Electrical Energy Consumption Forecasting Research Articles

Machine learning for electric energy consumption forecasting: Application to the Paraguayan system

Abstract In this paper we address the problem of short-term electric energy prediction using a time series forecasting approach applied to data generated by a Paraguayan electricity distribution provider. The dataset used in this work contains data collected over a three-year period. This is the first time that these data have been used; therefore, a preprocessing phase of the data was also performed. In particular, we propose a comparative study of various machine learning and statistical strategies with the objective of predicting the electric energy consumption for a given prediction horizon, in our case seven days, using historical data. In this paper we have tested the effectiveness of the techniques with different historical window sizes. Specifically, we considered two ensemble strategies, a neural network, a deep learning technique and linear regression. Moreover, in this study, we tested whether the inclusion of meteorological data can help achieve better predictions. In particular, we considered data regarding temperature, humidity, wind speed and atmospheric pressure registered during the three-year period of data collection. The results show that, in general, the deep learning approach obtains the best results and that such results are obtained when meteorological data are also considered. Moreover, when meteorological data is used, a smaller historical window size is required to obtain precise predictions.

Short-Term Forecasting of Electricity Consumption Using Fuzzy Logic

The high demand for electricity in the production process at PT Semen Padang requires a system that can cope with various kinds of disturbances. The problem is that the need for electrical loads is dynamic, especially in the short term, allowing fluctuations between electrical loads at uncertain times. A short-term electric energy consumption forecasting method is needed to determine load growth and distributed power supply. This research aims to use a fuzzy logic algorithm to perform short-term electrical energy consumption forecasting and compare the forecasting results with the actual load at PT Semen Padang. The results showed that short-term load forecasting for seven days using the fuzzy Mamdani method, namely Smallest of Maximum (SoM), obtained a percentage MAPE value of 8.15%. Meanwhile, the Weight of Average (WoA) Sugeno defuzzification method gets a portion of the MAPE value of 9.51%. The Sugeno method is more accurate than the Mamdani method in short-term electricity load forecasting for PPI Indarung V PT Semen Padang. If based on the time category, then forecasting the electricity load on holidays is better than predicting on weekdays. However, when viewed in terms of per day, in Wednesday has the smallest average MAPE value of 5.05%.

Long-term forecast of electrical energy consumption with considerations for solar and wind energy sources

The long-term goals achievement of sustainable development based on consumption of electrical energy is possible through the use of renewable energy sources. The objectives of this study include (a) long-term forecasting of Electrical Energy consumption (EEC) for sample countries with developed and developing economies, and (b) analyzing of different scenarios based on the use of solar and wind energy sources with 1%, 2% and 3% of the EEC. Artificial neural network (ANN) modeling with socio-economics data of the energy balance sheet last 30 years (1990–2019) as input data contain Gross Domestic Product (GDP), Population (POP), Import (IMP), Export (EXP), and EEC are used in order to forecast the EEC in the long-term (2020–2050). The United States and the OECD as developed economies and China, India, and Iran as developing economies are the countries under study. The structure of ANN is optimized for long-term EEC forecasting based on PSO and E-PSO algorithms. For both types of inputs and economies, the results demonstrate that E-PSO – ANN model can be used by SRE-3% scenario, which finally leads to a reduction of 55% and 54% in the EEC and amount of CO2 emissions in average according to the Paris Agreement (PA) goals, respectively.

Industrial Electrical Energy Consumption Forecasting by using Temporal Convolutional Neural Networks

In conjunction with the 4th Industrial Revolution, many industries are implementing systems to collect data on energy consumption to be able to make informed decision on scheduling processes and manufacturing in factories. Companies can now use this historical data to forecast the expected energy consumption for cost management. This research proposes the use of a Temporal Convolutional Neural Network (TCN) with dilated causal convolutional layers to perform forecasting instead of conventional Long-Short Term Memory (LSTM) or Recurrent Neural Networks (RNN) as TCN exhibit lower memory and computational requirements. This approach is also chosen due to traditional regressive methods such as Autoregressive Integrated Moving Average (ARIMA) fails to capture non-linear patterns and features for multi-step time series data. In this research paper, the electrical energy consumption of a factory will be forecasted by implementing a TCN to extract the features and to capture the complex patterns in time series data such daily electrical energy consumption with a limited dataset. The neural network will be built using Keras and TensorFlow libraries in Python. The energy consumption data as training data will be provided by GoAutomate Sdn Bhd. Then, the historical data of economic factors and indexes such as the KLCI will be included alongside the consumption data for neural network training to determine the effects of the economy on industrial energy consumption. The forecasted results with and without the economic data will then be compared and evaluated using Weighted Average Percentage Error (WAPE) and Mean Absolute Percentage Error (MAPE) metrics. The parameters for the neural network will then be evaluated and fined tuned accordingly based on the accuracy and error metrics. This research is able create a CNN to forecast electrical energy consumption with WAPE = 0.083 & MAPE = 0.092, of a factory one (1) week ahead with a small scale dataset with only 427 data points, and has determined that the effects of economic index such as the Bursa Malaysia has no meaningful impact on industrial energy consumption that can be then applied to the forecasting of energy consumption of the factory.

