Multi step electricity price Forecasting using hybrid models based on wavelet decomposition and Deep learning neural network Optimized by metaheuristic Algorithms

Abstract

The major task in the deregulated power system is to efficiently forecast the long-term and short-term electricity prices for electricity market participants to economically manage the electric power. An increase in the addition of renewable energy sources and distributed generators to the electric grid of deregulated electricity market makes the time serious electricity price as a complex and nonlinear data. The addition of new generation and distribution companies in the deregulated electricity market increases the number of participants in the electricity market and makes the electricity market a competitive one. newlineThe primary contribution to this research work is electricity price forecasting using a hybrid back propagation and modified particle swarm optimization in which a modified particle swarm optimization technique is used to improve the performance of the backpropagation algorithm. Two different MLP are used for electricity price forecasting one MLP is with a single hidden layer and another MLP is with three hidden layers, both the neural networks are trained by BP and initial parameters such as weights between different layers, the bias of the layers, and activation function of each layer are selected by MPSO. Normally MLP trained by BP uses linear activation function for all layers and neurons, but in this case, we use three different functions namely linear function, sigmoid function, and tangent function as activation functions. This method is used to forecast the electricity price of the Austria and Northern Italy electricity market. Overall performance, convergence time, and convergence efficiency of the BP are greatly improved by independently selecting the activation function. newline