ANN algorithm-based online monitoring data anomaly analysis method for dissolved gas in transformer oil

Abstract

The invention provides an ANN algorithm-based online monitoring data anomaly analysis method and system for dissolved gas in transformer oil and a related device, and the method comprises the following steps: carrying out the preprocessing of an obtained real-time online anomaly detection result of a to-be-detected transformer, and obtaining a preprocessed anomaly detection result; taking the preprocessed anomaly detection result as the input of a pre-constructed prediction model, and carrying out reverse deduction to obtain a detection condition of a sensor monitoring item; identifying a key influence factor corresponding to the real-time online anomaly detection result based on actually collected sensor monitoring data and the detection condition; according to the method, intelligent monitoring and self-diagnosis in the detection process are achieved, the adaptive capacity of the system for processing complex working conditions is improved through data normalization and classification processing, finally, the accurate oil quality judgment is guaranteed, meanwhile, the false alarm rate is greatly reduced, and timely and reliable technical support is provided for transformer state evaluation and operation and maintenance decision making.

Description

Technical Field

[0001] This invention belongs to the field of power equipment monitoring and fault diagnosis technology, specifically involving an anomaly analysis method for online monitoring data of dissolved gases in transformer oil based on the ANN algorithm. Background Technology

[0002] In power systems, transformers, as core equipment for energy conversion and transmission, directly affect the safety and stability of the power grid. Transformer oil, as a crucial internal medium, not only serves as insulation and cooling but also reflects the transformer's health status through changes in its physicochemical properties. Therefore, real-time and accurate online monitoring of transformer oil is of great significance for preventing transformer failures and ensuring power system safety. In recent years, with the development of smart grids, online transformer oil monitoring systems have gradually become an important technical means in the power industry.

[0003] Although online transformer oil monitoring systems have made some progress, they still face many challenges and shortcomings in practical applications. The most prominent issues are inaccurate detection results and false alarms. Inaccurate results are mainly due to a combination of factors, including limited sensor accuracy, environmental interference, and the complexity of gas composition. Sensors may experience accuracy degradation, drift, or malfunction during long-term operation, causing the detection data to deviate from the true value. Simultaneously, changes in ambient temperature and humidity can also affect sensor performance, further exacerbating data inaccuracy. False alarms are often related to algorithm design, threshold settings, and equipment malfunctions. Existing fault diagnosis algorithms may not be fully adaptable to all fault conditions, leading to misjudgments under certain specific conditions. Furthermore, threshold settings require comprehensive consideration of various factors, such as equipment aging and operating environment; improper settings can easily trigger false alarms. In addition, hardware malfunctions, such as those of sensors and data acquisition modules, are also a significant cause of false alarms.

[0004] Given the aforementioned defects and shortcomings of existing detection technologies, it is particularly urgent and necessary to add a self-testing system to eliminate misjudgments and false alarms caused by inaccurate detection and sensor accuracy issues. Summary of the Invention

[0005] The purpose of this invention is to provide an anomaly analysis method for online monitoring data of dissolved gases in transformer oil based on the ANN algorithm. This method solves the above-mentioned shortcomings in the prior art, ensures that the system can accurately and reliably monitor dissolved gases in transformer oil during long-term operation, detect potential faults in a timely manner, and avoid false alarms caused by inaccurate detection or sensor accuracy issues.

[0006] To achieve the above objectives, the technical solution adopted by the present invention is as follows: In a first aspect, the present invention provides a method for anomaly analysis of online monitoring data of dissolved gases in transformer oil based on the ANN algorithm, comprising the following steps: The real-time online anomaly detection results of the transformer under test are preprocessed to obtain the preprocessed anomaly detection results. The preprocessed anomaly detection results are used as input to a pre-built prediction model to infer the detection conditions of the sensor monitoring project. Based on the actual collected sensor monitoring data and the detection conditions, the key influencing factors corresponding to the real-time online anomaly detection results are identified.

[0007] Preferably, the pre-built prediction model is specifically implemented using the following method: Obtain historical datasets, which include sensor monitoring project data and corresponding online detection result data; Using an artificial neural network algorithm, with the sensor monitoring project data as input and the online detection result data as output, a nonlinear prediction model from sensor monitoring conditions to online detection results is established through training, thereby obtaining a pre-constructed prediction model.

