An Approach to Intelligent Machining

Machining has traditionally been one of the major operations within most manufacturing systems and therefore, has been intensively investigated from many points of view over the past several decades. Many of the machining control techniques developed have not been successful, due primarily to inadequate sensor information and machining process models which do not sufficiently reflect the complexity of the process. In this paper, machining control is treated as a decision-making problem, using an approach similar to the way in which a human would make "wise" decisions. This approach can be concisely expressed in a decision-matrix format where the elements of the matrix represent the evaluation of alternatives with respect to appropriate criteria. The elements of the matrix are generated by several independent sensor-based models. Although the process models may not be individually accurate, in combination they may provide sufficient information to make intelligent decisions. A force-based model suitable for this approach has been developed. The model determines independent force components due to changes in the feed rate and due to tool wear.