Energy Demand Forecasting Research Articles

Energy Planning in West Java using Software LEAP (Long-range Energy Alternatives Planning)

The increase in global energy consumption is increasing every year. The increasing need for energy also in fact collides with the human need to create a clean and pollution-free environment. This condition requires us to look for alternative and renewable energy sources to support our daily lives. With the preparation of the General Provincial Energy Plan (GPEP) in the future, it is expected to strengthen the competitiveness of the West Java economy, fulfill basic needs in the energy sector and strengthen the sustainability of West Java development based on the environment and local wisdom. Preparation of the West Java Regional Energy General Plan (REGP). Data processing to predict the level of planning for electrical energy needs in West Java is processed using LEAP (Long-range Energy Alternatives Planning System) software. Demand is calculated based on the amount of electrical energy consumption activity and the amount of electrical energy consumption per activity (energy intensity). The year 2019 is the base year for the calculation. The results obtained from the prediction of electrical energy demand in 2019-2030 show a positive trend, namely an increase from 111.7 million Giga Joules to 147.9 million Giga Joules. The energy transformation in the fuel output of each power plant in West Java for the baseline scenario, in 2030 will reach 95.3 million Giga Joules. Meanwhile, energy generation in the Demand Side Management (DSM) scenario in 2019 is 6 (Thousand MW), with a capacity of 3,600 MW. Then there is an increase every year, until in 2030 it will reach 26.6 (Thousand MW), with a capacity of 5,800 MW. From this paper, the procedures and data results from calculations using the LEAP tool are expected to provide an overview and input for readers in making policy in the energy sector in particular and regional development planning in general.

Benefit Modeling and Analysis of Wind Power Generation under Social Energy Economy and Public Health.

Wind is a renewable energy source. Overall, using wind to produce energy has fewer effects on the environment than many other energy sources. Wind and solar energy provide public health and environmental benefits to the social. Wind turbines may also reduce the amount of electricity generation from fossil fuels, which results in lower total air pollution and carbon dioxide emissions. In order to better optimize the effect of social energy economic management and facilitate the multiobjective decision making of coordinated development of energy and socioeconomic environment, a modeling and analysis method of economic benefits of wind power generation based on deep learning is proposed. In this paper, based on the principle of deep learning, the evaluation indicators of wind power economic benefits are excavated, a scientific and reasonable economic benefit evaluation system is constructed, a wind power economic benefit analysis model supported by public management is constructed, and the steps of wind power economic benefit analysis are simplified. It is concluded that the modeling and analysis method of wind power economic benefits based on deep learning has high practicability in the actual application process, which is convenient for the prediction and analysis of energy demand for social and economic development.

Integration of solar energy supply on a smart distribution board based on IoT system

Modern electricity management of solar power supply at the smart distribution board is heavily dependent on the Internet of Things (IoT). It is still difficult to create practical, affordable intelligent systems for condition monitoring, protection and control at the smart distribution board of distribution networks. Smart grids use the capabilities of information and communication technologies, especially the Internet of Things, to improve the sustainability, quality and productivity of energy production and demand forecasting, as well as to reduce resource consumption and increase the integration of renewable energy sources. The performance of the developed system is evaluated for a variety of residential electrical loads with different energy consumption patterns on a smart distribution board. In this project, the photovoltaic tracking system as well as the cleaning system will be connected to the smart board. All the input data will be collected and sent via Bluetooth/WI-FI by a working prototype built on an Arduino board, which can independently collect and monitor the real-time energy and environmental conditions of solar plants. The mobile application is also used to allow individual users to monitor the system in real time, notify users of electrical system problems, display real-time energy consumption, and provide other important notifications and alerts to the remote user when there is a deviation of solar power production quality parameters from a predefined set of standard values. A number of improvements are also proposed for further integration with Internet of Things (IoT) and smart home technologies.

