Method, equipment, and system for diagnosing faults of transformer

Abstract

The invention provides a method, a system, computer equipment, and a computer readable storage medium for diagnosing faults of a transformer, and relates to the technical field of a power system. Thesystem comprises a historical data acquisition module, a classifier establishment module, a fault data acquisition module, and a diagnosis result output module. The historical data acquisition moduleis used to acquire historical fault data of the transformer, and the historical fault data comprises oil chromatographic data and meteorological data. The classifier establishment module is used to establish a combined classifier according to the historical fault data. The fault data acquisition module is used to acquire fault data of the transformer that needs to be diagnosed. The diagnosis result output module is used to output a fault diagnosis result of the transformer according to the combined classifier. According to the method, the system, the computer equipment, and the computer readable storage medium for diagnosing faults of the transformer, the combined classifier is established by combining historical meteorological data related to the transformer with the oil chromatographic data. The random forest algorithm is used to vote on classification results of the combined classifier to give a diagnosis result, thereby avoiding the inaccuracy of the diagnosis result of a single classifier, and improving the reliability of the diagnosis result.

Description

technical field [0001] The present invention relates to the technical field of power systems, in particular to the detection technology of power equipment in the power system, and specifically to a transformer fault diagnosis method, system, computer equipment and computer-readable storage medium. Background technique [0002] Transformer is the core pivotal equipment in the entire power system, which plays a very important role in power transmission and transformation, and can even determine the stability of the power system. Since the transformer needs to work continuously with heavy load, the probability of failure is relatively high, so timely and accurate diagnosis of transformer failure has important engineering application value. [0003] The inventors of the present application found that classification algorithms in the prior art are an important means of transformer fault diagnosis, and many classification algorithms have been used in transformer fault diagnosis, s...

AI Technical Summary

Problems solved by technology

However, a single classifier often has inaccurate diagnostic results and is prone to overfitting

In addition, since the concentration of various gases in the transformer insulating oil is closely related to the operation of the transformer, it is an important basis for diagnosing transformer faults. Therefore, in the prior art, the fault diagnosis of transformers is generally only based on the oil chromatographic data of the transformer, which reduces the Reliability of Diagnosis Results

Images