Forecasting of Market Clearing Volume Using Wavelet Packet-Based Neural Networks with Tracking Signals

In order to analyze the nature of electrical demand series in deregulated electricity markets, various forecasting tools have been used. All these models developed improve accuracy reliability model. Therefore, a Wavelet Packet Decomposition (WPD) was implemented decompose into subseries. Each subseries has forecasted individually with help features that series, and were chosen on basis mutual correlation among all-time lags using an Auto Correlation Function (ACF). Thus, this context, new hybrid WPD-based Linear Neural Network Tapped Delay (LNNTD) model, cyclic one-month moving window for one-year market clearing volume (MCV) proposed. The proposed model effectively two years (2015–2016) unconstrained MCV data collected from Indian Energy Exchange (IEX) 12 grid regions India. results presented by are better terms accuracy, yearly average MAPE 0.201%, MAE 9.056 MWh, coefficient regression (R2) 0.9996. Further, forecasts validated tracking signals (TS’s) which values TS’s lie within balanced limit between −492 6.83, universality carried out multiple steps-ahead up sixth step. It found powerful forecasting.