Estimation Of Electricity Demand Research Articles

A Neural Network Forecasting Approach for the Smart Grid Demand Response Management Problem

Demand response management (DRM) plays a crucial role in the prospective development of smart grids. The precise estimation of electricity demand for individual houses is vital for optimizing the operation and planning of the power system. Accurate forecasting of the required components holds significance as it can substantially impact the final cost, mitigate risks, and support informed decision-making. In this paper, a forecasting approach employing neural networks for smart grid demand-side management is proposed. The study explores various enhanced artificial neural network (ANN) architectures for forecasting smart grid consumption. The performance of the ANN approach in predicting energy demands is evaluated through a comparison with three statistical models: a time series model, an auto-regressive model, and a hybrid model. Experimental results demonstrate the ability of the proposed neural network framework to deliver accurate and reliable energy demand forecasts.

Review on Economic Analysis of Grid-tied Battery Storage Systems and Optimization Strategies

This comprehensive review paper provides an in-depth exploration of the landscape of energy management strategies within the context of emerging energy markets. It underscores the significance of effective energy management and optimization, particularly in light of the surging integration of renewable energy sources, growing sustainability concerns, and the dynamic shifts within the energy sector. This review extensively analyses three fundamental domains: electricity demand estimation, battery storage systems, and energy arbitrage methodologies placing a significant emphasis on the body of the research related to electricity storage optimization.Furthermore, it delves into the widespread strategies that are adopted in battery storage systems, which have assumed a pivotal role in energy storage and the equilibrium of supply and demand in the modern energy landscape. Nevertheless, amidst this extensive examination, a considerable research gap arises, specifically highlighting the lack of dedicated studies addressing energy arbitrage methods tailored to the unique attributes of emerging energy markets. This gap is particularly prominent in the context of Sri Lanka's growing energy market, where the prominence of renewable energy sources is on the rise.The review underscores that in order to fully harness the vast potential of renewable resources and optimize energy supply within this distinctive setting, it is crucial to develop and implement customized energy arbitrage strategies. These strategies should be designed to accommodate the evolving energy dynamics and the integration of renewable energy, that would thus enable energy markets to function effectively while maximizing economic benefits. Therefore, as the global energy landscape continues to transform, with a growing focus on sustainability and renewable energy, this paper sheds light on the need for innovative and tailored energy solutions in emerging markets.

Machine learning application for estimating electricity demand by municipality

In order to achieve efficient and low carbon energy systems, attention has been focused on locally distributed energy systems, but a detailed and accurate supply–demand database is essential for simulating and optimizing such local systems. Demand data are generally not maintained at small scales, and it is common to estimate electricity demand by prorating global data from large energy carriers. However, the accuracy of the proration method is questionable. The present study explores the idea of using machine learning as an alternative to the proration method for downscaling electricity demand data based on a neural network model. The input data consist of four categories: weekday/holiday determination data, temperature data, periodic signal data, and demand sector configuration data. The model shows a satisfying generalization capacity for downscaling data from regional to prefectural and municipal scales while capturing local characteristics that are not accounted with the proration method.

Emission reduction pathways for satisfying Türkiye's future electricity demand: A cross-country dynamic panel analysis with linear programming scenarios

This paper studies whether feasible and plausible pathways exist for Türkiye to meet its growing electricity demand while reducing its emissions, by relying more on renewables, instead of increasing the use of its local coal resources. Our quantitative analysis proceeds in two stages. In the first stage, we determine the future electricity demand of Türkiye from 2020 to 2040 with the use of a dynamic panel data model. A 41-country balanced panel data set that comprises of five-year interval data, between 1990 and 2015. In the second stage, we develop linear programming models to generate realistic and reasonable scenarios representing three probable future pathways to meet the econometrically estimated electricity demand. The scenarios we have designed are business-as-usual (BAU), which includes the nuclear power plant, minimize GHGs (minGHG), and maximize local resources (MaxLocal). The latter scenarios omit the possible completion of the nuclear power plant in the next ten to twelve years. The generated scenarios are compared in terms of investment requirements and CO2e emissions. The model results of minGHG and MaxLocal both show that the share of renewable generation should reach around 65% to satisfy the projected demand by 2040. However, the difference in CO2e emissions (Mton/TWh) between the two cases is enormous: 0.408 for MaxLocal vs. 0.180 for minGHG! Moreover, required annual investment under minGHG is USD 0.19 billion cheaper on average (per year), which corresponds to a 21-year cumulative difference of USD 3.99 billion in current dollars. Therefore, a secure low-carbon pathway with a lower investment requirement is possible for Türkiye without nuclear power or new coal plants, while also phasing-out the existing coal plants under a moderate transition plan with a minimum amount of stranded assets.

