Scheduling Of Power Plants Research Articles

Dynamic stability improvement with integrated power plant scheduling method based on moment of inertia

Dynamic stability is one of the important aspects of maintaining the stability of an electrical power system as a whole. Dynamic stability study is the ability of the electric power system to return to the equilibrium point after a relatively small disturbance suddenly occurs for a long time. In this paper, we offer a method of rescheduling the generator to improve system stability by looking at the critical clearing time (CCT). Changes in the CCT value are due to changes in the load on each bus. The modelling of the IEEE 14 bus system is carried out with the help of the Simulink Power System Analysis Toolbox (PSAT) 2.1.10, which is integrated with the MATLAB R. 2016a program. The simulation results show that the CCT value decreases as the fault location gets closer to the main generator.

Impact of coronavirus disease 2019 on electricity demand and the unit commitment problem: a long–short-term memory-based machine learning approach

Coronavirus disease 2019 (COVID-19) has affected many behaviours and aspects of society. Electricity consumption has been considerably affected by the pandemic, with significant effects on the electricity load demand profile. In this article, the impact of COVID-19 on electricity demand in the state of Florida is investigated through a novel machine learning technique. The LSTM technique shows good accuracy in forecasting the load profiles for all days studied (weekdays and weekends) and also before and during the pandemic. The UC problem is solved considering the load profiles, and the impact of COVID-19 on power plant scheduling is evaluated. The simulation results show an increase in residential demand for electricity at weekends, while both residential and commercial demand are reduced during weekdays. Therefore, the operating cost of a weekday in 2020 was lower than that in 2019, while the operating cost of a weekend was higher in 2020 than in 2019.

Optimal dispatch of multi-virtual power plants based on Grey Wolf Optimization algorithm

The day-ahead scheduling of virtual power plant (VPP) is of great significance for the development of distributed generations. To solve the problem of large-scale VPP scheduling, this paper proposes a multi-VPPs optimal dispatching strategy. In this model, the VPP is divided into multiple sub-VPPs, and each sub-VPP scheduled internally. Likewise, the power interaction between each sub-VPP at the contract price is allowed. This model can realize complementary power, and hence maximize economic benefit of VPP and also improve the level of new energy consumption. Then, to solve this nonlinear optimization problem, Grey Wolf Optimization (GWO) algorithm is employed. Finally, the proposed model is tested through a detailed case study, and results demonstrate that the GWO algorithm can improve the computational efficiency.

Risk-Based Electrical-Thermal Scheduling of a Large-Scale Virtual Power Plant Using Downside Risk Constraints for Participating in Energy and Reserve Markets

Risk-Based Electrical-Thermal Scheduling of a Large-Scale Virtual Power Plant Using Downside Risk Constraints for Participating in Energy and Reserve Markets

Techno-economic benefits deriving from optimal scheduling of a Virtual Power Plant: Pumped hydro combined with wind farms

Wind and solar are renewable sources characterized by a great potential but a variable and unpredictable nature. Characteristics that provoke mismatch between power supply and demand which in turns are the source of network control issues. Problems manageable with the installation of large-scale energy storage like Pumped Hydro Energy Storage (PHES). To this purpose, the paper presents a techno-economic analysis for assessing the benefits deriving from the hybridization of a seawater PHES (sPHES) plant with an in-operation wind farm. Plants work as a Virtual Power Plant (VPP) instead of two separate unit. The hybridization aims to avoid the unbalances of the wind farm, having negative impact on both the grid and the wind farm owner itself. The unbalances cause unpredictable deviations from the predicted production and, hence, power production fluctuations faced by the grid with curtailments and/or calls in operation of fossil-fuelled power plants. A VPP techno-economic optimization on a daily bases is performed using the ASD-PSO algorithm while the optimization aim is the revenue maximization. Results show that a VPP composed by a sPHES and a wind farm and with a single owner, is able to manage the wind farm production, improving the revenue of 6.8% compared to plants acting separately on the market. The VPP approach is also convenient in the case of a joint venture between the plants owners. But, in this case, the revenue is increased only of 0.4% compared to plants working separately. Finally, results also highlight that, in the Italian electricity market, a sPHES plant alone is not able to generate sufficient revenue to guarantee a reasonable payback time while coupling it with a wind farm increases incomes and working days and avoid cost and grid issues in managing the unpredictable wind farm production.

