the actual situation, and the embodiments of this specification do not make any restrictions on this. The preset stop conditions include but are not limited to the analysis loss value being less than or equal to the preset threshold value and the number of iterations reaching the preset number of iterations, wherein the preset threshold value and the preset number of iterations are selected according to the actual situation, and the embodiments of this specification do not make any restrictions on this.

[0123] In a possible implementation of the present specification, after calculating the analysis loss value, the analysis loss value can be compared with a preset threshold. Specifically, if the analysis loss value is greater than the preset threshold, it means that the difference between the sample fault analysis result and the predicted fault analysis result is large, and the fault analysis ability of the fault analysis model is poor. At this time, the model parameters of the fault analysis model can be adjusted, and the fault analysis model can be continuously trained until the analysis loss value is less than or equal to the preset threshold, indicating that the difference between the sample fault analysis result and the predicted fault analysis result is small, and the preset stop condition is met, and a fault analysis model that has completed training is obtained.

[0124] In another possible implementation of the present specification, in addition to comparing the size relationship between the analysis loss value and the preset threshold, the number of iterations may also be combined to determine whether the current fault analysis model has been trained. Specifically, if the analysis loss value is greater than the preset threshold, the model parameters of the fault analysis model are adjusted, and the fault analysis model is continuously trained until the preset number of iterations is reached, and the iteration is stopped to obtain a fault analysis model that has been trained.

[0125] By applying the scheme of the embodiments of this specification, the fault analysis model is trained according to the sample fault analysis results and the predicted fault analysis results. If the preset stop condition is not met, the fault analysis model is continuously trained until the preset stop condition is met, and the training is completed to obtain the fault analysis model. By continuously adjusting the model parameters of the fault analysis model, the fault analysis capability of the fault analysis model can be stimulated, making the fault analysis model more accurate.

[0126] See also Figure 7 , Figure 7 A flowchart of an information processing method based on a fault analysis model provided by an embodiment of the present specification is shown. The information processing method based on the fault analysis model is applied to a task platform and specifically includes the following steps: Step 702: Receive a model request sent by a terminal device.

[0127] Step 704: Based on the model request, determine a target fault analysis model from multiple fault analysis models, wherein the target fault analysis model is used in the execution process of the fault analysis method or the power equipment fault analysis method or the fault analysis model training method.

[0128] It should be noted that the target fault analysis model is a fault analysis model suitable for the target fault analysis scenario. The model request includes at least one of the scenario identifier of the target fault analysis scenario, the scenario input data of the target fault analysis scenario, and the model specification parameters. Based on the model request, there are multiple ways to determine the target fault analysis model from multiple fault analysis models. In the first possible implementation of this specification, based on the model request, the corresponding target fault analysis model can be searched from at least one fault analysis model included in the model library; in the second possible implementation of this specification, based on the model request, the target fault analysis model can be trained and obtained; in the third possible implementation of this specification, the target fault analysis model can be constructed based on the model request.

[0129] Exemplarily, based on the scenario identifier of the target fault analysis scenario, at least one pre-trained fault analysis model can be searched from the model library, and then based on the model specification parameters, an initial fault analysis model can be screened from at least one fault analysis model. Then, based on the scenario input data of the target fault analysis scenario, the screened initial fault analysis model can be trained to obtain a target fault analysis model suitable for user needs.

[0130] The solution of the embodiment of this specification is applied to obtain the target fault analysis model according to user needs, realize personalized model service, provide users with an efficient, flexible and easy-to-use model service method, and improve user experience.

[0131] In an optional embodiment of the present specification, the model request includes a scenario identifier of a target fault analysis scenario; and the above-mentioned determining the target fault analysis model from multiple fault analysis models based on the model request may include the following steps: Based on the scenario identifier of the target fault analysis scenario, a target fault analysis model suitable for the target fault analysis scenario is searched from a model library, wherein the model library stores a plurality of fault analysis models suitable for different fault analysis scenarios.

[0132] It should be noted that the scenario identifier refers to a unique or specific label used to distinguish different fault analysis scenarios. The model library is a database for storing and managing various pre-trained deep learning models. Multiple fault analysis models adapted to different fault analysis scenarios cover different fault analysis scenarios and requirements. The model library allows users to select appropriate fault analysis models according to their needs, or directly call fault analysis models for fault analysis through the application programming interface.

[0133] The various models stored in the model library that are applicable to different fault analysis scenarios are optimized for specific fault analysis environments. For example, based on the scenario identifier of the target fault analysis scenario, "pressure loss analysis of metering equipment", the target fault analysis model applicable to the pressure loss analysis scenario of metering equipment can be searched from the model library.