Forecasting Electricity Demand in Turkey Using Optimization and Machine Learning Algorithms

Medium Neural Networks (MNN), Whale Optimization Algorithm (WAO), and Support Vector Machine (SVM) methods are frequently used in the literature for estimating electricity demand. The objective of this study was to make an estimation of the electricity demand for Turkey’s mainland with the use of mixed methods of MNN, WAO, and SVM. Imports, exports, gross domestic product (GDP), and population data are used based on input data from 1980 to 2019 for mainland Turkey, and the electricity demands up to 2040 are forecasted as an output value. The performance of methods was analyzed using statistical error metrics Root Mean Square Error (RMSE), Mean Absolute Error (MAE), R-squared, and Mean Square Error (MSE). The correlation matrix was utilized to demonstrate the relationship between the actual data and calculated values and the relationship between dependent and independent variables. The p-value and confidence interval analysis of statistical methods was performed to determine which method was more effective. It was observed that the minimum RMSE, MSE, and MAE statistical errors are 5.325 × 10−14, 28.35 × 10−28, and 2.5 × 10−14, respectively. The MNN methods showed the strongest correlation between electricity demand forecasting and real data among all the applications tested.

Analyzing at-scale distribution grid response to extreme temperatures

Threats against power grids continue to increase, as extreme weather conditions and natural disasters (extreme events) become more frequent. Hence, there is a need for the simulation and modeling of power grids to reflect realistic conditions during extreme events conditions, especially distribution systems. This paper presents a modeling and simulation platform for electric distribution grids which can estimate overall power demand during extreme weather conditions. The presented platform’s efficacy is shown by demonstrating estimation of electrical demand for (1) Electricity Reliability Council of Texas (ERCOT) during winter storm Uri in 2021, and (2) alternative hypothetical scenarios of integrating Distributed Energy Resources (DERs), weatherization, and load electrification. In comparing to the actual demand served by ERCOT during the winter storm Uri of 2021, the proposed platform estimates approximately 34 GW of peak capacity deficit.11These numbers are consistent with state-of-the-art prediction results published in the literature (Gruber et al., 2022). For the case of the future electrification of heating loads, peak capacity of 78 GW (124% increase) is estimated, which would be reduced to 47 GW (38% increase) with the adoption of efficient heating appliances and improved thermal insulation. Integrating distributed solar PV and storage into the grid causes improvement in the local energy utilization and hence reduces the potential unmet energy by 31% and 40%, respectively.

A deep reinforced learning spatiotemporal energy demand estimation system using deep learning and electricity demand monitoring data

Tracking decarbonization effects requires a model for the identification of spatial energy demands on city facilities. However, most developing countries lack detailed discrete time and device-specific energy demand data. In this study, we installed multiple energy demand monitoring systems that could observe electricity demands at the device level in some residences in Bogor, Indonesia. The study aimed to estimate the time-series and equipment ratio of electricity consumption by households in the entire city based on the monitoring data. However, the number of households monitored was small, and therefore unlikely to be regarded as having a representative system. Therefore, we used questionnaire data to create monitored mimic data and increased the number of samples to estimate the energy demand characteristics of the entire city via spatial interpolation. In addition, we developed a reinforcement learning system for discrete time–electricity demand estimation systems for unmonitored households using a 5-step procedure; 1) Analyzing energy demand and its patterns from monitoring data, 2) Questionnaire-based surveying of households, 3) Estimation of energy demand and its patterns based on questionnaire responses in monitored households, 4) Development of a deep learning model that extends the results from (3) to unmonitored households using data fusion, and 5) Spatial interpolation of energy demand characteristics for all households in Bogor using a spatial statistics method. The spatial electricity demand of households was interpolated from GIS and high-resolution satellite data matching procedures. Based on this analysis, we developed an hourly energy demand prediction system that could be automatically improved by adding new data from the reinforced learning framework.

