Forecasting energy consumption in Taiwan using hybrid nonlinear models

The total consumption of electricity and petroleum energies accounts for almost 90% of the total energy consumption in Taiwan, so it is critical to model and forecast them accurately. For univariate modeling, this paper proposes two new hybrid nonlinear models that combine a linear model with an artificial neural network (ANN) to develop adjusted forecasts, taking into account heteroscedasticity in the model's input. Both of the hybrid models can decrease round-off and prediction errors for multi-step-ahead forecasting. The results suggest that the new hybrid model generally produces forecasts which, on the basis of out-of-sample forecast encompassing tests and comparisons of three different statistic measures, routinely dominate the forecasts from conventional linear models. The superiority of the hybrid ANNs is due to their flexibility to account for potentially complex nonlinear relationships that are not easily captured by linear models. Furthermore, all of the linear and nonlinear models have highly accurate forecasts, since the mean absolute percentage forecast error (MAPE) results are less than 5%. Overall, the inclusion of heteroscedastic variations in the input layer of the hybrid univariate model could help improve the modeling accuracy for multi-step-ahead forecasting.     

Real-time prediction intervals for intra-hour DNI forecasts

We develop a hybrid, real-time solar forecasting computational model to construct prediction intervals (PIs) of one-minute averaged direct normal irradiance for four intra-hour forecasting horizons: five, ten, fifteen, and 20 min. This hybrid model, which integrates sky imaging techniques, support vector machine and artificial neural network sub-models, is developed using one year of co-located, high-quality irradiance and sky image recording in Folsom, California. We validate the proposed model using six-month of measured irradiance and sky image data, and apply it to construct operational PI forecasts in real-time at the same observatory. In the real-time scenario, the hybrid model significantly outperforms the reference persistence model and provides high performance PIs regardless of forecast horizon and weather condition.     

Simulation models for the analysis of space heat consumption of buildings

This study develops and analyzes an original methodology for the simulation and prediction of space heating energy consumption in buildings connected to a district heating system, characterized by lack of individual control systems for end-users. The identification of the input parameters is based on both classical engineering equations and statistical analysis of collected data. Two main factors play important roles in the model: (i) climate and (ii) human behavior. Model validation was undertaken through the analysis of field data collected during the winter, via a monitoring system working in a partially-controlled district heating system. The comparison between the results obtained with the proposed model versus classical methods points out the possibility to implement, using the proposed methodology, management policies for a district that offer significant cost-effective energy savings opportunities.     

Forecasting of Turkey's net electricity energy consumption on sectoral bases

In this study forecast of Turkey's net electricity energy consumption on sectoral basis until 2020 is explored. Artificial neural networks (ANN) is preferred as forecasting tool. The reasons behind choosing ANN are the ability of ANN to forecast future values of more than one variable at the same time and to model the nonlinear relation in the data structure. Founded forecast results by ANN are compared with official forecasts.     

Modeling and forecasting of Turkey’s energy consumption using socio-economic and demographic variables

This study deals with the modeling of the energy consumption in Turkey in order to forecast future projections based on socio-economic and demographic variables (gross domestic product-GDP, population, import and export amounts, and employment) using artificial neural network (ANN) and regression analyses. For this purpose, four diverse models including different indicators were used in the analyses. As the result of the analyses, this research proposes Model 2 as a suitable ANN model (having four independent variables being GDP, population, the amount of import and export) to efficiently estimate the energy consumption for Turkey. The proposed model predicted the energy consumption better than the regression models and the other three ANN models. Thus, the future energy consumption of Turkey is calculated by means of this model under different scenarios. The predicted forecast results by ANN were compared with the official forecasts. Finally, it was concluded that all the scenarios that were analyzed gave lower estimates of the energy consumption than the MENR projections and these scenarios also showed that the future energy consumption of Turkey would vary between 117.0 and 175.4 Mtoe in 2014.     