[0134] By applying the solution of the embodiments of this specification, based on scenario requirements, the target fault analysis model suitable for the scenario is accurately found through scenario identification, making the fault analysis more accurate and more in line with the scenario, thereby improving user experience and fault analysis quality.

[0135] In an optional embodiment of the present specification, the model request includes scenario input data of a target fault analysis scenario; and the above-mentioned determining the target fault analysis model from multiple fault analysis models based on the model request may include the following steps: Determine an initial fault analysis model suitable for a target fault analysis scenario from among multiple fault analysis models; Based on the scenario input data of the target fault analysis scenario, the parameters of the initial fault analysis model are adjusted to obtain the target fault analysis model.

[0136] It should be noted that the initial fault analysis model refers to a model that is applicable to the target fault analysis scenario among multiple fault analysis models. The initial fault analysis model may not only be applicable to the target fault analysis scenario, but also to other fault analysis scenarios. It is a general fault analysis model that can be applied to different fault analysis scenarios. The initial fault analysis model can be used to process the fault analysis task of the target fault analysis scenario, but the effect may not be very good. At this time, the parameters of the initial fault analysis model can be adjusted based on the scenario input data of the target fault analysis scenario. For example, the initial fault analysis model can be optimized based on the scenario input data of the metering equipment pressure loss analysis scenario to obtain the target fault analysis model applicable to the metering equipment pressure loss analysis scenario. The scenario input data of the target fault analysis scenario can be understood as a sample set of the target fault analysis scenario (including sample fault analysis data, sample fault analysis strategy, and sample fault analysis results).

[0137] By applying the solution of the embodiments of this specification, based on scenario requirements, a general initial fault analysis model is further trained through scenario input data to obtain a target fault analysis model adapted to the scenario, so that the target fault analysis model is more in line with the scenario, thereby improving user experience and fault analysis quality.

[0138] In an optional embodiment of the present specification, the model request includes model specification parameters; and the above-mentioned determining a target fault analysis model from a plurality of fault analysis models based on the model request may include the following steps: Based on the model specification parameters, a corresponding target fault analysis model is searched from a model library, wherein the model library stores a plurality of fault analysis models with different model specification parameters.

[0139] It should be noted that model specification parameters refer to various parameters that define the structure and behavior of the model. These parameters can be roughly divided into two categories: model parameters (learnable parameters) and hyperparameters. Model parameters refer to parameters that are automatically adjusted by the back-propagation algorithm during model training, including but not limited to weight matrices (weights) and biases (biases). For example, in a simple fully connected layer, the weight matrix is ​​a two-dimensional tensor that connects the neurons in the input layer and the output layer; the bias term is a one-dimensional vector that provides an additional offset value for each output neuron. Hyperparameters refer to parameters set before starting model training to control the learning process and architecture of the model. Hyperparameters include but are not limited to learning rate and the number of neurons per layer (Number of Neurons per Layer), which should be selected according to actual conditions.

[0140] By applying the solution of the embodiments of this specification, the corresponding target fault analysis model can be accurately found based on the model specification parameters, thereby ensuring the efficient and stable operation of the target fault analysis model and improving the user experience.

[0141] In an optional embodiment of the present specification, after determining the target fault analysis model from multiple fault analysis models based on the model request, the following steps may also be included: Deploy a target fault analysis model, and build a fault analysis interface based on the target fault analysis model so that the terminal device schedules the target fault analysis model to perform the target fault analysis task.

[0142] It should be noted that the fault analysis interface is an interactive programming interface for the terminal device to schedule the target fault analysis model to perform target fault analysis task processing. The fault analysis interface is usually provided in the form of an application programming interface.

[0143] In practical applications, there are many ways to deploy the target fault analysis model. In one possible implementation, the target fault analysis model can be deployed on the cloud-side device using the infrastructure provided by the cloud service provider. In another possible implementation, the target fault analysis model can be deployed on the edge device using a lightweight framework. For example, the target fault analysis model can be deployed on a distributed system, and based on the target fault analysis model, a fault analysis interface can be constructed and provided to the terminal device, so that the terminal device can schedule the target fault analysis model to perform the target fault analysis task.

[0144] By applying the solution provided in the embodiments of this specification, deploying the target fault analysis model, and building a fault analysis interface based on the target fault analysis model, the terminal device can efficiently call the target fault analysis model and improve the processing quality and response speed of the target fault analysis task.

