Long-term Electricity Consumption Research Articles

A factor-based bottom-up approach for the long-term electricity consumption estimation in the Japanese residential sector

Recent years, the energy consumption and carbon emissions released from the residential sector have increased rapidly due to the improvement of living standards. Japan, as one of the most developed countries worldwide, is found to be the fourth largest CO2 emitter in the world. Meanwhile, Japan is currently promoting social electrification from varied aspects. And this action further brings the household residential consumption to the vital place for overall energy conservation and emission reduction plans. Although electricity consumption and prediction analysis have been widely discussed. However, previous studies mainly focused on the estimation and analysis of energy consumption at the national level, without enough discussion from prefecture-level insights. To bridge this knowledge gap, this study established a factor-based bottom-up model to estimate the electricity consumption and carbon emissions during the 2015–2040 periods in the Japanese residential sector, considering the prefectural characteristics including per capita gross domestic product, population size, household size, residential floor space area, lifestyle, weather condition, and transition effect of appliance in the future. Nine scenarios that combine three levels of household size and three levels of growth of per capita gross domestic product are taken into account to estimate the electricity consumption for space heating and cooling, water heating, cooking, and appliances. Results indicate that the total residential electricity consumption will reach a peak during 2020s. The total carbon emissions will keep decreasing by 51.14–72.16 Mt between 2015 and 2040. Based on the results of this paper, policy recommendations are given for the Japanese government.

Forecasting residential electricity consumption: a bottom-up approach for Brazil by region

In this paper, we forecast residential electricity consumption in Brazil considering the variability in consumption behavior in different regions. In order to do so, we use a bottom-up approach to estimate long-term electricity consumption that considers three technology-driven scenarios: one assuming reference efficiency development, another one based on the Brazilian government’s predictions, and the third one assuming high efficiency. The results show an increase of residential electricity consumption due to the increasing appliance ownership rate, especially air conditioning, as well as the increasing number of households. In addition, it is shown that 96 TWh of electricity could be saved in the highest efficiency scenario due to significant efficiency gains between 2016 and 2050. Further, the results indicate the southeast of the country has the highest potential for consumption reduction: by 27% in the most efficient scenario. This work is the first to use a bottom-up approach to forecast the residential electricity consumption by region and by technology-driven scenarios in Brazil. With this methodology, it is possible to indicate which types of appliances should be targeted by policy makers when designing energy efficiency incentives. Hence, this model can be used as a roadmap to achieve the desired efficiency gains in Brazil.

Accurate Deep Model for Electricity Consumption Forecasting Using Multi-Channel and Multi-Scale Feature Fusion CNN–LSTM

Electricity consumption forecasting is a vital task for smart grid building regarding the supply and demand of electric power. Many pieces of research focused on the factors of weather, holidays, and temperatures for electricity forecasting that requires to collect those data by using kinds of sensors, which raises the cost of time and resources. Besides, most of the existing methods only focused on one or two types of forecasts, which cannot satisfy the actual needs of decision-making. This paper proposes a novel hybrid deep model for multiple forecasts by combining Convolutional Neural Networks (CNN) and Long-Short Term Memory (LSTM) algorithm without additional sensor data, and also considers the corresponding statistics. Different from the conventional stacked CNN–LSTM, in the proposed hybrid model, CNN and LSTM extracted features in parallel, which can obtain more robust features with less loss of original information. Chiefly, CNN extracts multi-scale robust features by various filters at three levels and wide convolution technology. LSTM extracts the features which think about the impact of different time-steps. The features extracted by CNN and LSTM are combined with six statistical components as comprehensive features. Therefore, comprehensive features are the fusion of multi-scale, multi-domain (time and statistic domain) and robust due to the utilization of wide convolution technology. We validate the effectiveness of the proposed method on three natural subsets associated with electricity consumption. The comparative study shows the state-of-the-art performance of the proposed hybrid deep model with good robustness for very short-term, short-term, medium-term, and long-term electricity consumption forecasting.

Long-term electrical consumption forecasting using Artificial Neural Network (ANN)

Long-term forecasting of electricity energy consumption has become one of the main fields in the electricity sector in each country. This study aims to compare the accuracy of the results of the method used in this study with research conducted by the government to estimate electricity consumption in Indonesia. Many methods can be used to estimate electrical energy consumption such as statistic methods (Exponential Smoothing, ARIMA, Regression), fuzzy logic and artificial neural network algorithms. The method used in RUPTL is Simple E (Simple Econometric) and method used in this study is the Artificial Neural Network algorithm. The results of this research are data on estimates of electricity consumption in Indonesia for 2019-2025. This research is expected to prove the accuracy of the Artificial Neural Network (ANN) method to estimate electricity consumption in Indonesia.

