ABNORMALITY DETERMINATION DEVICE, SIGNAL FEATURE VALUE PREDICTOR, ABNORMALITY DETERMINATION METHOD, LEARNING MODEL GENERATION METHOD, AND LEARNING MODEL

摘要

Provided are an abnormality determination device capable of precisely determining if there are abnormalities according to an operation condition, and a signal feature value predictor, abnormality determination method, learning model generation method, and learning model. This abnormality determination device comprises: an input unit which receives as input a first signal output from a first sensor regarding the operation of a specified power transmission device and a second signal from a second sensor which is attached to sense abnormalities in the power transmission device; an operation condition identifying unit which identifies the operation condition of the power transmission device on the basis of the first signal; a feature value prediction unit which, according to the identified operation condition, predicts the feature value of the second signal output from the second sensor for the power transmission device in a normal state; and a determination unit which determines whether an abnormality has occurred on the basis of the feature value of the second signal predicted by the feature value prediction unit according to the operation condition of the power transmission device identified on the basis of the first signal in a determination period and the feature value of the second signal received as input by the input unit in the determination period.