Stochastic multi-objective optimal energy and reactive power dispatch considering cost, loading margin and coordinated reactive power reserve management

Abstract

In the recent years, modern wind farms (WF) are being increasingly integrated in the power systems. Thanks to the capability curve of doubly-fed induction generator (DFIG) wind turbines, these modern WFs are able to separately supply active and reactive powers. Accordingly, by increasing the penetration of this kind of sources, both voltage stability and reactive power reserve (RPR) management of power system can be affected. This issue makes it important to propose a new methodology for optimal energy and reactive power dispatch in a system with integrated modern WFs in order to effectively take into consideration the capabilities of modern WFs in different aspects of power system operation. Under this perspective, this paper proposes a stochastic multi-objective optimal dispatch (SMO-OD) problem in the presence of modern WFs to make a techno-economic framework for the system operators (SOs). The proposed SMO-OD problem aims at minimizing expected payments of energy and reactive power markets, and maximizing expected RPR and loading margin (LM) in the presence of wind power generation uncertainty. This paper also contributes to the relevant literature by formulating a model which determines the RPR of modern WFs to be included in the management of total RPR of the power system. Also, a coordinated RPR (CRPR) scheme is proposed by using this model. The proposed model benefits from a detailed reactive power output model of the DFIG wind turbines in order to handle real-world power system operation. Two studies are considered in the proposed SMO-OD problem: (i) SMO-OD with uncoordinated RPR provision (SMO-OD-URPR), and (ii) SMO-OD with the CRPR provision (SMO-OD-CRPR). The proposed SMO-OD problem is applied to the IEEE 30-bus test system. Simulation results show that using the proposed model, a proper trade-off is made between technical and economic aspects of power system operation, and considerable LM and RPR are provided. In the proposed SMO-OD-CRPR problem, the value of expected RPR increases compared to SMO-OD-URPR problem, and the DFIG-based WF has more expected RPR to inject to grid under contingency or sudden increased load. Moreover, by increasing the wind penetration level, not only does the value of expected payments of energy and reactive power markets reduce, but also the value of expected LM and RPR increases.