Hybridizing Deep Learning and Neuroevolution: Application to the Spanish Short-Term Electric Energy Consumption Forecasting

The electric energy production would be much more efficient if accurate estimations of the future demand were available, since these would allow allocating only the resources needed for the production of the right amount of energy required. With this motivation in mind, we propose a strategy, based on neuroevolution, that can be used to this aim. Our proposal uses a genetic algorithm in order to find a sub-optimal set of hyper-parameters for configuring a deep neural network, which can then be used for obtaining the forecasting. Such a strategy is justified by the observation that the performances achieved by deep neural networks are strongly dependent on the right setting of the hyper-parameters, and genetic algorithms have shown excellent search capabilities in huge search spaces. Moreover, we base our proposal on a distributed computing platform, which allows its use on a large time-series. In order to assess the performances of our approach, we have applied it to a large dataset, related to the electric energy consumption registered in Spain over almost 10 years. Experimental results confirm the validity of our proposal since it outperforms all other forecasting techniques to which it has been compared.

HousEEC: Day-Ahead Household Electrical Energy Consumption Forecasting Using Deep Learning

Short-term load forecasting is integral to the energy planning sector. Various techniques have been employed to achieve effective operation of power systems and efficient market management. We present a scalable system for day-ahead household electrical energy consumption forecasting, named HousEEC. The proposed forecasting method is based on a deep residual neural network, and integrates multiple sources of information by extracting features from (i) contextual data (weather, calendar), and (ii) the historical load of the particular household and all households present in the dataset. Additionally, we compute novel domain-specific time-series features that allow the system to better model the pattern of energy consumption of the household. The experimental analysis and evaluation were performed on one of the most extensive datasets for household electrical energy consumption, Pecan Street, containing almost four years of data. Multiple test cases show that the proposed model provides accurate load forecasting results, achieving a root-mean-square error score of 0.44 kWh and mean absolute error score of 0.23 kWh, for short-term load forecasting for 300 households. The analysis showed that, for hourly forecasting, our model had 8% error (22 kWh), which is 4 percentage points better than the benchmark model. The daily analysis showed that our model had 2% error (131 kWh), which is significantly less compared to the benchmark model, with 6% error (360 kWh).

Multiple Electric Energy Consumption Forecasting Using a Cluster-Based Strategy for Transfer Learning in Smart Building.

Electric energy consumption forecasting is an interesting, challenging, and important issue in energy management and equipment efficiency improvement. Existing approaches are predictive models that have the ability to predict for a specific profile, i.e., a time series of a whole building or an individual household in a smart building. In practice, there are many profiles in each smart building, which leads to time-consuming and expensive system resources. Therefore, this study develops a robust framework for the Multiple Electric Energy Consumption forecasting (MEC) of a smart building using Transfer Learning and Long Short-Term Memory (TLL), the so-called MEC-TLL framework. In this framework, we first employ a k-means clustering algorithm to cluster the daily load demand of many profiles in the training set. In this phase, we also perform Silhouette analysis to specify the optimal number of clusters for the experimental datasets. Next, this study develops the MEC training algorithm, which utilizes a cluster-based strategy for transfer learning the Long Short-Term Memory models to reduce the computational time. Finally, extensive experiments are conducted to compare the computational time and different performance metrics for multiple electric energy consumption forecasting on two smart buildings in South Korea. The experimental results indicate that our proposed approach is capable of economical overheads while achieving superior performances. Therefore, the proposed approach can be applied effectively for intelligent energy management in smart buildings.

