Artificial Neural Networks for Photovoltaic Power Forecasting: A Review of Five Promising Models

IEEE Access, Год журнала: 2024, Номер 12, С. 90461 - 90485

Опубликована: Янв. 1, 2024

Solar energy is largely dependent on weather conditions, resulting in unpredictable, fluctuating, and unstable photovoltaic (PV) power outputs. Thus, accurate PV forecasts are increasingly crucial for managing controlling integrated systems. Over the years, advanced artificial neural network (ANN) models have been proposed to increase accuracy of various geographical regions. Hence, this paper provides a state-of-the-art review five most popular ANN forecasting. These include multilayer perceptron (MLP), recurrent (RNN), long short-term memory (LSTM), gated unit (GRU), convolutional (CNN). First, internal structure operation these studied. It then followed by brief discussion main factors affecting their forecasting accuracy, including horizons, meteorological evaluation metrics. Next, an in-depth separate analysis standalone hybrid provided. has determined that bidirectional GRU LSTM offer greater whether used as model or configuration. Furthermore, upgraded metaheuristic algorithms demonstrated exceptional performance when applied models. Finally, study discusses limitations shortcomings may influence practical implementation

Язык: Английский

---

AI-Driven Innovations in Building Energy Management Systems: A Review of Potential Applications and Energy Savings

Energies, Год журнала: 2024, Номер 17(17), С. 4277 - 4277

Опубликована: Авг. 27, 2024

Despite the tightening of energy performance standards for buildings in various countries and increased use efficient renewable technologies, it is clear that sector needs to change more rapidly meet Net Zero Emissions (NZE) scenario by 2050. One problems have been analyzed intensively recent years operation much than they were designed to. This problem, known as gap, found many often attributed poor management building systems. The application Artificial Intelligence (AI) Building Energy Management Systems (BEMS) has untapped potential address this problem lead sustainable buildings. paper reviews different AI-based models proposed applications with intention reduce consumption. It compares evaluated reviewed papers presenting accuracy error rates model identifies where greatest savings could be achieved, what extent. review showed offices (up 37%) when employ AI HVAC control optimization. In residential educational buildings, lower intelligence existing BEMS results smaller 23% 21%, respectively).

Язык: Английский

---

A Multi-Step-Ahead Photovoltaic Power Forecasting Approach Using One-Dimensional Convolutional Neural Networks and Transformer

Electronics, Год журнала: 2024, Номер 13(11), С. 2007 - 2007

Опубликована: Май 21, 2024

Due to environmental concerns about the use of fossil fuels, renewable energy, especially solar is increasingly sought after for its ease installation, cost-effectiveness, and versatile capacity. However, variability in factors poses a significant challenge photovoltaic (PV) power generation forecasting, which crucial maintaining system stability economic efficiency. In this paper, novel muti-step-ahead PV forecasting model by integrating single-step multi-step forecasts from various time resolutions was developed. One-dimensional convolutional neural network (CNN) layers were used capture specific temporal patterns, with transformer improving leveraging combined outputs CNN. This combination can provide accurate immediate as well ability identify longer-term trends. Using DKASC-ASA-1A 1B datasets empirical validation, several preprocessing methods applied series experiments conducted compare performance other widely deep learning models. The framework proved be capable accurately predicting multi-step-ahead at multiple resolutions.

Язык: Английский

---

SolarFlux Predictor: A Novel Deep Learning Approach for Photovoltaic Power Forecasting in South Korea

Electronics, Год журнала: 2024, Номер 13(11), С. 2071 - 2071

Опубликована: Май 27, 2024

We present SolarFlux Predictor, a novel deep-learning model designed to revolutionize photovoltaic (PV) power forecasting in South Korea. This uses self-attention-based temporal convolutional network (TCN) process and predict PV outputs with high precision. perform meticulous data preprocessing ensure accurate normalization outlier rectification, which are vital for reliable analysis. The TCN layers crucial capturing patterns energy data; we complement them the teacher forcing technique during training phase significantly enhance sequence prediction accuracy. By optimizing hyperparameters Optuna, further improve model’s performance. Our incorporates multi-head self-attention mechanisms focus on most impactful features, thereby improving In validations against datasets from nine regions Korea, outperformed conventional methods. results indicate that is robust tool systems’ management operational efficiency can contribute Korea’s pursuit of sustainable solutions.

