Locational Marginal Price Projection using a Novel RBFNN Approach in Spot Power Market

Abstract

The emerging trend of restructuring in the electric power industry has spawned a need of changed pricing mechanism, which can provide correct economic incentives and also facilitate the physical operation of the power network. Nodal /locational pricing mechanism in spot power market is an efficient tool for congestion management, as it provides implicit capacity allocation through bids for production/consumption at a specific location. In this regard optimal nodal prices can be decomposed into two components, locational marginal prices (LMP) and nodal congestion prices (NCP). The conventional methods for computing these prices are no more suitable for spot electricity market as these are too slow and case sensitive methods. This paper presents soft computing based approach for projecting locational marginal prices in spot power market by developing a novel Radial Basis Function Neural Network (RBFNN). The effectiveness of the proposed approach is examined on 6-bus system and RTS 24-bus system and is found computationally efficient.

Different ANN Models for Short-Term Electricity Price Forecasting

“…Among many existing tools, ANN has received much attention because of its clear model, easy implementation, and good performance. Many ANN based Price forecasting models were proposed in literature [7][8][9][10][11][12]. A novel non-iterative technique based on weighted least squares state estimation for short term load forecasting of Distribution Systems is proposed in [13].…”