High-speed Wire Edm Wire Cutting Process Simulation

At present, Machining process parameters on the effect of forecast and automatically optimize is a key issue. Solve the difficult choice of process parameters, the key is to establish a reasonable and effective WEDM model, and apply this model to achieve the overall effect of the scope of the forecast process.As a new type of neural network modeling tools, Neural networkis a neuron by a number of large-scale cross-link the composition of parallel distributed information processing system with self-organizing, self-learning and good fault tolerance, system complexity can be used to simulate realistic input and output relations, has a strong non-linear modeling capability. In this paper, combining the characteristics of artificial neural networks using WEDM process simulation model.By WEDM process analysis of the law, In this paper, the impact of process to select the main effect of four factors: the thickness of the workpiece, the average gap current, pulse width, pulse interval as the input parameters in order to constitute a sample of RBF network training. Select the processing speed, surface roughness of the two performance indicators as the objective function RBF network. Universal adoption of the second rotation design number of tests designed to determine the pilot program. RBF networks use a sample database to the new study, the corresponding process will be the effect of prediction model. At the same time,use GA to optimize processing parameters, it can be concluded under the conditions of the corresponding processing parameter optimization model.