Generator set fault report generation method and system based on tree structure

Abstract

The invention discloses a generator set fault report generation method and system based on a tree structure, and relates to the field of data analysis, and the method comprises the steps: collecting the time sequence operation data of a generator set according to a report generation instruction, and enabling the time sequence operation data to be connected with the report generation instruction, so as to obtain a first fault data set; obtaining six types of fault sub-data sets; and a tree structure fault data set is constructed based on the six types of fault sub-data sets, and the tree structure fault data set is used for fault analysis so as to feed back a generator set fault report to a terminal of the report generation instruction. Through the scheme of the invention, diversified and complicated fault conditions can be diagnosed and processed more effectively at the same time, and the accuracy and real-time performance of fault diagnosis are improved.

Description

Technical Field

[0001] The present application relates to the field of data analysis, and in particular to a method and system for generating a generator set fault report based on a tree structure. Background Art

[0002] In modern industrial production, generator sets are important energy supply equipment, and their stable operation is crucial to ensuring the continuity and safety of the entire production process. However, since generator sets may be subject to various external and internal influences during operation, failures are inevitable. In this context, effectively generating generator set fault reports can not only detect and diagnose problems in a timely manner, but also play an irreplaceable role in preventing future failures and guiding maintenance personnel to perform precise maintenance. Therefore, the generation method of generator set fault reports is particularly important. At present, the commonly used methods for generating generator set fault reports include manual inspection and automated monitoring systems. Manual inspection usually relies on the experience and intuition of engineers or technicians, while automated monitoring systems collect real-time operating data of generator sets through sensors and data acquisition equipment, such as temperature, pressure, current and other parameters, and judge the abnormal state of the equipment through preset logic or simple algorithms. However, these methods have many shortcomings. Manual inspection is not only inefficient, but also easily affected by personal subjective judgment; although the automated system improves the detection efficiency, it can usually only judge faults based on preset fixed parameters, which is difficult to adapt to complex and changeable fault conditions, and may lead to inaccurate fault diagnosis.

[0003] In view of the limitations of existing methods, the industry has begun to explore more efficient and intelligent solutions. For example, fault diagnosis technology based on machine learning can learn fault patterns by analyzing historical data, so that it can make accurate judgments when encountering unknown faults. In addition, fault prediction technology based on big data analysis can integrate multiple operating parameters and environmental factors, and predict the possibility of faults through complex data analysis models, thereby achieving preventive maintenance. However, these methods still have problems such as strong data dependence, high model training costs, and lack of real-time performance.

[0004] Therefore, there is an urgent need for a technical solution that can more effectively diagnose and handle diverse and complex fault conditions at the same time and improve the accuracy and real-time performance of fault diagnosis. Summary of the invention

[0005] In order to solve the deficiencies of the prior art, the embodiment of the present application provides a method and system for generating a fault report of a generator set based on a tree structure. The present application solves the technical problems that the prior art has poor ability to handle diversified and complex fault conditions, and the accuracy and real-time performance of fault diagnosis are insufficient.

[0006] An embodiment of the present application provides a method for generating a generator set fault report based on a tree structure, comprising: collecting the timing operation data of the generator set according to a report generation instruction, the timing operation data being used to connect the report generation instruction to obtain a first fault data set; obtaining a target fault unit number according to an output result of the first fault classification operation of the first fault data set, the target fault unit number being used to connect the first fault data set to obtain a second fault data set; performing a second fault classification operation on the second fault data set to obtain a target fault generator number; acquiring a historical fault data set, and filtering the historical fault data set using the target fault generator number to obtain a third fault data set; performing a third fault classification operation on the third fault data set to obtain six categories of fault sub-data sets; and constructing a tree-structured fault data set based on the six categories of fault sub-data sets, the tree-structured fault data set being used to perform fault analysis to feed back the generator set fault report to the terminal of the report generation instruction.

