Tool Breakage Detection in Milling Operations using a

Negative-Selection Algorithm.

Dipankar Dasgupta and Stephanie Forrest

Dept. of Computer Science

University of New Mexico

Albuquerque, NM 87131

Technical Report No. CS95-5

August 15, 1995

Abstract

Detection of tool breakage is very important for automated machining operations. This paper presents a negative-selection algorithm for tool breakage detection. The method is inspired by the defense mechanism of the immune system, which discriminates between self and non-self. Here self is defined to be normal cutting operations and non-self is any deviation beyond allowable variation of the cutting force. The proposed algorithm is illustrated with a simulation study of milling operations and the performance of the algorithm in detecting the occurrence of tool breakage is reported. The negative-selection algorithm detected tool breakage in all the test cases.

1 Introduction

Manufacturers are always looking for ways to improve productivity without compromising on quality of manufacturing processes. To this end, much attention has been directed towards automated manufacturing. In drilling or high-speed milling industries, on-line monitoring of the tool breakage is a key component in unmanned machining operations.

In most milling industries, a reliable and effective tool breakage detection technique is required to respond to unexpected tool failure [1]. In particular, such a monitoring technique is necessary to prevent possible damage to the workpiece and the machine tool or to avoid production of defective parts and possible overloading of tools. The normal operation of a milling cutter is often characterized from the measurements of some parameters that are correlated with tool wear. It is essential to detect the occurrence of abnormal events as quickly