COST MINIMIZATION STATISTICAL METHOD FOR ELECTRICITY DEMAND FORECASTING USING MULTI OBJECTIVE LEARNING MODEL
The multi-objective learning model used in this research to forecast power demand offers a novel approach with good accuracy while minimizing computational costs. Modern energy consumption prediction methods have limitations in their capacity to estimate energy usage because of a variety of issues like weather and occupants' dynamic behavior. The suggested method uses a hybrid strategy to predict data using a sequential model, combining the straightforward RNN and LSTM models. The approach entails preprocessing the time series data, independent model training, and weighted average ensemble technique integrating the results. The results show that the suggested technique is superior to the current state-of-the-art methodologies. The performance of the proposed method is evaluated using MAPE and RMSE metrics on the New York Independent System Operator (NYISO) dataset. The suggested approach is anticipated to contribute to better power management and cooperation between the power grid and building energy use, resulting in more effective and sustainable energy consumption.
Recurrent Neural Networks (RNNs), Long Short-Term Memory (LSTM), Energy Consumption Prediction (ECP), Mean Absolute Percentage Error (MAPE) and Root Mean Squared Error (RMSE)