[0008] Preferably, the pre-built prediction model includes a nonlinear regression prediction sub-model and a multi-classification sub-model, wherein: The nonlinear regression prediction sub-model includes an input layer, two hidden layers, and an output layer. The activation functions of the two hidden layers are tanh and ReLU, respectively; the activation function of the output layer is tanh. The multi-class sub-model includes an input layer, two hidden layers, and an output layer. The activation function of the two hidden layers is ReLU, and the activation function of the output layer is softmax.

[0009] Preferably, based on the actual collected sensor monitoring data and the detection conditions, the key influencing factors corresponding to the real-time online anomaly detection results are identified. The specific method is as follows: The actual collected sensor monitoring data and detection conditions are compared to calculate the error of each sensor monitoring item; Based on the aforementioned error identification, key factors affecting online detection results are identified.

[0010] Preferably, the key influencing factors affecting the online detection results are identified based on the aforementioned errors, and the specific method is as follows: If the error of a certain sensor monitoring item exceeds a preset threshold, then that item is determined to be a key influencing factor causing the detection abnormality; If the errors of all sensor monitoring items are below the preset threshold, then the key influencing factor for the abnormal online detection results is the abnormal oil quality of the transformer under test.

[0011] Secondly, the present invention provides an online monitoring data anomaly analysis system for dissolved gases in transformer oil based on the ANN algorithm, comprising: The detection result preprocessing unit is used to preprocess the real-time online anomaly detection results of the transformer under test to obtain the preprocessed anomaly detection results. The detection condition prediction unit is used to take the pre-processed abnormal detection results as input to the pre-built prediction model and inversely deduce the detection conditions of the sensor monitoring project. The influencing factor identification unit is used to identify the key influencing factors corresponding to the real-time online anomaly detection results based on the actual collected sensor monitoring data and the detection conditions.

[0012] Thirdly, the present invention provides an electronic device including a processor and a memory, wherein the memory stores computer instructions, and when the computer instructions are executed by the processor, the electronic device performs the method described thereon.

[0013] Fourthly, the present invention provides a computing device cluster, comprising at least one computing device, each computing device including a processor and a memory; The processor of the at least one computing device is configured to execute instructions stored in the memory of the at least one computing device to cause the cluster of computing devices to perform the method according to the method.

[0014] Fifthly, the present invention provides a computer program product, the computer program product including computer-executable instructions, which, when executed, implement the method described.

[0015] In a sixth aspect, the present invention provides a computer-readable storage medium storing computer-executable instructions that, when executed by a processor, implement the method described herein.

[0016] Compared with the prior art, the beneficial effects of the present invention are: This invention provides a method for analyzing anomalies in online monitoring data of dissolved gases in transformer oil based on the ANN algorithm. By constructing a dual self-checking system that includes sensor self-checking and algorithm self-checking, it significantly improves the accuracy and reliability of online chromatographic detection of transformer oil. This invention subdivides the detection process into four units: oil circulation, pretreatment, detection, and calibration. It monitors seven key influencing factors in real time, achieving comprehensive status tracking and anomaly identification throughout the detection process. The sensor self-checking part, through a multi-level calibration mechanism including theoretical degassing comparison, periodic standard gas injection, and immediate standard gas detection in case of anomalies, can quickly determine whether data anomalies originate from a fault in the detection system itself, effectively avoiding false alarms caused by sensor drift, environmental interference, and other factors. The algorithm self-checking part innovatively introduces artificial neural networks, establishing nonlinear regression prediction models and multi-classification models. By modeling the complex relationship between detection conditions and results in a "black box," it can reverse-engineer theoretical detection conditions when data anomalies occur and compare them with actual monitoring data, accurately locating the main factors causing the anomalies and greatly improving the pertinence and interpretability of fault diagnosis. This method not only enables intelligent monitoring and self-diagnosis of the detection process, but also improves the system's adaptability to complex operating conditions through data normalization and classification. Ultimately, while ensuring accurate oil quality judgment, it significantly reduces the false alarm rate, providing timely and reliable technical support for transformer condition assessment and operation and maintenance decisions. Detailed Implementation

[0017] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0018] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.