A Novel Accurate and Fast Converging Deep Learning-Based Model for Electrical Energy Consumption Forecasting in a Smart Grid

Energy consumption forecasting is of prime importance for the restructured environment of energy management in the electricity market. Accurate energy consumption forecasting is essential for efficient energy management in the smart grid (SG); however, the energy consumption pattern is non-linear with a high level of uncertainty and volatility. Forecasting such complex patterns requires accurate and fast forecasting models. In this paper, a novel hybrid electrical energy consumption forecasting model is proposed based on a deep learning model known as factored conditional restricted Boltzmann machine (FCRBM). The deep learning-based FCRBM model uses a rectified linear unit (ReLU) activation function and a multivariate autoregressive technique for the network training. The proposed model predicts future electrical energy consumption for efficient energy management in the SG. The proposed model is a novel hybrid model comprising four modules: (i) data processing and features selection module, (ii) deep learning-based FCRBM forecasting module, (iii) genetic wind driven optimization (GWDO) algorithm-based optimization module, and (iv) utilization module. The proposed hybrid model, called FS-FCRBM-GWDO, is tested and evaluated on real power grid data of USA in terms of four performance metrics: mean absolute percentage deviation (MAPD), variance, correlation coefficient, and convergence rate. Simulation results validate that the proposed hybrid FS-FCRBM-GWDO model has superior performance than existing models such as accurate fast converging short-term load forecasting (AFC-STLF) model, mutual information-modified enhanced differential evolution algorithm-artificial neural network (MI-mEDE-ANN)-based model, features selection-ANN (FS-ANN)-based model, and Bi-level model, in terms of forecast accuracy and convergence rate.

Combination of short-term load forecasting models based on a stacking ensemble approach

Building electric energy consumption forecasting is essential in establishing an energy operation strategy for building energy management systems. Because of recent developments of artificial intelligence hardware, deep neural network (DNN)-based electric energy consumption forecasting models yield excellent performances. However, constructing an optimal forecasting model using DNNs is difficult and time-consuming because several hyperparameters must be determined to obtain the best combination of neural networks. The determination of the number of hidden layers in the DNN model is challenging because it greatly affects the forecasting performance of the DNN models. In addition, the best number of hidden layers for one situation or domain is often not optimal for another domain. Hence, many efforts have been made to combine multiple DNN models with different numbers of hidden layers to achieve a better forecasting performance than that of an individual DNN model. In this study, we propose a novel scheme for the combination of short-term load forecasting models using a stacking ensemble approach (COSMOS), which enables the more accurate prediction of the building electric energy consumption. For this purpose, we first collected 15-min interval electric energy consumption data for a typical office building and split them into training, validation, and test datasets. We constructed diverse four-layer DNN-based forecasting models based on the training set and by considering the input variable configuration and training epochs. We selected optimal DNN parameters using the validation set and constructed four DNN-based forecasting models with various numbers of hidden layers. We developed a building electric energy consumption forecasting model using the test set and sliding window-based principal component regression for the calculation of the final forecasting value from the forecasting values of the four DNN models. To demonstrate the performance of our approach, we conducted several experiments using actual electric energy consumption data and verified that our model yields a better prediction performance than other forecasting methods.

Short-Term Building Electrical Energy Consumption Forecasting by Employing Gene Expression Programming and GMDH Networks

Over the past decade, energy forecasting applications not only on the grid side of electric power systems but also on the customer side for load and demand prediction purposes have become ubiquitous after the advancements in the smart grid technologies. Within this context, short-term electrical energy consumption forecasting is a requisite for energy management and planning of all buildings from households and residences in the small-scale to huge building complexes in the large-scale. Today’s popular machine learning algorithms in the literature are commonly used to forecast short-term building electrical energy consumption by generating an abstruse analytical expression between explanatory variables and response variables. In this study, gene expression programming (GEP) and group method of data handling (GMDH) networks are meticulously employed for creating genuine and easily understandable mathematical models among predictor variables and target variables and forecasting short-term electrical energy consumption, belonging to a large hospital complex situated in the Eastern Mediterranean. Consequently, acquired results yielded mean absolute percentage errors of 0.620% for GMDH networks and 0.641% for GEP models, which reveal that the forecasting process can be accomplished and formulated simultaneously via proposed algorithms without the need of applying feature selection methods.

