Chromatographic Fault Diagnosis Method of Transformer Oil Based on Niche Genetic Algorithm

Abstract

The invention discloses a transformer oil chromatogram fault diagnosis method based on an ecological niche genetic algorithm. The method comprises 1) for concrete problems, selecting suitable coding for a fault set, and inputting data to generate an initial population, and then calculating the individual fitness, and finally re-queuing the individuals in the population according to the magnitude of the fitness; 2) performing selection, cross and mutation operation on the formed initial population; and 3) after the step 2), performing ecological niche elimination operation on the population, recalculating the fitness to finally select the chromosome with the maximum fitness, that is, obtaining the combination of fault types, and completing the transformer oil chromatogram fault diagnosis based on the ecological niche genetic algorithm. The transformer oil chromatogram fault diagnosis method based on an ecological niche genetic algorithm utilizes the ecological niche genetic algorithm to analyze the gas characteristic signal in the fault oil and establish the corresponding relation between the oil chromatogram characteristic parameters and the fault types, can realize determination of the transformation operation fault, and has the advantages of being efficient and quick and being high in the adaptive learning capability.

Description

technical field [0001] The invention belongs to the technical field of on-line monitoring methods for smart grids, and in particular relates to a transformer oil chromatography fault diagnosis method based on a niche genetic algorithm. Background technique [0002] Transformers play an indispensable role in the power grid. They are the core of energy conversion and transmission, and also the key hub equipment in the first defense system of power grid security. Transformer faults will not only bring economic losses, but also may cause panic and inconvenience to people due to blackouts. Therefore, fault diagnosis of transformers is required for the development of smart grids. [0003] There are many existing methods for transformer fault diagnosis, among which the research on the analysis of dissolved gas in oil is particularly important. It involves various artificial intelligence algorithms, such as: BP neural network provides a better structural system for transformer faul...

AI Technical Summary

Problems solved by technology

It involves various artificial intelligence algorithms, such as: BP neural network provides a better structural system for transformer fault diagnosis, but it has the disadvantages of slow training convergence and easy to fall into local minimum points

Expert systems can effectively simulate fault diagnosis by human experts to complete the fault diagnosis process, but there are also many technical problems such as difficulties in knowledge acquisition, uncertainty reasoning, and self-learning difficulties.

[0004] Fuzzy control can use precise mathematical tools to clarify fuzzy concepts or natural language, so as to reasonably quantify fault phenomena, etc., but due to certain human factors in the determination process of membership functions and fuzzy rules, it lacks convincing objective basis

Images