
http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
(A) New Congestion Cost Allocation Method in a Uniform Pricing Model
정해성 서울대학교 공과대학원 2004 국내박사
전력계통의 여러 제약은 혼잡을 유발하며 이로 인해 추가적인 혼잡비용이 발생한다. 효율적으로 전력시장을 운영하기 위해서는 혼잡비용이 시장 참여자들에게 적절한 신호를 줄 수 있어야 한다. 단일요금제를 채택한 시장에서 혼잡으로 인해 증발 또는 감발하는 발전기들은 이에 대한 보상을 받으며 그 비용은 소비자들에게 부가정산금을 통하여 회수된다. 그러나 이러한 방식은 시장참여자들에게 적절한 가격신호를 줄 수 없다. 본 논문에서는 단일요금제에서 혼잡발생 시 시장참여자에게 적절한 가격 신호를 줄 수 있는 방법을 제안하였다. 혼잡비용은 혼잡을 유발하는 제약들로 인한 것이므로 각 제약들이 혼잡비용에 미치는 영향을 파악할 수 있다면 혼잡비용을 각 제약들에게 분배할 수 있다. 그러나 비용에 영향을 미치는 발전량, 부하, 선로 그리고 전압 등은 복잡하게 관련되어 있어 각각의 영향을 명료하게 분석해 내는 것은 어려운 일이다. 이러한 비선형성으로 인한 혼잡비용의 분배의 어려움을 극복하기 위해 약간의 가정을 도입한다면 그 비용을 분배할 수 있다. 특히 혼잡에 의해 혼잡이 유발되는 경우에 각 제약조건이 혼잡비용에 미치는 영향의 패턴이 변하게 되며 이를 고려하기 위해서는 혼잡비용과 제약조건 사이의 단계적인 처리가 필요하다. 본 연구에서 제안하는 방법은 먼저 단계별로 혼잡을 분류하며 각 단계에서 여러 개의 제약이 혼잡 비용에 미치는 영향을 선적분을 이용하여 분석하였다. 제안한 방법을 IEEE 118-bus 계통에 대해 사례 연구를 수행하였고 혼잡비용이 혼잡선로 인근의 부하들과 혼잡으로 인해 감발하는 발전기들에게 적절하게 분배됨을 볼 수 있었다. Competitive environment of the electricity markets requires wide access to networks that connect customers and suppliers. Transmission networks play an important role in the power system operation. Congestion is a special phenomenon of a power system network and has been a hot issue in the electricity restructuring and the transmission pricing. In a general pool model, electric power sellers submit bids to the Independent System Operator (ISO). The ISO computes an unconstrained market clearing price by an unconstrained dispatch based on biddings. If the unconstrained dispatch results in transmission violations, the ISO should coordinate the generation schedule in order to remove the congestion. This change of dispatch may cause to the increase of the cost and this increment of the cost is defined as the congestion cost, which should be recovered from market participants. Many methods on the congestion cost allocation have been proposed and implemented in various markets. Among them, nodal pricing has been widely accepted in many power markets to resolve congestion problems. The concept of the nodal pricing was proposed by Schweppe et al. and further developed by Hogan. While the nodal pricing is very effective for operation, it may cause merchandising surplus problem. The uplift is simple and it can handle the merchandising surplus problem. In this scheme, customers pay the congestion cost by a pro rata method. This uplift is effective for cost recovery but it cannot give efficient price signals to customers. The game theoretic approach for congestion cost allocation was also proposed. This method recovers the exact amount of the congestion cost using the integral. However, the concept of the nodal pricing makes it difficult to apply to a general pool model and requires complex calculation. Singh et al. suggested a useful method using the Lagrange multiplier associated with the transmission constraints to allocate the congestion cost. In this method, the congestion cost is allocated to the congested lines according to the Lagrange multipliers of transmission constraints based on a dc power flow and the market participants who use the congested lines pay the associated congestion costs. However this dc power flow cannot consider voltage limits, but this limitation can be overcome by adopting a full ac power flow. The critical problem is that the cost allocation can be distorted in case congestion in special area can cause congestion in other area because it ignores the sequence of violating constraints. This dissertation proposes a multi-stage method for the congestion cost allocation in a pool model. Generally, the set of active constraints in the constrained dispatch may not be equal to those limits in the unconstrained dispatch. Congestion cost comes from preventing violation of constraints. Thus the constraints which do not violate limits in the unconstrained dispatch but become active in the constrained dispatch can distort congestion cost allocation. In the proposed multi-stage method, the active constraints in the constrained dispatch are discriminated and grouped by the sequence of violating limits. Therefore, the effect of the active constraints which are within limits in the unconstrained dispatch can be evaluated fairly. The relationship between congestion and constraints is analyzed and derived using the line integral based on an ac power flow so that the effects of voltage limit and transmission capacity can be calculated to allocate congestion cost.