[0145] See also Figure 8 , Figure 8 A schematic diagram of the structure of a task platform provided by an embodiment of the present specification is shown, wherein the task platform 800 includes a request interface 802 and a response unit 804; The request interface 802 is used to receive a model request sent by a terminal device, wherein the model request includes a scenario identifier of a target fault analysis scenario, scenario input data of the target fault analysis scenario, and at least one of a model specification parameter; The response unit 804 is used to determine a target fault analysis model from multiple fault analysis models based on the model request, wherein the target fault analysis model is used in the execution process of the fault analysis method or the power equipment fault analysis method or the fault analysis model training method.

[0146] In an optional embodiment of the present specification, the task platform further includes a fault analysis interface, and the fault analysis interface is constructed based on the target fault analysis model; The fault analysis interface is used for terminal equipment to schedule and execute target fault analysis tasks.

[0147] By applying the solution of the embodiments of this specification, the task platform adapts to user needs to obtain the target fault analysis model, realizes personalized model services, provides users with an efficient, flexible and easy-to-use model service platform, and improves user experience.

[0148] The above is a schematic scheme of a task platform of this embodiment. It should be noted that the technical scheme of the task platform and the technical scheme of the information processing method based on the fault analysis model belong to the same concept. For details not described in detail in the technical scheme of the task platform, please refer to the description of the technical scheme of the information processing method based on the fault analysis model.

[0149] Corresponding to the above-mentioned fault analysis method embodiment, this specification also provides a fault analysis device embodiment, Fig. 9 FIG. 1 is a schematic diagram showing the structure of a fault analysis device provided by an embodiment of the present specification. Fig. 9 As shown, the device comprises: A first acquisition module 902 is configured to acquire fault description information of a target device; The first determination module 904 is configured to construct a plurality of tasks to be processed based on the fault description information, and determine processing strategies for the tasks to be processed, wherein the plurality of tasks to be processed include a fault analysis task and at least one data generation task, and the processing strategies include a fault analysis strategy for the fault analysis task and a data generation strategy for the at least one data generation task; A first processing module 906 is configured to process at least one data generation task according to a data generation strategy to obtain fault analysis data; The second processing module 908 is configured to process the fault analysis data and the fault analysis strategy using the fault analysis model to obtain the fault analysis result of the target device.

[0150] Optionally, the first determination module 904 is further configured to obtain task generation prompt information, wherein the task generation prompt information is used to guide the task planning model to determine multiple tasks to be processed and processing strategies based on the fault description information; the task generation prompt information and the fault description information are input into the task planning model to obtain multiple tasks to be processed and processing strategies.

[0151] Optionally, the data generation strategy is used to indicate the task processing engine and reference data type of the data generation task; the first processing module 906 is further configured to obtain task reference data based on the reference data type; call the task processing engine to process at least one data generation task according to the task reference data to obtain fault analysis data.

[0152] Optionally, the device further includes: a screening module configured to screen out a task processing engine from a plurality of candidate processing engines according to engine description information and data generation strategies of the plurality of candidate processing engines.

[0153] Optionally, the target device includes an electrical device, the data generation task includes a phase angle generation task, the task processing engine includes a phase angle calculation engine, and the task reference data includes reference electrical parameters; the first processing module 906 is further configured to call the phase angle calculation engine, process the phase angle generation task according to the reference electrical parameters, and obtain the phase angle of the electrical device.

[0154] Optionally, the second processing module 908 is further configured to construct fault analysis prompt information according to the fault analysis strategy, wherein the fault analysis prompt information is used to guide the fault analysis model to analyze the fault analysis data to obtain the fault analysis result; and input the fault analysis prompt information and the fault analysis data into the fault analysis model to obtain the fault analysis result of the target device.

[0155] Optionally, the second processing module 908 is further configured to obtain a fault determination condition corresponding to the fault analysis strategy, and construct fault analysis prompt information according to the fault determination condition.

[0156] Optionally, the second processing module 908 is further configured to obtain the primary side fault determination condition and the secondary side fault determination condition when the fault analysis strategy is the power equipment voltage loss analysis strategy, and construct fault analysis prompt information based on the primary side fault determination condition and the secondary side fault determination condition.

[0157] Optionally, the device also includes: a fourth acquisition module, configured to obtain device association information of the target device; a second processing module 908, further configured to input the fault analysis prompt information, device association information and fault analysis data into the fault analysis model to obtain the fault analysis result of the target device.