A Hybrid Approach for Energy Consumption Forecasting With a New Feature Engineering and Optimization Framework in Smart Grid

Electric energy consumption forecasting enables distribution system operators to perform efficient energy management by flexibly engaging energy consumers under the intelligent demand-response program in the smart grid (SG). With this motivation, in this paper, a fast and accurate hybrid electrical energy forecasting (FA-HELF) framework is developed. The proposed framework integrates two modules with support vector machine (SVM) based forecaster. These modules are data pre-processing and feature engineering, and modified enhanced differential evolution (mEDE) based optimizer. First, feature selection algorithms like random forests and relief-F are combined to devise a hybrid feature selection algorithm to alleviate redundancy. Secondly, for feature extraction, a radial basis Kernel-based principal component analysis algorithm is employed to eliminate the dimensionality reduction problem. Finally, to conduct accurate and fast electrical energy consumption forecasting, the mEDE based optimizer is integrated with the SVM based forecaster. The resulting FA-HELF framework is tested on publicly available independent system operator New England (ISO-NE) control area hourly load data. The results demonstrate that the FA-HELF framework is robust and shows significant improvements when compared to other benchmark frameworks in terms of accuracy and convergence speed.

Empirical mode decomposition based hybrid ensemble model for electrical energy consumption forecasting of the cement grinding process

Forecasting the electrical energy consumption of the cement grinding process remains a difficult task due to the intrinsic complexity and irregularity of its time series. To solve this difficulty and improve the prediction accuracy, a novel hybrid model is proposed based on the “decomposition-prediction-integration” methodology. The hybrid model integrates empirical mode decomposition (EMD), moving average filter (MAF), least squares support vector regression (LSSVR), and quadratic exponential smoothing (QES). And it is suitable for non-stationary time series. The proposed model is tested using hourly electrical energy consumption data of one cement grinding process in China. EMD is first applied to decompose the original data series into a limited number of independent intrinsic mode functions (IMFs) and a trend component. Then, MAF is used to reduce the high frequency noises in the IMFs. Next LSSVR is adopted to predict different IMFs, and QES is utilized to predict the trend component. At last, the predicted IMFs and trend component are summed to formulate an ensemble forecast for the original series. The performance of the EMD-LSSVR&QES hybrid model is compared with five other forecasting models. Experiment results indicate that the proposed hybrid ensemble model can give full play to the advantages of each algorithm and outperform other forecasting models.

Long-term electrical energy consumption forecasting for developing and developed economies based on different optimized models and historical data types

The objectives of this study are (a) development of optimized regression and ANN (artificial neural network) models for EEC (electrical energy consumption) forecasting based on several optimization methodologies, (b) examination of the effects of different historical data types on accuracy of EEC forecasting, and (c) long-term EEC forecasting for Iran and the U.S. as developing and developed economies, respectively. For long-term EEC forecasting for 2010–2030, the two types of historical data used in this study include, (i) EEC and (ii) socio-economic indicators, namely, gross domestic product, energy imports, energy exports, and population, for 1967–2009 period. For both types of economies, the results demonstrate that using historical data of socio-economic indicators leads to more accurate EEC forecasting than those of EEC, when IPSO (improved particle swarm optimization) is used for optimal design of ANN for EEC forecasting. It is found that for developing and developed economies, forecasted EEC trends are significantly different, as expected, and IPSO–ANN model can be utilized to forecast long-term EEC up to 2030 with mean absolute percentage error of 1.94 and 1.51% for Iran and the U.S., respectively.

Monthly Electric Energy Consumption Forecasting Using Multiwindow Moving Average and Hybrid Growth Models

Monthly electric energy consumption forecasting is important for electricity production planning and electric power engineering decision making. Multiwindow moving average algorithm is proposed to decompose the monthly electric energy consumption time series into several periodic waves and a long-term approximately exponential increasing trend. Radial basis function (RBF) artificial neural network (ANN) models are used to forecast the extracted periodic waves. A novel hybrid growth model, which includes a constant term, a linear term, and an exponential term, is proposed to forecast the extracted increasing trend. The forecasting results of the monthly electric energy consumption can be obtained by adding the forecasting values of each model. To test the performance by comparison, the proposed and other three models are used to forecast China's monthly electric energy consumption from January 2011 to December 2012. Results show that the proposed model exhibited the best performance in terms of mean absolute percentage error (MAPE) and maximal absolute percentage error (MaxAPE).