MILP Optimized Management of Domestic PV-Battery Using Two Days-Ahead Forecasts

This paper proposes an Energy Management System (EMS) for domestic PV-battery applications with the aim of reducing the absolute net energy exchange with the utility grid by utilizing the two days-ahead energy forecasts in the optimization process. A Mixed-Integer Linear Programming (MILP) exploits two days-ahead energy demand and PV generation forecasts to schedule the day-ahead battery energy exchange with both the utility grid and the PV generator. The proposed scheme is tested using the real data of the Active Office Building (AOB) located in Swansea University, UK. Performance comparisons with state-of-the-art and the commercial EMS currently running at the AOB reveal that the proposed EMS increases the self-consumption of PV energy and at the same time reduces the total energy cost. The absolute net energy exchange with the grid and the total operating costs are reduced by 121% and 54% compared to the state-of-the-art and 194% and 8% when compared to the commercial EMS over a six-month period. Furthermore, the results show that the proposed method can reduce the energy bill by up to 46% for the same period compared to the state-of-the-art. The paper also investigates the effect of using different objective functions on the performance of the EMS and shows that the proposed EMS operate more efficiently when it is compared with another cost function that directly promotes reducing the absolute net energy exchange.

Evolutionary Machine Learning-Based Energy Consumption Prediction for the industry

In the digitalization of industry and the industry 4.0 environment, it is important to master the accurate forecasting of energy demand in order to guarantee the continuity of production service as well as to improve the reliability of the electrical system while promoting energy efficiency strategies in the industrial sector. This paper proposes machine learning models to predict the energy consumption demand in an industrial plant, which takes into account the at-tributes that directly the consumption. The proposed models in this work include Multiple Linear Regression (MLR), Decision Tree (DT), Recurrent Neural Networks (RNN) and Gated Recurrent United (GRU), which are compared according to their performances criteria which help to find the best forecasting models. Basing on simulation results, it is proven that the MLR approach is the best forecasting method.

Realizacja regionalnej polityki energetycznej na podstawie planowania

The document presents general directions of the energy policy of Poland and the EU, principles of the regional energy policy based on the energy law and energy planning in the municipality in the light of the objectives of the European Parliament. Energy planning offers the possibility to forecast energy demand and coordinate the implementation of the EP at different levels. At the municipal level, there are great opportunities to increase energy efficiency, while reducing gas emissions and using renewable energy sources.

Forecasting Electricity Consumption and Production in Smart Homes through Statistical Methods

• Methods for forecasting energy demand and production are proposed. • Predictions contribute to balance and smoothen the electricity intake from the power grid. • Experimental evaluation is performed on data recorded on a real energy-management system. • The ARIMA and TBATS models are configured and then compared with other forecasting methods. Over the last years, a steady increase in both domestic electricity consumption and in the adoption of personal clean energy production systems has been observed worldwide. By analyzing energy consumption and production on photovoltaic panels mounted in a house, this work focuses on finding patterns in electrical energy consumption and devising a predictive model. Our goal is to find an accurate method to predict electrical energy consumption and production. Being able to anticipate how consumers will use energy in the near future, homeowners, companies and governments may optimize their behavior and the import and export of electricity. We evaluated the ARIMA and TBATS statistical prediction methods and compared them with other models on datasets from a household equipped with photovoltaics and an energy management system. The evaluation results have shown a mean absolute error of 73.62 Watts for the TBATS model, which is far better than the one obtained with neural forecasting methods.

Attention Mechanism-Based Transfer Learning Model for Day-Ahead Energy Demand Forecasting of Shopping Mall Buildings

Attention Mechanism-Based Transfer Learning Model for Day-Ahead Energy Demand Forecasting of Shopping Mall Buildings

Combined multi-objective optimization and agent-based modeling for a 100% renewable island energy system considering power-to-gas technology and extreme weather conditions