Thermal performances study on a façade-built-in two-phase thermosyphon loop for passive thermo-activated building system

Pipe-embedded building envelopes, as a burgeoning active insulation technique that has its pipe circuits inside the building envelope, are of particular interest to architects and engineers due to their excellent performances and invisibility characteristic. However, performances of this pump-driven technique highly rely on long-term electricity consumption, resulting in degradation of system efficiency in building load reduction and supplementary heating. In the present study, a facade-built-in two-phase thermosyphon loop (TPTL) for passive thermo-activated building system (i.e. TPTL-embedded building envelope) is proposed to turn the pump-driven heat injection process of active insulation technique into non-pump-driven. Experiments on a prototype are conducted to explore operational characteristics of the TPTL whose condenser is operated in the thermal boundary of Dirichlet and dynamic thermal response characteristics of the composite building envelope after coupling TPTL at different heat loads and filling ratios. Results show that the filling ratio has a critical impact on thermal resistances and startup behavior of TPTL, and the optimal filling ratio (around 116%) is not a fixed value at different operating conditions. Besides, condenser thermal resistance occupies the largest proportion among the system thermal resistance due to the heat accumulation phenomenon, therefore, the interlayer material also has a great impact on heat injection and release of TPTL-embedded building envelope. In addition, reverse startup and operation under specific test condition are observed and discussed accordingly. Overall, these results and conclusions are useful for the system design, operation control, and further investigations on the façade-built-in TPTL and the TPTL-embedded building envelope.

Long-term electricity consumption forecasting based on expert prediction and fuzzy Bayesian theory

Long-term electricity consumption (EC) forecasting is a very important part for the expansion planning of power system. Instead of point forecasting, based on fuzzy Bayesian theory and expert prediction, a novel long-term probability forecasting model is proposed to predict the Chinese per-capita electricity consumption (PEC) and its variation interval over the period 2010–2030. The special model structure can improve the reliability and accuracy of expert prediction through econometric methodology. It contains three components: fuzzy relation matrix, prior prediction, and fuzzy Bayesian formula. To contend with the long-term uncertainty, the prior prediction is implemented to combine the advantages of expert's experience with other time-based methods from the perspective of probability. With the utilization of fuzzy technique, the multiple effects of influencing factors (IFs) on PEC can be expressed as a fuzzy relation matrix. It can rule the results of prior prediction to obey the long-run equilibrium relationship of natural evolution thorough probability calibration. To demonstrate its efficiency and applicability, the result of this method is compared with that of other 6 approaches and 4 agencies. The case study shows that the proposed methodology has higher accuracy and adaptability.

Forecasting Uganda’s Net Electricity Consumption Using a Hybrid PSO-ABC Algorithm

Long-term electricity consumption forecasting is used by decision makers to make decisions regarding system expansion planning. Over the past decade, research on electricity consumption forecasting has reported results as point forecasts. Specifically for long-term forecasting, point forecasts are of little interest because it is hard to use them to assess the financial risk associated with system expansion versus demand variability and forecasting uncertainty. In this study, firstly we use power and quadratic forms to model Uganda’s net electricity consumption using population, gross domestic product, number of subscribers and average electricity price as variables in the forecasting models. We optimize the parameters of the forecasting models using a hybrid algorithm based on particle swarm optimization and artificial bee colony algorithms. Secondly we model the forecast residuals using simple linear regression to obtain 90% significance level lower and upper error bounds. The lower and upper error bounds were used to construct predication intervals for Uganda’s net electricity consumption up to year 2040. We accessed the accuracy and performance of the models using MAPE and $$R^{2}$$ . We obtained 1.4387 and 1.1741% for MAPE and 0.9920 and 0.9948 for $$R^{2}$$ , respectively, for power and quadratic models. According to our results, in year 2040 Uganda’s electricity consumption will be between [35,471.5, 36,317.6] GWh for the power model indicating an annual average increase of 10.1–11.3%. For the quadratic model consumption is expected to be between [47,443.1, 48,289.4] GWh which indicates an average annual increase of 11.2–12.4%.

A Hybrid Modeling of Long-Term Electricity Consumption Forecasting Based on ARIMA and ANN: The Case of Thailand with Projection

Modeling and forecasting of electricity consumption can provide reliable guidance for power operation and planning in developing countries such as Thailand. In this study, formulates the effects of two different historical data type is modeled by auto regressive integrated moving averaged (ARIMA) and artificial neural network (ANN) based on population and gross domestic product per capita (GDP). The derived model is validated by various statistical approaches such as the determination coefficient. Additionally, the performances of the derived model are assessed using mean absolute percentage error (MAPE), root mean square error (RMSE) and mean absolute error (MAE). Three scenarios are used for forecasting Thailand’s electricity consumption in 2011 – 2015. The simulation results are validated by actual data sets observed from 1993 to 2010. Empirical results showed that the proposed method has higher accuracy compared to single ARIMA and artificial intelligence based models.

