Research On Optimization Control Strategy Of Elevator Group

In large buildings, the single elevator can no longer meet the requirements oftransportation efficiency, service quality, and so on. As a result, elevator group controldispatching has gradually become the mainstream technology for passenger flowoptimal transportation. This paper takes elevator group control system composed of fourelevators as the research object, carrying out the study of group control algorithm andcontrol strategy. The main work of the paper includes:1. This paper analyzes dynamic characteristics, traffic pattern and performanceevaluation index of elevator group control system. On this basis, discusses the necessityand feasibility about introducing genetic algorithm into elevator group control system.The objective evaluation function of group control system is regarded as weightedcombination of passengers’ waiting time, riding time and system energy consumption.The weight of each index in different traffic pattern can be achieved by fuzzy reasoningmethod, thus determining the best-match objective function with current traffic pattern.2. Combined with actual condition of the building, the thesis establishes passengerflow analysis model with mathematical statistics theory including Poisson distribution,Monte-carlo sampling method, and so on. In a period of time, we can obtain thepassenger flow data such as arrival time, origin floors and destination floors for allpassengers.3. In this paper, genetic algorithm is used to search the best solution for objectiveevaluation function, which realizes the multi-objective optimization control. In order tospeed up search efficiency, the best individual preservation strategy is applied in geneticoperating process. And then the adaptive algorithm is proposed, in which the fitnesscould be adjusted with crossover rate and mutation rate, increasing convergence speed of the algorithm.4. By means of Matlab, design and implementation of elevator dispatching methodare completed based on the improved genetic algorithm. Its feasibility is also proved.On this basis, simulation software of elevator group control system is developed withVisual Basic language. Users could set simulation parameters of building-elevator,passenger flow generation, and so on. Both the basic genetic algorithm and theimproved genetic algorithm are executed respectively on this software platform. Thesimulation results show that more reasonable scheme for elevator dispatching can beacquired with multi-objective combinational optimization method based on theimproved genetic algorithm. This method effectively improves operation efficiency ofsystem and passengers’ satisfaction.