Power Consumption Forecasting Research Articles

Investment factor’s role in forecasting of regional energy consumption (by the example of the Irkutsk region)

The article examines a methodological approach to regional forecasting of power consumption. It is also analyzed the role of the investment factor in forecasting the economic development of the country and regions. By the example of the Irkutsk region it is estimated the prospective level of power consumption, which is necessary to predict the rational development of the region’s fuel and energy complex in terms of social and economic as well as environmental indicators.

AEBIS: AI-Enabled Blockchain-Based Electric Vehicle Integration System for Power Management in Smart Grid Platform

A Virtual Power Plant (VPP) is a network of distributed power generating units, flexible power consumers, and storage systems. A VPP balances the load on the grid by allocating the power generated by different linked units during periods of peak load. Demand-side energy equipment, such as Electric Vehicles (EVs) and mobile robots, can also balance the energy supply-demand when effectively deployed. However, fluctuation of the power generated by the various power units makes the supply power balance a challenging goal. Moreover, the communication security between a VPP aggregator and end facilities is critical and has not been carefully investigated. This paper proposes an AI-enabled, blockchain-based electric vehicle integration system, named AEBIS for power management in a smart grid platform. The system is based on an artificial neural-network and federated learning approaches for EV charge prediction, in which the EV fleet is employed as a consumer and as a supplier of electrical energy within a VPP platform. The evaluation results show that the proposed approach achieved high power consumption forecast with $\mathit {R^{2}}$ score of 0.938 in the conventional training scenario. When applying a federated learning approach, the accuracy decreased by only 1.7%. Therefore, with the accurate prediction of power consumption, the proposed system produces reliable and timely service to supply extra electricity from the vehicular network, decreasing the power fluctuation level. Also, the employment of AI-chip ensures a cost-efficient performance. Moreover, introducing blockchain technology in the system further achieves a secure and transparent service at the expense of an acceptable memory and latency cost.

Special Issue “Intelligent Control in Energy Systems”

The editor of this special issue on “Intelligent Control in Energy Systems” have made an attempt to publish a book containing original technical articles addressing various elements of intelligent control in energy systems. The response to our call had 60 submissions, of which 27 were published submissions and 33 were rejections. This book contains 27 technical articles and one editorial. All have been written by authors from 15 countries (China, Netherlands, Spain, Tunisia, United States of America, Korea, Brazil, Egypt, Denmark, Indonesia, Oman, Canada, Algeria, Mexico, and Czech Republic), which elaborated several aspects of intelligent control in energy systems. It covers a broad range of topics including fuzzy PID in automotive fuel cell and MPPT tracking, neural network for fuel cell control and dynamic optimization of energy management, adaptive control on power systems, hierarchical Petri Nets in microgrid management, model predictive control for electric vehicle battery and frequency regulation in HVAC systems, deep learning for power consumption forecasting, decision tree for wind systems, risk analysis for demand side management, finite state automata for HVAC control, robust μ-synthesis for microgrid, and neuro-fuzzy systems in energy storage.

A Novel Ensemble Method for Electric Vehicle Power Consumption Forecasting: Application to the Spanish System

The use of electric vehicle across the world has become one of the most challenging issues for environmental policies. The galloping climate change and the expected running out of fossil fuels turns the use of such non-polluting cars into a priority for most developed countries. However, such a use has led to major concerns to power companies, since they must adapt their generation to a new scenario, in which electric vehicles will dramatically modify the curve of generation. In this paper, a novel approach based on ensemble learning is proposed. In particular, ARIMA, GARCH and PSF algorithms' performances are used to forecast the electric vehicle power consumption in Spain. It is worth noting that the studied time series of consumption is non-stationary and adds difficulties to the forecasting process. Thus, an ensemble is proposed by dynamically weighting all algorithms over time. The proposal presented has been implemented for a real case, in particular, at the Spanish Control Centre for the Electric Vehicle. The performance of the approach is assessed by means of WAPE, showing robust and promising results for this research field.

