Multi-criterion-based transformer fault diagnosis method and related device

Abstract

The invention discloses a multi-criterion-based transformer fault diagnosis method and a related device, and belongs to the technical field of power equipment fault diagnosis, and the method comprises the steps: obtaining the vibration signal time sequence data of a to-be-diagnosed transformer, and carrying out the fault diagnosis of the to-be-diagnosed transformer based on the vibration signal time sequence data and a pre-trained vibration signal fault diagnosis model, obtaining a first diagnosis result of the to-be-diagnosed transformer; based on the first diagnosis result, obtaining latest data of dissolved gas in oil during operation of the to-be-diagnosed transformer, inputting the latest data of dissolved gas in oil during operation of the to-be-diagnosed transformer into a pre-trained oil chromatography fault diagnosis model, and obtaining a second diagnosis result of the to-be-diagnosed transformer; and obtaining a final fault diagnosis result according to the first diagnosis result and the second diagnosis result. The technical problems of single diagnosis criterion and easy missed judgment and misjudgment at present are solved.

Description

Technical Field

[0001] The invention belongs to the technical field of power equipment fault diagnosis, and in particular relates to a transformer fault diagnosis method based on multiple criteria and a related device. Background Art

[0002] The power transformer is one of the key equipment in the power system. Its operating condition will directly affect the safe operation of the power system. Once a fault occurs, it will have a significant impact and harm to the power system and end users. Therefore, studying transformer fault diagnosis technology has important practical significance for the safe operation of the power system.

[0003] Due to the complex operating environment of transformers, existing monitoring devices are usually special monitoring devices that can only monitor gas in transformer oil, partial discharge and vibration. In addition, as the complexity of transformers continues to increase, it is not sufficient to rely solely on direct observation and single state quantity analysis to estimate the operating status of the transformer and diagnose faults. It is easy to miss or misjudge, and it is impossible to achieve intelligent state analysis.

[0004] Currently, there is no method that can achieve all-round and multi-dimensional condition monitoring of transformers by only one means.

[0005] For a long time, transformer fault diagnosis at home and abroad has been mainly achieved through regular maintenance. Regular maintenance means that the transformer is overhauled according to the maintenance items and maintenance cycles specified in the "Guidelines for Maintenance of Power Transformers" (DL / 573-2010). The disadvantages of regular maintenance are also obvious, that is, the operation mode and health status of the transformer are not taken into account. In addition, the conventional transformer insulating oil test items and cycles still have great limitations. The various transformer insulating oil tests required by the regulations cannot timely and accurately detect the signs of power equipment accidents.

[0006] It is worth noting that when oil-filled equipment such as transformers fails, it is often accompanied by abnormal insulation chromatography data and abnormal vibration or abnormal noise of the equipment body. Practice has shown that when oil-filled equipment such as transformers fails, it will cause changes in its internal temperature, pressure and gas composition, and produce various abnormal noise problems.

[0007] Existing transformer oil chromatography monitoring is divided into online and offline. The monitoring frequency of offline and online data is often several months, so faults cannot be discovered in time, which may affect equipment safety and the normal operation of the power system. However, if the return time of oil chromatography data is shortened, due to the lack of stability of online monitoring equipment data and the accuracy of diagnostic methods, false alarms will occur, resulting in unnecessary shutdown and maintenance; if offline detection is used, a lot of manpower and material resources will be wasted. At the same time, a single oil chromatography data cannot fully reflect the status of the transformer.

[0008] Therefore, a new fault diagnosis method is needed to realize the diagnosis and analysis of the operating status of transformer equipment. Summary of the invention

[0009] The purpose of the present invention is to provide a transformer fault diagnosis method and related devices based on multiple criteria to solve the technical problem that the current diagnosis method cannot balance the timely detection of faults and the missed and misjudged faults.