Hybrid intra-hour DNI forecasts with sky image processing enhanced by stochastic learning

We propose novel smart forecasting models for Direct Normal Irradiance (DNI) that combine sky image processing with Artificial Neural Network (ANN) optimization schemes. The forecasting models, which were developed for over 6 months of intra-minute imaging and irradiance measurements, are used to predict 1 min average DNI for specific time horizons of 5 and 10 min. We discuss optimal models for low and high DNI variability seasons. The different methods used to develop these season-specific models consist of sky image processing, deterministic and ANN forecasting models, a genetic algorithm (GA) overseeing model optimization and two alternative methods for training and validation. The validation process is carried over by the Cross Validation Method (CVM) and by a randomized training and validation set method (RTM). The forecast performance for each solar variability season is evaluated, and the models with the best forecasting skill for each season are selected to build a hybrid model that exhibits optimal performance for all seasons. An independent testing set is used to assess the performance of all forecasting models. Performance is assessed in terms of common error statistics (mean bias and root mean square error), but also in terms of forecasting skill over persistence. The hybrid forecast models proposed in this work achieve statistically robust forecasting skills in excess of 20% over persistence for both 5 and 10 min ahead forecasts, respectively.     

Determination and modelling of energy consumption in wheat production using neural networks: “A case study in Canterbury province, New Zealand”

This study was conducted on irrigated and dryland wheat fields in Canterbury in the 2007-2008 harvest year based on an extensive process of data collection involving a questionnaire and interviews. Total energy consumption in wheat production was estimated at 22,566 MJ/ha. On average, fertilizer and electricity were used more than other energy sources, at around 10,654 (47%) and 4870 (22%) MJ/ha, respectively. The energy consumption for wheat production in irrigated and dryland farming systems was estimated at 25,600 and 17,458 MJ/ha, respectively.In this study, several direct and indirect factors have been identified to create an artificial neural networks (ANN) model to predict energy use in wheat production. The final model can predict energy consumption based on farm conditions (size of crop area), farmers’ social considerations (level of education), and energy inputs (N and P use and irrigation frequency), and it predicts energy use in Canterbury arable farms with an error margin of ±12% (±2900 MJ/ha). Furthermore, comparison between the ANN model and a Multiple Linear Regression (MLR) model showed that the ANN model can predict energy consumption relatively better than the MLR multiple model on the selected training set and validation set.     

Forecasting annual gross electricity demand by artificial neural networks using predicted values of socio-economic indicators and climatic conditions: Case of Turkey

In this work, the annual gross electricity demand of Turkey was modeled by multiple linear regression and artificial neural networks as a function population, gross domestic product per capita, inflation percentage, unemployment percentage, average summer temperature and average winter temperature. Among these, the unemployment percentage and the average winter temperature were found to be insignificant to determine the demand for the years between 1975 and 2013. Next, the future values of the statistically significant variables were predicted by time series ANN models, and these were simulated in a multilayer perceptron ANN model to forecast the future annual electricity demand. The results were validated with a very high accuracy for the years that the electricity demand was known (2007–2013), and they were also superior to the official predictions (done by Ministry of Energy and Natural Resources of Turkey). The model was then used to forecast the annual gross electricity demand for the future years, and it was found that, the demand will be doubled reaching about 460 TW h in the year 2028. Finally, it was concluded that the approach applied in this work can easily be implemented for other countries to make accurate predictions for the future.     

Prediction of net energy consumption based on economic indicators (GNP and GDP) in Turkey

The most important theme in this study is to obtain equations based on economic indicators (gross national product—GNP and gross domestic product—GDP) and population increase to predict the net energy consumption of Turkey using artificial neural networks (ANNs) in order to determine future level of the energy consumption and make correct investments in Turkey. In this study, three different models were used in order to train the ANN. In one of them (Model 1), energy indicators such as installed capacity, generation, energy import and energy export, in second (Model 2), GNP was used and in the third (Model 3), GDP was used as the input layer of the network. The net energy consumption (NEC) is in the output layer for all models. In order to train the neural network, economic and energy data for last 37 years (1968–2005) are used in network for all models. The aim of used different models is to demonstrate the effect of economic indicators on the estimation of NEC. The maximum mean absolute percentage error (MAPE) was found to be 2.322732, 1.110525 and 1.122048 for Models 1, 2 and 3, respectively. R² values were obtained as 0.999444, 0.999903 and 0.999903 for training data of Models 1, 2 and 3, respectively. The ANN approach shows greater accuracy for evaluating NEC based on economic indicators. Based on the outputs of the study, the ANN model can be used to estimate the NEC from the country's population and economic indicators with high confidence for planing future projections.     