[0019] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0020] As used in this application specification and the appended claims, the term "if" may be interpreted, depending on the context, as "when," "once," "in response to determination," or "in response to detection." Similarly, the phrase "if determined" or "if detected [the described condition or event]" may be interpreted, depending on the context, as meaning "once determined," "in response to determination," "once detected [the described condition or event]," or "in response to detection [the described condition or event]."

[0021] Furthermore, in the description of this application and the appended claims, the terms "first," "second," "third," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0022] References to "one embodiment" or "some embodiments" as described in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0023] Example 1 The method for analyzing anomalies in online monitoring data of dissolved gases in transformer oil based on the ANN algorithm provided in this embodiment includes the following steps: The acquired historical dataset is normalized to obtain a processed historical dataset, which includes sensor monitoring project data and corresponding online detection result data. Using an artificial neural network algorithm, with the sensor monitoring project data as input and the online detection result data as output, a nonlinear prediction model from sensor monitoring conditions to online detection results is established through training, thereby obtaining a pre-constructed prediction model; The real-time online anomaly detection results of the transformer under test are preprocessed to obtain the preprocessed anomaly detection results. The preprocessed anomaly detection results are used as input to a pre-built prediction model to obtain the detection conditions for the sensor monitoring project. Based on the actual collected sensor monitoring data and the detection conditions, the key influencing factors corresponding to the real-time online anomaly detection results are identified. The key influencing factors are oil temperature, oil circulation volume, degassing process temperature, actual degassing volume, degassing oscillation amplitude, degassing oscillation frequency or oscillation time.

[0024] In this embodiment, by tracking seven main factors that significantly affect the online detection results of oil chromatography, the detection data is compared and analyzed in real time. This allows for the timely detection and correction of outliers or deviations in the data, thereby improving the accuracy of the detection results and reducing false alarms. Furthermore, by optimizing the fault diagnosis algorithm and threshold setting scheme, the adaptability and accuracy of the algorithm can be further improved, reducing the false alarm rate.

[0025] Example 2 Based on Example 1, this example provides a method for anomaly analysis of online monitoring data of dissolved gases in transformer oil based on the ANN algorithm. This example utilizes an artificial neural network algorithm to construct a nonlinear prediction model from sensor monitoring conditions to online detection results. The model is trained using the sensor monitoring data as input and the online detection result data as output to obtain a pre-constructed prediction model. Specifically: A nonlinear prediction model was constructed, consisting of two multilayer perceptron neural network (MLP) models: a nonlinear regression prediction model (NPM) and a multi-classification model (MCM). Both models used algorithms to optimize and update their parameters. The NPM structure consists of an input layer with 6 nodes; two hidden layers with 13 and 15 nodes respectively, using tanh and ReLU activation functions; and an output layer with 1 node and tanh activation function. The maximum number of training iterations is 5000. Mean squared error (MSE) and R-Square (R) are used. 2 (Equation) is used as the model evaluation method; the closer the MSE value is to 0, the better the R... 2 The closer the value is to 1, the better the model's nonlinear regression prediction ability.

[0026] (1) (2) In the formula: The number of data points; For the test results; Output the detection result value; This represents the average of the test results.

[0027] The MCM structure consists of an input layer with 6 nodes; two hidden layers with 17 and 20 nodes respectively, both using ReLU activation; and an output layer with 6 nodes and softmax activation. The maximum training iterations are 6000. Categorical cross-entropy (CCE) and categorical accuracy are used as model evaluation methods. A smaller CCE value and a categorical accuracy closer to 1 indicate a more reliable multi-class classification model. (3) In the formula: The number of data points; For the number of process conditions; It is a condition variable, either 0 or 1; For the sample Process conditions The predicted probability.

[0028] The detection results from the historical dataset are discretized into six sets: (0.95,1]; (0.9,0.95]; (0.8,0.9]; (0.6,8]; (0.3,0.6]; [0,0.3], with class labels of 0, 1, 2, 3, 4, and 5, respectively. The discretized datasets are used for training and evaluation of MCM.