МОДЕЛИ ПРОГНОЗИРОВАНИЯ ПОТРЕБЛЕНИЯ ЭЛЕКТРОЭНЕРГИИ С УЧЕТОМ ВЛИЯНИЯ МЕТЕОФАКТОРОВ

The purpose of the study is to assess the effect of meteorological factors on electricity consumption. The object of the study is the power system of the Astrakhan region. As the initial data, a sample was compiled of the data of 12 months of 2015 and 2016. One of the main indicators in planning the work of the energy association and the energy system is the magnitude of the forecasts of the expected power consumption (electricity and power consumption) as a whole for the system, groups and individual consumers, the nodes of the electrical scheme. The magnitude of the forecast of power consumption is a reference indicator for the subsequent planning of energy balances, power and calculation of electrical regimes. The need for accurate forecasting is due to technological and economic reasons. Accurate calculations to ensure the optimal distribution of the load between stations, from an economic point of view, facilitate sales of economically viable operations for the purchase and sale of electricity. There has been given analysis of energy consumption depending on temperature changes and the amount of precipitation during the accounting period. Calculations are made for the deviations of each indicator. The graphs of the dependences of power consumption on the amount of precipitation and temperature are plotted. For these dependencies, polynomial equations are compiled. For a constant relationship between power consumption and the temperature of the calculated correlation coefficients. The correlation coefficient makes it possible to determine how proportional the variability of the two variables is. Taking into account the set of meteorological factors makes it possible to increase the effectiveness of the predicted changes in the energy sector.

Big Data based Service Platform and Its Typical Applications of Electric Power Industries

The smart power grid is a national immediate demand project, as well as a challenging research issue in power sector. With the emerging and rapidly developing of big data technologies, smart power services based on big data became increasingly significant. By investigating the related work and requirements in power sector, a general service platform framework in electric power industries was proposed, in which big data oriented processing modules are illustrated for power data flow. Finally, the associated typical power services are presented for China Southern Power Grid, where it was considered as an experimental base to carry out several power big data applications, as their results show that the power big data service platform can preferably improve the performance in line loss calculation, real-time power equipment detection and power consumption forecasting.

System analysis of power consumption by nonferrous metallurgy enterprises on the basis of rank modeling of individual technocenosis castes

To increase energy efficiency at non-ferrous metallurgy enterprises, an integrated system approach for estimation of electricity consumption is needed. The paper presents the results of a rank analysis of the power consumption of individual castes of process equipment on the basis of an integrated energy survey of the enterprise. A methodology for constructing mathematical models for calculating and predicting electric power consumption for all castes of the ranked H-distribution of technocenosis has been developed. For the first time, according to the established regularity of the H-distribution, a mathematical model for predicting power consumption has been developed, including a quantitative analysis of the energy characteristics of consumers by individual castes of technocenosis. A retrospective check of the relative error in the prediction of electricity consumption showed that for the model it does not exceed 2%, which is significantly lower than the relative error of the prediction for a number of models of other types. The received model is recommended for use in the automated system of dispatching control of power consumption for the purposes of short-term forecasting of electric power consumption at industrial enterprises of non-ferrous metallurgy.

Выбор метода для задач краткосрочного прогнозирования электропотребления собственных нужд ТЭЦ

Выбор метода для задач краткосрочного прогнозирования электропотребления собственных нужд ТЭЦ

Integrated approach to the planning of energy consumption by non-traction railway consumers

Non-traction railway load consumes a significant amount of electricity. Russian Railways supplies electricity not only to its structural units but also to other consumers. Private houses, individual entrepreneurs and small production facilities located near the railway are powered by the RZD. It is very important for the company to plan consumption for non- traction needs. The subject of the study in the paper is a systematic approach to the planning of power consumption using the mathematical apparatus of artificial neural networks, correlation analysis and the method of expert assessments. The method of expert assessments allows identifying the most significant factors that affect the consumption of electricity. It is necessary to attract experienced professionals in the field of electricity, working at a particular enterprise. They are able to determine with a high degree of accuracy those factors that have a significant impact on the consumption of the organization. Correlation analysis allows you to mathematically check the degree of influence of a single factor on the resulting value. The apparatus of artificial neural networks allows building a forecast of power consumption, taking into account the influence of external factors. The authors propose to use a systematic approach to the planning of power consumption. It is necessary to combine three tools: the method of expert assessments, correlation analysis and artificial neural networks. The combination of these tools will improve the accuracy of power consumption planning and, as a result, will lead to increased economic efficiency due to the rational consumption of electricity.

