ARIMA forecasting of primary energy demand by fuel in Turkey

Forecasting of energy demand in emerging markets is one of the most important policy tools used by the decision makers all over the world. In Turkey, most of the early studies used include various forms of econometric modeling. However, since the estimated economic and demographic parameters usually deviate from the realizations, time-series forecasting appears to give better results. In this study, we used the Autoregressive Integrated Moving Average (ARIMA) and seasonal ARIMA (SARIMA) methods to estimate the future primary energy demand of Turkey from 2005 to 2020. The ARIMA forecasting of the total primary energy demand appears to be more reliable than the summation of the individual forecasts. The results have shown that the average annual growth rates of individual energy sources and total primary energy will decrease in all cases except wood and animal–plant remains which will have negative growth rates. The decrease in the rate of energy demand may be interpreted that the energy intensity peak will be achieved in the coming decades. Another interpretation is that any decrease in energy demand will slow down the economic growth during the forecasted period. Rates of changes and reserves in the fossil fuels indicate that inter-fuel substitution should be made leading to a best mix of the country's energy system. Based on our findings we proposed some policy recommendations.     

Sectoral energy consumption in Turkey

Turkey expects a very large growth in energy demand, especially for electricity and natural gas. Today, Turkey’s energy production meets nearly 48% of the total primary energy demand. Total primary energy demand will reach 98 Mtoe in 2001 and 308 Mtoe in 2020. Import of primary energy will reach 226 Mtoe and production of primary energy will increase 81 Mtoe in 2020. As seen, Turkey is an importer country for primary energy. Turkey’s indigenous energy sources are limited, and the country is heavily dependent on the import of primary energy from abroad. The growth of Turkey’s industry is giving rise to a substantial increase in energy demand. In this paper, the primary energy production and sectoral consumption in Turkey is investigated. Further, a sectoral energy demand projection in Turkey is given until 2020.     

A seasonal hybrid procedure for electricity demand forecasting in China

Electricity is a special energy which is hard to store, so the electricity demand forecasting in China remains an important problem. This paper aims at developing an improved hybrid model for electricity demand in China, which takes the advantages of moving average procedure, combined method, hybrid model and adaptive particle swarm optimization algorithm, known as MA-C-WH. It is designed for making trend and seasonal adjustments which simultaneously presents the electricity demand forecasts. Four actual electricity demand time series in China power grids are selected to illustrate the proposed MA-C-WH model, and one existing seasonal autoregressive integrated moving average model (SARIMA) is selected to compare with the proposed model using the same data series. The results of popular forecasting precision indexes show that our proposed model is an effective forecasting technique for seasonal time series with nonlinear trend.     

Prediction of daily peak electricity demand in South Africa using volatility forecasting models

Daily peak electricity demand forecasting in South Africa using a seasonal autoregressive integrated moving average (SARIMA) model, a SARIMA model with generalized autoregressive conditional heteroskedastic (SARIMA–GARCH) errors and a regression-SARIMA–GARCH (Reg-SARIMA–GARCH) model is presented in this paper. The GARCH modeling methodology is introduced to accommodate the possibility of serial correlation in volatility since the daily peak demand data exhibits non-constant mean and variance, and multiple seasonality corresponding to weekly and monthly periodicity. The proposed Reg-SARIMA–GARCH model is designed in such a way that the predictor variables are initially selected using a multivariate adaptive regression splines algorithm. The developed models are used for out of sample prediction of daily peak demand. A comparative analysis is done with a piecewise linear regression model. Results from the study show that the Reg-SARIMA–GARCH model produces better forecast accuracy with a mean absolute percent error (MAPE) of 1.42%.     

Short term and medium term power distribution load forecasting by neural networks

Load forecasting is an important subject for power distribution systems and has been studied from different points of view. In general, load forecasts should be performed over a broad spectrum of time intervals, which could be classified into short term, medium term and long term forecasts. Several research groups have proposed various techniques for either short term load forecasting or medium term load forecasting or long term load forecasting. This paper presents a neural network (NN) model for short term peak load forecasting, short term total load forecasting and medium term monthly load forecasting in power distribution systems. The NN is used to learn the relationships among past, current and future temperatures and loads. The neural network was trained to recognize the peak load of the day, total load of the day and monthly electricity consumption. The suitability of the proposed approach is illustrated through an application to real load shapes from the Turkish Electricity Distribution Corporation (TEDAS) in Nigde. The data represents the daily and monthly electricity consumption in Nigde, Turkey.     

