基于人工免疫系统的多目标函数优化

为克服传统遗传算法退化和早熟等缺点，同时降低优化算法的复杂度，提出基于人工免疫系统（Artificial Immune System， AIS）实现无约束多目标函数的优化。使用随机权重法和自适应权重法计算种群个体的适应值，使Pareto最优解均匀分布的同时，加快算法的收敛；通过引入人工免疫系统的三个基本算子：克隆、超变异和消亡，保持种群的多样性；在进化种群外设立Pareto 解集，保存历代的近似最优解。使用了两个典型的多目标检测函数验证了该算法的有效性。优化结果表明，基于AIS的多目标优化算法可使进化种群迅速收敛到Pareto前沿，并能均匀分布，是实现多目标函数优化的有效方法。

Multi-objective optimization based on artificial immune system

To overcome shortcomings of traditional GA(Genetic Algorithm) and reduce the complexity of optimizing process,Artificial Immune System(AIS) was introduced to optimize multiple conflicting objective functions.Both random weight method and adaptive weight method were used to calculate individual fitness,which not only made solutions distributed on the front of the Pareto solution equably,but also let this algorithm convergent more quickly;AIS accelerated the convergent speed and kept the group diversity;a new method was used to set up a Pareto set to keep the approximate Pareto solutions in every evolving population.Two typical multi-objective testing functions validated the algorithm at the end of the article.Results indicate that the approach is competitive and it can be considered a viable alternative to solve multi-objective optimization problems.