階層的ベイズ推定に基づくGaussian Processを用いた地点別限界価格予測法

カテゴリ: 部門大会

論文No: 26

グループ名: 【B】平成28年電気学会電力・エネルギー部門大会

発行日: 2016/09/05

タイトル(英語): An Efficient Method for LMP Forecasting with Gaussian Process of Hierarchical Bayesian Estimation

著者名: 森 啓之(明治大学),中野 郁(明治大学)

著者名(英語): Hiroyuki Mori|Kaoru Nakano

キーワード: 電力価格予測|地点別限界価格|階層的ベイズ推定|マハラノビスカーネル|クラスタリングガウシアンプロセス,Electricity Price Forecasting,Locational Marginal Price,Hierarchical Bayesian Estimation,Mahalanobis Kernel,ClusteringGaussian Process

要約(日本語): In this paper, a new method is proposed for Locational Marginal Pricing (LMP) forecasting in power systems. The marginal cost is required to supply electricity to incremental loads in case where a certain node increases power demands in a balanced power system. LMP plays a key role to maintain economic efficiency in power markets in a way that electricity flows from a low-cost to high-cost area and the transmission network congestion is alleviated. The power market players are interested in maximizing the profits and minimizing the risks through selling and buying electricity. Thus, it is important to obtain accurate information on electricity pricing forecasting in advance. This paper proposes the Gaussian Process (GP) technique in which hierarchical Bayesian estimation is introduced to express the model parameter as the probabilistic variables. The advantage is that the model accuracy of GP is better than others. GP is integrated with DA clustering to improve the performance of GP. Also, this paper makes use of the Mahalanobis kernel in GP rather than the Gaussian one so that GP is generalized to approximate nonlinear systems. EPSO is applied to GP to determine parameters of the kernel function. The proposed method is successfully applied to real data of ISO New England in USA.

PDFファイルサイズ: 457 Kバイト