Uni-Variate and Multi-Variate Short-Term Household Electricity Consumption Prediction Using Machine Learning Technique

Background: Electricity consumption prediction plays an important role in conservation, development, and future planning. Accurate prediction model has various field applications in real-life scenarios, future electricity demand estimation, performance evaluation of current time, fault detection, efficient energy production, resource-saving, and many more. In this paper, a CNN based short term building electricity consumption prediction model is developed and tested for two different types of datasets that can perform weekly prediction. Two different datasets are used to check how the algorithm behaves on different datasets i.e., what are the impacts dataset has on prediction accuracy. Errors were calculated using MAE and RMSE. Objective: The objective of the study is to develop an electricity consumption prediction (ECP) model for a univariate and multivariate dataset using CNN and LSTM network and to find that how the correlation and independency of features affect the electricity prediction task. Methods: The proposed electricity consumption model is built using the deep CNN andLSTM network and is trained and tested using the univariate and multivariate time series dataset thus the two experiments have been performed and are named as U-ECPCL (Univariate- Electricity Consumption Prediction using CNN and LSTM) and M-ECPCL (Multivariate- Electricity Consumption Prediction using CNN and LSTM) respectively. Results: The model predicts accurately with few errors with MAE of 0.251 and RMSE of 0.66 for univariate dataset and MAE of 4.36 and RMSE of 11.53 for a multivariate dataset. Conclusion: The model predicts accurately with few errors and if the prediction error of univariate and multivariate are compared then it is concluded that the univariate model outperforms the multivariate model.

Value at risk approach to producer's best response in an electricity market with uncertain demand

We deal with several sources of uncertainty in electricity markets. The independent system operator (ISO) maximizes the social welfare using chance constraints to hedge against discrepancies between the estimated and real electricity demand. We find an explicit solution to the ISO problem and use it to tackle the problem of a producer. In our model, production, as well as the income of a producer, are determined based on the estimated electricity demand predicted by the ISO, which is unknown to producers. Thus, each producer is hedging against the uncertainty of the prediction of the demand using the value-at-risk approach. To illustrate our results, a numerical study of a producer's best response given a historical distribution of both estimated and real electricity demand is provided.

Ex-post analysis of energy subsidy removal through integrated energy systems modelling

Energy subsidies can incentivise the overconsumption of energy resources and contribute to other economic or social distortions. In this paper, an ex-post analysis is presented that explores the extent to which electricity subsidy reform could have reduced Iran's energy demand during the period 1984–2017. It also quantifies the techno-economic and environmental benefits that could have been achieved through such reforms. A time-varying econometric model is linked to an energy systems optimisation model. The former estimates electricity demand under different subsidy removal scenarios, and the latter identifies the cost-optimal generation mix to meet the demand. The results of cost-optimal transition pathways under subsidy removal scenarios are compared with the real-world energy system development during the study horizon. The comparison reveals that the subsidy reform could have reduced the total cumulative electricity consumption by 22%. Renewable share in power generation could have increased from 5% to 15%. Moreover, the reform combined with a cost-optimal generation pathway would have saved $69 billion and avoided 944 million tons of CO2 emissions. The analysis also shows that every five-year delay in subsidy removal causes about 100 million tons of additional CO2 emissions. Finally, the paper presents lessons learnt for future energy modelling.

Household electricity demand estimation: a regression discontinuity design approach

Household electricity demand estimation: a regression discontinuity design approach

Analysis of Malaysia electricity demand and generation by 2040

Malaysia as an emerging country, increasing population, gross domestic product (GDP) growth and enhanced access to electricity lead to an expanding of demand. The crucial parameters to determine future energy demand and generation projections are GDP, population growth rates and weather implications due to climate change. The study aims to forecast the future trends based on the historical values and also to project the future electricity demand and generation. The electricity demand and generation growth evaluated based on 2 main elements which are population growth and weather parameters (maximum temperature and rainfall). The future trends are forecasted based on the historical values of population and weather parameters. There is 152.9% of population growth in 32 years. The population will keep on developing yet with the lower rate. The GDP trend and the population growth mirrors the pattern of emissions. The findings from Statistical Downscaling Model (SDSM) analysis shows that the rainfall distribution will diminish while the temperature will expand that depict the climate change impact as time passes by. In 2020, the most extreme temperature recorded is 31.7 °C while in 2040, the estimated greatest temperature is 32.3 °C. There will be a 0.6 °C increase in temperature in 20 years. The demand in 2040 will be expanded 50.3% more than demand in 2020. The estimated electricity demand per capita will continue expanding because of the augmentation of the populace and the significance of electricity in daily activities. The pattern shows that electricity demand and generation in Malaysia will be expanding massively year by year.

