Numerical control machine tool cutter wear monitoring method based on deep learning

Abstract

The invention belongs to the technical field of numerical control machine tool cutter wear monitoring, and particularly discloses a numerical control machine tool cutter wear monitoring method based on deep learning. The method comprises following steps that a three-phase current signal of a numerical control machine tool spindle motor is collected, a current signal corresponding to a cutter to bemonitored is cut out from the three-phase current signal, the current signal is segmented and is divided into M segments of current signals, and root-mean-square value of each segment of current signal can be calculated; each segment of current signal is subject to structured treatment, the structured current signals are input into a sparse automatic coding network to be trained, and a final coding vector obtained after training each time serves as a feature vector output; K value between feature vectors is calculated, and according to the K value, a K value fitted curve is obtained, and according to the K value fitted curve, monitoring of the cutter wear is achieved. Various cutter wear states under different machining conditions can be rapidly and accurately recognized, and the beneficial effects of being high in monitoring and diagnosis precision, high in monitoring real-time performance and high in adaptation are achieved.

Description

technical field [0001] The invention belongs to the technical field of tool wear monitoring of CNC machine tools, and more specifically relates to a method for monitoring tool wear of CNC machine tools based on deep learning. Background technique [0002] Tool wear monitoring of CNC machine tools refers to that in the process of product processing, the computer judges and predicts whether the tool is worn or not by detecting the signal changes of various sensors. The essence of the tool wear monitoring process is a pattern recognition process. A complete tool wear monitoring system consists of research objects (tools), processing conditions, sensor networks, signal processing, feature extraction, and pattern recognition. [0003] Due to the inevitable wear of the tool during the processing, it will directly affect the utilization rate of the machine tool and the quality of the workpiece. The light one will cause the quality of the processed workpiece to decline, and the seve...

AI Technical Summary

Problems solved by technology

[0004] After decades of development, tool monitoring technology has reached a certain level in breadth and depth, but so far there is no method that can be applied to different processing conditions and monitor various tool wear. The scope of application of some methods is limited, and they are far from meeting the requirements of automation and intelligence, and there are still certain limitations in practical application, such as insufficient real-time monitoring and false positives.

Images