A bidding strategy based on the Fuzzy Markov decision process and fuzzy‐c‐means for a competitive electric power market

Abstract

A bidder tends to win a bid over a horizon involving multiple transaction periods in a competitive electricity market. Therefore, a strategic bid should consider operation limits (e.g., line flow constraints and ramping rates for generators), current/forecast system loads and price information. A bidding strategy using the Fuzzy Markov Decision Process (FMDP) and Fuzzy‐c‐means (FCM) is proposed in this paper. The system states and the bidding strategies for a bidder are defined first. Because there are a great number of states, the states are clustered into several groups depending on the strategy number using FCM. The crisp transition probabilities are transformed into fuzzy transition probabilities using FMDP. Linear programming is used to solve the 1–0 problem: a state corresponds to a bidding strategy. A 30‐bus system is used to illustrate the applicability of the proposed method.

Key words:

• bidding strategy
• electric power market
• fuzzy clustering
• fuzzy Markov decision process

Notes

Corresponding author. (Tel: [phone]; Fax: [phone]; Email: [email protected])