Research On Optimization Of RGV’s Dwell Point Policy In Planar Mobile Stereo Garage

With the increase of car ownership,the problems of vehicle congestion and difficult parking are becoming more and more serious.The emergence of automatic three-dimensional garage has well alleviated such problems,not only increased the utilization of resources,but also brought convenience to people’s parking.Plane mobile three-dimensional garages occupy a high market share with the advantages of high capacity and high automation level.The research on the control strategy of its rail guide vehicle(RGV)is helpful to improve the operation efficiency of the garage.Among them,the dwell point policy is related to the response time of task execution and affects customer satisfaction.Designing an appropriate dwell point policy is very important to reduce the storage(retrieval)time and improve the efficiency of Plane mobile three-dimensional garage.Based on the research of many scholars on dwell point policy,the research contents of this dissertation are as follows:(1)The actual model of plane mobile three-dimensional garage is constructed,the RGV operation environment model is established by using the topological map method,and the operation process of vehicle access is analyzed.According to the position transfer of RGV and the cooperation of elevator,the operation process of vehicle access is divided into two operation stages of no-load and load,six different scenarios,and a mathematical model with the shortest service time of current task is established to determine the mean efficiency evaluation index from the perspectives of customers and operators;(2)The design idea of dwell point policy based on the analysis of customer access characteristics is determined,and the distribution of garage customer arrival process and parking time is studied.Based on the historical parking data of a commercial three-dimensional garage in Xi’an,the non-homogeneous Poisson arrival process of customer arrival is proved by K-S test,and the average arrival rate function of customers is obtained.The non-homogeneous Poisson process of customer arrival is proved to be closer to the reality by goodness of fit test.In addition,the distribution of customer parking time is fitted,and it is obtained that the vehicle parking time follows the normal distribution,By comparing the simulation statistics of the error between the number of storage(retrieval)customers and the actual number of storage(retrieval)customers in each hour of the day,the results show that the customer access arrival process replaced by the non-homogeneous Poisson arrival process and the customer parking time subject to normal distribution is consistent with the reality;(3)Use the fitting results of customer arrival process and parking time to predict the arrival time of customer access vehicles,take a certain proportion of predicted outbound customers in chronological order,take the intersection of the path from idle RGV to each predicted outbound customer vehicle as the outlet dwell point,determine the dwell point policy designed in this paper,and analyze the application scope of the design strategy.On this basis,increase the consideration of RGV load balance,The double objective function of minimizing the sum of RGV service time and each RGV load is established and solved by artificial bee colony algorithm.The results show that the dwell point policy designed in this paper is suitable for the low level of customer arrival rate,conforms to the basic principle of dwell point design,and the load balance of each RGV has been significantly improved when the average waiting time of customers does not change much;(4)In the process of parking space allocation,under the condition of dynamic operation,the conflict of each RGV path and the coupling relationship between the next task position and the parking space allocation strategy are analyzed to realize the joint optimization of each scheduling strategy of plane mobile three-dimensional garage.Based on the design of standby space strategy,the parking space allocation strategy under the condition of dynamic operation is designed and solved by simulated annealing algorithm,Compare the performance of the design allocation strategy and the nearest parking space allocation strategy in the dynamic operation process.The results show that in the dynamic operation,the efficiency indexes of the system under the nearby parking space allocation strategy increase significantly,and when the average waiting time of customers is approximately equal,the other efficiency indexes under the parking space allocation strategy designed in this thesis is better,which verifies the effectiveness of the designed parking space allocation strategy,and provides an idea for the research and optimization of relevant scheduling strategies of three-dimensional garage.