Combinatorial Optimization Method for Operation of Pumping Station with Adjustable Blade and Variable Speed Based on Experimental Optimization of Subsystem

Abstract

A decomposition-dynamic programming aggregation method based on experimental optimization for subsystem was proposed to solve mathematical model of optimal operation for single pumping station with adjustable blade and variable speed. Taking minimal daily electric cost as objective function and water quantity pumped by units as coordinated variable, this model was decomposed into several submodels of daily optimal operation with adjustable blade and variable speed for single pump unit which was solved by experimental optimization. The constructed aggregation model took water quantity pumped by each pump unit as decision variable and discrete values of water quantity pumped by pumping station as state variable and was solved by one-dimensional dynamic programming. Taking operation of typical pumping station as a study case, optimal operation with adjustable blade and variable speed, respectively, had an average cost saving of 4.19%, 22.15%, and 29.86% compared with operation with fixed blade angle and constant speed under 100%, 80%, and 60% load, which also had a remarkable saving amplitude of 15.85% and 24.18%, respectively, corresponding to 80% load and 60% load compared with operation with adjustable blade and constant speed. Meanwhile, the proposed method has provided a new way for solving complex nonlinear mathematical models with 3 decision variables.