Optimal scheduling of flexible thermal power plants with lifetime enhancement under uncertainty

• The proposed method can limit equipment damage in different scenarios. • Damage can be reduced 40% with a revenue loss of less than 1%. • Different damage mechanisms in the equipment can be included in the method. • Variable uncertainty in power demand and price can be included. • Adequate selection of material design temperature reduces deterioration. Renewable energy sources have been the focal point to decarbonise the power sector. The large deployment of these intermittent power generation units requires mechanisms to balance the grid. Thermal power plants can provide this service by increasing the number of start-ups, shut-downs, and intraday ramps at the expense of higher deterioration in critical equipment, including high-pressure steam drums, turbine rotors and blades, and high-temperature heat exchangers and pipes. This work proposes a method to formulate the power generation scheduling of thermal power plants as a stochastic optimisation problem with limitations on the maximum damage in critical components. This method models the uncertainty associated with intermittent power generation from renewable sources with a scenario-tree whilst computing the deterioration of the equipment in each scenario to limit the maximum damage. Scheduling of a flexible natural gas combined cycle demonstrated how this methodology can reduce the deterioration of the superheating heat exchanger of the power plant with minimum detriment in power generation and revenue. Furthermore, the effect of the design temperature of the material on the total damage was analysed for a broad range of temperatures and operating profiles, showing how adequate selection of design temperature can reduce the deterioration of the equipment and enhance its lifetime.

Evaluation of Communication Infrastructures for Distributed Optimization of Virtual Power Plant Schedules

With the transition towards renewable energy resources, the impact of small distributed generators (DGs) increases, leading to the need to actively stabilize distribution grids. DGs may be organized in virtual power plants (VPPs), where DGs’ schedules must be coordinated to enable the VPP to act as a single plant. One approach to solving this problem is using multi-agent systems (MAS) to offer autonomous, robust, and flexible control methods. The coordination of such systems requires communication between agents. The time required for this depends on communication characteristics, determined by the underlying communication infrastructure. In this paper, we investigate communication influences for the wireless technologies CDMA450 and LTE Advanced on the fully distributed optimization heuristic COHDA, which is used to perform optimized scheduling for a VPP. The use case under consideration is the adaptation of schedules to provide flexibility for regional congestion management for delivery on a regionalized ancillary service market (rAS). We investigate the scalability of the VPP and the effects of disturbances in the communication infrastructure. The results show that the optimization duration of COHDA can be influenced by the underlying communication infrastructure and that this optimization method is applicable to a limited extent for product delivery of rASs.

A Probabilistic Ensemble Prediction Method for PV Power in the Nonstationary Period

Due to the large number of grid connection of distributed power supply, the existing scheduling methods can not meet the demand gradually. The proposed virtual power plant provides a new idea to solve this problem. The photovoltaic power prediction provides the data basis for the scheduling of the virtual power plant. Prediction intervals of photovoltaic power is a powerful statistical tool used for quantifying the uncertainty of photovoltaic power generation in power systems. To improve the interval prediction accuracy during the non-stationary periods of photovoltaic power, this paper proposes a probabilistic ensemble prediction model, which combines the modules of data preprocessing, non-stationary period discrimination, feature extraction, deterministic prediction, uncertainty prediction, and optimization integration into a general framework. More specifically, in the non-stationary period discrimination module, the method of discriminating the difference of the power ratio difference is introduced and applied for identifying the non-stationary period of the data of photovoltaic output; in the deterministic point prediction module, a stacking- long-short-term memory neural network model is used for point forecasts; in the uncertainty interval prediction module, a BAYES neural network is introduced for probabilistic forecasts; in the optimization integration module, an optimization algorithm named Non-dominated Sorting Genetic Algorithm-II is applied for integrating and optimizing the results of the point forecast and probabilistic forecast. The proposed model is tested using two photovoltaic outputs and weather data measured from a grid-connected photovoltaic system. The results show that the proposed model outperforms conventional forecast methods to predict short-term photovoltaic power outputs and associated uncertainties. The interval width is reduced by 10–20%, and the prediction accuracy is improved by at least 10%; this can be a useful tool for photovoltaic power forecasting.

