Prediction of the eigenfrequency of industrial robots based on the ANN model

Abstract

When industrial robots are used in various machining processes, higher requirements for processing accuracy and performance stability are put forward. Low-frequency chatter occurs in the machining process. Because of the articulated mechanism of industrial robots, the low frequency varies in the whole workspace. This is related to the eigenfrequency of industrial robots. To investigate the low-frequency distribution, the first eigenfrequency of an industrial robot is obtained through the impact modal experiment, which is used to evaluate the machining field and robot pose. To obtain the eigenfrequency distribution of the workspace with limited experimental data, an artificial neural network (ANN) is used to establish a prediction model. The structure and prediction accuracy of the ANN model are discussed.