Islands are constrained by geographical conditions in terms of energy delivery. Due to weak connections with the mainland and the power grid, the diversity of island energy demand leads to high economic costs and environmental pollution issues. This study proposes a 100% renewable island energy system, which integrates with power-to-gas, combined cooling, heating and power, and desalination technologies to supply electricity, heating, cooling, gas and fresh water to the local residents. A comprehensive approach for energy demand prediction, system design and dispatch optimization, as well as system evaluation is proposed. For energy demand prediction, agent-based modeling is used to simulate the demand of electricity, heating, cooling, gas and fresh water for the case study community on the island. The k-means clustering and scenario tree are further adopted to generate representative stochastic scenarios, which are applied to capture the uncertainty of energy demand. A multi-objective optimization model is developed to optimize the system design and scheduling strategy simultaneously. In order to demonstrate the effectiveness of the proposed approach and to evaluate the obtained optimal solutions for the case study, different objectives and extreme weather conditions are specifically considered. The optimal solution obtained shows that compared to battery storage, a 2.5% annual cost reduction can be achieved by using power-to-gas technology for energy storage. The findings also suggest that extreme weather conditions can be coped with by increasing the capacity of biogas generation, desalination, and energy storage equipment, thereby improving the resilience of the island energy system.

A dynamic interactive optimization model of CCHP system involving demand-side and supply-side impacts of climate change. Part I: Methodology development

Combined cooling, heating and power (CCHP) system, as a superior energy-provision form of public building, is capable of achieving flexible and stable energy provision with high energy-utilization efficiency and low pollutant emission. However, some difficulties exist in operating such a system, due to its intrinsic multi-period, multi-factor and multi-layer features. In addition, the fluctuation in weather elements under climate change exacerbates the inaccuracy of energy demand prediction and facilities’ power output calculation, leading to imbalanced energy supply and demand. To tackle this issue, a dynamic interactive model combining user-demand prediction, energy-provision calculation and operational collaborative optimization was developed. It attempts to combine the regional climate simulation (PRECIS), demand prediction (TRNSYS), equipment output calculation (mechanism modeling) and collaborative optimization (LINGO) into a general framework. The specific operation processes include: (i) utilize PRECIS model to identify the variations in temperature and radiation under climate change; (ii) exploit TRNSYS software to predict the users’ demand of targeted hospital in the future; (iii) establish a mechanism simulation model for gas turbine and estimate power output under extreme meteorological condition; (iv) incorporate the results generated by processes (ii) and (iii) into formulated operation optimization model of CCHP system; (v) generate optimal energy provision scheme adapted to climate change. This dynamic interactive model comprehensively considers the interactions at all aspects involved into operation management of CCHP system and improves the economy and adaptability of operation pattern. This study mainly addressed the problem background and model formulation. A detailed case study will be discussed in another follow-up work.

Forecasting energy demand in isolated rural communities: A comparison between deterministic and stochastic approaches

Off-grid renewable energy grids will contribute to the achievement of SDG 7 on universal energy access, especially in off-grid communities. But the scarcity of resources for development aid requires very tight designs that minimize the cost of investment and operation of the mini-grid. To do this, the future energy demand of the community must be analysed very well, which is very difficult due to the lack of previous data. In the literature, two main approaches to analyse future energy demand, deterministic and stochastic methodologies. In this article, we compare both methodologies for a real case study in Honduras and discuss their advantages and disadvantages. Although the deterministic approach requires less information and less mathematical processing, it generates less accurate results. In contrast, the stochastic approach consumes more resources but gives more realistic results for the correct design of the mini-grid. In conclusion, the deterministic approach methods can be useful for the early stages of the project, when the investment is being sized. But for the advanced phases of the project, when the installation is being designed, the stochastic approach is recommended.

