Industrial equipment remaining useful life prediction method and system and electronic equipment

Abstract

The invention relates to an industrial equipment remaining useful life prediction method and system and electronic equipment. The method comprises the following steps: a, performing normalization processing on original vibration signal data of equipment; b, after feature expansion is conducted on the normalized vibration signal data in an empirical mode decomposition mode, extracting data featuresof the vibration signal data; c, constructing a time sequence convolution network according to the extracted data features; and d, outputting a residual effective life prediction result of the equipment by using the time sequence convolution network. According to the method, the data characteristics of the original signals are decomposed, extracted and enriched through the empirical model, and then the sequential convolutional neural network is used for training and predicting to obtain the residual effective service life prediction model, so that the prediction speed and prediction precisionof the residual life of the industrial equipment can be greatly improved, and the method has realizability in the actual manufacturing process.

Description

Technical field [0001] This application belongs to the technical field of equipment failure prediction, and particularly relates to a method, system and electronic equipment for predicting the remaining effective life of industrial equipment. Background technique [0002] In the industrial production process, the aging process of equipment is inevitable. In order to remain competitive, industrial production enterprises must keep their production equipment in good working conditions for a long time, and need to improve the availability, stability and safety of the equipment while reducing equipment maintenance costs, and equipment failure prediction becomes a key link. Accurate equipment failure prediction can provide equipment maintenance personnel with equipment safety warnings in advance. Maintenance personnel can determine equipment maintenance time in advance based on the warnings, reduce the scrap rate due to equipment failures, shorten the maintenance cycle, and thus greatl...

AI Technical Summary

Problems solved by technology

Although this type of method can provide relatively accurate results, the real equipment system is usually nonlinear, and the aging mechanism of the equipment is usually random and difficult to obtain in the form of an analysis model. In actual working conditions, it is difficult to establish a Models can accommodate complex environmental noise and degradation mechanisms

[0008] 2. In the existing data-driven methods, support vector machine (SVM) modeling is used to predict the remaining effective service life of industrial equipment, but for large-scale training data and multi-classification problems, SVM consumes a lot of running memory and computing time, resulting in less accurate and slower predictions

In addition, convolutional neural network (CNN) modeling analysis is also used to predict the remaining effective life, but the original vibration signal data collected from industrial equipment is essentially a time-series operating state data, convolutional neural network (CNN) ) cannot solve the time series prediction very well, resulting in low prediction accuracy. On the other hand, if the original signal is directly applied to the convolutional neural network (CNN), the model is difficult to converge and the prediction speed is slow

Images