Customer Load Demand Research Articles

Long-term microgrid expansion planning with resilience and environmental benefits using deep reinforcement learning

Microgrid plays an increasingly important role to enhance power resilience and environmental protection regarding greenhouse gas emission reduction through the widespread applications of distributed and renewable energy. Because of the steady growth of load demand, the strict power resilience requirements and the pressing need of carbon emission reduction, microgrid expansion planning considering those factors has become a currently topical topic. In this study, a new framework for long-term microgrid expansion planning, in which a microgrid serves as a backup power system in the event of main grid outages from the perspectives of economy, resilience and greenhouse gas emission, is proposed. Deep reinforcement learning method is used to solve this dynamic and stochastic optimization problem by taking into account various uncertainties and constraints for the long-range planning. Case studies of 20-year microgrid expansion planning using actual data are conducted. The simulation results demonstrate the effectiveness of the proposed framework on reducing greenhouse gas emissions and total cost including economic losses resulting from power grid outages, investment and operating cost of microgrid entities. In addition, the impact of customer load demand and microgrid entities price on optimal planning policies is discussed. The results demonstrate that microgrid expansion planning can be effectively adapted to different levels of load demand and different scenarios of price changes under the proposed framework. This work is helpful for decision makers to implement cost-effective and power resilient microgrid expansion planning with greenhouse gas emission reduction benefits in the long term.

Optimum sizing of economic and grid-outage resilient solar array considering the forecasting of solar project cost and customer load demand

Solar array, as a kind of renewable energy, is becoming popular because it helps reduce carbon emission, and its capability of providing power resilience during grid outages. In this work, long short term memory networks are utilized to forecast solar array installation cost and customer load demand. Then, a simulation-based approach is developed to optimize solar array investment from both power resilience and economic perspectives. Net present value of total system cost, achieved power resilience and the levelized cost of electricity are jointly considered for evaluating solar array investment options. Case studies are conducted for the households in Texas based on the real data of solar irradiation and grid outages. The results illustrate that both solar array size and its installation time influence the optimization of solar array investment. Customers can gain economic benefits and grid-outage power resilience by investing optimal sized solar arrays after a specific time. The proposed method is able to provide insights about decision making of future investment of solar array.

Coordinated Multilayer Control for Energy Management of Grid-Connected AC Microgrids

This paper proposes a coordinated multilayer control strategy for the energy management (EM) of grid-connected ac microgrids. The strategy predicts the customer's load demand and the photovoltaics (PV) power generation for a day-ahead EM. It utilizes the PV power generations and the bidirectional energy transactions from electric vehicles and battery storage to provide a combined response for load support. The system also predicts any uncertainties in customers demand and power generations and implements day-ahead precautionary measures to tackle that uncertainty. Two different prediction strategies are used, autoregressive moving average and artificial neural networks, and their performances for a day-ahead EM are investigated. The reference power from the tertiary layer EM is sent to the local controllers for power regulation at the inverter level. Additionally, the varying power output reference obtained from a day-ahead EM is classified into the slow, medium, and fast variations. The performance of the local controller employed in the interfacing grid-connected three-phase inverter is tested during the above-mentioned power reference variations. The total harmonic distortion incorporating a moving window technique is calculated for the ac output current during each class of power variations over a day.

Electricity Forecasting Using Multi-Stage Estimators of Nonlinear Additive Models

French electricity load forecasting has encountered major changes during the past decade. These changes are, among other things, due to the opening of the electricity market and the economic crisis, which require the development of new automatic time adaptive prediction methods. The advent of innovating technologies also needs the development of some automatic methods because thousands or tens of thousands of time series have to be studied. In this paper we adopt for prediction a semi-parametric approach based on additive models. We present an automatic procedure for explanatory variable selection in an additive model and show how to correct middle term forecasting errors for short term forecasting. First, we consider an application to the EDF customer load demand which is typical of a load demand at an aggregate level. The goal of the application is to select variables from a large explanatory variables dictionary. The second application presented is an application on load demand of GEFCom 2012 competition, which we consider as a local application, where a major difficulty is to select some meteorological stations.

Hybrid renewable energy integration (HREI) system for subtropical climate in Central Queensland, Australia

Global warming encourages interest worldwide to integrate large quantities of renewable energy into the power grid as these sources are free from greenhouse gas emissions as well as assisting to reduce the energy crisis. Demand for energy in the Capricornia region of Queensland, Australia is increasing due to expansion of the coal industry and the availability of nearby iconic tourist attractions such as the Great Barrier Reef. Therefore, to reduce the energy crisis as well as minimise global warming, a hybrid renewable energy integration system was developed to facilitate installation of significant renewable energy generation capacity into the grid in the subtropical climate of Central Queensland. The proposed hybrid renewable energy integration system comprises a prediction model that forecasts solar and wind generation in advance; a techno-economic model that analyses the techno-economic and environmental prospects of renewable energy; and a load management system by which utilities can manage customer load demand efficiently. It has been shown that a subtropical climatic region has significant potentialities for substantially increased use of renewable energy which can not only contribute to the reduction of global warming, but also reduce energy generation costs and help ameliorate the energy crisis.