ESTIMATION OF POTENTIAL OF ENERGY SAVING IN THE BUDGET SPHERE BASED ON THE RANGE ANALYSIS

The article outlines the issues of increasing energy efficiency and energy security of Ukraine, presents the technical energy system, and considers the problems of optimization of energy efficiency management. The research levels in the field of energy saving are presented and analyzed. The methods of mathematical statistics and technocenological methods of energy consumption modeling are presented. A normal distribution law (often called the Gauss law) and rank analysis, which is understood as the method of research technocenosis, involves optimization based on the criteria of the form of hyperbolic H-distributions, is considered. The procedures of interval estimation, parametric rationing, forecasting and standardization of consumption of resources are resulted. Rank analysis is presented in procedures such as: difflex analysis (performed at the interval evaluation interval), GZ-analysis (performed at the stage of forecasting) and ASR-analysis (carried out at the stage of valuation). The mathematical apparatus of the technocenological approach is represented by three models of hyperbolic H distribution: sighting, ranked and ranked by parameter. The algorithm of the rank analysis for the forecast of power consumption is developed. The procedure of technocenological optimization is described. Studies of cenoses are reduced to their systematic description of the hierarchical system of indicators (which is necessary for the identification of the cenosis) and to the structural cenological description. The paper presents the ranked parametric distribution of technocenoses for the consumption of electric energy by budgetary organizations and institutions of Solomyansky district of Kyiv in 2016, and the analytical dependence of this technocenose for the subsequent optimization procedure on the parameter of electric power consumption is calculated.

Business Intelligence in Facility Management: Determinants and Benchmarking Scenarios for Improving Energy Efficiency

ABSTRACTThis study investigates data sources for Business Intelligence solutions, which provide a foundation for holistic analyses, optimization, and forecasting of electrical power consumption within Facility Management for retail facilities networks. A multiple case study among Polish subsidiaries of Sweden, France, and UK-based companies along with a domestic Polish has been conducted. The authors identify, verify, and organize the determinants augmenting an energy efficiency-oriented decision-making process and introduce a set of potential benchmarking scenarios for managing it.

Day-Ahead Short-Term Forecasting Electricity Load via Approximation

The method of short-term forecasting of a power consumption which can be applied to short-term forecasting of power consumption is offered. The offered model is based on sinusoidal function for the description of day and night cycles of power consumption. Function coefficients – the period and amplitude are set up is adaptive, considering dynamics of power consumption with use of an artificial neural network. The presented results are tested on real retrospective data of power supply company. The offered method can be especially useful if there are no opportunities of collection of interval indications of metering devices of consumers, and the power supply company operates with electrical supply points. The offered method can be used by any power supply company upon purchase of the electric power in the wholesale market. For this purpose, it is necessary to receive coefficients of approximation of sinusoidal function and to have retrospective data on power consumption on an interval not less than one year.

The Increase of Power Efficiency of Underground Coal Mining by the Forecasting of Electric Power Consumption

In article the analysis of electric power consumption and problems of power saving on coal mines are considered. Nowadays the share of conditionally constant costs of electric power for providing safe working conditions underground on coal mines is big. Therefore, the power efficiency of underground coal mining depends on electric power expense of the main technological processes and size of conditionally constant costs. The important direction of increase of power efficiency of coal mining is forecasting of a power consumption and monitoring of electric power expense. One of the main approaches to reducing of electric power costs is increase in accuracy of the enterprise demand in the wholesale electric power market. It is offered to use artificial neural networks to forecasting of day-ahead power consumption with hourly breakdown. At the same time use of neural and indistinct (hybrid) systems on the principles of fuzzy logic, neural networks and genetic algorithms is more preferable. This model allows to do exact short-term forecasts at a small array of input data. A set of the input parameters characterizing mining-and-geological and technological features of the enterprise is offered.

Power Consumption Short Term Forecast Using Signal Auto Regressive Method

Short term forecast of power consumption plays a critical role in optimal exploiting of power system. Economic performance and reliability of a network significantly depends to load forecast's accuracy. This forecast is used for load management and planning for unit commitment (UC) and it has especial complexity due to influence of multiple nonlinear relations between daily periodic variations and load consumption changes. 30–60 minutes forecasts are employed extensively in power distribution network. This study aims to perform short term forecasts by using linear auto regressive (AR) modelling. In this paper, collected data from a 63 kV distribution station is modelled by a aforementioned method. It is demonstrated that AR method can model data properly using auto correlation functions residual error and sum of squares criteria. Since data is non-stationary, performance of auto regressive integrated moving average (ARIMA) is investigated and optimal rank of model and the best data length to perform modelling are presented. Regarding to forecasts, appropriate model's rank for AR and ARIMA are 20 and (1, 1, 0) respectively. In AR model forecast error is ±10% which is equal to 38.1 and in ARIMA method forecast mse is ±1.21% and equals to 0.7277 and mse value for neural network method is 0.952 that totally ARIMA method shows better performance than the other.