Energy demand and production forecasting in Pakistan

Pakistan has been in severe energy crises since the year 2004. Major reasons behind energy crises are lack of use of modeling tools in power planning and policy development, dependence on imported energy sources and poor governance. In this paper, electricity demand forecasting for Pakistan up to the year 2030 and a proposal for utilizing domestic energy resources, such as, coal, natural gas, and solar resources available in Baluchistan province of Pakistan for the electricity needs of the country are presented. Long-range Energy Alternative Planning (LEAP) software is used to develop energy demand model for Pakistan that forecasts energy demand under two scenarios, i.e., baseline scenario and energy conservation scenario for the period from 2018 to 2030. Electrical power generation capacity of coal, natural gas and solar resources is also calculated considering the capacity factor of each type of energy asset using MATLAB. This work suggests that Pakistan's energy demand forecast is 399 TWh under the baseline scenario and 312 TWh under the energy conservation scenario. The power generation potential of domestic energy assets available in Baluchistan is 500.041 TWh in total, which includes Natural Gas 392.768 TWh, Solar 84.935 TWh and Coal 22.338 TWh. This research would help relevant government departments in Pakistan for power capacity development as per the required energy demand and alleviate energy crises in Pakistan.

Long-Term Energy Demand Forecasting Based on a Systems Analysis

—Energy demand forecasting plays a key role in solving the majority of problems connected with determining the economic and energy development prospects. In view of high inertia and capital intensity of energy generation facilities, the changes in the energy consumption structure and rates should be considered for a sufficiently long-term future of no less than 15 years. This adds much difficulty to the development of such forecasts, because it is necessary to take into account the possible results of technological progress in different sectors of the economy and changes in its structure. The suggested approach to energy demand forecasting is based on the systems analysis methods. Energy consumers are disaggregated by economic sectors, country regions, energy carriers, and their utilization areas. As a result, it becomes possible to take into account the possible future technological and structural changes in the economy, its differences in different regions, mutual replacements of energy carriers, and energy saving. The most important feature of the approach is that it involves the separation of economic and energy variables. Economic variables determine the development scales of economic sectors, and energy variables determine the energy consumption intensities in them. The separation of variables helps obtain essentially more accurate forecasts owing to the possibility of taking into account the differences in the variation trends of these variables in the forecast period. For forecasting the economic variables, a nonlinear conditionally dynamic model of interrelations between the economy and energy is used. The energy variables are forecasted on the basis of their links with economic factors, e.g., with cumulative investments in the fixed capital of the considered economic sectors. The discussed approach was implemented as the totality of adaptive simulation models united into the EDFS computing system. The approach was repeatedly and quite successfully used to forecast the energy demand for an up to 10–20-year horizon in solving various problems. The gained experience has demonstrated the correctness of applying the approach for solving long-horizon forecast problems. To illustrate the capacities of the approach, the electric energy demand forecast for the period of up to 2040 for the basic version of the social and economic development of Russia’s economy prepared on the basis of this approach is presented.

Predicting energy demand of residential buildings: A linear regression-based approach for a small sample size

Abstract The key design strategies that reduce the energy demand of buildings are not present in most thermal codes in many countries. Therefore, modeling techniques offer an alternative to combine the architects' modus operandi with the energy efficiency in the early stages of architectural design and with higher speed and precision. However, a review of the scientific literature using modeling techniques shows that most researchers use a relatively large sample of thermal simulations. This paper proposes a simplified method based on the linear regression modeling technique and considers a relatively smaller sample of thermal simulations. A total of 6 key building design strategies were identified, related to the urban context, building envelope, and shape factor. A simulation protocol containing 60 possible combinations was designed by random selection. In the present study, the Pleiades software was used to estimate the annual energy demand for heating and cooling for a typical dwelling in a humid climate zone. A parametric study and sensitivity analysis to identify the most efficient parameters was performed in SPSS 21. The resulting model predicts the annual energy demand with an accuracy of 93.7%, a root mean square error (RMSE) of 5.88, and a scatter index (SI) of 8.59%. The models performed could efficiently and quickly assist architects while designing the buildings in the architectural practice.

