Abstract: To further avoid the problem of premature convergence and search stolidity in the process of resolving function optimization, a differential evolution clonal selection algorithm based on the clonal selection theory is proposed which can enhance the operating efficiency and the convergence rate of the differential evolution algorithm and then can both select the best individual and guarantee the population diversity. The simulation experimental results show that the result of the multiple hump function is accurate and the convergence speed is fast.

Key words: numerical optimization, premature convergence, differential evolution, clonal selection algorithm

摘要： 为了解决函数优化过程中的“早熟收敛”和“搜索迟钝”问题，将差分演化算法与克隆选择算法进行了结合，提出了一种新的差分演化克隆选择算法。该算法将克隆选择操作引入到差分演化算法中，达到了既能够选出最好个体又能够保证种群多样性的效果。实验结果表明该算法在多峰值函数优化问题中，具有求解精度较高，收敛速度较快等优点。

关键词: 函数优化, 早熟收敛, 差分演化算法, 克隆选择算法