Research On Multi-objective Intelligent Optimal Scheduling Strategy Of Elevator Group Control System

The development of artificial intelligent provides a vast space for elevator group control system(EGCS). To follow the modernization city and intelligent architecture, integrated intelligence EGCS has become the recent development direction. Swarm intelligent algorithm(SI), is produced by simulating the biological self-organization of gregarious colony, and many research achievements demonstrate its unique advantages and wide application prospect. To improve EGCS and enlarge the application of SI, this paper deeply discusses and researches the scheduling strategy based on particle swarm intelligent algorithm. The innovation work in this paper is presented as follows:1) EGCS is a complex nonlinear system that includes multiple control objectives. Considering the service quality,service quantity and energy-saving, using multi-objective optimization, the multi-objective optimization model of EGCS is built which combines AWT,ART,LWP,CRD,RNC.2) After analyzing and studying the characteristics of particle swarm optimization algorithm (PSO), a new scheduling strategy is proposed based on PSO with linearly decreasing weight (LDW-PSO), using its cooperative search and information sharing mechanism. Comparing with the AWT scheduling strategy under the different traffic mode, LDW-PSO shows better scheduling results.3) By introducing the simulated annealing (SA) algorithm to the PSO, a hybrid algorithm PSO-SA is proposed which is able to change the weakness of PSO easily fall in the local best solution by using SA's probability selection mechanism and global convergence. Comparing with the LDW-PSO scheduling strategy under the different traffic mode, PSO-SA algorithm shows better optimization results. Meanwhile, it demonstrates that combining different complementary algorithms could reach better result, it is also a research direction of artificial intelligence optimization algorithm.