[0158] Optionally, the apparatus further includes: a generating module configured to generate a fault diagnosis report of the target device according to the fault description information, the processing strategies of the plurality of tasks to be processed, the fault analysis data and the fault analysis results.

[0159] Optionally, the device also includes: an adjustment module, configured to receive result feedback information sent by the client, wherein the result feedback information is information that the client feeds back on the fault analysis results based on analysis requirements; construct model optimization data based on the result feedback information; and use the model optimization data to adjust parameters of the fault analysis model to obtain an adjusted fault analysis model.

[0160] By applying the scheme of the embodiments of this specification, a plurality of tasks to be processed are constructed based on the fault description information, so that a fault description information processing task is decomposed into a plurality of tasks to be processed, so that various complex fault analysis situations can be flexibly responded to, and the flexibility and efficiency of the fault analysis process are improved. By processing each data generation task, the fault analysis data required for the fault analysis process is fully generated, the comprehensiveness and accuracy of the fault analysis are ensured, and the fault analysis data and fault analysis strategy are automatically processed using the fault analysis model, so that the fault analysis results can be generated efficiently and accurately.

[0161] The above is a schematic scheme of a fault analysis device of this embodiment. It should be noted that the technical scheme of the fault analysis device and the technical scheme of the above fault analysis method belong to the same concept, and the details not described in detail in the technical scheme of the fault analysis device can be referred to the description of the technical scheme of the above fault analysis method.

[0162] Corresponding to the above-mentioned power equipment fault analysis method embodiment, this specification also provides a power equipment fault analysis device embodiment, Fig.10 FIG. 1 is a schematic diagram showing the structure of a power equipment fault analysis device provided by an embodiment of the present specification. Fig.10 As shown, the device comprises: The second acquisition module 1002 is configured to acquire fault description information of the electric power equipment; The second determination module 1004 is configured to construct a plurality of tasks to be processed based on the fault description information, and determine processing strategies for the tasks to be processed, wherein the plurality of tasks to be processed include a fault analysis task and at least one data generation task, and the processing strategies include a fault analysis strategy for the fault analysis task and a data generation strategy for the at least one data generation task; The third processing module 1006 is configured to process at least one data generation task according to the data generation strategy to obtain fault analysis data; The fourth processing module 1008 is configured to process the fault analysis data and the fault analysis strategy using the fault analysis model to obtain the fault analysis result of the power equipment.

[0163] By applying the scheme of the embodiment of this specification, by constructing multiple tasks to be processed based on fault description information, it is possible to decompose a fault description information processing task into multiple tasks to be processed, so that various complex power equipment fault analysis situations can be flexibly responded to, and the flexibility and efficiency of the fault analysis process are improved. By processing each data generation task, the fault analysis data required for the fault analysis process is fully generated, ensuring the comprehensiveness and accuracy of the power equipment fault analysis, and the fault analysis model is used to automatically process the fault analysis data and the fault analysis strategy, so that the fault analysis results of the power equipment can be generated efficiently and accurately.

[0164] The above is a schematic scheme of a power equipment fault analysis device of this embodiment. It should be noted that the technical scheme of the power equipment fault analysis device and the technical scheme of the power equipment fault analysis method described above belong to the same concept, and the details not described in detail in the technical scheme of the power equipment fault analysis device can be found in the description of the technical scheme of the power equipment fault analysis method described above.

[0165] Corresponding to the above-mentioned fault analysis model training method embodiment, this specification also provides a fault analysis model training device embodiment, Fig.11 FIG. 1 is a schematic diagram showing the structure of a fault analysis model training device provided by an embodiment of the present specification. Fig.11 As shown, the device comprises: The third acquisition module 1102 is configured to acquire a sample set, wherein the sample set includes sample fault analysis data, a sample fault analysis strategy, and a sample fault analysis result, the sample fault analysis data is obtained by processing at least one sample data generation task based on the sample data generation strategy, and the sample fault analysis strategy, at least one sample data generation task, and the sample data generation strategy are obtained based on sample fault description information of the sample target device; An input module 1104 is configured to input sample fault analysis data and sample fault analysis strategies into a fault analysis model to obtain a predicted fault analysis result; The training module 1106 is configured to train the fault analysis model according to the sample fault analysis results and the predicted fault analysis results to obtain a trained fault analysis model.

[0166] By applying the scheme of the embodiments of this specification, the fault analysis model is trained according to the sample fault analysis results and the predicted fault analysis results. If the preset stop condition is not met, the fault analysis model is continuously trained until the preset stop condition is met, and the training is completed to obtain the fault analysis model. By continuously adjusting the model parameters of the fault analysis model, the fault analysis capability of the fault analysis model can be stimulated, making the fault analysis model more accurate.