[0007] In a possible implementation, before connecting the report generation instruction, the method further includes: converting the time series operation data from the time domain space to the frequency domain space; wherein the conversion method includes: Where X[k] represents the kth frequency component in the frequency domain, x[n] represents the nth sample in the time domain, and N represents the total number of samples.

[0008] In a possible implementation, a historical fault data set is obtained, and the historical fault data set is filtered using the target fault generator number to obtain a third fault data set, including: when the target fault generator number does not exist in the historical fault data set, deleting the data in the historical fault data set where the unit number of the generator is different from the unit number of the target fault generator to obtain a retained data set, otherwise deleting the data in the historical fault data set where the generator number is different from the target fault generator number to obtain a retained data set; and, fault phenomenon identification is performed on the second fault data set, and the result of the fault phenomenon identification is used to be spliced ​​with the retained data set to obtain a third fault data set.

[0009] In a possible implementation, the fault sub-dataset categories include: winding fault sub-dataset, magnetic field fault sub-dataset, insulation fault sub-dataset, bearing fault sub-dataset, cooling system fault sub-dataset and control system fault sub-dataset; wherein the faults in the winding fault sub-dataset include winding short circuit, open circuit or grounding; the faults in the magnetic field fault sub-dataset include magnetic field coil damage or insufficient magnetic field excitation; the faults in the insulation fault sub-dataset include insulation aging or insulation breakdown; the faults in the bearing fault sub-dataset include bearing damage, overheating or poor lubrication; the faults in the cooling system fault sub-dataset include cooling water leakage or cooling fan failure; the faults in the control system fault sub-dataset include voltage regulator failure or overload protection failure.

[0010] In a possible implementation, a tree-structured fault data set is constructed based on the six categories of fault sub-data sets, and the tree-structured fault data set is used for fault analysis to feed back the generator set fault report to the terminal of the report generation instruction, including: setting the target faulty generator as a root node, and setting the six categories of fault sub-data sets as six child nodes respectively to construct a first tree-structured fault data set; wherein the relationship between each child node is a brother relationship, and the relationship between each child node and the parent node is a parent-child relationship; and, using a fault probability model to obtain the probability value of each child node in turn to perform fault analysis.

[0011] In a possible implementation, a fault probability model is used to obtain the probability value of each child node in turn to perform fault analysis, including: reordering each child node in a preset order according to the corresponding probability value; comparing the reordered child nodes with the target threshold in turn, and when the child node is less than the target threshold, deleting the current child node and the child node that follows the current child node in the sequence to obtain a second tree-structured fault data set; and, based on the probability values ​​and Gaussian distribution of the child nodes in the second tree-structured fault data set, obtaining the root node fault probability value.

[0012] In a possible implementation, obtaining the root node failure probability value based on the probability values ​​of the child nodes in the second tree structure fault data set and the Gaussian distribution includes: Among them, P root represents the failure probability of the root node, P i represents the failure probability of a child node, w i represents the influence weight of each child node on the failure probability of the root node, μ represents the mean of the Gaussian distribution, σ represents the standard deviation, and erf represents the error function, which is used to calculate the cumulative probability of the corresponding Gaussian distribution.

[0013] In a possible implementation, after obtaining the root node fault probability value, it also includes: filling the target fault generator number of the root node, the root node fault probability value, the category of the child node in the second tree structure fault data set and the corresponding probability value into the prompt word template, and the prompt word template is used to input the large language model to obtain the generator set fault report that is fed back to the terminal of the report generation instruction.

[0014] The embodiment of the present application also provides a system for generating a generator set fault report based on a tree structure, comprising: an instruction unit, a collection unit and an analysis unit; wherein the instruction unit is used to input a report generation instruction; the collection unit is used to collect the timing operation data of the generator set according to the report generation instruction, and the timing operation data is used to connect the report generation instruction to obtain a first fault data set; the analysis unit is used to obtain a target fault unit number according to the output result of the first fault data set in the first fault classification operation, and the target fault unit number is used to connect the first fault data set to obtain a second fault data set; perform a second fault classification operation on the second fault data set to obtain a target fault generator number; obtain a historical fault data set, and use the target fault generator number to filter the historical fault data set to obtain a third fault data set; perform a third fault classification operation on the third fault data set to obtain six categories of fault sub-data sets; and, based on the six categories of fault sub-data sets, construct a tree-structured fault data set, and the tree-structured fault data set is used to perform fault analysis to feed back the generator set fault report to the terminal of the report generation instruction.

