Study of the load on best value based on the minimum wear in sliding electric contact system

As in Catenary-Pantograph system, the amount of wear caused by the sliding contact is affected by the contact force. When the load is smaller, there will be erosion resistance wear; at the other hands, there will be mechanism wear when the load is bigger. So some experiments using carbon sliding panel infiltrated with copper have been taken to find the relationship between the load and wear. In these experiments, there were a lot of data at the condition of different values of current and velocity. And then the data will be trained using the Bp neural-network method, whose input is the load and output is the wear data. Also black-box model should be validated using the other data. Base on this, an optimize arithmetic will be designed on the black-box model to find the best value of the load, and that will make the wear minimal.