[0029] Example 3 Based on Example 1, this example provides a method for anomaly analysis of online monitoring data of dissolved gases in transformer oil based on the ANN algorithm, which normalizes the acquired historical dataset using the following formula: (4) In the formula: Input data; The data is after normalization; This represents the maximum value of the input data. The minimum value of the input data; The maximum value of the normalized data is 1; The minimum value of the normalized data is 0.

[0030] Example 4 Based on Example 1, this example provides a method for analyzing anomalies in online monitoring data of dissolved gases in transformer oil based on the ANN algorithm. This method identifies key influencing factors corresponding to the real-time online anomaly detection results based on the actual collected sensor monitoring data and the detection conditions. The specific method is as follows: The actual collected sensor monitoring data and detection conditions are compared to calculate the error of each sensor monitoring item; Based on the aforementioned error identification, key factors affecting online detection results are identified.

[0031] Example 5 Based on Example 1, this example provides a method for analyzing anomalies in online monitoring data of dissolved gases in transformer oil based on the ANN algorithm. This method identifies key influencing factors affecting the online detection results based on the aforementioned errors. The specific method is as follows: If the error of a certain sensor monitoring item exceeds a preset threshold, then that item is determined to be a key influencing factor causing the detection abnormality; If the errors of all sensor monitoring items are below the preset threshold, then the key influencing factor for the abnormal online detection results is the abnormal oil quality of the transformer under test.

[0032] Example 6 Based on Example 1, the method for anomaly analysis of online monitoring data of dissolved gases in transformer oil based on the ANN algorithm provided in this example further includes: When the real-time online anomaly detection result of the transformer under test is obtained, the sensor monitoring data corresponding to the real-time online anomaly detection result is obtained in real time before the real-time online anomaly detection result is preprocessed. A self-check is performed on the sensor monitoring data to determine whether the influencing factor of the real-time online anomaly detection result of the transformer under test is a sensor acquisition device malfunction. If any of the following data is abnormal, the influencing factor of the real-time online anomaly detection result of the transformer under test is determined to be a sensor acquisition device malfunction, specifically: Determine if the sensor monitoring data is abnormal; The theoretical degassing capacity is calculated based on sensor monitoring data. The theoretical degassing capacity is then compared with the actual degassing capacity. The comparison results are used to determine whether the degassing unit is abnormal. A standard gas is injected periodically for testing to determine if the detection system is malfunctioning. When online data is abnormal, standard gas is automatically injected immediately for detection, and the detection system can be quickly and accurately determined to determine whether there is a detection fault based on the standard gas detection results.

[0033] Through the above optimization measures, when abnormal online oil chromatography data is detected, the sensor self-test section can immediately determine and report the authenticity of the data based on the above steps and sensor detection results, further confirming whether the abnormality is due to transformer oil. Once a genuine fault symptom appears, the system will immediately issue an accurate alarm, providing maintenance personnel with timely and reliable fault information.

[0034] Example 7 The online monitoring data anomaly analysis system for dissolved gases in transformer oil based on the ANN algorithm provided in this embodiment includes: The detection result preprocessing unit is used to preprocess the real-time online anomaly detection results of the transformer under test to obtain the preprocessed anomaly detection results. The detection condition prediction unit is used to take the pre-processed abnormal detection results as input to the pre-built prediction model and inversely deduce the detection conditions of the sensor monitoring project. The influencing factor identification unit is used to identify the key influencing factors corresponding to the real-time online anomaly detection results based on the actual collected sensor monitoring data and the detection conditions.

[0035] Example 8 This embodiment also provides a computing device. The computing device includes a bus, a processor, a memory, and a communication interface. The processor, memory, and communication interface communicate with each other via the bus. The computing device can be a server or a terminal device. It should be understood that this application does not limit the number of processors and memory in the computing device.

[0036] A bus can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of representation, a bus can include a path for transmitting information between various components of a computing device (e.g., memory, processor, communication interfaces).