Integrated Alternative Source of Energy System with Grid Substation

This paper discusses the modeling, simulation and optimization of integrated alternative source of energy system with grid substation. When there is a lack of energy in grid substation to meet the demand of customer load then alternative source of energy compensates energy deficiency and after that if there is excess energy available then remaining energy is stored in battery bank or sells to grid according to need. In another case when required energy is not generated by alternative sources then remaining energy is purchased from the grid and if the grid substation hasn’t available these remaining energy to meet the customer load demand during the peak load then stored battery energy is also active. The size of integrated alternative source of energy system may vary from few hundred watts to ten or hundreds of MW. The possible sources of alternative energy like solar, wind and biomass are considered under this paper. Study location, Muzaffarpur district of Bihar (India), is taken because there is good potential of energy generation from alternative source of energy and still there is no any wind and solar energy power plant. The modeling and simulation of the system is done by using HOMER software and the software solve the optimization problem to minimize the total cost of system and maximum utilization of alternative source of energy.

Exploring the trade-off between competing objectives for electricity energy retailers through a novel multi-objective framework

Energy retailer is the intermediary between Generation Companies and consumers. In the medium time horizon, in order to gain market share, he has to minimize his selling price while looking at the profit, which is dependent on the revenues from selling and the costs to buy energy from forward contracts and participation in the market pool. In this paper, the two competing objectives are engaged proposing a new multi-objective framework in which a ε-constraint mathematical technique is used to produce the Pareto front (set of optimal solutions). The stochasticity of energy prices in the market and customer load demand are coped with the Lattice Monte Carlo Simulation (LMCS) and the method of the roulette wheel, which allow the stochastic multi-objective problem to be turned into a set of deterministic equivalents. The method performance is tested into some case studies.

신재생 에너지원 기반의 마이크로그리드 최적운영 방안

This paper presents an optimal operation of microgrid systems and considering a tie-lines capacities that concerned each grid. The microgrid system consists of a wind turbine, a diesel generator, and a fuel cell. An one day load profile and wind resource for wind turbine generator were used for the study. For the grid interconnection, tie-line capacities were applied as constraints. The capacity constraints of tie-lines in production cost analysis are very important issues in the operation and planning of microgrid. In optimization, the Harmony Search (HS) algorithm is used for solving the problem of microgrid system operation which a various generation resources are available to meet the customer load demand with minimum operating cost. The application of HS algorithm to optimal operation of microgrid proves its effectiveness to determine optimally the generating resources without any differences of load mismatch.

INTELLIGENTLY CONTROLLED SUPERCONDUCTING MAGNETIC ENERGY STORAGE FOR IMPROVED LOAD FREQUENCY CONTROL

In multi-area power systems with steam reheat constraint andgovernor dead band nonlinearity, besides maintaining the normalfrequency, the control engineer is faced with difficult problem ofcontinuous electromechanical oscillations which the various tie-linesare subjected to whenever there is a change in customer loaddemand.

Reliability modeling of distributed generation in conventional distribution systems planning and analysis

The primary objective of any electric utility company in the new competitive environment would be to increase the market value of the services it provides with the right amount of reliability, and at the same time, lower its costs for operation, maintenance, and construction of new facilities in order to provide lower rates for customers. The electric utility company will strive to achieve this objective via many different means, one of which is to defer the capital distribution facility requirements in favor of a distributed generation (DG) solution by an independent power producer (IPP) to meet the growing customer load demand. In this case, the distribution capital investment deferral credit received by the IPP will be dependent on the incremental system reliability improvement rendered by the DG solution. In other words, the size, location and the reliability of the DG will be based on the comparable incremental reliability provided by the distribution solution under considerations. This paper presents a reliability model for determining the DG equivalence to a distribution facility for use in distribution system planning studies in the new competitive environment.

A probabilistic approach to determining generation equivalence to a transmission facility

The electricity transmission system will be planned, operated and maintained by an independent transmission administrator in the new electric industry structure. The primary objective of any electricity transmission system administrator in the new competitive environment would be to increase the market value of the services it provides with the right amount of reliability and, at the same time, lower its costs for operation, maintenance and construction of new facilities in order to provide lower rates for customers. The electricity transmission administrator will strive to achieve this objective via many different means, one of which is to defer capital transmission facility requirements in favor of a generation solution by an independent generator to meet the growing customer load demand. In this case, the transmission capital investment deferral credit received by the independent generator will be dependent on the incremental system reliability improvement rendered by the transmission solution. In other words, the size, location and the reliability of the independent generator will be based on comparable incremental reliability provided by the transmission solution under consideration. This paper presents a probabilistic technique for determining generation equivalence to a transmission facility for use in transmission system planning studies in the new competitive environment.

Distribution circuit state estimation using a probabilistic approach

Past work on distribution circuit state estimation has focused on the adoption of a transmission state estimator approach, without necessarily accounting for the specific requirement of a distribution circuit-based analysis. On distribution circuits, typically, there are very few available real-time measurements, and thus, researchers have treated customer load demand estimates as pseudo-measurements in a weighted-least-squares formulation. This can lead to convergence problems and also, the approach effectively assumes that all bus load demands are normally distributed (Gaussian) which may not be valid on distribution circuits. This paper presents an alternative approach to distribution circuit state estimation using a probabilistic extension of the radial load flow algorithm while accounting for real-time measurements as solution constraint. The algorithm which takes advantage of the radial nature of distribution circuits also accounts for other issues specific to distribution circuits. Namely, the algorithm accounts for nonnormally distributed loads, incorporates the concept of load diversity (load correlation) and can interact with a load allocation routine. The effectiveness of the algorithm is illustrated through comparisons made with Monte Carlo simulations.