[0010] In order to achieve the above object, the present invention adopts the following technical solution: In a first aspect, the present invention provides a transformer fault diagnosis method based on multiple criteria, comprising: Obtaining vibration signal time series data of the transformer to be diagnosed; After processing the vibration signal time series data of the transformer to be diagnosed, the data is input into a pre-trained vibration signal fault diagnosis model to obtain a first diagnosis result of the transformer to be diagnosed; If the vibration signal is detected as no fault, the latest data of dissolved gas in oil of the transformer to be diagnosed is obtained, and the latest data of dissolved gas in oil of the transformer to be diagnosed is input into the pre-trained oil chromatography fault diagnosis model to obtain the second diagnosis result of the transformer to be diagnosed; If the vibration signal is detected as a fault state, the frequency of collecting dissolved gas data in oil is increased, and the latest dissolved gas data in oil of the transformer to be diagnosed is input into the pre-trained oil chromatography fault diagnosis model to obtain the second diagnosis result of the transformer to be diagnosed; A final fault diagnosis result is obtained according to the first diagnosis result and the second diagnosis result.

[0011] The present invention is further improved in that: After processing the vibration signal time series data of the transformer to be diagnosed, inputting the pre-trained vibration signal fault diagnosis model, obtaining the first diagnosis result of the transformer to be diagnosed includes: Perform outlier detection on the transformer vibration signal time series data and remove outliers; Converting the vibration signal time series data after removing abnormal values ​​into two-dimensional image data; The two-dimensional image data is input into a pre-trained vibration signal fault diagnosis model to obtain the fault property, and the vibration signal fault diagnosis model is a convolutional neural network.

[0012] The present invention is further improved in that: The interquartile range method is used to detect outliers in the time series data of transformer vibration signals.

[0013] The present invention is further improved in that: The pre-trained vibration signal fault diagnosis model is obtained by training in the following steps: Build a digital prototype of the transformer; The vibration signals of the transformer digital prototype in normal state and under different types of faults are collected. The vibration signals are taken as input and the corresponding fault types are taken as output. The convolutional neural network is trained to obtain the vibration signal fault diagnosis model.

[0014] The present invention is further improved in that: The step of inputting the latest data of dissolved gas in oil of the transformer to be diagnosed into the pre-trained oil chromatography fault diagnosis model to obtain the second diagnosis result of the transformer to be diagnosed comprises: Performing ratio calculation on the latest data of dissolved gas in oil of the transformer to be diagnosed to expand the data, obtaining characteristic data, and extracting the main components of the characteristic data; The principal component is input into a pre-trained oil chromatography fault diagnosis model to obtain the fault property.

[0015] The present invention is further improved in that: The pre-trained oil chromatography fault diagnosis model is obtained by the following steps: Constructing a PCA-ANN network model, wherein the PCA-ANN network model includes an input layer, a hidden layer and an output layer; Select transformer fault gas data, expand the transformer fault gas data to obtain feature data, and extract the main components of the feature data; The above is used as the input of the transformer fault diagnosis neural network model, and the fault properties corresponding to the main components of the transformer fault gas data are used as labels to train the transformer fault diagnosis neural network model to obtain a pre-trained oil chromatography fault diagnosis model.

[0016] The present invention is further improved in that: The obtaining of a final fault diagnosis result according to the first diagnosis result and the second diagnosis result comprises: If the first diagnostic result is normal and the second diagnostic result is output as normal, it is determined that the current state of the transformer is normal; If the first diagnosis result is a fault and the second diagnosis result is a fault, the transformer is judged to be in an abnormal state, the nature of the fault is determined by the first diagnosis result, and the type of the fault is determined by the second diagnosis result; If the first diagnosis result and the second diagnosis result are inconsistent, manual analysis and diagnosis are performed.

[0017] In a second aspect, the present invention provides a transformer fault diagnosis device based on multiple criteria, comprising: An acquisition module, used to acquire the vibration signal time series data of the transformer to be diagnosed and the latest dissolved gas in oil data of the transformer to be diagnosed; A vibration signal diagnosis module, used for inputting the vibration signal time series data of the transformer to be diagnosed into a pre-trained vibration signal fault diagnosis model to obtain a first diagnosis result of the transformer to be diagnosed; An oil chromatography diagnosis module, used for inputting the latest data of dissolved gas in oil of the transformer to be diagnosed into a pre-trained oil chromatography fault diagnosis model to obtain a second diagnosis result of the transformer to be diagnosed; The comprehensive diagnosis module is used to compare the first diagnosis result and the second diagnosis result, verify and supplement each other, and obtain the final fault diagnosis result.

