Power equipment fault detection and positioning method based on artificial intelligence reasoning fusion

Abstract

The invention discloses a power equipment fault detection and positioning method based on artificial intelligence reasoning fusion. The power equipment fault detection and positioning method comprisesthe steps: 1) acquiring monitoring information of different monitoring points of power equipment in a normal operation state; 2) setting faults, and acquiring monitoring information of different fault types, different fault positions and different monitoring points of the equipment; 3) taking the monitoring information obtained in the steps 1) to 2) as a training data set and the fault type and position as labels, and inputting the training data set, the fault type and the position into a deep convolutional neural network for training; 4) collecting monitoring data, performing verification classification by using the method in the step 3), and obtaining a probability value corresponding to each label; and 5) taking classification results of different labels as basic probability distribution values, taking different sensors as different evidences ek of decision fusion for a monitoring system consisting of a plurality of sensors, and performing fusion processing by utilizing a DS evidence theory to obtain a final fault diagnosis result. According to the invention, power equipment fault detection, fault type discrimination and fault positioning can be intelligently realized.

Description

technical field [0001] The invention relates to a method for detecting and locating electric equipment faults based on artificial intelligence reasoning fusion, in particular to a method for diagnosing electric equipment faults combined with deep convolutional neural networks and multi-source information fusion theory. Background technique [0002] With the advent of the Industry 4.0 era, future equipment will continue to develop in the direction of intelligence and integration. The connection between various industrial equipment is getting closer and closer, and the failure of any important part may lead to the collapse of the entire system. Therefore, it is very necessary to study intelligent fault diagnosis methods in depth. Traditional monitoring and diagnosis methods are poor in robustness, and it is difficult to meet the characteristics of large data capacity, multi-source heterogeneity, and high acquisition frequency of intelligent monitoring systems, and diagnostic ...

AI Technical Summary

Problems solved by technology

However, the above methods do not consider the fault location information, and the research on the deep learning positioning method in the fault diagnosis of power equipment is almost still blank.

Images