[0015] An embodiment of the present application also provides a device for generating a generator set fault report based on a tree structure, comprising: a processor, a memory, and a system bus; wherein the processor and the memory are connected via the system bus; the memory is used to store one or more programs, and the one or more programs include instructions, and when the instructions are executed by the processor, the processor executes the generation method described in the above embodiment.

[0016] In the method, system and device for generating a generator set fault report based on a tree structure as provided above, the embodiment of the present application constructs a tree-like logical structure to hierarchically classify the operating status and possible fault factors of the generator set, and each node represents a specific inspection or judgment step. Through a step-by-step refinement method, not only can the cause of the fault be accurately located, but also corresponding processing suggestions can be provided according to the specific circumstances of the fault. Compared with the traditional linear diagnostic process, this method is more in line with the actual needs of complex system fault diagnosis and can more effectively handle diversified and complex fault conditions. In addition, due to its highly modular and scalable characteristics, the tree-based method can easily integrate new detection technologies and fault data, and continuously optimize and update the fault judgment logic. This not only greatly improves the accuracy and real-time performance of fault diagnosis, but also reduces the dependence on professionals and improves the automation level of the system. BRIEF DESCRIPTION OF THE DRAWINGS

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

[0018] Figure 1 A schematic diagram of a flow chart of a method for generating a generator set fault report based on a tree structure provided in an embodiment of the present application;

[0019] Figure 2 A schematic diagram of a system for generating a generator set fault report based on a tree structure provided in an embodiment of the present application. DETAILED DESCRIPTION

[0020] Various exemplary embodiments of the present application will now be described in detail with reference to the accompanying drawings. It should be noted that unless otherwise specifically stated, the relative arrangement of components and steps, numerical expressions and numerical values ​​set forth in these embodiments do not limit the scope of the present application.

[0021] Those skilled in the art will appreciate that the terms "first", "second" and the like in the embodiments of the present application are only used to distinguish different steps, devices or modules, etc., and do not represent any specific technical meaning, nor do they represent the necessary logical order between them. It should also be understood that in the embodiments of the present application, "multiple" may refer to two or more, and "at least one" may refer to one, two or more. It should also be understood that for any component, data or structure mentioned in the embodiments of the present application, in the absence of explicit limitation or contrary revelation given in the context, it can generally be understood as one or more. In addition, the term "and / or" in the present application is only a kind of association relationship describing the associated objects, indicating that there can be three kinds of relationships, for example, A and / or B, which can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in the present application generally indicates that the associated objects before and after are an "or" relationship. It should also be understood that the description of each embodiment in the present application emphasizes the differences between the embodiments, and the same or similar parts can refer to each other. For the sake of brevity, they will not be repeated one by one.

[0022] At the same time, it should be understood that, for ease of description, the sizes of the various parts shown in the drawings are not drawn according to the actual proportional relationship. The following description of at least one exemplary embodiment is actually only illustrative and is by no means intended to limit the present application and its application or use. The techniques, methods and devices known to ordinary technicians in the relevant fields may not be discussed in detail, but where appropriate, the techniques, methods and devices should be considered as part of the specification. It should be noted that similar numbers and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further discussed in subsequent drawings.