[0018] In a third aspect, the present invention provides an electronic device, comprising: at least one processor; and, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute a transformer fault diagnosis method based on multiple criteria as described in any one of the first aspects of the present invention.

[0019] In a fourth aspect, the present invention provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements a transformer fault diagnosis method based on multiple criteria as described in any one of the first aspects of the present invention.

[0020] Compared with the prior art, the present invention has at least the following beneficial technical effects: Most of the existing transformer diagnosis methods are to perform fault diagnosis by collecting oil chromatography data offline or online, and the diagnosis basis is single. In addition, the offline data collection process is cumbersome, and the return frequency of online monitoring data is only about a few days, which makes it impossible to detect faults in time. In addition, online data has the problem of insufficient accuracy and stability. The present invention combines oil chromatography data diagnosis with vibration diagnosis, adds vibration data as a criterion on the original basis, and improves the reliability of fault diagnosis; and performs status diagnosis through real-time vibration data, and adjusts the return frequency of oil chromatography data in time according to the diagnosis results, ensuring timely detection of faults while avoiding unnecessary shutdowns for maintenance, saving manpower and material resources.

[0021] In addition, the present invention uses a digital prototype to collect fault samples, uses COMSOL finite element analysis software as an analysis tool, establishes a transformer digital prototype, realizes the simulation of vibration characteristics under different states (faults), studies the changes in vibration signals of transformers under different operating conditions, and provides support for the database of fault models. The use of this method to collect fault samples can simulate and analyze a variety of fault conditions in a short time without waiting for the manufacturing and testing process of the physical prototype, greatly reducing time and economic costs. The fault samples collected using the digital prototype have higher accuracy, and are used as training samples for model training, which enhances the accuracy of fault diagnosis model analysis.

[0022] The present invention performs transformer status analysis and fault diagnosis through fusion analysis of oil chromatography data and vibration data, which has high engineering application value. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 is a flow chart of a transformer fault diagnosis method; Figure 2 A structural block diagram of a transformer fault diagnosis device based on multiple criteria provided by an embodiment of the present invention; Figure 3 A schematic diagram of a module of a transformer fault diagnosis device based on multiple criteria provided by an embodiment of the present invention; Figure 4 A block diagram of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0024] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0025] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0026] The present invention is further described in detail below in conjunction with the accompanying drawings: Example 1 Reference Figure 1 This embodiment provides a transformer fault diagnosis method based on multiple criteria, comprising the following steps: Among them, the multi-criteria fault diagnosis model is mainly a multi-criteria model in which the vibration fault diagnosis model and the oil chromatography fault diagnosis model verify and complement each other.

[0027] S1. Obtain the time series data of the vibration signal of the transformer as the input sample of the vibration signal fault diagnosis model, perform fault diagnosis based on the convolutional neural network based on the collected vibration data, and detect the fault type of the transformer.

[0028] Among them, the fault types include: no fault, loose core clamp and winding clamp fault, core multi-point grounding fault, winding deformation fault, metal spike partial discharge fault, overexcitation fault and overload fault.

[0029] S2. If the vibration signal is detected as normal, the latest data of dissolved gas in oil of the transformer to be diagnosed is obtained as the input sample of the oil chromatography fault diagnosis model, and data-driven fault diagnosis is also performed on the collected data of dissolved gas in oil to detect the relevant fault properties; If the vibration signal is detected as a fault state, a signal is sent to increase the oil chromatogram acquisition frequency and shorten the cycle to 1 hour, so that the oil chromatogram data can be transmitted back in time and the transformer fault diagnosis can be performed based on the oil chromatogram data.

[0030] S3. Compare the diagnosis results of the vibration fault diagnosis model and the oil chromatography fault diagnosis model, verify and complement each other, and obtain the final result.

[0031] In some embodiments, step S1 is divided into the following steps: S101. Data cleaning: use the interquartile range (IQR) method to detect outliers on the original input vibration signal time series data and remove outliers: IQR method: Calculate the quartiles of the vibration signal time series data (the first quartile Q1 and the third quartile Q3), and determine the interquartile range IQR, IQR = Q3 - Q1. Set the data range to [Q1 - 1.5IQR, Q3 +1.5IQR]. The vibration signal time series data outside this range is considered abnormal data. Remove the abnormal data to obtain the cleaned data.