Optimization of an artificial neural network dedicated to the multivariate forecasting of daily global radiation

This paper presents an application of Artificial Neural Networks (ANNs) to predict daily solar radiation. We look at the Multi-Layer Perceptron (MLP) network which is the most used of ANNs architectures. In previous studies, we have developed an ad-hoc time series preprocessing and optimized a MLP with endogenous inputs in order to forecast the solar radiation on a horizontal surface. We propose in this paper to study the contribution of exogenous meteorological data (multivariate method) as time series to our optimized MLP and compare with different forecasting methods: a naïve forecaster (persistence), ARIMA reference predictor, an ANN with preprocessing using only endogenous inputs (univariate method) and an ANN with preprocessing using endogenous and exogenous inputs. The use of exogenous data generates an nRMSE decrease between 0.5% and 1% for two stations during 2006 and 2007 (Corsica Island, France). The prediction results are also relevant for the concrete case of a tilted PV wall (1.175 kWp). The addition of endogenous and exogenous data allows a 1% decrease of the nRMSE over a 6 months-cloudy period for the power production. While the use of exogenous data shows an interest in winter, endogenous data as inputs on a preprocessed ANN seem sufficient in summer.     

Artificial neural network analysis of world green energy use

This paper focuses on the analysis of world green energy consumption through artificial neural networks (ANN). In addition, the consumption is also analyzed of world primary energy including fossil fuels such as coal, oil and natural gas. A feed-forward back-propagation ANN is used for training and learning processes by taking into consideration data from the literature of world energy consumption from 1965 to 2004. Also, an ANN approach for forecasting world green energy consumption to the year 2050 is presented, and the consumption equations for different energy sources are derived. The environmental aspects of green energy and fossil fuels are discussed in detail. The resulting ANN-based equation curve profiles verify that the available economic reserves of fossil fuel resources are limited, and become “depleted” in the near future. It is expected that world green energy consumption will reach almost 62.74 EJ by 2010, and be on average 32.29% of total energy use between 2005 and 2025. However, world green energy and natural gas consumption will continue increasing after 2050, while world oil and coal consumption are expected to remain relatively stable after 2025 and 2045, respectively. The ANN approach appears to be a suitable method for forecasting energy consumption data, should be utilized in efforts to model world energy consumption.     

The stationary of energy consumption for Turkish disaggregate data by employing linear and nonlinear unit root tests

This study is the first attempt to investigate the stationary of energy consumption for Turkish disaggregates data by employing linear and non-linear unit root tests extending from 1970 to 2006. It is concluded that the linearity is rejected in 4 cases in 7 Turkish sectors. In addition, when LM (Lagrange multiplier) tests that account for at most two structural breaks are taken into account for residential, industrial and agricultural where energy consumption follow a linear behavior, the unit root null is rejected. This means that energy demand management policies designed to shrink energy consumption will instead have a transitory impact on the energy consumption and return it toward its original trend path. On the other hand, it’s concluded that the transportation, non-energy uses & other, final energy consumption and cycle & energy sector’s energy consumption are non-stationary which means that any shock to energy consumption is likely to be permanent and energy demand management policies will have a permanent impact on these sectors.     