[0023] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0024] Figure 1 The flowchart of the method for generating a generator set fault report based on a tree structure provided in an embodiment of the present application is as follows. Figure 1As shown, at step S101, according to the report generation instruction, the time series operation data of the generator set is collected, and the time series operation data is used to connect the report generation instruction to obtain a first fault data set. It should be understood that when reporting and analyzing the fault of the generator set, it is usually necessary to collect and process a large amount of time series operation data. These time series operation data reflect the various operating states and parameter changes of the generator set over a period of time. However, the original time series data usually exists in the time domain space, that is, in the form of time as the horizontal coordinate and the measured value as the vertical coordinate. Converting the time series data from the time domain space to the frequency domain space can provide a more comprehensive and in-depth perspective for fault analysis.

[0025] Frequency domain analysis is a commonly used signal processing technique that reveals the energy distribution of signals at different frequency components by converting time domain signals into frequency domain representation. In generator set fault analysis, frequency domain analysis can help identify potential fault sources such as abnormal vibration and electromagnetic noise, and provide fault feature values ​​for subsequent fault identification and classification.

[0026] In an embodiment of the present application, a discrete Fourier transform (DFT) is used to convert time domain data to frequency domain. Before connecting the report generation instruction, it also includes: converting the time series operation data from the time domain space to the frequency domain space; wherein the conversion method includes:

[0027]

[0028] Among them, X[k] represents the kth frequency component in the frequency domain, x[n] represents the nth sample in the time domain, and N represents the total number of samples. It is worth noting that before performing discrete Fourier transform, the original time series data usually needs to be preprocessed, such as removing the DC component, applying a window function, etc., to reduce spectrum leakage and improve spectrum resolution. At the same time, due to the large amount of calculation of discrete Fourier transform, the fast Fourier transform (FFT) algorithm is usually used in practical applications to speed up the calculation.

[0029] After the conversion from time domain to frequency domain, the frequency domain characteristic data of the generator set during operation will be obtained. Next, by connecting the report generation instruction with these frequency domain characteristic data, the first fault data set can be obtained. The report generation instruction can include information such as the template, requirements, and format of the fault report, which is combined with the frequency domain characteristic data to facilitate subsequent fault identification and report generation.

[0030] At step S102, a target faulty unit number is obtained according to the output result of the first fault classification operation of the first fault data set, and the target faulty unit number is used to connect the first fault data set to obtain a second fault data set. It should be understood that the faulty unit number generally refers to a code or identifier used to uniquely identify a unit. A unit refers to a system or device composed of multiple components or devices, such as a generator set, a refrigeration unit, etc. Each faulty unit will be assigned a unique number for tracking and managing related fault data.

[0031] The first fault data set refers to the original fault data collected initially. These data usually come from various sensors, monitoring systems or manual records, reflecting the operating status and fault information of the unit at a specific time point or time range. The first fault data set may include the measured values ​​of parameters such as temperature, pressure, vibration, current, and related fault codes, fault descriptions and other information.

[0032] In the first fault classification operation, the first fault data set is usually analyzed and processed using machine learning or artificial intelligence algorithms. The purpose is to identify potential fault modes, fault types or fault root causes. The output of this operation may be a classification label for each fault instance, a fault probability distribution or other forms of analysis results.

[0033] According to the output result of the first fault classification operation, it can be determined which fault instances belong to the same target fault unit. For example, if multiple fault instances are classified as the same fault type and these instances are from the same unit, then their unit numbers can be determined as the target fault unit numbers.

[0034] The role of the target fault unit number is to connect the fault data belonging to the same unit in the first fault data set to form the second fault data set. The purpose of this step is to aggregate and organize related fault data to provide more complete and centralized data support for subsequent fault diagnosis, cause analysis and maintenance decisions.

[0035] In the second fault data set, fault data generated by different time points, different sensors or different components are all associated with the same unit number. This makes it easy to track the overall fault history of the unit, analyze the evolution of the fault, and perform more comprehensive fault mode identification and fault prediction. At the same time, the second fault data set also provides important data basis for the preventive maintenance, fault diagnosis and repair decision-making of the unit.