Прогнозирование объемов энергопотребления в зависимости от исходной информации

The transition to market relations between power consumers and power supply systems leads to stricter requirements to all market participants. Therefore, a power sales company has to face a severe competition in the power retail market and to solve a problem of an efficient distribution of power acquired in the wholesale market. A forecast value of power consumption is a reference indicator for further planning the rated power values required for response to power consumer demand and minimizing the power production and transportation cost. An inaccurate forecast results in a shortage or an excess of purchased power and makes the company buy or sell electricity at a disadvantageous price. The problem of forecasting power consumption can be solved based on data supplied by a power sales company. For this purpose, a forecast of power consumption with a minimum error takes into account meteorological factors, too. Forecasts with different databases are considered. The studies have revealed a clear link between meteorological factors and power consumption, which is expressed in the correlation coefficient. The most effective forecasting model is that with a great number of different input databases.

Determination of the Minimal Error of Power Consumption Forecast by Regression Analysis

A procedure of estimation of the non-regular component of a power consumption schedule is described in this paper. A value of the non-regular component enables to evaluate the minimal error of daily-ahead power consumption forecast based on regression mathematical models. Evaluating calculations are given by the example of 10 regional power systems of Russia.

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.

Prediction of Electric Power Consumption Based on the Improved GM(1, 1)

Based on the electric power consumption data in 2001-2010, this paper discusses GM (1, 1) model and its improved model in the application of power consumption forecasting. Due to the traditional Grey Model itself has certain defects, we grouped the original sequence according to the degree of deviation first, and then combined with nonlinear GM (1, 1, α) to improve the traditional GM (1, 1) model. Through the relative error testing and the posterior testing, this paper made a comparative analysis to the traditional GM (1, 1) model and the improved GM. Example of Beijing shows that the improved model had good accuracy; it had a good application value in the actual prediction system. DOI: http://dx.doi.org/10.11591/telkomnika.v11i8.3104

SARX Model Application for Industrial Power Demand Forecasting in Brazil

<p>The objective of this paper is to propose the application of the SARX model to arrive at industrial power consumption forecasts in Brazil, which are critical to support decisionmaking in the energy sector, based on technical, economic and environmentally sustainable grounds. The proposed model has a seasonal component and considers the influence of exogenous variables on the projection of the dependent variable and utilizes an autoregressive process for residual modeling so as to improve its explanatory power. Five exogenous variables were included: industrial capacity utilization, industrial electricity tariff, industrial real revenues, exchange rate, and machinery and equipment inflation. In addition, the model assumed that power forecast was dependent on its own time lags and also on a dummy variable to reflect 2009 economic crisis. The study used 84 monthly observations, from January 2003 to December 2009. The backward method was used to select exogenous variables, assuming a 0.10 descriptive value. The results showed an adjusted coefficient of determination of 93.9% and all the estimated coefficients were statistically significant at a 0.10 descriptive level. Forecasts were also made from January to May 2010 at a 95% confidence interval, which included actual consumption values for this period. The SARX model has demonstrated an excellent performance for industrial power consumption forecasting in Brazil.</p>

Evolving Takagi–Sugeno fuzzy model based on switching to neighboring models

In this paper, we propose a new online identification approach for evolving Takagi–Sugeno (TS) fuzzy models. Here, for a TS model, a certain number of models as neighboring models are defined and then the TS model switches to one of them at each stage of evolving. We define neighboring models for an in-progress (current) TS model as its fairly evolved versions, which are different with it just in two fuzzy rules. To generate neighboring models for the current model, we apply specially designed split and merge operations. By each split operation, a fuzzy rule is replaced with two rules; while by each merge operation, two fuzzy rules combine to one rule. Among neighboring models, the one with the minimum sum of squared errors – on certain time intervals – replaces the current model.To reduce the computational load of the proposed evolving TS model, straightforward relations between outputs of neighboring models and that of current model are established. Also, to reduce the number of rules, we define and use first-order TS fuzzy models whose generated local linear models can be localized in flexible fuzzy subspaces. To demonstrate the improved performance of the proposed identification approach, the efficiency of the evolving TS model is studied in prediction of monthly sunspot number and forecast of daily electrical power consumption. The prediction and modeling results are compared with that of some important existing evolving fuzzy systems.