Novel effects of demand side management data on accuracy of electrical energy consumption modeling and long-term forecasting

Worldwide implementation of demand side management (DSM) programs has had positive impacts on electrical energy consumption (EEC) and the examination of their effects on long-term forecasting is warranted. The objective of this study is to investigate the effects of historical DSM data on accuracy of EEC modeling and long-term forecasting. To achieve the objective, optimal artificial neural network (ANN) models based on improved particle swarm optimization (IPSO) and shuffled frog-leaping (SFL) algorithms are developed for EEC forecasting. For long-term EEC modeling and forecasting for the U.S. for 2010–2030, two historical data types used in conjunction with developed models include (i) EEC and (ii) socio-economic indicators, namely, gross domestic product, energy imports, energy exports, and population for 1967–2009 period. Simulation results from IPSO-ANN and SFL-ANN models show that using socio-economic indicators as input data achieves lower mean absolute percentage error (MAPE) for long-term EEC forecasting, as compared with EEC data. Based on IPSO-ANN, it is found that, for the U.S. EEC long-term forecasting, the addition of DSM data to socio-economic indicators data reduces MAPE by 36% and results in the estimated difference of 3592.8 MBOE (5849.9TWh) in EEC for 2010–2030.

Electric Energy Consumption Forecasting Based on Economic Variables in Nakhonratchasima, Thailand

Electric energy is vital for social and economic development. The electric energy consumption forecasting plays an important role for energy management and allocation of resources in the future. In this paper, the influence of economic variables on the annual electric energy consumption in Nakhonratchasima has been investigated. Models are developed by using multiple regression analysis. It is founded that the electric energy consumption correlated with four economic variables, which are gross provincial product (GPP), GPP per capita, the energy sales of liquid petroleum gas (LPG) and high speed diesel usages. The historical electric energy consumption and all variables for the period 2002–2010 have been analyzed in 10 models. The study proposed 5 models for electric energy prediction in 2011. In conclusion, the effective model has been selected by comparison of adjusted R2, mean absolute error (MAE) and root mean squared error (RMSE) of the proposed models. Model 3 is acceptable in relation to electric energy consumption forecasting, with adjusted R2 and RMSE equal to 0.9915 and 1.54% respectively. The results indicate that the model using GPP and diesel usages as variables has strong ability to predict future annual electric energy consumption with 4,202,326,368 kWh in 2011.

Fuzzy Wavelet Neural Networks for City Electric Energy Consumption Forecasting

In view of the defects of the prediction model based on neural network, such as when doing prediction of nonlinear sequence, it is likely to fall into local hypo-strong point, and the rate of training is very slow. This paper presents a fuzzy wavelet neural network (FWNN) approach for annual electricity consumption in high energy consumption city. It is claimed that, due to high fluctuations of energy consumption in high energy consumption cities, conventional regression models do not forecast energy consumption correctly and precisely. Although ANNs have been typically used to forecast short term consumptions, this paper shows that it is a more precise approach to forecast annual electricity consumption. Furthermore, the FWNN approach based on ANN is used to show it can estimate the annual consumption with less error. Actual data from high energy consuming (intensive) from 1983 to 2003 is used to illustrate the applicability of the FWNN approach. This is the first study to present an algorithm based on the ANN and wavelet for forecasting long term electricity consumption in high energy consuming city. The prediction effect of wavelet neural network prediction model is proved in matlab7.0 simulation environment. A better prediction result is gained, and the defect of falling into local hypo-strongpoint is overcome, at the same time, the rate of training is raised compared with the prediction model based on artificial neural network. The calculation result shows that the presented model is effective.

Influence of artificial neural network structure on quality of short-term electric energy consumption forecast

The influence of an artificial neural network structure on the quality of short-term electric energy consumption forecasts is described. The problem of short-term electric energy consumption forecasting is very important in the daily activity of electricity distribution companies and other industrial enterprises. Accurate electric energy consumption forecast enables optimisation of the costs of electric energy purchase and planning of technological processes. There are many different forecasting methods on the market. The most accurate methods are those based on an artificial neural network technique. An attempt to investigate the influence of a one-directional ANN structure (number of hidden layers and number of neurons in layers) on the short-term forecast quality is presented. The quality of an ANN learning process is also investigated. Negative phenomena observed during the learning process, especially the phenomena of an artificial neural network overlearning, are shown. Computational experiments have been executed on the test problems. Proper data from electricity distribution companies have been taken into account. Results obtained show that the ANN structure has an essential influence on the quality of electric energy consumption forecasts.