期刊
IEEE TRANSACTIONS ON POWER SYSTEMS
卷 36, 期 2, 页码 1594-1604出版社
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TPWRS.2020.3019828
关键词
Economics; Security; Power systems; Generators; Couplings; Electricity supply industry; Convergence; Coordinated optimization; multi-area economic dispatch (maed); optimality condition; redundant constraints; system reduction
资金
- National Key R&D Program of China [2020YFB0905900]
- National Natural Science Foundation of China [51777102]
This paper proposes a multi-area coordination framework based on condensed system representation (CSR), which can optimize multi-area systems by fixing non-marginal units and eliminating redundant security constraints without the need for iterative information exchange.
The coordinated multi-area economic dispatch enables optimal resource utilization in a larger spatial range. To overcome drawbacks such as heavy communication burden, convergence failure, and weak scalability of iterative coordination methods, this paper proposes a fully non-iterative multi-area coordination framework. The proposed framework is based on a novel system reduction technique termed as the condensed system representation (CSR). The CSR makes external equivalence of the dispatch problem of each area by exploiting power system operating characteristics, i.e., only a small number of generators may become marginal units and only a minority of security constraints may be active. An algorithm based on the optimality condition analysis is developed to identify the CSR by fixing non-marginal units to their output bounds, eliminating redundant security constraints, and making Norton equivalence of the internal network. With CSRs submitted by local areas, the multi-area system can be optimized without iterative information exchange. Case studies based on a 3-area 9-bus system verify the effectiveness of the CSR-based coordination framework. Larger-scale test systems are constructed to validate the computational efficiency, robustness, and scalability of the proposed method.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据