Analysis Of Elevator Traffic Flow Based On Five Tipical Periods

With the increasing of high-rise buildings, the vertical transport has drawn moreand more attentions. However, the research of vertical traffic is mainly concentratedon the elevator group scheduling methods. Traffic flow forecasting, the verticaltransport's another area, is often neglected. Because traffic flow prediction of buildingis a necessary condition for elevator group control simulation and configuration ofbuildings` elevator selection, so forecasting of traffic flow can lead to less waste ofconstruction and social resources, solve problems of low elevator service quality. Toaddress these problems, by summing up valuable prediction methods, this paperproposes own ideas and methods.According to the purpose and application areas of elevator traffic flow prediction,this thesis firstly propose that traffic flow forecast should be divided into twocategories: firstly , the traffic flow forecast before the building`s construction, whichis mainly used in elevator group control simulation system before installation, whichin turn provides the basis for the rational selection and configurations; Secondly,traffic flow forecasting after completion of buildings, it is mainly for elevator groupcontrol method of online or offline amendment, which in turn provides more highquality passenger services and plays an energy-saving effection. This article comesfrom a state fund project which aimes at the first kind of elevator traffic flow analysis.To realize the elevator traffic flow prediction basd on the Monte Carlo methods,this paper combises good interface expression of the VB programming software withexcellent computing performance of the MATLAB software to constitute theframework of the prediction system. At the same time, on the basis of careful study oftraffic flow characteristics of the construction, this paper divides one day's passengerflow into five periods: up peak, down peak, lunch up / down time, randomconventional time and idle time. This system combines the five typical periods withprediction system to reach the final elevator traffic flow prediction.The design method has better forecasting accuracy than traditional methods, andprovides some quality data for the next selection configurations. This paper proposesthe classification of elevator traffic flow prediction, which provides direction for thenext elevator traffic flow prediction. Of course, in this design there are many inadequacies which can be improved byfurther researches in future.