Day-ahead electricity price forecasting using back propagation neural networks and weighted least square technique

This paper proposes the day-ahead electricity price forecasting using the artificial neural networks (ANN) and weighted least square (WLS) technique in the restructured electricity markets. Price forecasting is very important for online trading, e-commerce and power system operation. Forecasting the hourly locational marginal prices (LMP) in the electricity markets is a very important basis for the decision making in order to maximize the profits/benefits. The novel approach proposed in this paper for forecasting the electricity prices uses WLS technique and compares the results with the results obtained by using ANNs. To perform this price forecasting, the market knowledge is utilized to optimize the selection of input data for the electricity price forecasting tool. In this paper, price forecasting for Pennsylvania-New Jersey-Maryland (PJM) interconnection is demonstrated using the ANNs and the proposed WLS technique. The data used for this price forecasting is obtained from the PJM website. The forecasting results obtained by both methods are compared, which shows the effectiveness of the proposed forecasting approach. From the simulation results, it can be observed that the accuracy of prediction has increased in both seasons using the proposed WLS technique. Another important advantage of the proposed WLS technique is that it is not an iterative method.     

Day-ahead price forecasting of electricity markets by a new feature selection algorithm and cascaded neural network technique

With the introduction of restructuring into the electric power industry, the price of electricity has become the focus of all activities in the power market. Electricity price forecast is key information for electricity market managers and participants. However, electricity price is a complex signal due to its non-linear, non-stationary, and time variant behavior. In spite of performed research in this area, more accurate and robust price forecast methods are still required. In this paper, a new forecast strategy is proposed for day-ahead price forecasting of electricity markets. Our forecast strategy is composed of a new two stage feature selection technique and cascaded neural networks. The proposed feature selection technique comprises modified Relief algorithm for the first stage and correlation analysis for the second stage. The modified Relief algorithm selects candidate inputs with maximum relevancy with the target variable. Then among the selected candidates, the correlation analysis eliminates redundant inputs. Selected features by the two stage feature selection technique are used for the forecast engine, which is composed of 24 consecutive forecasters. Each of these 24 forecasters is a neural network allocated to predict the price of 1 h of the next day. The whole proposed forecast strategy is examined on the Spanish and Australia’s National Electricity Markets Management Company (NEMMCO) and compared with some of the most recent price forecast methods.     

Households’ access to electricity and energy consumption pattern in Nigeria

Access to substantial quantity and quality energy infrastructures is essential to rapid and sustainable economic development. Access to modern energy services directly contributes to economic growth and poverty reduction through the creation of wealth. No country can develop and sustain beyond subsistence means without having at least minimum access to energy services for the larger portion of its population. The present study examines the households’ access to modern energy (electricity) services and pattern of energy consumption in Nigeria. It was found that the access to modern form of energy in the country is very low despite the country's abundant energy endowment. Greater proportions (over 40%) of Nigerian households do not have access to electricity and still depend largely on traditional forms of energy (e.g., firewood and kerosene) as energy sources.     

Artificial neural networks for energy analysis of office buildings with daylighting

An artificial neural network (ANN) model was developed for office buildings with daylighting for subtropical climates. A total of nine variables were used as the input parameters – four variables were related to the external weather conditions (daily average dry-bulb temperature, daily average wet-bulb temperature, daily global solar radiation and daily average clearness index), four for the building envelope designs (solar aperture, daylight aperture, overhang and side-fins projections), and the last variable was day type (i.e. weekdays, Saturdays and Sundays). There were four nodes at the output layer with the estimated daily electricity use for cooling, heating, electric lighting and total building as the output. Building energy simulation using EnergyPlus was conducted to generate daily building energy use database for the training and testing of ANNs. The Nash–Sutcliffe efficiency coefficient for the ANN modelled cooling, heating, electric lighting and total building electricity use was 0.994, 0.940, 0.993, and 0.996, respectively, indicating excellent predictive power. Error analysis showed that lighting electricity use had the smallest errors, from 0.2% under-estimation to 3.6% over-estimation, with the coefficient of variation of the root mean square error ranging from 3% to 5.6%.     