New definition of levelized cost of energy storage and its application to reversible solid oxide fuel-cell

Renewable energy installation capacity has rapidly increased in recent years. Subsequently, developing and commercializing large energy storage systems (ESSs) has become an important research objective. To evaluate development and compare between different ESSs, levelized cost of energy storage (LCOES) has been used. However, current LCOES often includes cost of electricity production in total investment, and neglects off-design characteristic of ESSs. Therefore, we proposed a new definition for the LCOES (LCOES2) to resolve these problems. An ESS using RSOFC coupled with waste steam was investigated for ESS efficiency characterization. The proposed ESSs and LCOES calculations were applied to South Korea case. The hourly average solar power data for each month was used. Hydrogen production and consumption were matched for sizing ESS and estimating electricity demand profile. The results show that, conventional LCOES, which considers charging electricity cost, overestimated the cost by 7.7% and 14.8% compared to the LCOES2, at RSOFC stack cost of $700/kW and $225/kW, respectively. The constant part-load efficiency model resulted in lower LCOES values, by up to 6.3%, due to higher overall round-trip efficiency. The new LCOES definition, considering variable part-load efficiency, was proved to be an efficiency-sensitive and reliable cost indicator.

Sectoral Electricity Demand and Direct Rebound Effects inNew Zealand

This paper is one of the limited studies to investigate rebound effects in sectoral electricity consumption and the specific case of New Zealand. New Zealand, like other OECD economies, has aimed for energy efficiency improvements and reduced electricity consumption from 9.2 MWh per capita in 2010 to 8.6 MWh per capita in 2015. However, following a significant decline since 2010, electricity consumption in the main New Zealand sectors is increasing. Energy conservation could play an important role in meeting the growing demand for electricity but rebound effects can affect the effectiveness of conservation policies. We decompose the sectoral electricity prices to capture the asymmetric demand response to electricity price changes and estimate electricity demand elasticity during 1980 and 2015 to estimate the sectoral rebound effects. We find partial rebound effects of 54% and 23% in the industrial and commercial sectors respectively while we find no rebound effect at the aggregate level. The rebound effect is insignificant in the residential sector. These findings lead to policy recommendations for sector specific energy conservation measures and policies.

The M-LED platform: advancing electricity demand assessment for communities living in energy poverty

Globally about 800 million people live without electricity at home, over two thirds of which are in sub-Saharan Africa. Planning electricity access infrastructure and allocating resources efficiently requires a careful assessment of the diverse energy needs across space, time, and sectors. Because of data scarcity, most country or regional-scale electrification planning studies have however assumed a spatio-temporally homogeneous (top-down) potential electricity demand. Poorly representing the heterogeneity in the potential electricity demand across space, time, and energy sectors can lead to inappropriate energy planning, inaccurate energy system sizing, and misleading cost assessments. Here we introduce M-LED, a Multi-sectoral Latent Electricity Demand geospatial data processing platform to estimate electricity demand in communities that live in energy poverty. The platform shows how big data and bottom-up energy modelling can be leveraged together to represent the potential electricity demand with high spatio-temporal and sectoral granularity. We apply the methodology to Kenya as a country-study and devote specific attention to the implications for water-energy-agriculture-development interlinkages. A more detailed representation of the demand-side in large-scale electrification planning tools bears a potential for improving energy planning and policy.