Sample average approximation for risk-averse problems: A virtual power plant scheduling application

In this paper, we address the decision-making problem of a virtual power plant (VPP) involving a self-scheduling and market involvement problem under uncertainty in the wind speed and electricity prices. The problem is modeled using a risk-neutral and two risk-averse two-stage stochastic programming formulations, where the conditional value at risk is used to represent risk. A sample average approximation methodology is integrated with an adapted L-Shaped solution method, which can solve risk-neutral and specific risk-averse problems. This methodology provides a framework to understand and quantify the impact of the sample size on the variability of the results. The numerical results include an analysis of the computational performance of the methodology for two case studies, estimators for the bounds of the true optimal solutions of the problems, and an assessment of the quality of the solutions obtained. In particular, numerical experiences indicate that when an adequate sample size is used, the solution obtained is close to the optimal one.

An adaptive ensemble predictive strategy for multiple scale electrical energy usages forecasting

Electrical energy prediction plays an important role in energy management, power plant scheduling, peak demand and grid security conflict. To deal with prediction scenarios at multiple energy and time scales, an adaptive ensemble model strategy with wider applicability is proposed. The strategy consists of three parts. Firstly, the data preprocessing part is composed of expert knowledge, recursive feature elimination (RFE) and fuzzy c-means clustering (FCM). RFE and FCM methods are used for feature identification and data clustering for different energy usage patterns. Secondly, the primary prediction part consists of a model library containing five commonly used data-driven models. To improve their prediction accuracies, key model parameters are optimized by swarm evolution algorithms (EAs). An algorithm package containing three EAs is combined with the model library. The choice of algorithm for each model depends on the comparison of accuracy in specific prediction scenarios. In the third part, a linear regression model provides the final result based on primary predictors’ outputs. Its weights are also optimized by EAs from the algorithm package.To verify the performance of this strategy, four case studies are carried out representing different energy and time scales’ prediction scenarios. Case A is a benchmark case from the first energy prediction competition organized by American Society of Heating Refrigerating and Air-conditioning Engineer (ASHRAE). Case B perform hourly electrical load prediction of a whole building from University of Wyoming, USA. Case C is a city-scale daily electrical load forecasting (Yizheng City, Jiangsu Province, China), and in Case D, national wide monthly electrical load prediction of the United States is carried out. Results of the four cases indicate that: (1) The proposed ensemble strategy performs best in all four case studies compared with five primary predictors. Compared to the best primary predictors in four case studies, the accuracies of the proposed model increase by 3% to 17% (MAPE); (2) Compared with previous reported ensemble models, the accuracy of the proposed model also increase by 11% (MAPE). The adaptive prediction framework is applicable and has the potential to become a general predictive strategy for energy predictions.

Daily peak-based short-term demand prediction using backpropagation combined to chi-squared distribution

Abstract An efficient and economic scheduling of power plants relies on an accurate demand forecast especially for the short-term due to its tight relation to power markets and trading operations in interconnected power systems. A slight deviation of load prediction from real demand could engender the start-up of a conventional power station which could be either time-consuming or requiring expensive combustible, a deviation that could interfere as well with renewables intermittency and demand response strategies. Hence, load forecasting still a challenging subject because of the various transformations that the energy sector undergoes and that directly impact the demand profile shape. Therefore, conceiving dynamic load demand forecast approaches will permit utilities save money in different vertical structures and regulation schemes. In this paper, we propose a novel approach for short-term demand prediction valid for normal and special days to address the impact of climate changes along with events occurrence on forecast accuracy. This approach is based on the prediction of hourly loads, established on the daily peak load prediction using backpropagation combined to chi-squared method for weighting historical data to enhance the training process. Obtained results from extensive testing on the Moroccan’s power system confirm the strength of the developed approach, that improved the forecast accuracy by a range of 1.1–4% compared to the existing methods.

Robust stochastic optimal dispatching method of multi-energy virtual power plant considering multiple uncertainties

In recent years, with the rapid development of the energy Internet and the deepening of the complementary coupling of various energy sources, the concept of multi-energy virtual power plant comes into being. At the same time, insufficient research on optimal scheduling of multi-energy virtual power plants under multiple uncertainties. Here we propose a robust stochastic optimal dispatching method to solve the scheduling problem under multiple uncertainties. For the source side uncertainties, the uncertain set of cardinalities with a robust adjustable coefficient is adopted to describe the output of wind turbines and photovoltaics. For the load side uncertainties, the Wasserstein generative adversarial network with gradient penalty is used to generate electric, thermal, cooling, and natural gas load scenarios, and the K-medoids clustering is used to get typical scenes. A two-stage robust stochastic optimal model of the min-max-min structure was established. Based on the dual transformation theory and the column constraint generation algorithm, the original model was solved alternately. Finally, the effectiveness of the proposed model and algorithm is verified by simulation analysis. The proposed method can get the scheduling scheme with the lowest operating cost in the worst scenario and is conducive to reducing the overall scheduling cost of the system.

