Construction and Simulation of Failure Distribution Model for Cycloidal Gears Grinding Machine

To increase reliability, reduce machine tool failure, and shorten maintenance time, serval failure distribution models are discussed based on the cycloidal gears form grinding. A fault data expansion algorithm model radial basis function (RBF) neural network is proposed to accurately evaluate reliability of grinding machines. The uses a self-organizing clustering learning determine RBF centers constants network. It trains using algorithm’s output imports randomly generated cumulative obtain simulation data. numerical example researching presented which shows that estimated value mean time between failures (MTBF) 909.20h, 0.4874. Finally, tree analysis-analytic hierarchy process (FTA-AHP) used analyze gear machine. evaluation indexes obtained by analyzing corresponding characteristic function, proves valid. has important guiding significance in evaluating performance large computer control (CNC)