[0037] The processor may include any one or more of the following: central processing unit (CPU), graphics processing unit (GPU), tensor processing unit (TPU), application specific integrated circuit (ASIC), field-programmable gate array (FPGA), microprocessor (MP), or digital signal processor (DSP).

[0038] Memory can include volatile memory, such as random access memory (RAM). Processors can also include non-volatile memory. volatile memory, such as read-only memory (ROM). ROM (memory only), flash memory, hard disk drive (HDD), or solid state drive (SSD).

[0039] The memory stores executable program code, which the processor executes to implement the functions of the aforementioned units, thereby achieving, for example, the method described in Embodiment 1. That is, the memory may store instructions for the methods and functions relating to the computing device in any of the above embodiments.

[0040] The communication interface uses transceiver modules such as, but not limited to, network interface cards and transceivers to enable communication between computing devices and other devices or communication networks.

[0041] Example 9 This embodiment also provides a computing device cluster. The computing device cluster includes at least one computing device. The computing device can be a server, such as a central server, an edge server, or a local server in a local data center. In some embodiments, the computing device can also be a terminal device such as a desktop computer, a laptop computer, or a smartphone.

[0042] The computing device cluster includes at least one computing device. The memory of one or more computing devices in the computing device cluster may store the same instructions for performing the methods and functions related to the computing devices in any of the above embodiments.

[0043] In some possible implementations, the memory of one or more computing devices in the computing device cluster may also store partial instructions for performing the methods and functions of the computing devices involved in any of the above embodiments. In other words, a combination of one or more computing devices can jointly execute the instructions for performing the methods and functions of the computing devices.

[0044] It should be noted that the memory in different computing devices within a computing device cluster can store different instructions, which are used to execute parts of the device's functions.

[0045] In some possible implementations, one or more computing devices in a computing device cluster can be connected via a network. This network can be a wide area network (WAN) or a local area network (LAN), etc. Two computing devices are connected to each other via the network. Specifically, they connect to the network through communication interfaces in each computing device.

[0046] Embodiments of this disclosure also provide a computer program product containing instructions that, when run on a computer, cause the computer to perform the methods and functions related to a computing device in any of the above embodiments.

[0047] Example 10 This embodiment also provides a computer-readable storage medium storing computer instructions that, when executed by a processor, cause the processor to perform the methods and functions of the computing device involved in any of the above embodiments.

[0048] Generally, the various embodiments of this disclosure can be implemented in hardware or dedicated circuitry, software, logic, or any combination thereof. Some aspects can be implemented in hardware, while others can be implemented in firmware or software, which can be executed by a controller, microprocessor, or other computing device. Although various aspects of the embodiments of this disclosure are shown and described as block diagrams, flowcharts, or represented using some other illustration, it should be understood that the blocks, apparatuses, systems, techniques, or methods described herein can be implemented as, as non-limiting examples, in hardware, software, firmware, dedicated circuitry or logic, general-purpose hardware or controllers or other computing devices, or some combination thereof.

[0049] Example 11 This embodiment provides at least one computer program product tangibly stored on a non-transitory computer-readable storage medium. The computer program product includes computer-executable instructions, such as instructions included in program modules, which execute in a device on a target real or virtual processor to perform the processes / methods of the above-described reference scheme. Typically, program modules include routines, programs, libraries, objects, classes, components, data structures, etc., that perform specific tasks or implement specific abstract data types. In various embodiments, the functionality of program modules can be combined or divided among program modules as needed. The machine-executable instructions for the program modules can execute within a local or distributed device. In a distributed device, the program modules can reside in both local and remote storage media.

[0050] Computer program code used to implement the methods of this disclosure may be written in one or more programming languages. This computer program code may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus, such that when executed by the computer or other programmable data processing apparatus, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be performed. The program code may be executed entirely on a computer, partially on a computer, as a stand-alone software package, partially on a computer and partially on a remote computer, or entirely on a remote computer or server.

[0051] In the context of this disclosure, computer program code or related data may be carried on any suitable carrier to enable a device, apparatus, or processor to perform the various processes and operations described above. Examples of carriers include signals, computer-readable media, and so on. Examples of signals may include electrical, optical, radio, sound, or other forms of propagation signals, such as carrier waves, infrared signals, etc.