[0167] The above is a schematic scheme of a fault analysis model training device of this embodiment. It should be noted that the technical scheme of the fault analysis model training device and the technical scheme of the above-mentioned fault analysis model training method belong to the same concept, and the details not described in detail in the technical scheme of the fault analysis model training device can be found in the description of the technical scheme of the above-mentioned fault analysis model training method.

[0168] Corresponding to the above-mentioned information processing method embodiment based on the fault analysis model, this specification also provides an information processing device embodiment based on the fault analysis model, Fig.12 FIG. 1 is a schematic diagram showing a structure of an information processing device based on a fault analysis model provided by an embodiment of the present specification. Fig.12 As shown, the device is applied to a task platform and includes: The receiving module 1202 is configured to receive a model request sent by a terminal device; The third determination module 1204 is configured to determine a target fault analysis model from multiple fault analysis models based on the model request, wherein the target fault analysis model is used in the execution process of the fault analysis method or the power equipment fault analysis method or the fault analysis model training method.

[0169] Optionally, the model request includes a scenario identifier of a target fault analysis scenario; the third determination module 1204 is further configured to search a target fault analysis model suitable for the target fault analysis scenario from a model library based on the scenario identifier of the target fault analysis scenario, wherein the model library stores multiple fault analysis models suitable for different fault analysis scenarios.

[0170] Optionally, the model request includes scenario input data of the target fault analysis scenario; the third determination module 1204 is further configured to determine an initial fault analysis model suitable for the target fault analysis scenario from multiple fault analysis models; based on the scenario input data of the target fault analysis scenario, adjust the parameters of the initial fault analysis model to obtain the target fault analysis model.

[0171] Optionally, the model request includes model specification parameters; the third determination module 1204 is further configured to search for a corresponding target fault analysis model from a model library based on the model specification parameters, wherein the model library stores a plurality of fault analysis models with different model specification parameters.

[0172] The solution of the embodiment of this specification is applied to obtain the target fault analysis model according to user needs, realize personalized model service, provide users with an efficient, flexible and easy-to-use model service method, and improve user experience.

[0173] The above is a schematic scheme of an information processing device based on a fault analysis model of this embodiment. It should be noted that the technical scheme of the information processing device based on the fault analysis model and the technical scheme of the information processing method based on the fault analysis model belong to the same concept, and the details not described in detail in the technical scheme of the information processing device based on the fault analysis model can be referred to the description of the technical scheme of the information processing method based on the fault analysis model.

[0174] Fig.13 A structural block diagram of a computing device 1300 provided in one embodiment of the present specification is shown.

[0175] The computing device 1300 includes: a memory 1310 and a processor 1320; the memory 1310 is used to store computer programs / instructions, and the processor 1320 is used to execute computer programs / instructions. When the computer program / instructions are executed by the processor 1320, the steps of the above-mentioned fault analysis method or power equipment fault analysis method or fault analysis model training method or information processing method based on the fault analysis model are implemented.

[0176] In one or more embodiments of the present specification, the computing device may be understood as an integrated intelligent terminal, including but not limited to a server, a desktop computer, a personal computer, an all-in-one model machine, a mobile phone, a tablet computer or other portable intelligent terminal, etc., and the computing device may be pre-installed with the model described in the above embodiments of the present application.

[0177] Specifically, the computing device can pre-set multiple types of models, including but not limited to models in the fields of natural language processing, visual processing, speech processing, code processing, multimodal task processing, etc., so as to provide a variety of model choices. In different product forms, the computing device can support one or more model usage methods, including but not limited to model training, model calling, model fine-tuning, model deployment, model reasoning and application, etc. In some product forms, the computing device also supports model management, including but not limited to multi-type model management (supporting the management of multiple types of models such as discriminants and generative models), model version control (supporting the control of different model versions), model evaluation (based on model evaluation tools, evaluating the performance and effect of the model), etc. In other product forms, the computing device can also create applications based on models, provide application programming interface (API, Application Programming Interface) calling capabilities, and can call models to the created applications through API interfaces, while providing application management tools to achieve management and monitoring of applications.

[0178] Furthermore, the computing device can also include data management (supporting the creation and management of model tuning data sets), a training center (providing rich training resources to help users learn and master artificial intelligence technology), and basic management and control capabilities (providing enterprise-level basic management and control capabilities to ensure the security and efficient operation of the system). Through the above functions, a comprehensive, integrated artificial intelligence development, training, deployment and application device is provided.