[0036] It is worth noting that in actual applications, the format, data type and storage method of the first fault data set and the second fault data set may be different. Some systems may directly use the unit number as part of the first fault data set, while some systems require additional data processing steps to associate the unit number.

[0037] In step S103, a second fault classification operation is performed on the second fault data set to obtain a target fault generator number.

[0038] At step S104, a historical fault data set is obtained, and the historical fault data set is filtered using the target fault generator number to obtain a third fault data set. In one embodiment, it may include: when the target fault generator number does not exist in the historical fault data set, deleting the data whose generator unit number in the historical fault data set is different from the unit number of the target fault generator to obtain a retained data set, otherwise deleting the data whose generator number in the historical fault data set is different from the target fault generator number to obtain a retained data set; and, performing fault phenomenon identification on the second fault data set, and the result of the fault phenomenon identification is used to splice with the retained data set to obtain a third fault data set.

[0039] At step S105, the third fault classification operation is performed on the third fault data set to obtain six categories of fault sub-data sets. The fault sub-data set categories include: winding fault sub-data set, magnetic field fault sub-data set, insulation fault sub-data set, bearing fault sub-data set, cooling system fault sub-data set and control system fault sub-data set; wherein the faults in the winding fault sub-data set include winding short circuit, open circuit or grounding; the faults in the magnetic field fault sub-data set include magnetic field coil damage or insufficient magnetic field excitation; the faults in the insulation fault sub-data set include insulation aging or insulation breakdown; the faults in the bearing fault sub-data set include bearing damage, overheating or poor lubrication; the faults in the cooling system fault sub-data set include cooling water leakage or cooling fan failure; the faults in the control system fault sub-data set include voltage regulator failure or overload protection failure.

[0040] In step S106, a tree-structured fault data set is constructed based on the six categories of fault sub-data sets, and the tree-structured fault data set is used for fault analysis to feed back the generator set fault report to the terminal of the report generation instruction.

[0041] In one embodiment, the target faulty generator is set as the root node, and the six categories of fault sub-datasets are correspondingly set as six child nodes to construct a first tree-structured fault data set; wherein the relationship between each child node is a brother relationship, and the relationship between each child node and the parent node is a parent-child relationship; and, the fault probability model is used to sequentially obtain the probability value of each child node to perform fault analysis. Specifically, each child node is reordered in a preset order according to the corresponding probability value; the reordered child nodes are sequentially compared with the target threshold, and when the child node is less than the target threshold, the current child node and the child nodes after the current child node in the sequence are deleted to obtain a second tree-structured fault data set; and, based on the probability values ​​and Gaussian distribution of the child nodes in the second tree-structured fault data set, the root node fault probability value is obtained. Further, it may include:

[0042]

[0043] Among them, P root represents the failure probability of the root node, P i represents the failure probability of a child node, w i represents the influence weight of each child node on the failure probability of the root node, μ represents the mean of the Gaussian distribution, σ represents the standard deviation, and erf represents the error function, which is used to calculate the cumulative probability of the corresponding Gaussian distribution.

[0044] In addition, after obtaining the root node fault probability value, it also includes: filling the target fault generator number of the root node, the root node fault probability value, the category of the child node in the second tree structure fault data set and the corresponding probability value into the prompt word template, and the prompt word template is used to input the large language model to obtain the generator set fault report that is fed back to the terminal of the report generation instruction.

[0045] Figure 2 A schematic diagram of a system for generating a fault report of a generator set based on a tree structure provided in an embodiment of the present application. It should be understood that the system shown in the figure is exemplary and not restrictive. This means that the system architecture involved is not limited to a specific form or design, but is presented as an example. In other words, the architecture shown in the figure can be regarded as a way of expression to clearly describe related concepts and relationships, and does not exclude other forms of architecture. Therefore, when interpreting the architecture in the picture, it should be understood that the model is flexible and diverse, and its purpose is to provide an exemplary description rather than a restrictive provision on a specific form.

