采用落后个体更新策略的换热网络优化方法

Heat Exchanger Network Optimization Method Using Backward Individual Update Strategy

针对强制进化随机游走算法(RWCE)优化换热网络的过程中存在部分个体优化质量长期处于落后状态的现象,提出了一种落后个体更新策略的RWCE算法。通过周期性考察所有个体性能变化,并实时计算当前所有个体的平均性能,建立了一种落后个体识别机制。对落后个体采取强制更新策略,强制改变其原先的优化路径,增强了种群活力,提高了算法的全局搜索能力。将该算法应用于无分流换热网络的20股流和16股流实例中,年综合费用较改进前的最优结果分别下降了1 005和19 084$/a,表明该策略的有效性,由此证明改进后的算法能有效改善因落后个体所导致的优化质量下降的问题。

Aiming at the phenomenon that some individuals’ optimization quality is lagging for a long time in the process of forced evolution random walk algorithm optimization of heat exchange network,a RWCE algorithm for lagging individual update strategy is proposed.The algorithm establishes a backward individual identification mechanism by periodically examining the performance changes of all individuals and calculating the average performance of all current individuals in real time.At the same time,a compulsory update strategy is adopted for backward individuals to forcibly change their original optimization path,which enhances the population vitality and improves the global search ability of the algorithm.Finally,the algorithm was applied to the 20 SP and 16 SP instances of the splitless heat exchange network,and the annual comprehensive cost was decreased by 1 005 and 19 084 $/a,respectively,indicating the strategy effectiveness.This indicates that the improved algorithm can effectively improve the problem of optimization quality degradation caused by backward individuals.