Phase unbalance and PAR constrained optimal active and reactive power scheduling of Virtual Power Plants (VPPs)

In near future, the integration of unknown and unpredictable quantity of Electric Vehicles (EVs) can violate reliable and quality power service of distribution system. In spite of that, ability of simultaneously regulate active and reactive power by Plug-in-Hybrid Electric Vehicles (PHEVs) in quick response time, without affecting the batteries can help to ensure reliable and quality power service in distribution system. This paper investigates the optimal active and reactive power scheduling of PHEVs and Distributed Generations (DGs) in Virtual Power Plants (VPPs), considering unbalance and Peak-to-Average Ratio (PAR) constraints. To compute stochastic model of PHEVs, we considers dynamic nature of driving pattern based on NHTS 2017 data. The proposed approach is implemented on IEEE-25 bus unbalanced distribution system. The developed planning and operational investigation of VPPs also presents dependencies of cost and losses in terms of Unbalance factor and PAR.

A stochastic short-term scheduling of virtual power plants with electric vehicles under competitive markets

This paper presents a risk-averse stochastic framework for short-term scheduling of virtual power plants (VPPs) in a competitive environment considering the potential of activating electric vehicles (EVs) and smart buildings in demand response (DR) programs. In this framework, a number of EV Parking Lots (PLs) which are under the jurisdiction of the VPP and its rivals are considered that compete to attract EVs through competitive offering strategies. On the other hand, EVs' owners try to choose a cheaper PL for EVs' charging to reduce payment costs. Therefore, the objective of EVs owners can be in conflict with the objective of PLs that provide services for EVs under each VPP. In this regard, the decision-making problem from the VPP's viewpoint should be formulated as a bi-level optimization model, in which in the upper-level, the VPP profit should be maximized and in the lower-level, procurement costs of EVs and other responsive loads should be minimized, simultaneously. To solve the proposed bi-level problem, it is transformed into a traceable mixed-integer linear programming (MILP) problem using duality theory and Karush-Kahn-Tucker (KKT) optimality conditions. The proposed model is tested on a practical system and several sensitivity analyses are carried out to confirm the capability of the proposed bi-level decision-making framework.

Risk-averse probabilistic framework for scheduling of virtual power plants considering demand response and uncertainties

In this paper, a risk-based stochastic framework is presented for short-term energy and reserve scheduling of a virtual power plant (VPP) considering demand response (DR) participation. The VPP comprises several dispatchable generation units, battery energy storage systems (BESSs), wind power units, and flexible loads. The proposed scheduling framework is formulated as a risk-constrained stochastic program to maximize the VPP’s profit considering uncertainties of loads, wind energy and electricity prices as well as N-1 contingencies. The proposed model considers both supply and demand-sides capability for providing and deploying reserves in order to optimize the use of resources while satisfying N-1 security and other constraints. Moreover, the effect of risk-aversion on decision making of the VPP in the offering/bidding power and required reserve services is investigated by implementing conditional value-at-risk (CVaR) in the optimization model. The proposed scheme is implemented on a test VPP and the energy and reserve scheduling with and without DR participants is addressed in detail through a numerical study. Moreover, the effects of the operator’s risk-averse behavior on the VPP energy and reserve management and its security indices are investigated.

Short-term optimal scheduling of hydro-thermal power plants using artificial bee colony algorithm

This research paper reports the implementation of short-term scheduling of hydro-thermal power plants by using an artificial bee colony algorithm. Short-term hydro-thermal scheduling is a type of economic dispatch problem in which thermal power plants are dispatched at the optimized operating point to reduce the fuel cost and to achieve in parallel the maximum cost-benefit from hydel power plants. The power system, considered in this study, is assumed to run optimally and the transmission losses have also been taken into account. The artificial bee colony algorithm proves itself to be most suited for this particular problem as it results in the minimum cost in the shortest time compared with those obtained from previously applied techniques.