COMPARING TIME SERIES FORECASTING METHODS TO ESTIMATE WIND SPEED IN KIRIKKALE REGION

Due to the non-storable nature of electric energy, short-term and long-term electricity generation and consumption forecast are critical to keeping electricity market in balance. In addition, the production estimate of wind energy is parallel to the estimate of wind speed. Since wind speed forecasts includes seasonal and time-dependent trends, time series forecasting methods produce successful results in wind energy forecasting. However, choosing the most appropriate time series forecasting method for short-term and long-term production forecasts is of special importance. In this study, short-term and long-term wind speed estimations were made for the wind turbine at Kirikkale University by using Exponential Smoothing (ES) and ARMA (Auto Regressive Moving Average) methods. The most suitable methods for forecasting short-term and long-term wind speed have been determined with the obtained results.

Long-term electrical energy consumption formulating and forecasting via optimized gene expression programming

This study formulates the effects of two different historical data types on electrical energy consumption of ASEAN-5 counties. On this basis, optimized GEP (gene expression programming) is applied to precisely formulate the relationships between historical data and electricity consumption. The optimized GEP is a more recent extension of GEP with high probability of finding closed-form solution in mathematical modeling without prior knowledge about the nature of the relationships between variables. This merit is provided by balancing the exploration of solution structure and exploitation of its appropriate weighting factors through use of a robust and efficient optimization algorithm in learning process of GEP. To assess the applicability and accuracy of the proposed method, its estimates are compared with those obtained from ANN (artificial neural network), SVR (support vector regression), ANFIS (adaptive neuro-fuzzy inference system), rule-based data mining algorithm, GEP, linear and quadratic models optimized by PSO (particle swarm optimization), CSA (cuckoo search algorithm) and BSA (backtracking search algorithm). The simulation results are validated by actual data sets observed from 1971 until 2011. The results confirm the higher accuracy of the proposed method as compared with other artificial intelligence based models. Future estimations of electrical energy consumption in ASEAN-5 countries are projected up to 2030 according to rolling-based forecasting procedure.

A Small-Sample Adaptive Hybrid Model for Annual Electricity Consumption Forecasting

Annual electricity consumption forecasting is one of the important foundations of power system planning. Considering that the long-term electricity consumption curves of developing countries usually present approximately exponential growth trends and linear and accelerated growth rate trends may also appear in certain periods, this paper first proposes a small-sample adaptive hybrid model (AHM) to extrapolate the above curves. The iterative trend extrapolation equation of the proposed model can simulate the linear, exponential, and steep trends adaptively at the same time. To estimate the equation parameters using small samples, the partial least squares (PLS) and iteration starting point optimization algorithms are suggested. To evaluate forecasting performance, the artificial neural network (ANN), grey model (GM), and AHM are used to forecast electricity consumption in China from 1991 to 2014, and then the results of these models are compared. Analysis of the forecasting results shows that the AHM can overcome stochastic changes and respond quickly to changes in the main electricity consumption trend because of its specialized equation structure. Overall error analysis indicators also show that AHM often obtains more precise forecasting results than the other two models.

Forecasting long-term electricity demand for cooling of Singapore’s buildings incorporating an innovative air-conditioning technology

In an effort to accurately plan for investment on energy production and distribution, this paper proposes a long-term electricity consumption forecasting model for buildings’ cooling by employing a high energy conservative scenario. The key aspect of the high energy conservative scenario is to adopt an innovative adsorbent-based dehumidifier and an indirect evaporative cooling (AD-IEC) technology as opposed to conventional mechanical vapor compression system. Bottom-up equations were developed to identify the cooling load and electricity consumption of both residential and non-residential buildings for the period 2002–2013. Based on the time-series electricity consumption, a multiple linear regression model is developed to forecast electricity demand for the future period of 2014–2030. It is found that the electricity demands for cooling in the building sectors account for 31±2% of the total electricity consumption in Singapore, This study concluded that the high conservative scenario realizes the best potential of electricity saving of 21,096GWh until 2030. Using a CO2 emission factor of 4.49×10−4metrictons CO2/kWh, the total carbon footprint saving from all power plants is estimated to be 9491,264t of CO2. This work evolves a new forecasting methodology to predict buildings’ cooling energy consumption involving the use of novel cooling technologies.

