Large-scale grouping optimizing method of parts based on feed characteristic

Abstract

The invention discloses a large-scale grouping optimizing method of parts based on a feeding characteristic. The large-scale grouping optimizing method of parts comprises the specific steps of: (1) grouping the parts based on the similarity characteristic to obtain different similar part groups, and putting each part into a similar part group; (2) regrouping the parts based on the feed fit characteristic, wherein each regrouped similar part group corresponds to a feed optimizing subproblem; and (3) optimizing and compensating the parts, dynamically compensating the distribution of groups of parts in the optimizing process, and finally combining the optimization results of all the groups to obtain a feeding scheme of the original problem. The large-scale grouping optimizing method of parts can be used for grouping parts adaptively, obtaining reasonable and stable part groups, avoiding influence of a specific clustering algorithm on the grouping result, and alleviating the contradiction of the algorithm between time efficiency and material utilization rate.

Description

technical field [0001] The invention relates to a large-scale part grouping optimization method based on blanking characteristics. The method can alleviate the contradiction between algorithm time efficiency and material utilization rate through part grouping optimization, and obtain a stable blanking scheme with high material utilization rate. Background technique [0002] The optimal blanking problem is an NP-complete problem with the highest computational complexity. The increase in feasible combinations brought about by the increase in the number of parts is explosive, resulting in low efficiency in the algorithm search time, and it is easy to fall into a local optimal solution, resulting in loss The practice of exchanging material utilization rate for time solution efficiency has been difficult to meet the application needs of enterprises, resulting in the contradiction between the two issues of time efficiency and material utilization rate in the algorithm. Therefore, ...

AI Technical Summary

Problems solved by technology

(2) Detachability of blanking problem

[0003] Existing research on parts grouping optimization blanking mainly focuses on how to obtain reasonable and stable parts grouping, and most of them use various clustering algorithms to realize the similarity grouping of parts. Due to the limitations of problems such as local optimal solutions, the grouping results are more dependent on the specific clustering algorithm, and the adaptability to part data changes and the stability of the grouping results are not ideal.

Images