Artificial Intelligence–Based Loss Allocation Algorithm in Open Access Environments

Abstract

In deregulated power systems, market participants prefer to have a fast and accurate estimation of their loss quota for any transaction before confirming their transaction in the market. This helps participants increase their own benefit. This paper presents a fast artificial intelligence–based incremental transmission loss allocation (ITLA) algorithm for determining the loss quota of any transaction and participant entity in open access environments. As a feature selection technique, the decision tree (DT) method is applied in order to define the transactions with inconsiderable impact on the loss quota of each market participant. Then, using only the effective transactions, an artificial neural networks (ANN) is trained to estimate the loss quota of each transaction in the market. Applying the DT significantly reduces the input dimension of the ANN, and thus it reduces the training time and improves the accuracy of the loss quota estimated by ANN. The market participants can employ the proposed artificial intelligence-based ITLA algorithm to estimate their own loss quota even in large scale power systems with large amount of possible transactions. One attractive result of the proposed algorithm is that it is implemented in multilateral open access environments including some trader entities and bilateral markets as well. The proposed algorithm is computationally efficient and provides solution on a real-time application. Therefore, it can be based on ex-ante loss allocation methods in the future to improve market efficiency. The proposed algorithm is applied on IEEE-RTS test system and simulation results show the algorithm’s effectiveness.