[0046] Specifically, in an embodiment of the present application, a knowledge graph-driven large-model business intelligence decision-making question-answering system includes: an instruction unit, a collection unit and an analysis unit; wherein the instruction unit is used to input a report generation instruction; the collection unit is used to collect the timing operation data of the generator set according to the report generation instruction, and the timing operation data is used to connect the report generation instruction to obtain a first fault data set; the analysis unit is used to obtain a target fault unit number according to the output result of the first fault data set in the first fault classification operation, and the target fault unit number is used to connect the first fault data set to obtain a second fault data set; perform a second fault classification operation on the second fault data set to obtain a target fault generator number; obtain a historical fault data set, and use the target fault generator number to filter the historical fault data set to obtain a third fault data set; perform a third fault classification operation on the third fault data set to obtain six categories of fault sub-data sets; and, based on the six categories of fault sub-data sets, construct a tree-structured fault data set, and the tree-structured fault data set is used to perform fault analysis to feed back the generator set fault report to the terminal of the report generation instruction.

[0047] Furthermore, an embodiment of the present application also provides a device for generating a generator set fault report based on a tree structure, comprising: a processor, a memory, and a system bus; the processor and the memory are connected via the system bus; the memory is used to store one or more programs, and the one or more programs include instructions, which, when executed by the processor, enable the processor to execute any one of the above-mentioned generation methods.

[0048] Furthermore, an embodiment of the present application also provides a computer program product, which, when executed on a terminal device, enables the terminal device to execute any one of the above-mentioned generation methods.

[0049] It can be known from the description of the above implementation mode that those skilled in the art can clearly understand that all or part of the steps in the above-mentioned embodiment method can be implemented by means of software plus a necessary general hardware platform. Based on such an understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product can be stored in a storage medium such as ROM / RAM, a disk, an optical disk, etc., including several instructions for enabling a computer device (which can be a personal computer, a server, or a network communication device such as a media gateway, etc.) to execute the methods described in the various embodiments of the present application or certain parts of the embodiments.

[0050] It should be noted that the various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments, and the same or similar parts between the various embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part description.

[0051] It should also be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0052] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for generating a generator set fault report based on a tree structure, characterized in that: include: According to the report generation instruction, collecting the time sequence operation data of the generator set, wherein the time sequence operation data is used to connect the report generation instruction to obtain a first fault data set; According to the output result of the first fault classification operation of the first fault data set, a target fault unit number is obtained, and the target fault unit number is used to connect the first fault data set to obtain a second fault data set; performing a second fault classification operation on the second fault data set to obtain a target fault generator number; Acquire a historical fault data set, and filter the historical fault data set using the target faulty generator number to obtain a third fault data set; performing a third fault classification operation on the third fault data set to obtain six types of fault sub-data sets; as well as, A tree-structured fault data set is constructed based on the six categories of fault sub-data sets, and the tree-structured fault data set is used for fault analysis to feed back the generator set fault report to the terminal of the report generation instruction.

2. The generation method according to claim 1, characterized in that: in, Before connecting the report generation instruction, it also includes: The time series operation data is converted from the time domain space to the frequency domain space; wherein the conversion method includes: Where X[k] represents the kth frequency component in the frequency domain, x[n] represents the nth sample in the time domain, and N represents the total number of samples.

3. The generation method according to claim 1, characterized in that: The process of obtaining a historical fault data set and filtering the historical fault data set using the target fault generator number to obtain a third fault data set includes: When the target fault generator number does not exist in the historical fault data set, the data whose generator unit number in the historical fault data set is different from the unit number of the target fault generator is deleted to obtain a retained data set, otherwise the data whose generator number in the historical fault data set is different from the target fault generator number is deleted to obtain a retained data set; and, Fault phenomenon identification is performed on the second fault data set, and the result of the fault phenomenon identification is used to be spliced ​​with the reserved data set to obtain a third fault data set.