[0032] S102, data processing, converting the cleaned one-dimensional vibration signal time series data into two-dimensional image data using Gram angle field, wavelet analysis or Markov transformation field.

[0033] S103, inputting the two-dimensional image data into a vibration signal fault diagnosis model, where the vibration signal fault diagnosis model is a convolutional neural network.

[0034] In some embodiments, the data set for training the vibration signal fault diagnosis model comes from a transformer digital prototype.

[0035] Transformer digital prototype: Taking a typical three-phase oil-immersed transformer as an example, COMSOL finite element analysis software is used as an analysis tool to build a transformer digital prototype and conduct electromagnetic-structural-force multi-physics field coupling analysis. The rated capacity, winding form, transformation ratio, number of coil turns, etc. of the transformer can be set in the simulation according to the actual situation. Based on the typical oil-immersed transformer digital model, the transformer is simulated with no fault, loose core clamp and winding clamp fault tests, core multi-point grounding test, winding deformation test, metal spike partial discharge test, overexcitation and overload test. Vibration signals under different working conditions are obtained as training data sets for the model.

[0036] For S2, the fault diagnosis method based on oil chromatography data is a neural network-based fault diagnosis method, which takes the dissolved gas in the oil as input and the fault property as output to build a neural network model. The following is just an example of a specific method, and other methods in the industry can also be applied here.

[0037] S201: The collected data of dissolved gas in oil of the transformer to be diagnosed are expanded by ratio operation to obtain characteristic data, and then the characteristic data are processed by principal component analysis method to eliminate the correlation between vectors in the characteristic data, reduce the dimension of the input vector, and extract the principal components that affect the output prediction value; and the transformer fault diagnosis neural network model is constructed by using the fitting ability of the neural network to realize automatic fault diagnosis. The transformer fault diagnosis neural network model is Principal component analysis- Artificial neural network (Principal Component Analysis-Artificial Neural Network, PCA –ANN) network model; S202: Training of transformer fault diagnosis neural network model: The transformer fault diagnosis neural network model adopts a three-layer structure, that is, the first layer is the input layer, the second layer is the hidden layer, and the third layer is the output layer. The input of the input layer is the principal component that affects the transformer fault prediction value extracted by principal component analysis, and the output layer is the five fault properties of the transformer, namely no fault, medium and low temperature overheating, high temperature overheating, low energy discharge, and high energy discharge; the principal components (80% of the data, i.e., training data) of the transformer fault gas data (which can be obtained from the standard library or public information) after dimensionality reduction are selected as the input of the transformer fault diagnosis neural network model, and the fault properties corresponding to the principal components are used as labels to train the transformer fault diagnosis neural network model to obtain the oil chromatography fault diagnosis model.

[0038] S203: Input the principal component in step S201 into the oil chromatography fault diagnosis model to obtain the fault property.

[0039] For S2, the transmission medium between the oil chromatography device and the detection host is a wireless network module, a 4G / 5G module, which is transmitted to the detection host using the UDP / TCP protocol.

[0040] S3. Comprehensive evaluation: If the output of the vibration signal fault diagnosis model a in S1 is normal and the output of the oil chromatography fault diagnosis model b in S2 is normal, then the current state of the transformer is judged to be normal; if the output of the vibration signal fault diagnosis model a in S1 is faulty and the output of the oil chromatography diagnosis model b in S2 is faulty, then the current state of the transformer is judged to be abnormal, the possible fault nature is the output of model b, and the specific fault type is the output of model a; if there is inconsistency, the data will be transmitted back to the data center for manual analysis and diagnosis.

[0041] Among them, the fault properties include: no fault, medium and low temperature overheating, high temperature overheating, low energy discharge, high energy discharge, and partial discharge.

[0042] Example 2 See also Figure 2 In this embodiment, a transformer fault diagnosis device based on multiple criteria is provided, including an oil chromatography detection device, a vibration sensor and a detection host.

[0043] The oil chromatography detection device is an online monitoring device for multi-component gases dissolved in transformer oil based on gas chromatography technology. It is generally installed near the transformer mailbox to collect oil chromatography data.

[0044] The vibration sensor uses an accelerometer, which is installed on the transformer base, core, and winding to collect the transformer vibration signal.

