Electricity load forecasting and feature extraction in smart grid using neural networks

Accurate load forecasting is of great significance to ensure the safety and stable operation of the smart grid's power system. In this paper, we study this problem of making an accurate short-term load prediction for an electricity system. Due to its inherent nonlinear properties, our proposed solution is based on the nonlinear model, and the power load sequence is also random, non-stationary and periodic. The original historical load sequence is decomposed into a series of modal functions by using the variational mode decomposition (VMD) technique, and a load forecasting model is established for each modal function. A load forecasting model combining the beetle swarm optimisation(BSO) algorithm and extreme learning machine(ELM) is put forward. The BSO finds the optimal input weight and hidden layer threshold in the ELM network. In the end, the VMD-BSO-ELM prediction model is established. The experimental results demonstrate that the proposed method is better than other individual methods and their combinations.

Comparison of echo state network and feed-forward neural networks in electrical load forecasting for demand response programs

A novel neural network technique for electric load forecasting based on weather compensation is presented. The proposed method is a nonlinear generalisation of Box and Jenkins approach for nonstationary time-series prediction. A nonlinear autoregressive integrated (NARI) model is identified to be the most appropriate model to include the weather compensation in short-term electric load forecasting. A weather compensation neural network based on an NARI model is implemented for one-day ahead electric load forecasting. This weather compensation neural network can accurately predict the change of electric load consumption of the coming day. The results, based on Hong Kong Island historical load demand, indicate that this methodology is capable of providing more accurate load forecast with a 0.9% reduction in forecast error.

In this paper, an artificial neural network (ANN) based methodology is proposed to forecast electricity load and price. The performance of an ANN forecast model depends on appropriate input parameters. Parameter tuning of ANN is very important to increase the accuracy of electricity price and load prediction. This is done using mutual information and decision tree. After selecting best features for forecasting, these features are given to forecasting engine working on principles of recurrent neural network (RNN). For simulations, dataset is taken from national electricity market (NEM), Australia. Results show that the methodology has increased the accuracy of electricity load and price forecast. Whereas, the error rate of forecasting is lower than the other models for electricity load and price.

Day-ahead load and wind power forecasts provide useful information for operational decision making, but they are imperfect and forecast errors must be offset with operational reserves and balancing of (real time) energy. Procurement of these reserves is of great operational and financial importance in integrating large-scale wind power. We present a probabilistic method to determine net load forecast uncertainty for day-ahead wind and load forecasts. Our analysis uses data from two different electric grids in the US with similar levels of installed wind capacity but with large differences in wind and load forecast accuracy, due to geographic characteristics. We demonstrate that the day-ahead capacity requirements can be computed based on forecasts of wind and load. For 95% day-ahead reliability, this required capacity ranges from 2100 to 5700 MW for ERCOT, and 1900 to 4500 MW for MISO (with 10 GW of installed wind capacity), depending on the wind and load forecast values. We also show that for each MW of additional wind power capacity for ERCOT, 0.16–0.30 MW of dispatchable capacity will be used to compensate for wind uncertainty based on day-ahead forecasts. For MISO (with its more accurate forecasts), the requirement is 0.07–0.13 MW of dispatchable capacity for each MW of additional wind capacity.

The evolution of conventional power systems into smart grids demands intelligent and efficient energy management solutions. Accurate load forecasting is essential for optimal generation scheduling, demand-side management, and system reliability. However, traditional methods often fail to capture the nonlinear and time-varying nature of electrical loads. This work proposes a machine learning–based approach for local load forecasting and grid stability enhancement. Techniques such as Artificial Neural Networks (ANN) and Support Vector Machines (SVM) are utilized to analyze historical load and weather data for improved prediction accuracy. In addition, a predictive control strategy is implemented to maintain system stability by mitigating voltage fluctuations, frequency deviations, and load imbalances. The proposed model dynamically adjusts control actions such as demand response and load shedding. Simulation results demonstrate reduced forecasting errors (MAE, RMSE) and improved voltage and frequency regulation, ensuring enhanced reliability, efficiency, and stable operation of modern smart grid systems. Keywords: Smart Grid; Machine Learning; Load Forecasting; Grid Stability; Predictive Control.

With the large-scale development of smart grids and the integration of renewable energy, the operational complexity and load volatility of power systems have increased significantly, placing higher demands on the accuracy and timeliness of electricity load forecasting. However, existing methods struggle to capture the nonlinear and volatile characteristics of load sequences, often exhibiting insufficient fitting and poor generalization in peak and abrupt change scenarios. To address these challenges, this paper proposes a deep learning model named CGA-LoadNet, which integrates a one-dimensional convolutional neural network (1D-CNN), gated recurrent units (GRUs), and a self-attention mechanism. The model is capable of simultaneously extracting local temporal features and long-term dependencies. To validate its effectiveness, we conducted experiments on a publicly available electricity load dataset. The experimental results demonstrate that CGA-LoadNet significantly outperforms baseline models, achieving the best performance on key metrics with an R2 of 0.993, RMSE of 18.44, MAE of 13.94, and MAPE of 1.72, thereby confirming the effectiveness and practical potential of its architectural design. Overall, CGA-LoadNet more accurately fits actual load curves, particularly in complex regions, such as load peaks and abrupt changes, providing an efficient and robust solution for short-term load forecasting in smart grid scenarios.

