Method for predicting residual life of gear based on LSTM-AON

Abstract

The invention relates to a gear residual life prediction method based on LSTM-AON, and belongs to the field of big data and intelligent manufacturing. According to the invention, the method comprisesthe steps: designing a novel LSTM-AON neural network to predict the residual life of the gear; based on an LSTM network, improving a hidden layer of the LSTM-AON network, so the LSTM-AON network selects an interlayer position where an element with the maximum attention coefficient is located as a hierarchical position by calculating the attention coefficients of related elements and labels; guiding the division of the hierarchical structure by attention and is determined by an element most similar to the label; and embedding the tree-shaped hierarchical structure into the recurrent neural network through the attention degree of the information so as to mine the inter-layer sequence information ignored by the traditional network. Compared with a traditional LSTMs network, the LSTM-AON network adopted by the method is high in calculation speed, small in calculation amount and higher in precision.

Description

technical field [0001] The invention belongs to the field of big data and intelligent manufacturing, and relates to a method for predicting the remaining life of gears based on LSTM-AON. Background technique [0002] Gears are widely used in mechanical equipment and are one of the most widely used mechanical parts. Gears have unique advantages such as high transmission efficiency, compact structure, good transmission smoothness, large carrying capacity, and long service life, which make them have strong and lasting vitality. Under complex working conditions and environments, gears are prone to failure, which may lead to disasters in machine operation and even endanger personal safety. This is especially true for large or very large equipment, such as hydroelectric generators, mining conveying machinery, helicopter power transmission systems, heavy machine tools, etc. The life prediction of in-service gears can effectively determine the maintenance time of equipment, improv...

AI Technical Summary

Problems solved by technology

However, the traditional LTSM network does not analyze the gear degradation data comprehensively enough, and does not mine the order information of the neural network, so the life prediction ability is not good.

Images