The application of seasonal latent variable in forecasting electricity demand as an alternative method

In this study, we used ARIMA, seasonal ARIMA (SARIMA) and alternatively the regression model with seasonal latent variable in forecasting electricity demand by using data that belongs to “Kayseri and Vicinity Electricity Joint-Stock Company” over the 1997:1–2005:12 periods. This study tries to examine the advantages of forecasting with ARIMA, SARIMA methods and with the model has seasonal latent variable to each other. The results support that ARIMA and SARIMA models are unsuccessful in forecasting electricity demand. The regression model with seasonal latent variable used in this study gives more successful results than ARIMA and SARIMA models because also this model can consider seasonal fluctuations and structural breaks.     

Ant colony optimization approach to estimate energy demand of Turkey

This paper attempts to shed light on the determinants of energy demand in Turkey. Energy demand model is first proposed using the ant colony optimization (ACO) approach. It is multi-agent systems in which the behavior of each ant is inspired by the foraging behavior of real ants to solve optimization problem. ACO energy demand estimation (ACOEDE) model is developed using population, gross domestic product (GDP), import and export. All equations proposed here are linear and quadratic. Quadratic_ACOEDE provided better-fit solution due to fluctuations of the economic indicators. The ACOEDE model plans the energy demand of Turkey until 2025 according to three scenarios. The relative estimation errors of the ACOEDE model are the lowest when they are compared with the Ministry of Energy and Natural Resources (MENR) projection.     

The forecast of motor vehicle,energy demand and CO2 emission from Taiwan's road transportation sector

The grey forecasting model, GM(1,1) was adopted in this study to capture the development trends of the number of motor vehicles, vehicular energy consumption and CO₂ emissions in Taiwan during 2007–2025. In addition, the simulation of different economic development scenarios were explored by modifying the value of the development coefficient, a, in the grey forecasting model to reflect the influence of economic growth and to be a helpful reference for realizing traffic CO₂ reduction potential and setting CO₂ mitigation strategies for Taiwan. Results showed that the vehicle fleet, energy demand and CO₂ emitted by the road transportation system continued to rise at the annual growth rates of 3.64%, 3.25% and 3.23% over the next 18 years. Besides, the simulation of different economic development scenarios revealed that the lower and upper bound values of allowable vehicles in 2025 are 30.2 and 36.3 million vehicles, respectively, with the traffic fuel consumption lies between 25.8 million kiloliters to 31.0 million kiloliters. The corresponding emission of CO₂ will be between 61.1 and 73.4 million metric tons in the low- and high-scenario profiles.     

Using machine learning tools for forecasting natural gas consumption in the province of Istanbul

Commensurate with unprecedented increases in energy demand, a well-constructed forecasting model is vital to managing energy policies effectively by providing energy diversity and energy requirements that adapt to the dynamic structure of the country. In this study, we employ three alternative popular machine learning tools for rigorous projection of natural gas consumption in the province of Istanbul, Turkey's largest natural gas-consuming mega-city. These tools include multiple linear regression (MLR), an artificial neural network approach (ANN) and support vector regression (SVR). The results indicate that the SVR is much superior to ANN technique, providing more reliable and accurate results in terms of lower prediction errors for time series forecasting of natural gas consumption. This study could well serve a useful benchmarking study for many emerging countries due to the data structure, consumption frequency, and consumption behavior of consumers in various time-periods.     

Electric load forecasting by seasonal recurrent SVR (support vector regression) with chaotic artificial bee colony algorithm

Support vector regression (SVR), with hybrid chaotic sequence and evolutionary algorithms to determine suitable values of its three parameters, not only can effectively avoid converging prematurely (i.e., trapping into a local optimum), but also reveals its superior forecasting performance. Electric load sometimes demonstrates a seasonal (cyclic) tendency due to economic activities or climate cyclic nature. The applications of SVR models to deal with seasonal (cyclic) electric load forecasting have not been widely explored. In addition, the concept of recurrent neural networks (RNNs), focused on using past information to capture detailed information, is helpful to be combined into an SVR model. This investigation presents an electric load forecasting model which combines the seasonal recurrent support vector regression model with chaotic artificial bee colony algorithm (namely SRSVRCABC) to improve the forecasting performance. The proposed SRSVRCABC employs the chaotic behavior of honey bees which is with better performance in function optimization to overcome premature local optimum. A numerical example from an existed reference is used to elucidate the forecasting performance of the proposed SRSVRCABC model. The forecasting results indicate that the proposed model yields more accurate forecasting results than ARIMA and TF-ε-SVR-SA models. Therefore, the SRSVRCABC model is a promising alternative for electric load forecasting.     

