Research On Assembly Sequence Planning Based On Differential Mutation Bat Algorithm

Product assembling is the final stage and the most critical link throughout themanufacturing process, the level of the establishment of assemble process willdirectly affect productivity and cost of product. Among them, assembly sequenceplanning is the core in assembly process and it also has been proven to be an NPcombinatorial optimization problem. How to find out the reasonable andcost-effective assembly sequence has become a hot topic in the field of productassembly research. This paper takes the product assembly sequence planning as theobject of study, has conducted the thorough research on the evaluation index, theassembly information extraction technology and optimization method, providetechnical support for the realization of the automatic planning of assembly sequence.The paper mainly completed the following research work:To reduce the complexity of the assembly problem, the assembly process hasbeen simplified reasonably. For lack of a complete assembly sequence evaluationindex of the past, continuity and stability have been added to the evaluation system asthe same as the geometric feasibility, retargetability and polymerizability, and hasproposed five assembly information matrix suitable for computer storage andexpression of each index is calculated to quantify.Has researched on the assembly information extraction based on Solidworkssecondary development, by generating assembly tree to achieve the extraction ofassembly and parts feature information, and proposed automatic extraction methodsand specific processes of each assembly information matrix.Based on the imperfection of the bat algorithm, bring in the variation and theselection mechanism in the differential evolution algorithm, differential mutation batalgorithm has been proposed. The bat speed and position coding method andalgorithm for various related operations have been redefined, and have researched theapplication step of the differential mutation bat algorithm in the assembly sequenceplanning. Experimental results show that the differential mutation bat algorithm isbetter than the basic algorithm and genetic algorithm.Based on the proposed assembly sequence planning method, using C++programming language, has researched the assembly sequence planning prototypesystem development by Solidworks platform, and has verified the correctness of the assembly sequence planning method by assembling examples, to lay the foundationfor realizing the automation of assembly sequence planning.