Abstract
The advancement in electrical load forecasting techniques with new algorithms offers reliable solutions to operators for operational cost reduction, optimum use of available resources, effective power management, and a reliable planning process. The focus is to develop a comprehensive understanding regarding the forecast accuracy generated by employing a state of the art optimal autoregressive neural network (NARX) for multiple, nonlinear, dynamic, and exogenous time varying input vectors. Other classical computational methods such as a bagged regression tree (BRT), an autoregressive and moving average with external inputs (ARMAX), and a conventional feedforward artificial neural network are implemented for comparative error assessment. The training of the applied method is realized in a closed loop by feeding back the predicted results obtained from the open loop model, which made the implemented model more robust when compared with conventional forecasting approaches. The recurrent nature of the applied model reduces its dependency on the external data and a produced mean absolute percentage error (MAPE) below 1%. Subsequently, more precision in handling daily grid operations with an average improvement of 16%–20% in comparison with existing computational techniques is achieved. The network is further improved by proposing a lightning search algorithm (LSA) for optimized NARX network parameters and an exponential weight decay (EWD) technique to control the input error weights.
Highlights
Short Term Load Forecasting has ended up one of the major research fields in power system designing
The results show a notable development when compared with statistical methods i.e., ARMAX and state space, Decision Tree i.e., bagged regression tree (BRT) and conventional feedforward artificial neural network (ANN) FitNet method without exogenous inputs
Based on best mean absolute percentage error (MAPE) fit on the available data, an open loop NARX model is chosen for a forecasting purpose
Summary
Short Term Load Forecasting has ended up one of the major research fields in power system designing. A power system operates by coordinating numerous stakeholders who can be influenced by an inaccurate forecasting estimate: Generation planning required a 24–48 h demand forecast in order to allocate the power resources economically [1]. Ordinary load variations are not linear and, non-linear networks have demonstrated high success in generating accurate short-term estimates. The focus lies on short-term load forecasting utilizing ANN. A unique and improved NARXNN based recurrent load forecaster is developed using the lighting search algorithm, which determines the optimal value for the number of hidden layer neurons, feedback delays, and input delays. This paper develops a recurrent neural network (RNN) with an LSA optimization algorithm to increase the robustness and intelligence of the forecasting method. LSA is considered and explained in the proposed solution
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