[0179] Fig.14 A structural block diagram of an electronic device 1400 provided according to an embodiment of the present specification is shown.

[0180] The memory 1410 and the processor 1420 are connected via a bus 1430; The memory 1410 is used to store computer programs / instructions, and the processor 1420 is used to execute computer programs / instructions. When the computer program / instructions are executed by the processor 1420, the steps of the above-mentioned fault analysis method or power equipment fault analysis method or fault analysis model training method or information processing method based on the fault analysis model are implemented.

[0181] Specifically, the components of the electronic device 1400 include but are not limited to a memory 1410 and a processor 1420. The processor 1420 and the memory 1410 may be connected via a bus 1430.

[0182] The electronic device 1400 may also include an access device 1440 that enables the electronic device 1400 to communicate with a database 1450 storing data via one or more networks 1460. Examples of these networks include a public switched telephone network (PSTN), a local area network (LAN), a wide area network (WAN), a personal area network (PAN), or a combination of communication networks such as the Internet. The access device 1440 may include one or more of any type of network interface (e.g., a network interface card (NIC)) of wired or wireless, such as an IEEE 802.11 wireless local area network (WLAN) wireless interface, a world microwave interconnection access (Wi-MAX) interface, an Ethernet interface, a universal serial bus (USB) interface, a cellular network interface, a Bluetooth interface, a near field communication (NFC) interface, and the like.

[0183] In one embodiment of this specification, the above components of the electronic device 1400 and Fig.14 Other components not shown in the figure may also be connected to each other, for example, via a bus. It should be understood that Fig.14 The electronic device structure block diagram shown is only for the purpose of illustration, and is not intended to limit the scope of this specification. Those skilled in the art can add or replace other components as needed.

[0184] The electronic device 1400 may be any type of stationary or mobile electronic device, including a mobile computer or mobile computing device (e.g., a tablet computer, a personal digital assistant, a laptop computer, a notebook computer, a netbook, etc.), a mobile phone (e.g., a smart phone), a wearable electronic device (e.g., a smart watch, smart glasses, etc.), or other types of mobile devices, or a stationary electronic device such as a desktop computer or a PC. The electronic device 1400 may also be a mobile or stationary electronic device.

[0185] The above is a schematic scheme of an electronic device of this embodiment. It should be noted that the technical scheme of the electronic device and the technical schemes of the above-mentioned fault analysis method, power equipment fault analysis method, fault analysis model training method, and information processing method based on the fault analysis model belong to the same concept. For details not described in detail in the technical scheme of the electronic device, please refer to the description of the technical scheme of the above-mentioned fault analysis method or power equipment fault analysis method or fault analysis model training method or information processing method based on the fault analysis model.

[0186] An embodiment of the present specification also provides a computer-readable storage medium storing a computer program / instruction, which, when executed by a processor, implements the steps of the above-mentioned fault analysis method or power equipment fault analysis method or fault analysis model training method or information processing method based on a fault analysis model.

[0187] The above is a schematic scheme of a computer-readable storage medium of this embodiment. It should be noted that the technical scheme of the storage medium and the technical scheme of the above-mentioned fault analysis method, power equipment fault analysis method, fault analysis model training method and information processing method based on the fault analysis model belong to the same concept, and the details not described in detail in the technical scheme of the storage medium can be referred to the description of the technical scheme of the above-mentioned fault analysis method or power equipment fault analysis method or fault analysis model training method or information processing method based on the fault analysis model.

[0188] An embodiment of the present specification also provides a computer program product, including a computer program / instruction, which, when executed by a processor, implements the steps of the above-mentioned fault analysis method or power equipment fault analysis method or fault analysis model training method or information processing method based on a fault analysis model.

[0189] The above is a schematic scheme of a computer program product of this embodiment. It should be noted that the technical scheme of the computer program product and the technical schemes of the above-mentioned fault analysis method, power equipment fault analysis method, fault analysis model training method, and information processing method based on fault analysis model belong to the same concept. For details not described in detail in the technical scheme of the computer program product, please refer to the description of the technical scheme of the above-mentioned fault analysis method or power equipment fault analysis method or fault analysis model training method or information processing method based on fault analysis model.

[0190] The above is a description of a specific embodiment of the specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0191] The computer instructions include computer program codes, which may be in source code form, object code form, executable files or some intermediate forms, etc. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium, etc. It should be noted that the content contained in the computer-readable medium may be appropriately increased or decreased according to the requirements of patent practice. For example, in some regions, according to patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.

