A new short-term energy price forecasting method based on wavelet neural network

Abstract

A wavelet neural network (WNN) is proposed for short-term price forecasting (STPF) in electricity markets. Back propagation algorithm is used for training the wavelet neural network for prediction. Weights in the back propagation algorithm are usually initialised with small random values. If the random initial weights happen to be far from a suitable solution or near a poor local optimum, training may take a long time or get trapped in the local optimum. In this paper, we show that WNN has acceptable prediction properties compared to other forecasting techniques. We investigated proper weight initialisations of WNN, and proved that it attains a superior prediction performance. Finally, we used a two-step correlation analysis algorithm for input selecting. This algorithm selects the best relevant and non-redundant input features for WNN. Our model is examined for MCP prediction of the Spanish market and LMP forecasting in PJM (Pennsylvania, New Jersey and Maryland) market for the year 2002 and 2006 respectively.

Electric Water Boiler Energy Prediction: State-of-the-Art Review of Influencing Factors, Techniques, and Future Directions

“…Their results showed that the hybrid model significantly outperformed the individual models. Keynia et al proposed a model that uses the wavelet transform and neural networks to predict the electric water boilers' energy consumption [178]. The wavelet transform is used to decompose the load series into different frequency load demand components, and neural networks are used to predict each decomposed series.…”

“…This allows for the comparison of prediction accuracy across different datasets. Furthermore, most deterministic and traditional probabilistic and data-driven stochastic methods use MAPE and RMSE to evaluate their models [6,178,179]. Lower MAPE values indicate better prediction accuracy, facilitating easy model comparisons.…”

“…Table 6 presents the confusion matrix (MC) classification used to evaluate the performance of a predictive model, which consists of four main metrics: True Positives (TP); True Negatives (TN); False Positives (FP); and False Negatives (FN) [178]. Subsequently, some commonly used metrics derived from the confusion matrix (MC) are as follows:…”

“…Accuracy(ACC) = These metrics offer a thorough assessment of the model's performance, encompassing several characteristics like overall accuracy, the capacity to correctly identify positive cases, and the trade-off between precision and recall [178,181];…”

A New ANN Approach for Time Series Analysis

“…; Q is the number of hidden nodes, and f and g are the transfer functions that are often used as a logistic function 7 .…”