Short-term forecasting of crude petroleum and natural gas production

This paper details the Box-Jenkins approach to forecasting time series and applies it to short-term natural gas marketed production and crude petroleum production in the United States. After establishing the efficacy of the approach for forecasting the two series of interest, monthly forecasts for 1978 are made. The results indicate that natural gas production in 1978 will increase by 2·8 per cent over the 1977 level while crude petroleum production will fall by 4·0 per cent.     

Forecasting risks of natural gas consumption in Slovenia

Efficient operation of modern energy distribution systems often requires forecasting future energy demand. This paper proposes a strategy to estimate forecasting risk. The objective of the proposed method is to improve knowledge about expected forecasting risk and to estimate the expected cash flow in advance, based on the risk model. The strategy combines an energy demand forecasting model, an economic incentive model and a risk model. Basic guidelines are given for the construction of a forecasting model that combines past energy consumption data, weather data and weather forecast. The forecasting model is required to estimate expected forecasting errors that are the basis for forecasting risk estimation. The risk estimation strategy also requires an economic incentive model that describes the influence of forecasting accuracy on the energy distribution systems’ cash flow. The economic model defines the critical forecasting error levels that most strongly influence cash flow. Based on the forecasting model and the economic model, the development of a risk model is proposed. The risk model is associated with critical forecasting error levels in the context of various influential parameters such as seasonal data, month, day of the week and temperature. The risk model is applicable to estimating the daily forecasting risk based on the influential parameters. The proposed approach is illustrated by a case study of a Slovenian natural gas distribution company.     

A trend fixed on firstly and seasonal adjustment model combined with the ε-SVR for short-term forecasting of electricity demand

Short-term electricity demand forecasting has always been an essential instrument in power system planning and operation by which an electric utility plans and dispatches loading so as to meet system demand. The accuracy of the dispatching system, derived from the accuracy of demand forecasting and the forecasting algorithm used, will determines the economic of the power system operation as well as the stability of the whole society. This paper presents a combined ε-SVR model considering seasonal proportions based on development tendencies from history data. We use one-order moving averages to produce a comparatively smooth data series, taking the averaging period as the interval that can effectively eliminate the seasonal variation. We used the smoothed data series as the training set input for the ε-SVR model and obtained the corresponding forecasting value. Afterward, we accounted for the previously removed seasonal variation. As a case, we forecast northeast electricity demand of China using the new method. We demonstrated that this simple procedure has very satisfactory overall performance by an analysis of variance with relative verification and validation. Significant reductions in forecast errors were achieved.     

Long term forecasting of natural gas production

Natural gas is an important energy source for power generation, a chemical feedstock and residential usage. It is important to analyse the future production of conventional and unconventional natural gas. Analysis of the literature determined conventional URR estimates of 10,700–18,300 EJ, and the unconventional gas URR estimates were determined to be 4250–11,000 EJ. Six scenarios were assumed, with three static where demand and supply do not interact and three dynamic where it does. The projections indicate that world natural gas production will peak between 2025 and 2066 at 140–217 EJ/y (133–206 tcf/y). Natural gas resources are more abundant than some of the literature indicates.     

An analysis of industrial demand for natural gas

Several analysts have proposed the theory that there is a frustrated demand for natural gas in the United States. According to this theory, the natural gas curtailments of the 1970s have convinced industrial users that natural gas is not reliably available. The users are willing to buy fuel oil at a premium and store it to assure a reliable energy supply. If this theory is correct, as the supply of natural gas becomes more reliable, natural gas should be able to recapture the market share it lost to fuel oil.To investigate this theory, a state-level data base on demand for fuel oil and natural gas by all manufacturing sectors for 1971 and 1974 to 1979 was developed. Then a model was developed to explain manufacturers' demand for fuel oil and natural gas during this period on the basis of fuel price and the availability of natural gas. Rather than a frustrated demand for natural gas, we find an excess demand for natural gas in several states. Fuel oil may be able to capture the market from natural gas in these states.     

