Solving hydro unit commitment problems with multiple hydraulic heads based on a two-layer nested optimization method

doi:10.1016/j.renene.2021.02.126

Return

Solving hydro unit commitment problems with multiple hydraulic heads based on a two-layer nested optimization method

2021-07-01

PRE

DOI：10.1016/j.renene.2021.02.126

Original

Original request for help

Save

Abstract

En 中文

Optimizing the hydro unit commitment (HUC) has the potential to increase water use efficiency. How-ever, the large number of variables, along with some nonlinear and non-concave constraints, present a huge challenge to the efficient solution of the HUC problem. For hydropower stations including multiple power plants that have different hydraulic heads, the problem is even more difficult. In this paper, we address this complex HUC problem using a two-layer nested optimization method. First, the total load demand is allocated to each power plant using a novel distribution rate. Thus, in the outer layer, a cuckoo search algorithm optimizes the distribution rate and each unit?s on/off status, while in the inner layer, the load demand of each plant is allocated to online units through dynamic programming. China?s Gezhouba Hydropower Station is selected for a case study. It is shown that: (1) the optimized water consumption is, on average, 1.71% lower than that of actual operations; (2) the optimized status of units vary smoothly with fewer startup/shutdown events; and (3) the optimization performance in non-flood seasons is better than that in flood seasons. These findings verify that the two-layer nested optimization method is an effective means of solving HUC problems for hydropower stations including multiple hydro plants with different hydraulic heads. ? 2021 Elsevier Ltd. All rights reserved.

Keywords:

Hydro unit commitment

Hydropower station

Hydraulic head

Two-layer nested optimization method

AI Summary

Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.