Flexible Expert System for Automated On-Line Diagnosis of Tool Condition

摘要

An increasing number of Flexible Manufacturing Systems (FMS) have been installedin Europe during the past few years. The general experience is that the availability of the FMS is not as high as was originally expected and especially their unmanned use during three shifts has not been successful. One of the major problems is the deterioration or failure of the tools. To develop a system to address this, a wide cutting test and analysis program for tool wear was performed. The test program covered both shank end and end mills together with twist drills and tread taps. For monitoring the tool wear, a number of monitoring methods such as vibration, acoustic emission, sound, spindle power and current, axial force, torque were tested. The relations between the analysed signals and tool wear form a basis for the diagnosis rules that are used in an diagnostic expert system module. An expert system for automated on-line diagnosis of tool wear of different types of tools was built using a new approach. In this approach, the faults are described in a fault tree database and the corresponding features of condition monitoring signals together with the machine status information are described in a symptom tree database. Using a rule synthesizer program the information gathered in the databases is automatically converted to expert system code.