Essentially, Electricity Load Forecasting is an approximation of upcoming active loads from a variety of load buses before the active loads occur. Also, it is an important factor for power system energy management. Accurate and precise load forecasting can help to reduce the capacity of the power system, to make unit commitment decisions, and to increase the dependability of power systems. Hence, this paper presents a generalized regression Neural Network (GRNN) based approach for Short Term Load Forecasting (STLF). The results showed that the performance of GRNN with 30 neurons is better of short-term load forecasting in comparison with 10 neurons. For 10 neurons, the Mean Absolute Percentage Error (MAPE) was 2.10% and Mean Absolute Error (MAE) was 306.21 MWh. However, for 30 neurons it was observed that MAPE is 1.81% and MAE 268.48 MWh.

Load forecasting is used primarily to predict future loads for a particular system over a given time period. Short-term loads are typically treated as variable elements influenced by factors such as historical load information and weather datasets, including precipitation, wind speed, and temperature. A precise forecasting with an individual model is almost impossible. The primary challenge for utility companies worldwide is accurately forecasting energy consumption. Accurate short-term load forecasting (STLF) is a cornerstone of smart grid operation, enabling demand-side management (DSM), demand response programs, and efficient integration of distributed energy resources. This study proposes a fuzzy time series (FTS)-based methodology for residential electricity consumption forecasting at hourly, daily, and weekly scales. By addressing overfitting during data partitioning and refining the fuzzification process, our approach improves prediction accuracy compared to traditional FTS models. Simulation results using real consumption data demonstrate up to 40% improvement in hourly forecasting and up to 58% and 84% improvements in daily and weekly forecasts, respectively. These results highlight the potential of FTS-based models to enhance residential demand forecasting, reduce peak-demand uncertainty, and support grid operators in achieving more resilient, flexible, and sustainable smart grid systems.

Power load forecasting is an important part of the power system planning, and accurate load forecasting can provide necessary basis data for the dispatcher, which is extremely important in the planning and operation of power system. Neural network can approximate any nonlinear mapping with arbitrary precision, and its distributed information storage and processing structure have a certain fault tolerant. So neural network is suitable for complex system modeling, and it can be used as the main method to predict photovoltaic power operation state variables. This paper uses the neural network algorithm to establish photovoltaic power system's load forecast model, and except the generate historical data, meteorological forecast information is added to the algorithm, then the model is trained and tested, the high-precision prediction results can show the effectiveness of the algorithm.

Forecasting of electrical load is extremely important for the effective and efficient operation of any power system. Good forecasts results help in minimizing the risk in decision making and reduces the costs of operating the power plant. This work focuses on the short-term load forecast of the 132/33KV transmission sub-station at Port-Harcourt, Nigeria, using the Artificial Neural Network (ANN). It provides accurate week-ahead load forecast using hourly load data of previous weeks. ANN has three sections namely; input, processing and output sections. There are four input parameters for the input section which are historical hourly load data (in MW), time of the day (in hours), days of the week and weekend while the output parameter after the processing (i.e. training, validation and test) is the next week hourly load predicted for the entire system. The technique used is the artificial neural network with the aid of MATLAB software. It was proven to be a good forecast method as it resulted in R-value of 0.988 which gives a mean absolute deviation (MAD) of 0.104 and mean squared error (MSE) of 0.27. Keywords : Load forecast, transmission substation, artificial neural network, power system JEL Classifications: C63, L94, L98, Q48 DOI: https://doi.org/10.32479/ijeep.8629

The forecasting of short term electricity load plays a vital role in power system. It is essential for the power system's reliable, secure, and cost-effective functioning. This paper contributes significantly for enhancing the accuracy of short term electricity load forecasting. It presents a hybrid forecasting model called Gated Recurrent Unit with Ensemble Empirical Mode Decomposition and Boruta feature selection (EBGRU). It is a hybrid model that addresses the non-stationary, non-linearity and noisy issues of the time series input by using Ensemble Empirical Mode Decomposition (EEMD). It also addresses overfitting and curse of dimensionality issues of load forecasting by identifying the pertinent features using Boruta wrapper feature selection. It effectively handles the uncertainty and temporal dependency characteristics of load and forecasts the future load using deep learning based Gated Recurrent Unit (GRU). The proposed EBGRU model is experimented by using European and Australian Electricity load datasets. The temperature has high correlation with load demand. In this study, both load and temperature features are considered for the accurate short term load forecasting. The experimental outcome demonstrates that the proposed EBGRU model outperforms other deep learning models such as RNN, LSTM, GRU, RNN with EEMD and Boruta (EBRNN) and LSTM with EEMD and Boruta (EBLSTM).

The accurate short-term load forecasting is essential for the efficient power system operation and the system marginal price decision of the electricity market. So far, errors of load forecasting for Chuseok Holiday are very big compared with forecasting errors for the other special days. In order to improve the accuracy of load forecasting for Chuseok Holiday, selection of input data, the daily normalized load patterns and load forecasting model are investigated. The efficient data selection and daily normalized load pattern based on fuzzy linear regression model is proposed. The proposed load forecasting method for Chuseok Holiday is tested in recent 5 years from 2006 to 2010, and improved the accuracy of the load forecasting compared with the former research.

Accurate and reliable load forecasting is necessary to ameliorate energy management. Short-term load forecast plays a crucial role in economic and secure system operation. This paper presents a practical method for short-term electric load forecast problem using an artificial neural network with a powerful Levenberg-Marquardt training algorithm approach. The applications of real load from Ontario, Canada with hourly load, daily load, and weekly load predictions have been successfully achieved. Both visual comparison and statistical test are discussed and analyzed to validate training and testing phases of the neural network.

Accurate and efficient electrical load forecasting is a vital pre-requisite for power system planning and distribution as it can result in cost saving and better operating conditions. Object of this study focuses on the various factors affecting load forecasting. This paper attempts to consider the effects of these factors to highlight their role in operation and planning. A good understanding of the influence of the factors on the pattern of load forecasting helps to reduce the gap between demand and supply of electrical power leading to higher profit margin.