Machine Learning for Short-Term Load Forecasting in Smart Grids

Abstract

A smart grid is the future vision of power systems that will be enabled by artificial intelligence (AI), big data, and the Internet of things (IoT), where digitalization is at the core of the energy sector transformation. However, smart grids require that energy managers become more concerned about the reliability and security of power systems. Therefore, energy planners use various methods and technologies to support the sustainable expansion of power systems, such as electricity demand forecasting models, stochastic optimization, robust optimization, and simulation. Electricity forecasting plays a vital role in supporting the reliable transitioning of power systems. This paper deals with short-term load forecasting (STLF), which has become an active area of research over the last few years, with a handful of studies. STLF deals with predicting demand one hour to 24 h in advance. We extensively experimented with several methodologies from machine learning and a complex case study in Panama. Deep learning is a more advanced learning paradigm in the machine learning field that continues to have significant breakthroughs in domain areas such as electricity forecasting, object detection, speech recognition, etc. We identified that the main predictors of electricity demand in the short term: the previous week’s load, the previous day’s load, and temperature. We found that the deep learning regression model achieved the best performance, which yielded an R squared (R2) of 0.93 and a mean absolute percentage error (MAPE) of 2.9%, while the AdaBoost model obtained the worst performance with an R2 of 0.75 and MAPE of 5.70%.

Data-Driven Short-Term Load Forecasting for Multiple Locations: An Integrated Approach

“…Existing studies demonstrated the effectiveness of DL-based models such as ANN, CNN, and LSTM compared to conventional ML-based algorithms [10,16]. RNN is a type of DL algorithm that utilizes feedback connection, enabling it to transfer information from one timestep to another.…”

“…The ANN method demonstrated lower error (MAPE) and a regression factor (R) closer to 1, indicating superior performance compared to ARIMA. Ibrahim et al [16] utilized various ML and deep learning (DL) algorithms, including XG-Boost, AdaBoost, SVR, and ANN, for 24 h ahead predictions, with ANN exhibiting superior performance in terms of MAPE, RMSE, and R 2 , despite longer training times and higher computational expenses for DL algorithms.…”

Security Baseline for Substation Automation Systems

“…These problems are then addressed by many researchers, where we can mention [ 48 ], who presented a dynamic prediction of retail electricity prices in the Smart Grid based on the Stackelberg model. Then, [ 49 ], where the authors discuss the possibility of using ML and DL in refining the power generation forecast, and similarly [ 50 , 51 , 52 ] discuss the stability and security of the Smart Grid concerning short-term load. In conclusion, Smart Grids are rapidly replacing conventional grids on a global scale.…”

Improved Secure Encryption with Energy Optimization Using Random Permutation Pseudo Algorithm Based on Internet of Thing in Wireless Sensor Networks

“…Rewards that are given out after state transitions are used to address the convergence problem. The transition mechanism considers different device energy levels and deep learning methods to make the changes in energy consumptions [42][43][44]. Various schemes are proposed for perfect secret sharing mechanism in CRT [45].…”