[0192] It should be noted that, for the above-mentioned method embodiments, for the sake of simplicity of description, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the embodiments of this specification are not limited by the order of the actions described, because according to the embodiments of this specification, some steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the embodiments of this specification.

[0193] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0194] The preferred embodiments of this specification disclosed above are only used to help explain this specification. The optional embodiments do not describe all the details in detail, nor do they limit the invention to only the specific implementation methods described. Obviously, many modifications and changes can be made according to the content of the embodiments of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the embodiments of this specification, so that technicians in the relevant technical field can well understand and use this specification. This specification is only limited by the claims and their full scope and equivalents.

Claims

1. A fault analysis method, comprising: Get the fault description information of the target device; Based on the fault description information, construct multiple tasks to be processed, and determine processing strategies for the tasks to be processed, wherein the multiple tasks to be processed include a fault analysis task and at least one data generation task, and the processing strategy includes a fault analysis strategy for the fault analysis task and a data generation strategy for the at least one data generation task; Processing the at least one data generation task according to the data generation strategy to obtain fault analysis data; The fault analysis data and the fault analysis strategy are processed using a fault analysis model to obtain a fault analysis result of the target device.

2. The method according to claim 1, wherein the constructing a plurality of tasks to be processed based on the fault description information and determining the processing strategies of the tasks to be processed comprises: Acquire task generation prompt information, wherein the task generation prompt information is used to guide the task planning model to determine the multiple tasks to be processed and the processing strategy based on the fault description information; The task generation prompt information and the fault description information are input into the task planning model to obtain the multiple tasks to be processed and the processing strategy.

3. The method according to claim 1, wherein the data generation strategy is used to indicate a task processing engine and a reference data type of the data generation task; The step of processing the at least one data generation task according to the data generation strategy to obtain fault analysis data includes: Based on the reference data type, obtaining task reference data; The task processing engine is called to process the at least one data generation task according to the task reference data to obtain fault analysis data.

4. The method according to claim 3, wherein before calling the task processing engine and processing the at least one data generation task according to the task reference data to obtain the fault analysis data, the method further comprises: The task processing engine is selected from the plurality of candidate processing engines according to the engine description information of the plurality of candidate processing engines and the data generation strategy.

5. The method according to claim 3, wherein the target device comprises an electric power device, the data generation task comprises a phase angle generation task, the task processing engine comprises a phase angle calculation engine, and the task reference data comprises a reference electrical parameter; The calling of the task processing engine to process the at least one data generation task according to the task reference data to obtain fault analysis data includes: The phase angle calculation engine is called to process the phase angle generation task according to the reference electrical parameters to obtain the phase angle of the electric power equipment.

6. The method according to claim 1, wherein the using a fault analysis model to process the fault analysis data and the fault analysis strategy to obtain a fault analysis result of the target device comprises: According to the fault analysis strategy, construct fault analysis prompt information, wherein the fault analysis prompt information is used to guide the fault analysis model to analyze the fault analysis data to obtain the fault analysis result; The fault analysis prompt information and the fault analysis data are input into the fault analysis model to obtain the fault analysis result of the target device.

7. The method according to claim 6, wherein constructing fault analysis prompt information according to the fault analysis strategy comprises: A fault determination condition corresponding to the fault analysis strategy is obtained, and the fault analysis prompt information is constructed according to the fault determination condition.

8. The method according to claim 7, wherein the acquiring the fault determination condition corresponding to the fault analysis strategy and constructing the fault analysis prompt information according to the fault determination condition comprises: When the fault analysis strategy is a power equipment voltage loss analysis strategy, a primary side fault determination condition and a secondary side fault determination condition are obtained, and the fault analysis prompt information is constructed according to the primary side fault determination condition and the secondary side fault determination condition.

9. The method according to claim 6, before inputting the fault analysis prompt information and the fault analysis data into the fault analysis model to obtain the fault analysis result of the target device, further comprising: Acquire device association information of the target device; The step of inputting the fault analysis prompt information and the fault analysis data into the fault analysis model to obtain the fault analysis result of the target device includes: The fault analysis prompt information, the device association information and the fault analysis data are input into the fault analysis model to obtain the fault analysis result of the target device.

10. The method according to any one of claims 1 to 9, wherein after the fault analysis model is used to process the fault analysis data and the fault analysis strategy to obtain the fault analysis result of the target device, the method further comprises: A fault diagnosis report of the target device is generated according to the fault description information, the processing strategies of the multiple tasks to be processed, the fault analysis data and the fault analysis results.

