Robust Multi-Step Predictor for Electricity Markets with Real-Time Pricing

Sachin Kahawala,Damminda Alahakoon,Rashmika Nawaratne,Valeriy Vyatkin,Seppo Sierla,Evgeny Osipov,Andrew Jennings,Daswin De Silva

doi:10.3390/en14144378

Sachin Kahawala, Damminda Alahakoon + Show 6 more

Open Access

https://doi.org/10.3390/en14144378

Copy DOI

Abstract

Real-time electricity pricing mechanisms are emerging as a key component of the smart grid. However, prior work has not fully addressed the challenges of multi-step prediction (Predicting multiple time steps into the future) that is accurate, robust and real-time. This paper proposes a novel Artificial Intelligence-based approach, Robust Intelligent Price Prediction in Real-time (RIPPR), that overcomes these challenges. RIPPR utilizes Variational Mode Decomposition (VMD) to transform the spot price data stream into sub-series that are optimized for robustness using the particle swarm optimization (PSO) algorithm. These sub-series are inputted to a Random Vector Functional Link neural network algorithm for real-time multi-step prediction. A mirror extension removal of VMD, including continuous and discrete spaces in the PSO, is a further novel contribution that improves the effectiveness of RIPPR. The superiority of the proposed RIPPR is demonstrated using three empirical studies of multi-step price prediction of the Australian electricity market.

Highlights

We evaluate Robust Intelligent Price Prediction in Real-time (RIPPR) on three experiments conducted on five different datasets of EPF for the state of New South Wales (NSW), Australia
We report the empirical evaluation of RIPPR in terms of the following performance metrics, mean absolute error (MAE), root mean square error (RMSE), mean absolute percentage error (MAPE) and mean squared error (MSE)
We propose a novel Artificial Intelligence (AI) based approach for electricity price forecasting that addresses the challenges of accuracy, robustness and realtime multi-step prediction

Summary

Introduction

Real-time pricing is emerging as a solution for coordinating renewable generation with other intelligent energy resources [3], such as flexible loads [4], battery storages [5] and electric vehicles [6]. The transformation of residential and commercial buildings into prosumers with local renewable generation is one driver for such short interval real-time pricing markets [12]. Elma et al [13] proposed a domestic prosumer operating at five min intervals, rescheduling or curtailing loads according to forecasted local photovoltaic generation and real-time electricity prices. Mbungu et al [14] presented a similar approach for a commercial building prosumer with photovoltaic generation and battery storage; the proposed real-time pricing scheme is built on top of a time-of-use pricing scheme.

Methods

Results

Conclusion