Computer aided optimum design of friction materials using uniform design

摘要

The application of Uniform design in friction material formulation and a multi-layer back propagation artificial neural network (BPANN) methods of Data Handling are investigated for modeling and prediction of the wear properties of composites. The friction and wear properties show that Uniform design can distribute the experiment points evenly so as to have fewer number of tests. With computer aided artificial neural network (ANN) modeling and optimization technology, the relationships between factors and properties can be predicted.