(A new evaluation function of elevator group control system (EGCS) is proposed, and the evaluation standard of elevator office is generally considered and corresponding fuzzy rules are established. Together, these rules are mapped to the fuzzy neural network for learning, which improves the intelligence of scheduling. In order to ensure the physical meaning of the weights of each layer of the neural network and the importance of improving the rules, a partial layer learning algorithm is proposed, and the center value and width learning algorithms are deduced.)
收藏
