Dependability evaluation of numerical control machine based on fuzzy neural networks

For realizing comprehensive and accurate dependability evaluation of the numerical control (NC) machine, a dependability evaluation model based on fuzzy neural network is proposed in this paper. The reliability indexes are taken as the evaluation indexes, and the dependability of NC machine is evaluated by the fuzzy comprehensive evaluation method. Then a dependability evaluation model based on the adaptive network based fuzzy inference system (ANFIS) is established. The reliability indexes are taken as the inputs of the model, and the fuzzy comprehensive evaluation results are taken as the outputs of the model. The hybrid arithmetic, which formed by the back propagation (BP) arithmetic and least square arithmetic, is taken as the learning arithmetic of the model. After 20 steps training, the training error of the model reduced to 1.4819 × 10^-6, and the membership functions of the inputs are auto-adjusted. The artificial factors in the evaluation process are avoided.