基于SA算法的行星齿轮减速器离散变量优化设计

考虑制造工艺要求，将所有设计变量均视为离散变量，包括一般离散变量和伪离散变量，并就这两种情况下状态产生函数的设计原理进行深入研究，解决了将模拟退火算法用于离散变量函数优化的关键技术问题，介绍了一种基于模拟退火算法的离散变量函数优化的新方法。行星齿轮传动中各齿轮的齿数受传动比条件、同轴条件和装配条件的限制而不能任意取值，齿轮的模数也要受国家标准的制约只能取一些离散值，用以数学规划理论为基础的经典约束优化方法求解效果很差，用基于模拟退火算法的离散变量优化设计方法则可以方便快捷地获得满足各方面要求的最优设计方案。

Study on discrete variables optimization design for heliocentric-type reducer based on simulated annealing algorithm

Taking process requirements into account, all design variables are considered as discrete ones including general discrete variables and dummy discrete variables. Further research of the design principle to generate functions in these two situations is carried out, which solves the key technical problem to apply SA algorithm in discrete variable function optimization. Then a new method of discrete variables function optimization based on SA algorithm is proposed, in planetary gear transmission, the number of gear teeth is constrained by transmission by transmission ratio, matching requirements and coaxial requirements. The module of gear can also select some discrete values due to the limits of national standard. The above leads to bad results through classical constraint optimization method based on mathematics planning theory. However, discrete variable optimization design method of SA can easily obtain optimum scheme which satisfies all requirements.