Modeling and integration of water desalination units in thermal unit commitment considering energy and water storage

In this paper, a new short-term daily scheduling model is proposed and formulated. In the proposed model, thermal power plant will be optimally scheduled in a system with water desalination units. For the thermal units, various cost terms including start-up, shut-down, and also fuel cost are considered and modeled. Also, thermal dynamics of the plants including power generation ramp-up and ramp-down, minimum up and down time limits, and start-up and shut-down ramps are considered. In addition, energy storage batteries as well as water storage tanks of the desalination units are modeled. The proposed model is a mixed integer linear programming (MILP) for simultaneous scheduling of thermal power plants and desalination units while considering transmission constraints. Power and water balance at the network buses in addition to the power flow constraints of the lines are considered. Various membrane and thermal desalination methods are modeled. The proposed model is implemented on a test case and the simulation results show a significant reduction in the system operating costs as well as improved performance of both the thermal power plants and the desalination unit. The proposed scheduling method can be used for the various future studies of joint power-water production and coordination.

Very-Short-Term Power Prediction for PV Power Plants Using a Simple and Effective RCC-LSTM Model Based on Short Term Multivariate Historical Datasets

Improving the accuracy of very-short-term (VST) photovoltaic (PV) power generation prediction can effectively enhance the quality of operational scheduling of PV power plants, and provide a reference for PV maintenance and emergency response. In this paper, the effects of different meteorological factors on PV power generation as well as the degree of impact at different time periods are analyzed. Secondly, according to the characteristics of radiation coordinate, a simple radiation classification coordinate (RCC) method is proposed to classify and select similar time periods. Based on the characteristics of PV power time-series, the selected similar time period dataset (include power output and multivariate meteorological factors data) is reconstructed as the training dataset. Then, the long short-term memory (LSTM) recurrent neural network is applied as the learning network of the proposed model. The proposed model is tested on two independent PV systems from the Desert Knowledge Australia Solar Centre (DKASC) PV data. The proposed model achieving mean absolute percentage error of 2.74–7.25%, and according to four error metrics, the results show that the robustness and accuracy of the RCC-LSTM model are better than the other four comparison models.

Techno-economic assessment of consumers' participation in the demand response program for optimal day-ahead scheduling of virtual power plants

In this paper, the consumers' participation in the demand response program for virtual power plant (VPP) is proposed. A proposed price-based scheduling strategy for optimal day-ahead scheduling of VPP is investigated to maximize the profit of the VPP composed of a set of distributed generation, dispatchable load, and wind turbines. For solving the optimal scheduling problem, a modified differential evolution algorithm (MDEA) is proposed. To achieve the work target, 33-kV, 18-bus test system is selected as a test system. The proposed algorithm leads to effective scheduling of VPP resources, especially for reserve power. To investigate the proposed methodology, the load is increased and the scheduling of VPP is investigated at the best and worst boundaries of VPP resources uncertainty. The best case happens at maximum load, high prices and maximum energy from renewables, while the worst scenario occurs at minimum load, low prices and minimum energy from renewables. In addition, the algorithm ensures the importance of VPP in playing a dual role as a demand/generation unit from the perspective of the upstream network in some time periods and highlights the vital effect of dispatchable loads on the management process at different conditions of the dispatchable load.

Cost and Emission Optimization in Power Plant Using the Single Echelon Economic Dispatch Model (SEED)

The electricity supply chain system consists of power plants, transmission and distribution. Differences in power plant characteristics cause differences in fuel costs and emissions. Inaccuracies in scheduling the plant have an impact on the surge in fuel use. This becomes the basis for the optimization of power plant scheduling. Optimization on the generator side is usually using the economic dispatch model. However, the Economic Dispatch Model does not consider optimization on the transmission network. Optimization on the transmission network is needed because there are losses in the transmission network. Losses will be even greater when the distance between the power plant and the customer is getting further away. So if distance can be minimized, fuel costs can be reduced. Minimizing the distance for shipping goods can use the transportation model, but the transportation model cannot be used for electricity distribution. This is because there are differences between goods and electricity. Therefore, this study proposes the Single Echelon Economic Dispatch (SEED) model. This model is a combination of the economic dispatch model and the transportation model. This model is able to do simultaneous optimization between the generator side and the transmission side. The model is applied to the Mahakam system in East Kalimantan, using all PLN’s power plants. Load consists of two types, namely low load and peak load. As a result, in peak load and low load conditions, the P11 power plant produces the biggest losses and emissions. This is because electricity production in P11 has to go through a longer path compared to other plants. The P11 is also a power plant that requires the highest fuel costs