Tunicate swarm algorithm-trained multi-layered perceptron for data centre energy demand forecasting and relative percentage contribution analysis of input parameters

PurposeEnergy management is critical to data centres (DCs) majorly because they are high energy-consuming facilities and demand for their services continue to rise due to rapidly increasing global demand for cloud services and other technological services. This projected sectoral growth is expected to translate into increased energy demand from the sector, which is already considered a major energy consumer unless innovative steps are used to drive effective energy management systems. The purpose of this study is to provide insights into the expected energy demand of the DC and the impact each measured parameter has on the building's energy demand profile. This serves as a basis for the design of an effective energy management system.Design/methodology/approachThis study proposes novel tunicate swarm algorithm (TSA) for training an artificial neural network model used for predicting the energy demand of a DC. The objective is to find the optimal weights and biases of the model while avoiding commonly faced challenges when using the backpropagation algorithm. The model implementation is based on historical energy consumption data of an anonymous DC operator in Cape Town, South Africa. The data set provided consists of variables such as ambient temperature, ambient relative humidity, chiller output temperature and computer room air conditioning air supply temperature, which serve as inputs to the neural network that is designed to predict the DC’s hourly energy consumption for July 2020. Upon preprocessing of the data set, total sample number for each represented variable was 464. The 80:20 splitting ratio was used to divide the data set into training and testing set respectively, making 452 samples for the training set and 112 samples for the testing set. A weights-based approach has also been used to analyze the relative impact of the model’s input parameters on the DC’s energy demand pattern.FindingsThe performance of the proposed model has been compared with those of neural network models trained using state of the art algorithms such as moth flame optimization, whale optimization algorithm and ant lion optimizer. From analysis, it was found that the proposed TSA outperformed the other methods in training the model based on their mean squared error, root mean squared error, mean absolute error, mean absolute percentage error and prediction accuracy. Analyzing the relative percentage contribution of the model's input parameters based on the weights of the neural network also shows that the ambient temperature of the DC has the highest impact on the building’s energy demand pattern.Research limitations/implicationsThe proposed novel model can be applied to solving other complex engineering problems such as regression and classification. The methodology for optimizing the multi-layered perceptron neural network can also be further applied to other forms of neural networks for improved performance.Practical implicationsBased on the forecasted energy demand of the DC and an understanding of how the input parameters impact the building's energy demand pattern, neural networks can be deployed to optimize the cooling systems of the DC for reduced energy cost.Originality/valueThe use of TSA for optimizing the weights and biases of a neural network is a novel study. The application context of this study which is DCs is quite untapped in the literature, leaving many gaps for further research. The proposed prediction model can be further applied to other regression tasks and classification tasks. Another contribution of this study is the analysis of the neural network's input parameters, which provides insight into the level to which each parameter influences the DC’s energy demand profile.

A novel Invasive Weed Optimization with levy flight for optimization problems: The case of forecasting energy demand

Energy is very important nowadays and it has become essential for human life. More amount of energy is necessary for a community with increased living standards and population. This causes an increase in energy consumption. Energy has become one of the most significant problems across the world today; therefore, it should be generated at the best level. Excessive energy generation makes countries lose money while less amount of energy generation causes crises. Therefore, countries need to adjust their energy demand optimally. It is possible to estimate energy demands of countries by using various applications. This study proposes a new improved algorithm for linear regression models to forecast the energy demand of Turkey. The selected algorithm is Invasive Weed Optimization (IWO) algorithm which has been developed with levy flight called LF-IWO. In the linear regression model, input parameters included data regarding Turkey’s gross domestic product (GDP), population, import and export. Turkey’s energy demand was estimated for these parameters by using the data between 1979 and 2011. The estimation results obtained from the model were compared with those of similar studies in the literature to measure the performance success of the developed algorithm.

