禁忌免疫网络算法及其在函数优化中的应用

Application of a Tabu immune network algorithm in function optimizations

基于人工免疫网络算法(aiNet),借鉴禁忌搜索算法的机制,提出一种禁忌人工免疫网络算法(TS-aiNet).在算法中引入禁忌表,禁忌那些在网络迭代中亲和度不再增加的细胞,并通过特赦准则赦免一些被禁忌的优良状态,增加一个记忆表,用于保存成熟的记忆细胞,改进了高斯变异方式,以保证多样化的有效搜索.通过对多个典型系统仿真分析该方法的收敛性,并与克隆选择算法和aiNet算法进行比较分析.结果表明,该算法在多模态搜索空间中具有更好的全局收敛性、稳定性和寻找极值点能力,能够克服早熟现象,是一种有效的全局优化搜索方法.

A Tabu search artificial immune algorithm （TS-aiNet） was developed based on the aiNet and Tabu search algorithms. It introduces a taboo list of cells whose affinities are to no longer increase in network iterations, and releases some excellent tabooed cells in line with amnesty criteria. A memory table is added to store mature memory cells. Moreover, expressions of Gaussian mutation for a diversity search in the process of global optimization were improved. Convergence analysis was performed with some typical systems and comparison was made with KLONALG and aiNet algorithms. The simulation results showed that the approach presented has better global convergent ability and stability in multi-modal search space, and can avoid prematurity effectively. So it is a global optimization algorithm with good feasibility and high efficiency.