FAULT PREDICTION METHOD, FAULT PREDICTION DEVICE AND FAULT PREDICTION PROGRAM

摘要

PROBLEM TO BE SOLVED: To easily and accurately find a fault sign of machinery equipment.;SOLUTION: A fault prediction method includes a collection step, an extraction step, a first learning step, a second learning step, an evaluation step, and a determination step. In the collection step, sensor data of a plurality of sensors provided in machinery equipment is collected. In the extraction step, a portion for a prescribed period of normalcy when the machinery equipment is in a normal state, of the sensor data is extracted. In the first learning step, a regression model which outputs a predictive value for sensor data at a reference date is generated. In the second learning step, an inter-sensor correlation model is generated. In the evaluation step, predictive errors at an evaluation date are acquired with respect to at least some of the sensors, and a degree of deviation from the normal state of the machinery equipment is calculated on the basis of an output value of the correlation model which is obtained by inputting the predictive errors to the correlation model. In the determination step, a fault sign of the machinery equipment is determined on the basis of the degree of deviation.;SELECTED DRAWING: Figure 2A;COPYRIGHT: (C)2018,JPO&INPIT