Machine learning-based energy management and power forecasting in grid-connected microgrids with multiple distributed energy sources

Scientific Reports, Journal Year: 2024, Volume and Issue: 14(1)

Published: Aug. 19, 2024

The growing integration of renewable energy sources into grid-connected microgrids has created new challenges in power generation forecasting and management. This paper explores the use advanced machine learning algorithms, specifically Support Vector Regression (SVR), to enhance efficiency reliability these systems. proposed SVR algorithm leverages comprehensive historical production data, detailed weather patterns, dynamic grid conditions accurately forecast generation. Our model demonstrated significantly lower error metrics compared traditional linear regression models, achieving a Mean Squared Error 2.002 for solar PV 3.059 wind forecasting. Absolute was reduced 0.547 0.825 scenarios, Root (RMSE) 1.415 1.749 power, showcasing model's superior accuracy. Enhanced predictive accuracy directly contributes optimized resource allocation, enabling more precise control schedules reducing reliance on external sources. application our resulted an 8.4% reduction overall operating costs, highlighting its effectiveness improving management efficiency. Furthermore, system's ability predict fluctuations output allowed adaptive real-time management, stress enhancing system stability. approach led 10% improvement balance between supply demand, 15% peak load 12% increase utilization enhances stability by better balancing mitigating variability intermittency These advancements promote sustainable microgrid, contributing cleaner, resilient, efficient infrastructure. findings this research provide valuable insights development intelligent systems capable adapting changing conditions, paving way future innovations Additionally, work underscores potential revolutionize practices providing accurate, reliable, cost-effective solutions integrating existing infrastructures.

Language: Английский

---

Advanced Automated Machine Learning Framework for Photovoltaic Power Output Prediction Using Environmental Parameters and SHAP Interpretability

Results in Engineering, Journal Year: 2025, Volume and Issue: unknown, P. 103838 - 103838

Published: Jan. 1, 2025

Language: Английский

---

Enhancing Solar Energy Conversion Efficiency: Thermophysical Property Predicting of MXene/Graphene Hybrid Nanofluids via Bayesian-Optimized Artificial Neural Networks

Results in Engineering, Journal Year: 2024, Volume and Issue: 24, P. 102858 - 102858

Published: Sept. 7, 2024

Language: Английский

---

Short-Medium-Term Solar Irradiance Forecasting with a CEEMDAN-CNN-ATT-LSTM Hybrid Model Using Meteorological Data

Applied Sciences, Journal Year: 2025, Volume and Issue: 15(3), P. 1275 - 1275

Published: Jan. 26, 2025

In recent years, the adverse effects of climate change have increased rapidly worldwide, driving countries to transition clean energy sources such as solar and wind. However, these energies face challenges cloud cover, precipitation, wind speed, temperature, which introduce variability intermittency in power generation, making integration into interconnected grid difficult. To achieve this, we present a novel hybrid deep learning model, CEEMDAN-CNN-ATT-LSTM, for short- medium-term irradiance prediction. The model utilizes complete empirical ensemble modal decomposition with adaptive noise (CEEMDAN) extract intrinsic seasonal patterns irradiance. addition, it employs encoder-decoder framework that combines convolutional neural networks (CNN) capture spatial relationships between variables, an attention mechanism (ATT) identify long-term patterns, long short-term memory (LSTM) network dependencies time series data. This has been validated using meteorological data more than 2400 masl region characterized by complex climatic conditions south Ecuador. It was able predict at 1, 6, 12 h horizons, mean absolute error (MAE) 99.89 W/m2 winter 110.13 summer, outperforming reference methods this study. These results demonstrate our represents progress contributing scientific community field environments high its applicability real scenarios.

Language: Английский

---

A review of PV power forecasting using machine learning techniques

Deleted Journal, Journal Year: 2025, Volume and Issue: unknown, P. 100058 - 100058

Published: Jan. 1, 2025

Language: Английский

---

Explainable ensemble learning framework for estimating corrosion rate in suspension bridge main cables

Results in Engineering, Journal Year: 2024, Volume and Issue: 23, P. 102723 - 102723

Published: Aug. 13, 2024

Language: Английский

---

Deep learning model for solar and wind energy forecasting considering Northwest China as an example

Results in Engineering, Journal Year: 2024, Volume and Issue: unknown, P. 102939 - 102939

Published: Sept. 1, 2024

Language: Английский

---

Explicit and Explainable Artificial Intelligent Model for Prediction of CO2 Molecular Diffusion Coefficient in Heavy Crude Oils and Bitumen

Results in Engineering, Journal Year: 2024, Volume and Issue: 24, P. 103328 - 103328

Published: Nov. 6, 2024

Language: Английский

---

Ultra-short-term global horizontal irradiance forecasting based on a novel and hybrid GRU-TCN model

Results in Engineering, Journal Year: 2024, Volume and Issue: 23, P. 102817 - 102817

Published: Sept. 1, 2024

The need for energy is increasing globally due to a several factors, including population growth and economic development. Achieving this demand in the face of global warming depletion fossil fuels requires use renewable energy. Photovoltaic one sources that widely used many nations across world. (PV) integration into grid has significant benefits environment economy, but at high penetration levels, its intermittent nature makes system stability difficult maintain. Accurate ultra-short-term horizontal irradiance forecasting necessary order guarantee most optimal photovoltaic power production sources. For GHI forecasting, novel GRU-TCN-based model proposed paper. It composed two neural networks: temporal convolutional network gated recurrent unit. After extracting features from time-series solar data using GRU, spatial are obtained correlation matrix different meteorological variables target neighbor position TCN. Univariate multivariate GRU-TCN models have been ultra short-term forecasting. This paper compares univariate with TCN, LSTM, GRU based on three evaluation metrics investigate how combinations affect accuracy one-step findings indicate adoption historical suitable obtain reliable 23.02 (W/m2) MAE as opposed best achieved 25.67 MAE. According results, outperforms other assessed offers practical alternative

Language: Английский

---

Real-Time Ultra Short-Term Irradiance Forecasting Using a Novel R-GRU Model for Optimizing PV Controller Dynamics

Results in Engineering, Journal Year: 2025, Volume and Issue: unknown, P. 105046 - 105046

Published: April 1, 2025

Language: Английский

---