Transformer state evaluation and fault detection method based on multi-source data fusion

Abstract

The invention belongs to the technical field of transformer fault type detection, and discloses a transformer state evaluation and fault detection method based on multi-source data fusion. Transformercurrent data are detected by using a current sensor corrected in a cyclic mode based on least square method. A voltmeter improving accuracy based on a remainder splitting algorithm is utilized to detect transformer voltage data. Transformer temperature data are detected by using a temperature sensor. A gas sensor performing temperature compensation based on a standard artificial bee colony algorithm is used for detecting concentration data of transformer fault characteristic gas. A data processing software is utilized to build a transformer fault model, and the transformer fault state is evaluated according to the detected data. An alarm or notification is given in time according to the evaluation results by using an alarm apparatus. The transformer state evaluation and fault detection method adopts a theory of a probability fuzzy set to process and analyze; the fault state of a transformer can be evaluated, the uncertainty of the characteristic value of the fault state of the transformer is reflected, and theoretical guidance is provided for the evaluation of the fault state of the transformer.

Description

technical field [0001] The invention belongs to the technical field of transformer fault type detection, in particular to a transformer state evaluation and fault detection method based on multi-source data fusion. Background technique [0002] Transformer is a device that uses the principle of electromagnetic induction to change AC voltage. The main components are primary coil, secondary coil and iron core (magnetic core). The main functions are: voltage conversion, current conversion, impedance conversion, isolation, voltage stabilization (magnetic saturation transformer), etc. According to the use, it can be divided into: power transformers and special transformers (electric furnace transformers, rectifier transformers, power frequency test transformers, voltage regulators, mining transformers, audio transformers, intermediate frequency transformers, high frequency transformers, impact transformers, instrument transformers, electronic transformers , reactors, transformer...

AI Technical Summary

Problems solved by technology

[0004] Existing transformer fault type detection cannot take into account the influence of equipment service age and state characteristic quantity maintenance measures, resulting in inaccurate detection data and untimely alarm; at the same time, traditional transformer fault state evaluation technologies do not fully consider the uncertainties of influencing factors The applicability, practicability and applicability of the calculation method are also difficult to be satisfied; the current data collected by the current sensor is inaccurate; the processing of decimal places in the voltmeter directly affects the accuracy of the data; the gas sensor will be affected by temperature. Changes Resulting in reduced test accuracy and stability

Images