改进适应度遗传算法在泵站优化调度中的应用

The Application of Improved Fitness Fine-tuned Genetic Algorithm in Pump Station Optimal Scheduling

为进一步优化泵站多机组联合运行方式、降低泵站能耗,建立了以泵站日耗电量最小为优化目标的调度模型。以遗传算法为模型求解算法,针对遗传算法存在的局部收敛问题,从改进算法适应度函数的角度,设计了基于遗传个体顺序的非线性适应度函数。通过在泵站的实际应用,对比发现:改进适应度遗传算法(FFGA)在算法稳定性上明显优于传统遗传算法(SGA);且FFGA法得出的泵站优化调度方案的日耗电量比SGA法和经验制定的调度方案的日耗电量分别低1.77%和8.07%,具有一定工程实践意义。

In order to further optimize the pump station multi-unit combined operation mode and to reduce the pump station energy consumption,a dispatching model is established to minimize the daily power consumption of the pump station in our study.Aiming at the local convergence problem of genetic algorithm,a nonlinear fitness function based on genetic individual order is designed from the perspective of improved fitness function of genetic algorithm.Through the practical application in pumping stations,it is found that the improved fitness genetic algorithm(FFGA)is obviously better than the traditional genetic algorithm SGA in algorithm stability.The daily power consumption of optimal dispatching scheme of pumping stations by using the FFGA method is 1.77%and 8.07%lower than the SGA method and the practical method,respectively.The results can be useful to practical engineering.