DYNAMIC CLUSTERING-BASED METHOD OF ESTABLISHING SUPPLY-AND-DEMAND EARLY WARNING MODEL IN ELECTRICITY MARKET

摘要

A dynamic clustering-based method of establishing a supply-and-demand early warning model in an electricity market. The method comprises: 1.1 selecting early warning indications before an electricity market transaction including: (1) declared space share; (2) price increase rate; (3) high quoted price ratio; (4) high quoted price success rate; 1.2 performing k-means method clustering according to capacity indicators capable of reflecting a market bidding position and strength and influencing market price, comprising the declared space share, the price increase rate, the high quoted price ratio, and the high quoted price success rate as determined in step 1.1, wherein an initial center is determined by an SPSS itself; 1.3 utilizing the k-means method to obtain a final category center difference and category center degree of deviation between each category, so as to establish a supply-and-demand early warning model in an electricity market.