Fast Cluster Optimization Method for Concentrated Solar Power Plants Toward Power Systems Under High Renewable Penetration

Abstract

With the development of power systems under high renewable penetration, the concentrating solar power (CSP) plant, which can provide both renewable energy and operational flexibility, have attracted more and more attention. However, most existing CSP modeling and optimization methods have the following limitations: applying the mixed integer linear programming model to the annual operation cost analysis of the long-term planning problem for large-scale systems is complicated. In this paper, we introduce a fast cluster optimization method for CSP plants toward power systems under high renewable energy penetration. Through this method, the CSP plants with similar operation characteristics in adjacent geographical areas are clustered and grouped, so as to optimize the group behavior, rather than the behavior of individual plant. The number of decision variables is greatly reduced and the mixed integer linear programming model is transformed into a linear optimization model by introducing additional linear (continuous) variables instead of integer variables, thus significantly reducing the computational complication and realizing the rapid calculation of long-term planning for large-scale power systems. We also apply the proposed model to collaborative optimization of power system planning and operation for a province in China to prove the effectiveness of the proposed model and the solution method.