论文部分内容阅读
针对同时存在整型变量及连续变量的换热网络优化问题,提出一种多子群协进化的粒子群算法。为了增强粒子群算法的全局搜索能力,将种群按精英个体、一般个体、较差个体划分为3个子群,针对每个子群的粒子进化状态提出不同的学习算子,用于丰富粒子的进化方式,增加种群多样性;同时建立协进化机制,动态地更新子群,以实现粒子之间的良性竞争,更好地引导粒子进化。采用结构优化策略处理整型变量,并与多子群协进化的粒子群算法结合,实现了连续变量与整型变量的同步优化。通过两个优化实例验证算法的性能,优化结果表明了新方法的有效性。
Aiming at the optimization problem of heat exchanger networks with integer variables and continuous variables, a multi-subgroup co-evolutionary particle swarm optimization algorithm is proposed. In order to enhance the global search capability of particle swarm optimization, the population is divided into three subgroups according to elite individuals, general individuals and poor individuals. Different learning operators are proposed according to the evolution state of particles in each subgroup, which is used to enrich the evolution of particles , To increase the population diversity. At the same time, a co-evolutionary mechanism was established to dynamically update subgroups to achieve healthy competition among particles and better guide particle evolution. The structural optimization strategy is used to deal with integer variables, and combined with multi-subgroup co-evolutionary particle swarm optimization to achieve the synchronous optimization of continuous variables and integer variables. The performance of the algorithm is verified by two optimization examples. The optimization results show the effectiveness of the new method.