Application of Sensitivity Pruning Neural Networks in Surface Roughness Prediction

The surface roughness is a key parameters in high speed machining and often hard to control. The prediction model for surface roughness was created based on artificial neural networks which have strong non-linear modeling ability. The sample data collection method was analyzed and BP neural networks was designed, but the traditional BP neural networks has many shortcomings like easily step into local minimum, with weak generalization ability and the middle layer neuron are hard to determine, so the sensitivity pruning algorithm applied. The simulation shows the method is effective and can provide a guidance to optimize cutting parameters and control surface quality.