Vision 2023: Feasibility analysis of Turkey's renewable energy projection

Electricity consumption of Turkey at the year 2023 is estimated to be around 530,000 GWh. Turkey plans to supply 30% or 160,000 GWh of this demand from renewable energy sources according to the recently avowed government agenda Vision 2023. However, the current installed renewable energy capacity is around 60,000 GWh. Detailed literature analysis showed that only wind and solar energy potential in Turkey can solely supply this demand. In this study, two different scenarios were generated to analyse the cost and environmental impacts of supplying this demand. Scenario 1, which is derived from the official Vision 2023 targets, suggests supplying this demand from wind, solar, geothermal energy and hydropower. The total projected cost based on Scenario 1 is estimated to be $31.000 billion and annual greenhouse gas emissions of 1.05 million tonnes of CO₂ equivalent. According to Scenario 2 or the contrary setup it is assumed that the required demand gap could not be supplied from new renewable energy investments but equally from coal and natural gas. The projected cost is estimated to be around $8.000 billion and annual greenhouse gas emissions at appalling 71.30 million tonnes of CO₂ equivalent. Assuming carbon tax at the year 2023 to be $50 per tonne of CO₂ emitted, supplying the demand from renewable energy sources according to Scenario 1 would generate savings worth nearly $2.175 billion from environmental taxes annually. Thus, making the payback time of the renewable energy investments less than 15 years.     

A hybrid procedure for energy demand forecasting in China

Energy consumption in China is continuously increasing. Accordingly, the present paper aims to develop a hybrid procedure for energy demand forecasting in China with higher precision. The mechanism of the affecting factors of China’s energy demand is investigated via path-coefficient analysis. The main affecting factors include gross domestic product, population, economic structure, urbanization rate, and energy structure. These factors are the inputs of the model with three forms: linear, exponential, and quadratic. To obtain better parameters, an improved hybrid algorithm called PSO-GA (particle swarm optimization-genetic algorithm) is proposed. This proposed algorithm differs from previous hybrids in the two ways. First, the GA and PSO approaches produce a hybrid hierarchy. Second, two information transfers are accomplished in the process. Results of this study show that China’s energy demand will be 4.70 billion tons coal equivalent in 2015. Furthermore, the proposed forecast method shows its superiority compared with single optimization methods, such as GA, PSO or ant colony optimization, and multiple linear regressions.     

Turkey’s short-term gross annual electricity demand forecast by fuzzy logic approach

This paper aims to forecast Turkey's short-term gross annual electricity demand by applying fuzzy logic methodology while general information on economical, political and electricity market conditions of the country is also given. Unlike most of the other forecast models about Turkey's electricity demand, which usually uses more than one parameter, gross domestic product (GDP) based on purchasing power parity was the only parameter used in the model. Proposed model made good predictions and captured the system dynamic behavior covering the years of 1970–2014. The model yielded average absolute relative errors of 3.9%. Furthermore, the model estimates a 4.5% decrease in electricity demand of Turkey in 2009 and the electricity demand growth rates are projected to be about 4% between 2010 and 2014. It is concluded that forecasting the Turkey's short-term gross electricity demand with the country's economic performance will provide more reliable projections. Forecasting the annual electricity consumption of a country could be made by any designer with the help of the fuzzy logic procedure described in this paper. The advantage of this model lies on the ability to mimic the human thinking and reasoning.     

A system dynamic model for production and consumption policy in Iran oil and gas sector

A system dynamic model is presented, which considers the feedback between supply and demand and oil revenue of the existing system in Iran considering different sectors of the economy. Also the export of the oil surplus and the injection of the gas surplus into the oil reservoirs are seen in the model by establishing a balance between supply and demand. In this model the counter-effects and existing system feedbacks between supply and demand and oil revenue can be seen considering different sectors of the economy. As a result, the effects of oil and gas policies in different scenarios for different sectors of Iran’s economy together with the counter-effects of energy consumption and oil revenue are examined. Three scenarios, which show the worst, base and ideal cases, are considered to find future trends of major variables such as seasonal gas consumption in power plants, seasonal injected gas in oil reservoirs, economic growth in the industrial sector, oil consumption in the transportation sector, industrial gas consumption and exported gas. For example, it is shown that the exported gas will reach between 500 and 620 million cubic-meter per day in different scenarios and export revenues can reach up to $500 billion by 2025.     

Exploring the production of natural gas through the lenses of the ACEGES model

Due to the increasing importance of natural gas for modern economic activity, and gas's non-renewable nature, it is extremely important to try to estimate possible trajectories of future natural gas production while considering uncertainties in resource estimates, demand growth, production growth and other factors that might limit production. In this study, we develop future scenarios for natural gas supply using the ACEGES computational laboratory. Conditionally on the currently estimated ultimate recoverable resources, the ‘Collective View’ and ‘Golden Age’ Scenarios suggest that the supply of natural gas is likely to meet the increasing demand for natural gas until at least 2035. The ‘Golden Age’ Scenario suggests significant ‘jumps’ of natural gas production – important for testing the resilience of long-term strategies.