[0045] Reference Figure 3 ,The detection host includes a data processing module and a diagnosis module.

[0046] The data processing module is used to clean and normalize the data obtained from the sensor or detection device. The data cleaning process has been introduced in detail above. Data preprocessing is to process the vibration signal and oil chromatography signal into input signals that can be recognized by the model. The vibration signal is a one-dimensional time series data, which is converted into two-dimensional image data through the Gram angle field. The oil chromatography data is simply normalized.

[0047] The diagnosis module is also divided into vibration signal diagnosis module, oil chromatography diagnosis module and comprehensive diagnosis module. The specific process has been introduced before.

[0048] A vibration signal diagnosis module, used for inputting the vibration signal time series data of the transformer to be diagnosed into a pre-trained vibration signal fault diagnosis model to obtain a first diagnosis result of the transformer to be diagnosed; An oil chromatography diagnosis module, used for inputting the latest data of dissolved gas in oil of the transformer to be diagnosed into a pre-trained oil chromatography fault diagnosis model to obtain a second diagnosis result of the transformer to be diagnosed; The comprehensive diagnosis module is used to compare the first diagnosis result and the second diagnosis result, verify and supplement each other, and obtain the final fault diagnosis result.

[0049] All relevant contents of each step involved in the aforementioned embodiment of a transformer fault diagnosis method can be referred to the functional description of a functional module corresponding to a transformer diagnosis device in an embodiment of the present invention, and will not be repeated here.

[0050] Example 3 Reference Figure 4, this embodiment provides an electronic device, which includes a processor and a memory, and the processor and the memory are connected through a bus; the memory is used to store a computer program, the computer program includes program instructions, and the processor is used to execute the program instructions stored in the computer storage medium. The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. It is the computing core and control core of the terminal, which is suitable for implementing one or more instructions, and is specifically suitable for loading and executing one or more instructions in a computer storage medium to implement a corresponding method flow or corresponding function; the processor described in the embodiment of the present invention can be used for the operation of a transformer fault diagnosis method based on multiple criteria. The bus can be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The bus can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 4 The fact that only one line is used in the diagram does not mean that there is only one bus or only one type of bus.

[0051] Example 4 This embodiment provides a storage medium, specifically a computer-readable storage medium (Memory), which is a memory device in an electronic device for storing programs and data. It is understandable that the computer-readable storage medium here can include both built-in storage media in the electronic device and, of course, extended storage media supported by the electronic device. The computer-readable storage medium provides a storage space, which stores the operating system of the terminal. In addition, one or more instructions suitable for being loaded and executed by the processor are also stored in the storage space, and these instructions can be one or more computer programs (including program codes). It should be noted that the computer-readable storage medium here can be a high-speed RAM memory or a non-volatile memory (non-volatile memory), such as at least one disk memory. The processor can load and execute one or more instructions stored in the computer-readable storage medium to implement the corresponding steps of a transformer fault diagnosis method based on multiple criteria in the above embodiment.

[0052] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0053] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0054] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1A function specified in one or more boxes.

[0055] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0056] Example 5 This embodiment provides a computer program product, including a non-volatile computer-readable storage medium, wherein the non-volatile computer-readable storage medium stores a computer program product, and when the computer program is executed by a processor, the steps of the method described in each embodiment of the present application are implemented.

[0057] 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 this application 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 relevant laws, regulations and standards of relevant countries and regions.

[0058] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0059] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A transformer fault diagnosis method based on multiple criteria, characterized in that: include: Obtaining vibration signal time series data of the transformer to be diagnosed; After processing the vibration signal time series data of the transformer to be diagnosed, the data is input into a pre-trained vibration signal fault diagnosis model to obtain a first diagnosis result of the transformer to be diagnosed; If the vibration signal is detected as no fault, the latest data of dissolved gas in oil of the transformer to be diagnosed is obtained, and the latest data of dissolved gas in oil of the transformer to be diagnosed is input into the pre-trained oil chromatography fault diagnosis model to obtain the second diagnosis result of the transformer to be diagnosed; If the vibration signal is detected as a fault state, the frequency of collecting dissolved gas data in oil is increased, and the latest dissolved gas data in oil of the transformer to be diagnosed is input into the pre-trained oil chromatography fault diagnosis model to obtain the second diagnosis result of the transformer to be diagnosed; A final fault diagnosis result is obtained according to the first diagnosis result and the second diagnosis result.