Short-term prediction of household electricity consumption: Assessing weather sensitivity in a Mediterranean area

Urban microclimatic variations, along with a rapid reduction of unit cost of air-conditioning (AC) equipments, can be addressed as some of the main causes of the raising residential energy demand in the more developed countries. This paper presents a forecasting model based on an Elman artificial neural network (ANN) for the short-time prediction of the household electricity consumption related to a suburban area. Due to the lack of information about the real penetration of electric appliances in the investigated area and their utilization profiles it was not possible to implement a statistical model to define the weather and climate sensitivities of appliance energy consumption. For this reason an ANN model was used to predict the household electric energy demand of the investigated area and to evaluate the influence of the AC equipments on the overall consumption.The data used to train the network were recorded in Palermo (Italy) and include electric current intensity and weather variables as temperature, relative humidity, global solar radiation, atmospheric pressure and wind speed values between June 1, 2002 and September 10, 2003.The work pointed out the importance of a thermal discomfort index, the Humidex index, for a simple but effective evaluation of the conditions affecting the occupant behaviour and thus influencing the household electricity consumption related to the use of heating, ventilation and air conditioning (HVAC) appliances. The prediction performances of the model are satisfying and bear out the ability of ANNs to manage incomplete and noisy data, solve nonlinear problems and learn complex underlying relationships between input and output patterns.     

Neural networks in forecasting electrical energy consumption: univariate and multivariate approaches

This paper presents an artificial neural network (ANN) approach to electric energy consumption (EEC) forecasting in Lebanon. In order to provide the forecasted energy consumption, the ANN interpolates among the EEC and its determinants in a training data set. In this study, four ANN models are presented and implemented on real EEC data. The first model is a univariate model based on past consumption values. The second model is a multivariate model based on EEC time series and a weather‐dependent variable, namely, degree days (DD). The third model is also a multivariate model based on EEC and a gross domestic product (GDP) proxy, namely, total imports (TI). Finally, the fourth model combines EEC, DD and TI. Forecasting performance measures such as mean square errors (MSE), mean absolute deviations (MAD), mean percentage square errors (MPSE) and mean absolute percentage errors (MAPE) are presented for all models. Copyright © 2002 John Wiley & Sons, Ltd.     

Modeling of electricity consumption in the Asian gaming and tourism center—Macao SAR,People's Republic of China

The use of electricity is indispensable to modern life. As Macao Special Administrative Region becomes a gaming and tourism center in Asia, modeling the consumption of electricity is critical to Macao's economic development. The purposes of this paper are to conduct an extensive literature review on modeling of electricity consumption, and to identify key climatic, demographic, economic and/or industrial factors that may affect the electricity consumption of a country/city. It was identified that the five factors, namely temperature, population, the number of tourists, hotel room occupancy and days per month, could be used to characterize Macao's monthly electricity consumption. Three selected approaches including multiple regression, artificial neural network (ANN) and wavelet ANN were used to derive mathematical models of the electricity consumption. The accuracy of these models was assessed by using the mean squared error (MSE), the mean squared percentage error (MSPE) and the mean absolute percentage error (MAPE). The error analysis shows that wavelet ANN has a very promising forecasting capability and can reveal the periodicity of electricity consumption.     

Electric load forecasting by using dynamic neural network

Electrical energy is fundamental for the wellbeing and for the economic development of any country. However, all countries must ensure access to essential resources and ensure the continuity of its supply. Due to the non-storable nature of electrical energy, the amount of consumed active power should always be equal the produced active power just to avoid power system frequency deviation problem. In order to keep the relationship production–consumption relation in compliance with different standards and to secure profitable operations of power system, electric load consumption must be predicted and controlled instantaneously. Several statistical and classical techniques are proposed in the literature but unfortunately all these methods are not accurate in a satisfactory manner. In this paper, a dynamic neural network is used for the prediction of daily power consumption. The suitability and the performance of the proposed approach is illustrated and verified with simulations on load data collected from French Transmission System Operator (RTE) website. The obtained results show that the accuracy and the efficiency are improved comparatively to conventional methods widely used in this field of research.