Forecast of Long-term Electricity Consumption of the Industrial Sub-sector of Pulp and Paper in Brazil Using a Bottom-up Approach

Abstract The subsector of pulp and paper presented in 2013 an electric power consumption of 9.3% of electricity consumption in the industrial sector in Brazil. It experienced a significant growth of production in the last 20 years, exceeding 160%, and their consumption of electricity increased by 130%, reaching the level of 19,594GWh. Due to its characteristics, with a known structure and production processes, and its importance in the country's industrial sector, this subsector was selected for study in this paper, to evaluate the behavior of its long-term annual electricity demand, at the horizon of 20 years ahead, in energy efficiency scenarios through bottom-up approach. This approach is data intensive and requires data with good quality; it consists of a hierarchical structure that enables to check how specific actions can affect the results at each level and its effect on higher levels.

Improved Electric Power Demand Forecasting by adapting the Weighted Average to the MISMO Strategy

Forecasting the annual long-term consumption of electrical energy in a country has remained for the Electrical Engineer quite a difficult problem to solve. As an important planning tool, the forecast of electrical energy consumption has to be as precise as possible. The most commonly employed method is that of scenario building. With the scenario method, consumption forecasting is done through the simulation of a sequence of events. The generated data cannot therefore be as stochastic as it is in reality. With the advent of the computer age, numerous other statistical methods for consumption forecasting have been developed. Prominent among them is the forecasting by machine learning with multiple-Input multiple-Output local learning strategy. The objective here is to obtain a forecast which is as precise as possible, while conserving the stochastic nature between the historical and the forecasted data. This article first presents the different strategies for long-term power consumption forecasting using Multiple-Input Multiple-Output local learning strategies. It then proposes, based on the work of earlier researchers, an approach that uses the weighted averages to improve on the level of precision obtained. Furthermore, it applies this new improved calculation method to forecast the power consumption specifically in Cameroon for horizon 2035, when the country aspires to become an emerging economy. The last part of this article utilizes historical data on the electricity consumption of some countries from the World Bank dataset to do a comparative study between the here newly proposed method and that used previously. The results show that, the new method using MISMO plus weighted average delivers more exact results for long-term electrical power consumption forecasts.

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.

An Adaptive-Network-Based Fuzzy Inference System-Data Envelopment Analysis Algorithm for Optimization of Long-Term Electricity Consumption, Forecasting and Policy Analysis: The Case of Seven Industrialized Countries

This article presents an adaptive-network-based fuzzy inference system (ANFIS)-data envelopment analysis (DEA) algorithm for improvement of long-term electricity consumption forecasting and analysis. Six models are proposed to forecast annual electricity demand. Six different membership functions and several linguistic variables are considered in building ANFIS. The proposed models consist of two input variables, namely, gross domestic product and population. All trained ANFIS are then compared with respect to mean absolute percentage error. To meet the best performance of the intelligent-based approaches, data are pre-processed (scaled) and finally our outputs are post-processed (returned to its original scale). DEA is used to optimize the electricity consumption as well as examine the behavior of electricity consumption. To show the applicability and superiority of the ANFIS-DEA algorithm, actual electricity consumption in the USA, Canada, Germany, United Kingdom (UK), Japan, France and Italy from 1980–2007 is considered. Electricity consumption is then forecasted up to 2015. The unique features of the ANFIS-DEA algorithm are: behavioral analysis and optimization in complex, non-linear and uncertain environments.

A new adaptive fuzzy inference system for electricity consumption forecasting with hike in prices

Large increase or hike in energy prices has proven to impact electricity consumption in a way which cannot be drawn from historical data, especially when price elasticity of demand is not significant. This paper proposes an integrated adaptive fuzzy inference system (FIS) to estimate and forecast long-term electricity consumption when prices experience large increase. To this end, first a novel procedure for construction and adaptation of Takagi–Sugeno fuzzy inference system (TS-FIS) is suggested. Logarithmic linear regressions are estimated with historical data and used to construct an initial first-order TS-FIS. Then, in the adaptation phase, expert knowledge is used to define new fuzzy rules which form a new secondary FIS for electricity forecasting. To show the applicability and usefulness of the proposed model, it is applied for forecasting of annual electricity consumption in Iran where removing energy subsidies has resulted in a hike in electricity prices. Gross domestic product (GDP), population and electricity price are three inputs for the initial TS-FIS. A questionnaire survey was conducted to collect the expert estimation on possible change in electricity per capita, change in electricity intensity and the ratio of GDP elasticity to population elasticity when price hikes. Based on the information collected, a fuzzy rule base is formed and used to construct the secondary FIS which is used for electricity forecasting until 2016. Furthermore, the performance of the proposed model of this paper is compared with three other models namely ANFIS, ANN and one-stage regression in terms of their mean absolute percentage error. The comparison shows a superior performance for the proposed FIS model.