2. A transformer fault diagnosis method based on multiple criteria according to claim 1, characterized in that: After processing the vibration signal time series data of the transformer to be diagnosed, inputting the pre-trained vibration signal fault diagnosis model to obtain the first diagnosis result of the transformer to be diagnosed includes: Perform outlier detection on the transformer vibration signal time series data and remove outliers; Converting the vibration signal time series data after removing abnormal values ​​into two-dimensional image data; The two-dimensional image data is input into a pre-trained vibration signal fault diagnosis model to obtain the fault property, and the vibration signal fault diagnosis model is a convolutional neural network.

3. A transformer fault diagnosis method based on multiple criteria according to claim 2, characterized in that: The interquartile range method is used to detect outliers in the time series data of transformer vibration signals.

4. A transformer fault diagnosis method based on multiple criteria according to claim 2, characterized in that: The pre-trained vibration signal fault diagnosis model is obtained by training in the following steps: Build a digital prototype of the transformer; The vibration signals of the transformer digital prototype in normal state and under different types of faults are collected. The vibration signals are taken as input and the corresponding fault types are taken as output. The convolutional neural network is trained to obtain the vibration signal fault diagnosis model.

5. The transformer fault diagnosis method based on multiple criteria according to claim 1 is characterized in that: The step of inputting the latest oil dissolved gas data of the transformer to be diagnosed into the pre-trained oil chromatography fault diagnosis model to obtain the second diagnosis result of the transformer to be diagnosed comprises: Performing ratio calculation on the latest data of dissolved gas in oil of the transformer to be diagnosed to expand the data, obtaining characteristic data, and extracting the main components of the characteristic data; The principal component is input into a pre-trained oil chromatography fault diagnosis model to obtain the fault property.

6. A transformer fault diagnosis method based on multiple criteria according to claim 5, characterized in that: The pre-trained oil chromatography fault diagnosis model is obtained by the following steps: Constructing a PCA-ANN network model, wherein the PCA-ANN network model includes an input layer, a hidden layer and an output layer; Select transformer fault gas data, expand the transformer fault gas data to obtain feature data, and extract the main components of the feature data; The above is used as the input of the transformer fault diagnosis neural network model, and the fault properties corresponding to the main components of the transformer fault gas data are used as labels to train the transformer fault diagnosis neural network model to obtain a pre-trained oil chromatography fault diagnosis model.

7. A transformer fault diagnosis method based on multiple criteria according to claim 5, characterized in that: Obtaining a final fault diagnosis result according to the first diagnosis result and the second diagnosis result includes: If the first diagnostic result is normal and the second diagnostic result is output as normal, it is determined that the current state of the transformer is normal; If the first diagnosis result is a fault and the second diagnosis result is a fault, the transformer is judged to be in an abnormal state, the nature of the fault is determined by the first diagnosis result, and the type of the fault is determined by the second diagnosis result; If the first diagnosis result and the second diagnosis result are inconsistent, manual analysis and diagnosis are performed.

8. A transformer fault diagnosis device based on multiple criteria, characterized in that: include: An acquisition module, used to acquire the vibration signal time series data of the transformer to be diagnosed and the latest dissolved gas in oil data of the transformer to be diagnosed; A vibration signal diagnosis module, used for inputting the vibration signal time series data of the transformer to be diagnosed into a pre-trained vibration signal fault diagnosis model to obtain a first diagnosis result of the transformer to be diagnosed; An oil chromatography diagnosis module, used for inputting the latest data of dissolved gas in oil of the transformer to be diagnosed into a pre-trained oil chromatography fault diagnosis model to obtain a second diagnosis result of the transformer to be diagnosed; The comprehensive diagnosis module is used to compare the first diagnosis result and the second diagnosis result, verify and supplement each other, and obtain the final fault diagnosis result.

9. An electronic device, characterized in that: include: at least one processor; as well as, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute a transformer fault diagnosis method based on multiple criteria as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, a transformer fault diagnosis method based on multiple criteria as described in any one of claims 1 to 7 is implemented.

Images