Long Short-Term Renewable Energy Sources Prediction for Grid-Management Systems Based on Stacking Ensemble Model

Energies, Journal Year: 2024, Volume and Issue: 17(13), P. 3145 - 3145

Published: June 26, 2024

The transition towards sustainable energy systems necessitates effective management of renewable sources alongside conventional grid infrastructure. This paper presents a comprehensive approach to optimizing by integrating Photovoltaic (PV), wind, and energies minimize costs enhance sustainability. A key focus lies in developing an accurate scheduling algorithm utilizing Mixed Integer Programming (MIP), enabling dynamic allocation resources meet demand while minimizing reliance on cost-intensive energy. An ensemble learning technique, specifically stacking algorithm, is employed construct robust forecasting pipeline for PV wind generation. model achieves remarkable accuracy with Root Mean Squared Error (RMSE) less than 0.1 short-term (15 min one day ahead) long-term (one week month predictions. By combining optimization methodologies, this research contributes advancing capable harnessing efficiently, thus facilitating cost savings fostering sustainability the sector.

Language: Английский

---

Short-Term Forecasts of Energy Generation in a Solar Power Plant Using Various Machine Learning Models, along with Ensemble and Hybrid Methods

Energies, Journal Year: 2024, Volume and Issue: 17(17), P. 4234 - 4234

Published: Aug. 24, 2024

High-quality short-term forecasts of electrical energy generation in solar power plants are crucial the dynamically developing sector renewable generation. This article addresses issue selecting appropriate (preferred) methods for forecasting from a plant within 15 min time horizon. The effectiveness various machine learning was verified. Additionally, proprietary ensemble and hybrid proposed examined. research also aimed to determine sets input variables predictive models. To enhance performance models, additional (feature engineering) were constructed. significance individual examined depending on model used. concludes with findings recommendations regarding preferred methods.

Language: Английский

---

Day-Ahead Operational Planning for DisCos Based on Demand Response Flexibility and Volt/Var Control

Energies, Journal Year: 2023, Volume and Issue: 16(20), P. 7045 - 7045

Published: Oct. 11, 2023

Considering the integration of distributed energy resources (DER) such as generation, demand response, and electric vehicles, day-ahead scheduling plays a significant role in operation active distribution systems. Therefore, this article proposes comprehensive methodology for short-term operational planning company (DisCo), aiming to minimize total daily cost. The proposed integrates on-load tap changers, capacitor banks, flexible loads participating response (DR) reduce losses manage congestion voltage violations, while considering costs associated with use controllable resources. Furthermore, forecast PV output load behind meter at MV/LV transformer level, net forecasting model using deep learning techniques has been incorporated. scheme is solved through an efficient two-stage strategy based on genetic algorithms dynamic programming. Numerical results modified IEEE 13-node system typical 37-node Latin American validate effectiveness methodology. obtained verify that, methodology, DisCo can effectively schedule its installations DR cost reducing robustly managing issues.

Language: Английский

---