Fault diagnosis device of transformer based on rough set theory and diagnosis method thereof

Abstract

The invention relates to a fault diagnosis device of a transformer based on rough set theory and a diagnosis method thereof, belonging to the field of power transformer fault diagnosis. The device comprises a dual-core module comprising a CPU module and a DSP module, and a redundancy dual detection module. The method adopting the device for diagnosis is as follows: according to the rough set theory, each condition attribute is arranged from small to big during carrying out continuous attribute discretization to an original information table, and the 'golden section ratio' of two adjacent ondition attribute values is adopted as a breakpoint. The invention has the advantages that the device adopts a dual-core module structure, thereby greatly improving the reliability, stability, flexibility and operation efficiency of a system; and the diagnosis method makes use of the information provided by data to simplify the data and calculate the minimum expression of knowledge, has strong fault-tolerant capability for incomplete information and high accuracy, and improves the rapidity of calculating data and information processing.

Description

Technical field: [0001] The invention belongs to the field of power transformer fault diagnosis, and in particular relates to a transformer fault diagnosis device and a diagnosis method based on the dissolved gas content in transformer oil and rough set theory calculation. Background technique: [0002] The power transformer is the key equipment of the power system, and its operation reliability is directly related to the safety and stability of the power system. Some reports point out that abnormal phenomena occur frequently in the operation of medium and large transformers, and even cause accidents. The safe operation of the power grid has caused serious threats. Therefore, it is of great practical significance to detect potential hidden troubles early through transformer fault diagnosis. [0003] Due to the complexity and variety of power transformer faults, and the causes of faults are very complicated and not obvious, it is very difficult to accurately judge the type of...

AI Technical Summary

Problems solved by technology

[0003] Due to the complexity and variety of power transformer faults, and the causes of faults are very complicated and not obvious, it is very difficult to accurately judge the type of power transformer fault and the location of the fault. With the rapid development of the power industry, power transformers are moving toward high The development in the direction of voltage and large capacity has made the failure rate of power transformers higher and higher, and the scope of influence has become larger and larger, resulting in greatly increased maintenance time and difficulties. At present, my country's electric power industry generally implements a regular maintenance system. Lack of pertinence in planning often leads to the double disadvantages of overhaul and underhaul

[0004] At present, the artificial intelligence methods for diagnosing transformer faults using dissolved gas analysis (DGA) in oil include expert systems, fuzzy mathematics and other means, and have made some progress. The expert system is based on professionals in the field. Knowledge and experience are used to reason and judge, and simulate the decision-making process of professionals to solve problems. The fuzzy theory method is mainly to solve problems by establishing a fuzzy relationship matrix in the fault symptom space and fault cause space based on expert experience. However, under the current conditions, experts The system cannot realize the function of self-expansion and renewal, and cannot fully adapt to the requirements of on-site diagnosis. Fuzzy mathematics is an effective tool for dealing with inaccurate information. The method of determining its membership function is relatively simple and lacks convincing objective basis. , it is of great significance to find an effective method to solve the problem of universality and uncertainty

Images