Comparison of Three Different Learning Methods of Multilayer Perceptron Neural Network for Wind Speed Forecasting

In the world, electric power is the highest need for high prosperity and comfortable living standards. The security of energy supply is an essential concept in national energy management. Therefore, ensuring the security of electricity supply requires accurate estimates of electricity demand. The share of electricity generation from renewables is significantly growing in the world. This kind of energy types are dependent on weather conditions as the wind and solar energies. There are two vital requirements to locate and measure specific systems to utilize wind power: modelling and forecasting of the wind velocity. To this end, using only 4 years of measured meteorological data, the present research attempts to estimate the related speed of wind within the Libyan Mediterranean coast with the help of ANN (artificial neural networking) with three different learning algorithms, which are Levenberg-Marquardt, Bayesian Regularization and Scaled Conjugate Gradient. Conclusions reached in this study show that wind speed can be estimated within acceptable limits using a limited set of meteorological data. In the results obtained, it was seen that the SCG algorithm gave better results in tests in this study with less data.

Predicting initial electricity demand in off-grid Tanzanian communities using customer survey data and machine learning models

Mini-grids are the lowest-cost solutions for electrifying many homes and businesses in rural communities with low energy access. Estimates of the electricity demand of unelectrified customers are a crucial input to selecting mini-grid sites, projecting revenue, and sizing system components to provide adequate capacity while minimizing capital costs. Typical customer survey-based demand estimates for these communities — where there are no historical data — are not reliable, typically overpredicting demand. Here, we test a data-driven approach to demand prediction using survey and smart meter data from 1378 Tanzanian mini-grid customers. We found that models incorporating customer survey data into their predictions consistently out-performed a baseline model that did not. Our best-performing model, the LASSO, predicted daily electricity demand with a median absolute error of 66% and 37% for individual connections and mini-grid sites, respectively. Quantitative measures of variable importance show that most survey data are not useful for estimating demand. These results suggest that surveys should prioritize thorough inventories of prospective customers' currently-owned appliances instead of detailed demographic information or self-reported habits and plans. Pairing shortened questionnaires with smart meter data from preexisting mini-grids can improve estimates of initial customer electricity demand significantly compared to standard field practices.

Fractal analysis for estimating electricity demand through the application of Hurst’s re-staging exponent

This article presents a methodology for fractal analysis using the Hurst re-staging exponent for the generation prediction of a hydroelectric power plant. Which is an innovative technique to address this problem since literature usually performs it through classical and lately Bayesian statistical processes. The importance of the work implies amalgamation of concepts from fractal theory, chaos, phase space, delay time and Hurst exponent, necessary in the analysis that will determine an estimate of the future behavior of a chaotic series. Thus, the problem is addressed from the qualitative analysis through the theory of chaos.

Elasticity estimation and forecasting: An analysis of residential electricity demand in Brazil

Spatiotemporal models to estimate electricity demand are scarce in the existing literature. In this paper, we compare three models to estimate elasticities and forecast demand for residential electricity in Brazil. The Dynamic Spatial Durbin Model presented the best goodness of fit, with results that confirm the need to consider spatial dependence in the Brazilian regions. The results showed temporal inertia, inelasticity of demand concerning price and income, and a significant impact of the temperature and the number of households connected to the grid. We conclude that omitting the spatiotemporal dynamic could lead to bias in the models used by Brazilian utilities.

Research on the method of electricity demand analysis and forecasting: the case of China

With economic expansion of China having moderated to a "New Normal" phase, concerns on the surplus supply of electricity have accelerated, especially in the northwest and northeast regions. The ongoing power system reform also brings various of uncertainties, posing challenge to power supply and demand balance. Therefore, the accurate estimation of electricity demand is still inevitable and urgent. In this paper, the causality relationship of electricity demand and the selected factors, namely GDP, population, energy structure, industrial structure and urbanization, is examined by using Stationary, Co-integration and Granger Causality Tests. Then the direct and indirect effects of these factors are investigated via a path-coefficient analysis. Finally, in these foundations, an improved Chicken Swarm Optimization based on Stimulated Annealing, namely Stimulated Annealing Chicken Swarm Optimization, is proposed to optimize the weighting factors of three forms of electricity demand models. The Stimulated Annealing Chicken Swarm Optimization not only inherits the advantages of the standard Chicken Swarm Optimization such as uncomplicated principle, handy implementation and robustness to control parameters, but also can avoid premature convergence to improve the ability of finding the best solution. Case study reveals that the Stimulated Annealing Chicken Swarm Optimization has better predictive ability than other benchmark algorithms.