Planning And Design Of Auto Store Intensive Storage System And Scheduling Optimization

Auto store intensive storage system is a new storage system which integrates intensive shelves and AGV carrying cars.Compared with the existing system,it has certain advantages in storage space utilization and operation efficiency.At present,scholars at home and abroad mainly focus on shuttle intensive storage system,including: batch or combination of storage tasks,optimization of storage AGV resource allocation,storage location zoning and scheduling path optimization.At present,there are few researches on "goods to people" storage system.This paper takes this as the research object,and studies the optimization of resource allocation,task allocation and path planning in the storage system.In the research process,some methods are improved adaptively.The main research contents are as follows:The Petri net model and Flexsim simulation model are established according to the operation process of the storage system,and the optimal configuration relationship between the number of AGV and the number of inbound and outbound work stations is analyzed.And through the establishment of mathematical model,the paper analyzes the optimal quantity calculation formula,as well as the impact on the optimal quantity allocation of the storage system AGV and operation platform when the shelf height and order frequency of the system change.The relationship between them is deduced by mathematical formula and verified by simulation software.Aiming at the problem of task assignment in AGV scheduling,a mathematical model is built.An improved adaptive multi group genetic algorithm is designed to solve the team problem.The main improvements are:adding the algorithm to judge the distribution of the generated initial solution,making the initial solution more evenly distributed in the solution space,ensuring the diversity of the algorithm and avoiding falling into the local optimal solution.Secondly,for the solution with low fitness,the probability of cross mutation is increased.If the solution does not reach the average fitness,another probability range is increased to speed up the convergence.Through an example,the feasibility and effectiveness of the improved multi population genetic algorithm are verified.This paper analyzes the path planning problem of storage multi AGV system,puts forward a two-stage A* algorithm based on topological graph weight,and designs a new heuristic function,which not only reduces the search amount of the algorithm to the path nodes,improves the algorithm speed,and reduces the possibility of AGV traffic conflict.The conflict type prediction mechanism is proposed for traffic conflict,and the conflict resolution strategy is designed according to different conflict types.In the multi AGV conflict,the greedy algorithm is designed to solve the strategy set to reduce the impact of the conflict strategy on the subsequent AGV traffic and minimize the waiting time cost of AGV conflict.The experimental scheme is designed to verify the improved A* algorithm and the conflict resolution strategy.