[0052] Computer-readable media can be any tangible medium that contains or stores programs for or relating to an instruction execution system, apparatus, or device, or a data storage device such as a data center containing one or more available media. Computer-readable media can be computer-readable signal media or computer-readable storage media. Computer-readable media can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. More detailed examples of computer-readable storage media include electrical connections with one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0053] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A method for anomaly analysis of online monitoring data of dissolved gases in transformer oil based on ANN algorithm, characterized in that, Includes the following steps: The real-time online anomaly detection results of the transformer under test are preprocessed to obtain the preprocessed anomaly detection results. The preprocessed anomaly detection results are used as input to a pre-built prediction model to infer the detection conditions of the sensor monitoring project. Based on the actual collected sensor monitoring data and the detection conditions, the key influencing factors corresponding to the real-time online anomaly detection results are identified.

2. The method for anomaly analysis of online monitoring data of dissolved gases in transformer oil based on ANN algorithm according to claim 1, characterized in that, Pre-built prediction models, specifically: Obtain historical datasets, which include sensor monitoring project data and corresponding online detection result data; Using an artificial neural network algorithm, with the sensor monitoring project data as input and the online detection result data as output, a nonlinear prediction model from sensor monitoring conditions to online detection results is established through training, thereby obtaining a pre-constructed prediction model.

3. The method for anomaly analysis of online monitoring data of dissolved gases in transformer oil based on ANN algorithm according to claim 1, characterized in that, The pre-built prediction model includes a nonlinear regression prediction sub-model and a multi-classification sub-model, wherein: The nonlinear regression prediction sub-model includes an input layer, two hidden layers, and an output layer. The activation functions of the two hidden layers are tanh and ReLU, respectively; the activation function of the output layer is tanh. The multi-class sub-model includes an input layer, two hidden layers, and an output layer. The activation function of the two hidden layers is ReLU, and the activation function of the output layer is softmax.

4. The method for anomaly analysis of online monitoring data of dissolved gases in transformer oil based on ANN algorithm according to claim 1, characterized in that, Based on the actual collected sensor monitoring data and the detection conditions, the key influencing factors corresponding to the real-time online anomaly detection results are identified. The specific method is as follows: The actual collected sensor monitoring data and detection conditions are compared to calculate the error of each sensor monitoring item; Based on the aforementioned error identification, key factors affecting online detection results are identified.

5. The method for anomaly analysis of online monitoring data of dissolved gases in transformer oil based on ANN algorithm according to claim 1, characterized in that, Based on the aforementioned error identification, the key influencing factors affecting online detection results are determined using the following method: If the error of a certain sensor monitoring item exceeds a preset threshold, then that item is determined to be a key influencing factor causing the detection abnormality; If the errors of all sensor monitoring items are below the preset threshold, then the key influencing factor for the abnormal online detection results is the abnormal oil quality of the transformer under test.

6. An online monitoring data anomaly analysis system for dissolved gases in transformer oil based on ANN algorithm, characterized in that, include: The detection result preprocessing unit is used to preprocess the real-time online anomaly detection results of the transformer under test to obtain the preprocessed anomaly detection results. The detection condition prediction unit is used to take the pre-processed abnormal detection results as input to the pre-built prediction model and inversely deduce the detection conditions of the sensor monitoring project. The influencing factor identification unit is used to identify the key influencing factors corresponding to the real-time online anomaly detection results based on the actual collected sensor monitoring data and the detection conditions.

7. An electronic device, characterized in that, It includes a processor and a memory, the memory storing computer instructions that, when executed by the processor, cause the electronic device to perform the method of any one of claims 1 to 5.

8. A computing device cluster, characterized in that, It includes at least one computing device, each computing device including a processor and memory; The processor of the at least one computing device is configured to execute instructions stored in the memory of the at least one computing device to cause the cluster of computing devices to perform the method according to any one of claims 1 to 5.

9. A computer program product, characterized in that, The computer program product includes computer-executable instructions that, when executed, implement the method of any one of claims 1 to 5.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions that, when executed by a processor, implement the method of any one of claims 1 to 5.