This paper describes a multitooth tool diagnosis application to be used in the car industry. The main focus is on the optimal variable selection (noise, vibration, temperature, and tool drives electrical power consumption) of the tool’s environment, and the signal processing by means of the segmentation of the electrical power consumption signals into groups of inserts according to the type of tools studied. The fault detection algorithms are based on statistical analysis of the spindle tool power consumption. Different statistical parameters are used in a change detection algorithm, while keeping in mind the need for a reliable and low-cost fault diagnosis system. Multitooth tools add an important degree of difficulty to the fault detection problem as opposed to simple tools because of the complexity introduced by the high number of inserts that the workpiece is machining at the same time, for different kinds of finishing and operations.
Tool breakage detection Multitooth tools Change detection Electrical power consumption
This is a preview of subscription content, log in to check access.