论文部分内容阅读
通过对目前几种常用仿生优化算法的研究与分析,在基本免疫算法的框架下用C++实现人工免疫算法;通过测试函数研究提高算法性能的关键因素。将疫苗接种、基因重组、动态分库操作、多重变异概率调节机制加入到算法中,并通过大量实验测试,提高了算法性能。并且通过探讨交叉操作对免疫算法的影响,提出一种比较适合免疫算法的交叉算子;进而得出了改进后算法的流程并验证了其性能。最终将改进后的算法应用于机械设计问题实例,并与相应文献比对优化结果,证明了改进后算法的优越性能。
Through the research and analysis of several commonly used biomimetic optimization algorithms, artificial immune algorithm is implemented in C ++ in the framework of the basic immune algorithm. The key factor to improve the performance of the algorithm is studied through the test function. The vaccination, gene recombination, dynamic sub-library operation, multiple mutation probability adjustment mechanism added to the algorithm, and through a large number of experimental tests to improve the performance of the algorithm. And by exploring the effect of crossover operation on the immune algorithm, a crossover operator is proposed which is more suitable for the immune algorithm. Then the flow of the improved algorithm is obtained and its performance is verified. Finally, the improved algorithm is applied to the case of mechanical design problems, and the optimization results are compared with the corresponding documents, which proves the superior performance of the improved algorithm.