Efficient Operation of Integrated Electrical-Water System for Wind Power Accommodation

Abstract

This paper investigates the operation of an integrated electrical-water system for wind power accommodation. The efficient operation can be considered as how well to handle the binary variables in a water distribution system (WDS) model. The optimization problem of WDS is NP-hard, and it often takes hours to find the optimal solution. Since wind power varies in minutes, long computation time may hinder the WDS from being fully used by power system to deal with the wind power fluctuation and uncertainties. This paper develops a three-step optimization model to answer the challenges of the WDS optimization problem. The solution hinges on the calculation of the binary variable values in advance. The first step is to construct a linear power-WDS co-optimization (LPWCO) model to find the optimal electrical demand of WDS. The second step develops a tank flow allocation based method and a greedy-based algorithm to determine the binary variable values of WDS. This allows the WDS optimization model to be efficiently solved by commercial solvers in the third step. The proposed three-step model converts the NP-hard optimization problem of WDS into a polynomial-time problem which can be solved quickly. This allows the wind power system to leverage the energy flexibility provided by WDS in the real-time operation. Results of case study show that, compared with existing methods, the proposed method can reduce the wind power curtailment and the operation cost by about 55% and 53%. More importantly, the solution time is reduced dramatically from hours to seconds, suitable for practical application.