Determination of cutting speed in machining using neural networks.

In machining, it is necessary to know elements such as geometry to be obtained, material to be machined, type of operation, machine tool, cutting tool, cutting depth, feed, cutting speed. These last three quantifiable elements are determined using tables, charts, dedicated computer software, or CAD / CAM system, offering a wide range of choices but lacking transparency and flexibility. The contribution of this article is to apply artificial intelligence techniques based on Artificial Neural Networks (ANN) to the development of a decision system for the choice of cutting parameters. To model the cutting speed, we use a ANN with back propagation algorithm. Experimental values from a source chart will be used to construct and establish an ANN to estimate the value of the cutting speed, using as data a number of influence parameters. The validity of the results obtained shows that this method can be successfully applied and that its use in the manufacturing domain can contribute to optimizing the cutting conditions by a more precise and faster choice of the cutting speed.