Research on multi-objective-multi-elevator coordinative strategy based on neural network

Abstract

With rapid development of industry, more and more attention has been paid to the effective control of elevator in highrising buildings. This paper studied the elevator control problem, proposed a multi-objective-multi-elevator cooperative scheduling strategy based on the BP neural network, constructed the BP neural network model with four dimensional indexes such as time, energy consumption, comfort level and efficiency, established the elevator's dispatching decision model and obtained the optimal strategy in each period of a day. Finally, the accuracy and feasibility of the model are verified by taking six ten-floor elevators as the specific control objects. The verification results shows that the proposed algorithm is accurate, efficient and flexible.