Energy demand prediction with machine learning supported by auto-tuning: a case study

The operational energy of buildings is making up one of the highest proportions of life-cycle carbon emissions. A more efficient operation of facilities would result in significant energy savings but necessitates computational models to predict a building’s future energy demands with high precision. To this end, various machine learning models have been proposed in recent years. These models’ prediction accuracies, however, strongly depend on their internal structure and hyperparameters. The time demand and expertise required for their finetuning call for a more efficient solution. In the context of a case study, this paper describes the relationship between a machine learning model’s prediction accuracy and its hyperparameters. Based on time-stamped recordings of outdoor temperatures and electricity demands of a hospital in Japan, recorded every 30 minutes for more than four years, using a deep neural network (DNN) ensemble model, electricity demands were predicted for sixty time steps to follow. Specifically, we used automatic hyperparameter tuning methods, such as grid search, random search, and Bayesian optimization. A single time step ahead, all tuning methods reduced the RSME to less than 50%, compared to non-optimized tuning. The results attest to machine learning models’ reliance on hyperparameters and the effectiveness of their automatic tuning.

Energy prediction for CNC machining with machine learning

Nowadays, the reduction of CO2 emissions by moving from fossil to renewable energy sources is on the policy of many governments. At the same time, these governments are forcing the reduction of energy consumption. Since large industries have been in the focus for the last decade, today also small and medium enterprises with production lot size one are increasingly being obliged to reduce their energy requirements in production. Energy-efficient CNC machine tools contribute to this goal. In machining processes, the machining strategy also has a significant influence on energy demand. For manufacturing of lot size one, the prediction of the energy demand of a machining strategy, before a part is manufactured plays a decisive role. In numerous previous studies, analytical models between the energy demand and the machining strategy have been developed. However, their accuracy depends largely on the parameterization of these models by dedicated experiments. In this paper, different machine learning algorithms, especially variations of the decision tree (’DecisionTree’, ’RandomForest’, boosted ’RandomForest’) are investigated for their ability to predict the energy demand of CNC machining operations based on real production data, without the need for dedicated experiments. As shown in this paper, the most accurate energy demand predictions can be achieved with the ’RandomForest’ algorithm.

SynergyGrids: blockchain-supported distributed microgrid energy trading

Growing intelligent cities is witnessing an increasing amount of local energy generation through renewable energy resources. Energy trade among the local energy generators (aka prosumers) and consumers can reduce the energy consumption cost and also reduce the dependency on conventional energy resources, not to mention the environmental, economic, and societal benefits. However, these local energy sources might not be enough to fulfill energy consumption demands. A hybrid approach, where consumers can buy energy from both prosumers (that generate energy) and also from prosumer of other locations, is essential. A centralized system can be used to manage this energy trading that faces several security issues and increase centralized development cost. In this paper, a hybrid energy trading system coupled with a smart contract named SynergyGrids has been proposed as a solution, that reduces the average cost of energy and load over the utility grids. To the best of our knowledge, this work is the first attempt to create a hybrid energy trading platform over the smart contract for energy demand prediction. An hourly energy data set has been utilized for testing and validation purposes. The trading system shows 17.8% decrease in energy cost for consumers and 76.4% decrease in load over utility grids when compared with its counterparts.

Interpretable Forecasting of Energy Demand in the Residential Sector

Energy demand forecasting is practiced in several time frames; different explanatory variables are used in each case to serve different decision support mandates. For example, in the short, daily, term building level, forecasting may serve as a performance baseline. On the other end, we have long-term, policy-oriented forecasting exercises. TIMES (an acronym for The Integrated Markal Efom System) allows us to model supply and anticipated technology shifts over a long-term horizon, often extending as far away in time as 2100. Between these two time frames, we also have a mid-term forecasting time frame, that of a few years ahead. Investigations here are aimed at policy support, although in a more mid-term horizon, we address issues such as investment planning and pricing. In this paper, we develop and evaluate statistical and neural network approaches for this mid-term forecasting of final energy and electricity for the residential sector in six EU countries (Germany, the Netherlands, Sweden, Spain, Portugal and Greece). Various possible approaches to model the explanatory variables used are presented, discussed, and assessed as to their suitability. Our end goal extends beyond model accuracy; we also include interpretability and counterfactual concepts and analysis, aiming at the development of a modelling approach that can provide decision support for strategies aimed at influencing energy demand.