Over the past decade, PV power plants have increasingly contributed to power generation. However, PV power generation widely varies due to environmental factors; thus, the accurate forecasting of PV generation becomes essential. Meanwhile, weather data for environmental factors include many missing values; for example, when we estimated the missing values in the precipitation data of the Korea Meteorological Agency, they amounted to ~16% from 2015–2016, and further, 19% of the weather data were missing for 2017. Such missing values deteriorate the PV power generation prediction performance, and they need to be eliminated by filling in other values. Here, we explore the impact of missing data imputation methods that can be used to replace these missing values. We apply four missing data imputation methods to the training data and test data of the prediction model based on support vector regression. When the k-nearest neighbors method is applied to the test data, the prediction performance yields results closest to those for the original data with no missing values, and the prediction model’s performance is stable even when the missing data rate increases. Therefore, we conclude that the most appropriate missing data imputation for application to PV forecasting is the KNN method.
Electric load forecasting for buildings is important as it assists building managers or system operators to plan energy usage and strategize accordingly. Recent increases in the adoption of advanced metering infrastructure (AMI) have made building electrical consumption data available, and this has increased the feasibility of data-driven load forecasting. Self-organizing map (SOM) has been successfully utilized to cluster a dataset into subsets containing similar data points. These subsets are then used to train the forecasting models to improve forecasting accuracy. However, some buildings may have insufficient data since newly installed monitoring devices such as AMI have no choice but to collect a limited amount of data. Using a clustering technique on small datasets could lead to overfitting when using forecasting models following an SOM network to be trained with clusters. This results in a relatively high generalization error. In this study, we propose to address this problem by employing the stacking ensemble learning method (SELM) that is well-known for its generalization ability. An experimental study was conducted using the electricity consumption data of an actual institutional building and meteorological data. Our proposed model outperformed other baseline models, which means it successfully mitigates the effect of overfitting.
With the high penetration level of wind power in power systems, wind power generation is necessary to participate in the generation scheduling to balance supply and demand. If a power system operator wants wind power generation to participate in generation scheduling, wind power should be forecasted accurately. Varying weather conditions constantly subject wind power generation to uncertainty and variability, making the determination of generating capacity difficult for system operators. Wind power forecasting should be corrected to reduce errors of generation scheduling. This study proposes a method to correct wind power forecasting by considering wind speed forecast error, which is represented in normal distribution. The biased mean of wind power forecast error is calculated using the analytical approach for wind speed forecast. The proposed method corrects wind power forecast by subtracting the biased mean from the wind power forecast error.
Integrated energy systems can provide a more efficient supply than individual systems by using resources such as cogeneration. To foster efficient management of these systems, the flexible operation of cogeneration resources should be considered for the generation expansion planning model to satisfy the varying demand of energy including heat and electricity, which are interdependent and present different seasonal characteristics. We propose an optimization model of the generation expansion planning for an integrated energy system considering the feasible operation region and efficiency of a combined heat and power (CHP) resource. The proposed model is formulated as a mixed integer linear programming problem to minimize the sum of the annualized cost of the integrated energy system. Then, we set linear constraints of energy resources and describe linearized constraints of a feasible operation region and a generation efficiency of the CHP resource for application to the problem. The effectiveness of the proposed optimization problem is verified through a case study comparing with results of a conventional optimization model that uses constant heat-to-power ratio and generation efficiency of the CHP resource. Furthermore, we evaluate planning schedules and total generation efficiency profiles of the CHP resource for the compared optimization models.
Multi-energy systems can integrate heat and electrical energy efficiently, using resources such as cogeneration. In order to meet energy demand cost-effectively in a multi-energy system, adopting appropriate energy resources at the right time is of great importance. In this paper, we propose an expansion planning method for a multi-energy system that supplies heat and electrical energy. The proposed approach formulates expansion planning as a mixed integer linear programming (MILP) problem. The objective is to minimize the sum of the annualized cost of the multi-energy system. The candidate resources that constitute the cost of the multi-energy system are fuel-based power generators, heat-only boilers, a combined heat and power (CHP) unit, energy storage resources, and a renewable electrical power source. We use a load-energy curve, instead of a load-duration curve, for constructing the optimization model, which is subsequently linearized using a Douglas-Peucker algorithm. The residual load-energy curve, for utilizing the renewable electrical power source, is also linearized. This study demonstrates the effectiveness of the proposed method through a comparison with a conventional linearization method. In addition, we evaluate the cost and planning schedules of different case studies, according to the configuration of resources in the multi-energy system.
This study proposes optimal day-ahead demand response (DR) participation strategies and distributed energy resource (DER) management in a residential building under an individual DR contract with a grid-system operator. First, this study introduces a DER management system in the residential building for participation to the day-ahead DR market. The distributed photovoltaic generation system (PV) and energy-storage system (ESS) are applied to reduce the electricity demand in the building and sell surplus energy on the grid. Among loads in the building, lighting (LTG) and heating, ventilation, and air conditioning (HVAC) loads are included in the DR program. In addition, it is assumed that a power management system of an electric vehicle (EV) charging station is integrated the DER management system. In order to describe stochastic behavior of EV owners, the uncertainty of EV is formulated based on their arrival and departure scenarios. For measuring the economic efficiency of the proposed model, we compare it with the DER self-consuming operation model without DR participation. The problem is solved using mixed integer linear programming to minimize the operating cost. The results in summer and winter are analyzed to evaluate the proposed algorithm’s validity. From these results, the proposed model can be confirmed as reducing operation cost compared to the reference model through optimal day-ahead DR capacity bidding and implementation.
As renewable energy penetration in power systems grows, adequate energy policies are needed to support the system’s operations with flexible resources and to adopt more sustainable energies. A peak-biased incentive for energy storage systems (ESS) using the Korean renewable portfolio standard could make power system operations more difficult. For the first time in the research, this study evaluates the effect of imposing a renewable energy certificate incentive in off-peak periods on mitigating wind power fluctuations. We design a coordinated model of a wind farm with an ESS to model the behavior of wind farm operators. Optimization problems are formulated as mixed integer linear programming problems to test the implementation of revenue models under Korean policy. These models are designed to consider additional incentives for discharging the ESS during off-peak periods. The effects of imposing the incentives on wind power fluctuations are evaluated using the magnitude of the renewable energy certificate (REC) multiplier.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.