基于免疫进化的野草优化随机搜索算法

An Invasive Weed Optimization Algorithm Based on Immune Evolution

针对野草优化随机搜索算法中存在的不成熟收敛问题和易陷入局部极值的缺陷,提出了一种基于免疫进化的野草优化随机搜索算法。该算法引入免疫进化理论对野草种群中的最优个体进行免疫进化迭代计算,并且充分利用最优个体引导不同野草个体进行局部搜索和全局搜索,能够有效避免算法陷入局部极值,并且以更高的精度逼近全局最优解。实验通过对4种典型Benchmark测试函数进行数值寻优曲线对比与平均最优解对比,结果表明,相比于遗传算法、粒子群优化算法与传统的野草优化随机搜索算法,该算法具有更好的寻优能力、稳定的效果和更快的收敛速度。

Aiming at the limitations of easily falling into local minimum and premature convergence in invasive weed optimization （IWO）,we propose a modified invasive weed optimization algorithm based on immune evolution. The theory of immune evolution is introduced intoIWO for immune and evolutionary iteration computation to the optimal solution that is applied to guide different weeds in global search andlocal search,which can be free from falling into the local optimum and be close to the global optimal solution with higher precision for thealgorithm. Through numerical optimization curve contrast and average optimal contrast with four kinds of typical Benchmark functions,theexperiments show that the proposed algorithm has better optimal searching ability and stability as well as faster convergence than those ofbasic IWO.