A Method and an Apparatus for Generating Degradation Data under Field Operating Conditions to Predict Remaining Useful Life Based on an Accelerated Life Testing Data

摘要

The present invention relates to a method for generating operation load data, capable of generating degradation data to predict the remaining useful life in a system in an actual operation environment using accelerated life testing data. According to an embodiment of the present invention, the method for generating operation load data to predict failure, using the accelerated life testing data includes: a step of generating multiple regression models by each load using an artificial neural network by receiving an accelerated life testing data set of at least three first loads among multiple loads; a step of calculating an inclination of an inverse power model by the multiple regression models about the received accelerated life testing data set of at least three first loads based on an average value of results calculated based on the accelerated life testing data of various degradation levels, and generating multiple sample y sections using Bayesian approach; a step of selecting at least three second loads which are the closest to the actual operation load among the multiple loads formed at constant intervals, and generating multiple regression models through an artificial neural network by the first load, the multiple regression models about the accelerated life testing data set, and the sample y section; and a step of generating the degradation data about an actual operation load based on one among the multiple regression models generated about the second load.