11. The method according to any one of claims 1 to 9, wherein after the fault analysis model is used to process the fault analysis data and the fault analysis strategy to obtain the fault analysis result of the target device, the method further comprises: Receiving result feedback information sent by the client, wherein the result feedback information is information provided by the client on the fault analysis result based on analysis requirements; According to the result feedback information, a model is constructed to optimize the data; The model optimization data is used to adjust the parameters of the fault analysis model to obtain an adjusted fault analysis model.

12. A method for analyzing a fault of an electric power device, comprising: Obtain fault description information of power equipment; Based on the fault description information, construct multiple tasks to be processed, and determine processing strategies for the tasks to be processed, wherein the multiple tasks to be processed include a fault analysis task and at least one data generation task, and the processing strategy includes a fault analysis strategy for the fault analysis task and a data generation strategy for the at least one data generation task; Processing the at least one data generation task according to the data generation strategy to obtain fault analysis data; The fault analysis model is used to process the fault analysis data and the fault analysis strategy to obtain a fault analysis result of the electric power equipment.

13. A fault analysis model training method, comprising: Acquire a sample set, wherein the sample set includes sample fault analysis data, a sample fault analysis strategy, and a sample fault analysis result, the sample fault analysis data is obtained by processing at least one sample data generation task based on a sample data generation strategy, and the sample fault analysis strategy, the at least one sample data generation task, and the sample data generation strategy are obtained based on sample fault description information of a sample target device; Inputting the sample fault analysis data and the sample fault analysis strategy into a fault analysis model to obtain a predicted fault analysis result; The fault analysis model is trained according to the sample fault analysis result and the predicted fault analysis result to obtain a trained fault analysis model.

14. An information processing method based on a fault analysis model, applied to a task platform, comprising: Receiving a model request sent by a terminal device; Based on the model request, a target fault analysis model is determined from a plurality of fault analysis models, wherein the target fault analysis model is used in the execution of the method according to any one of claims 1 to 11 or claim 12 or claim 13.

15. The method according to claim 14, wherein the model request includes a scenario identifier of a target fault analysis scenario; The step of determining a target fault analysis model from a plurality of fault analysis models based on the model request includes: Based on the scenario identifier of the target fault analysis scenario, a target fault analysis model adapted to the target fault analysis scenario is searched from a model library, wherein the model library stores a plurality of fault analysis models adapted to different fault analysis scenarios.

16. The method of claim 14, wherein the model request includes scenario input data of a target failure analysis scenario; The step of determining a target fault analysis model from a plurality of fault analysis models based on the model request includes: Determining, from the multiple fault analysis models, an initial fault analysis model adapted to the target fault analysis scenario; Based on the scenario input data of the target fault analysis scenario, the parameters of the initial fault analysis model are adjusted to obtain a target fault analysis model.

17. The method of claim 14, wherein the model request comprises model specification parameters; The step of determining a target fault analysis model from a plurality of fault analysis models based on the model request includes: Based on the model specification parameters, a corresponding target fault analysis model is searched from a model library, wherein the model library stores a plurality of fault analysis models with different model specification parameters.

18. A task platform, comprising a request interface and a response unit; The request interface is used to receive a model request sent by a terminal device, wherein the model request includes a scenario identifier of a target fault analysis scenario, scenario input data of the target fault analysis scenario, and at least one of a model specification parameter; The response unit is used to determine a target fault analysis model from multiple fault analysis models based on the model request, wherein the target fault analysis model is used in the execution process of the method according to any one of claims 1 to 11 or claim 12 or claim 13.

19. A computing device comprising: Memory and processor; The memory is used to store computer programs / instructions, and the processor is used to execute the computer programs / instructions. When the computer program / instructions are executed by the processor, the steps of the method described in any one of claims 1 to 17 are implemented.

20. An electronic device, comprising: A memory and a processor, wherein the memory and the processor are connected via a bus; The memory is used to store computer programs / instructions, and the processor is used to execute the computer programs / instructions. When the computer program / instructions are executed by the processor, the steps of the method described in any one of claims 1 to 17 are implemented.

21. A computer-readable storage medium storing a computer program / instruction, wherein the computer program / instruction, when executed by a processor, implements the steps of the method according to any one of claims 1 to 17.

22. A computer program product, comprising a computer program / instruction, which, when executed by a processor, implements the steps of the method according to any one of claims 1 to 17.

Images