4. The generation method according to claim 3, characterized in that: The fault sub-dataset categories include: winding fault sub-dataset, magnetic field fault sub-dataset, insulation fault sub-dataset, bearing fault sub-dataset, cooling system fault sub-dataset and control system fault sub-dataset; The faults in the winding fault sub-data set include winding short circuit, open circuit or grounding; The faults in the magnetic field fault sub-data set include damage to the magnetic field coil or insufficient magnetic field excitation; The faults in the insulation fault sub-data set include insulation aging or insulation breakdown; The faults in the bearing fault sub-dataset include bearing damage, overheating or poor lubrication; The faults in the cooling system fault sub-dataset include cooling water leakage or cooling fan failure; The faults in the control system fault sub-data set include voltage regulator fault or overload protection failure.

5. The generation method according to claim 4, characterized in that: The tree-structured fault data set is constructed based on the six types of fault sub-data sets, and the tree-structured fault data set is used for fault analysis to feed back the generator set fault report to the terminal of the report generation instruction, including: The target faulty generator is set as the root node, and the six categories of faulty sub-datasets are correspondingly set as six child nodes to construct a first tree-structured fault data set; wherein the relationship between each child node is a sibling relationship, and the relationship between each child node and the parent node is a parent-child relationship; and, The fault probability model is used to obtain the probability value of each child node in turn for fault analysis.

6. The generation method according to claim 5, characterized in that: The fault probability model is used to obtain the probability value of each child node in turn to perform fault analysis, including: Rearrange each child node in a preset order according to the corresponding probability value; Compare the reordered child nodes with the target threshold in sequence, and when the child node is less than the target threshold, delete the current child node and the child nodes after the current child node in the sequence to obtain a second tree structure fault data set; and, Based on the probability values ​​of the child nodes in the second tree structure fault data set and the Gaussian distribution, a root node fault probability value is obtained.

7. The generation method according to claim 6, characterized in that: Wherein, based on the probability values ​​and Gaussian distribution of the child nodes in the second tree structure fault data set, obtaining the root node fault probability value includes: Among them, P root represents the failure probability of the root node, P i represents the failure probability of a child node, w i represents the influence weight of each child node on the failure probability of the root node, μ represents the mean of the Gaussian distribution, σ represents the standard deviation, and erf represents the error function, which is used to calculate the cumulative probability of the corresponding Gaussian distribution.

8. The generation method according to claim 6, characterized in that: in, After obtaining the root node failure probability value, it also includes: The target faulty generator number of the root node, the root node fault probability value, the category of the child node in the second tree structure fault data set and the corresponding probability value are filled into the prompt word template, and the prompt word template is used to input the large language model to obtain the generator set fault report that is fed back to the terminal of the report generation instruction.

9. A system for generating fault reports of generator sets based on a tree structure, characterized in that: include: Instruction unit, collection unit and analysis unit; wherein, The instruction unit is used to input report generation instructions; The collection unit is used to collect the time sequence operation data of the generator set according to the report generation instruction, and the time sequence operation data is used to connect the report generation instruction to obtain a first fault data set; The analysis unit is used to obtain a target faulty unit number according to an output result of a first fault classification operation on the first fault data set, and the target faulty unit number is used to connect the first fault data set to obtain a second fault data set; perform a second fault classification operation on the second fault data set to obtain a target faulty generator number; obtain a historical fault data set, and use the target faulty generator number to filter the historical fault data set to obtain a third fault data set; perform a third fault classification operation on the third fault data set to obtain six categories of fault sub-data sets; and construct a tree-structured fault data set based on the six categories of fault sub-data sets, and the tree-structured fault data set is used to perform fault analysis to feed back a generator set fault report to the terminal of the report generation instruction.

10. A device for generating a fault report of a generator set based on a tree structure, characterized in that: include: A processor, a memory, and a system bus; wherein the processor and the memory are connected via the system bus; The memory is used to store one or more programs, and the one or more programs include instructions, and when the instructions are executed by the processor, the processor executes the generation method according to any one of claims 1 to 8.

Images