Язык: Английский

---

Deep learning-based optimization of energy utilization in IoT-enabled smart cities: A pathway to sustainable development

Energy Reports, Год журнала: 2024, Номер 12, С. 2946 - 2957

Опубликована: Сен. 6, 2024

Язык: Английский

---

MultiFuseYOLO: Redefining Wine Grape Variety Recognition through Multisource Information Fusion

Sensors, Год журнала: 2024, Номер 24(9), С. 2953 - 2953

Опубликована: Май 6, 2024

Based on the current research wine grape variety recognition task, it has been found that traditional deep learning models relying only a single feature (e.g., fruit or leaf) for classification can face great challenges, especially when there is high degree of similarity between varieties. In order to effectively distinguish these similar varieties, this study proposes multisource information fusion method, which centered SynthDiscrim algorithm, aiming achieve more comprehensive and accurate recognition. First, optimizes improves YOLOV7 model novel target detection called WineYOLO-RAFusion, significantly localization precision compared with YOLOV5, YOLOX, YOLOV7, are models. Secondly, building upon WineYOLO-RAFusion model, incorporated method into ultimately forming MultiFuseYOLO model. Experiments demonstrated outperformed other commonly used in terms precision, recall, F1 score, reaching 0.854, 0.815, 0.833, respectively. Moreover, improved hard Chardonnay Sauvignon Blanc increased from 0.512 0.813 0.533 0.775 Blanc. conclusion, offers reliable solution task identification, distinguishing visually varieties realizing high-precision identifications.

Язык: Английский

---

Advancing ensemble learning techniques for residential building electricity consumption forecasting: Insight from explainable artificial intelligence

PLoS ONE, Год журнала: 2024, Номер 19(11), С. e0307654 - e0307654

Опубликована: Ноя. 14, 2024

Accurate electricity consumption forecasting in residential buildings has a direct impact on energy efficiency and cost management, making it critical component of sustainable practices. Decision tree-based ensemble learning techniques are particularly effective for this task due to their ability process complex datasets with high accuracy. Furthermore, incorporating explainable artificial intelligence into these predictions provides clarity interpretability, allowing managers homeowners make informed decisions that optimize usage reduce costs. This study comparatively analyzes decision tree–ensemble augmented transparency interpretability building forecasting. approach employs the University Residential Complex Appliances Energy Prediction datasets, data preprocessing, decision-tree bagging boosting methods. The superior model is evaluated using Shapley additive explanations method within framework, explaining influence input variables decision-making processes. analysis reveals significant temperature-humidity index wind chill temperature short-term load forecasting, transcending traditional parameters, such as temperature, humidity, speed. complete source code have been made available our GitHub repository at https://github.com/sodayeong purpose enhancing precision system thereby promoting enabling replication.

Язык: Английский

---

A review of the influencing factors of building energy consumption and the prediction and optimization of energy consumption

AIMS energy, Год журнала: 2025, Номер 13(1), С. 35 - 85

Опубликована: Янв. 1, 2025

<p>Concomitant with the expeditious growth of construction industry, challenge building energy consumption has become increasingly pronounced. A multitude factors influence operations, thereby underscoring paramount importance monitoring and predicting such consumption. The advent big data engendered a diversification in methodologies employed to predict Against backdrop influencing operation consumption, we reviewed advancements research pertaining supervision prediction deliberated on more energy-efficient low-carbon strategies for buildings within dual-carbon context, synthesized relevant progress across four dimensions: contemporary state supervision, determinants optimization Building upon investigation three predictive were examined: (ⅰ) Physical methods, (ⅱ) data-driven (ⅲ) mixed methods. An analysis accuracy these revealed that methods exhibited superior precision actual Furthermore, predicated this foundation identified determinants, also explored prediction. Through an in-depth examination prediction, distilled pertinent accurate forecasting offering insights guidance pursuit conservation emission reduction.</p>

Язык: Английский

---

SolarNexus: A deep learning framework for adaptive photovoltaic power generation forecasting and scalable management

Applied Energy, Год журнала: 2025, Номер 391, С. 125848 - 125848

Опубликована: Апрель 11, 2025

Язык: Английский

---

A Meta-Survey on Intelligent Energy-Efficient Buildings

Big Data and Cognitive Computing, Год журнала: 2024, Номер 8(8), С. 83 - 83

Опубликована: Июль 30, 2024

The rise of the Internet Things (IoT) has enabled development smart cities, intelligent buildings, and advanced industrial ecosystems. When IoT is matched with machine learning (ML), advantages resulting enhanced environments can span, for example, from energy optimization to security improvement comfort enhancement. Together, ML technologies are widely used in particular, reduce consumption create Intelligent Energy-Efficient Buildings (IEEBs). In IEEBs, models typically analyze predict various factors such as temperature, humidity, light, occupancy, human behavior aim optimizing building systems. literature, many review papers have been presented so far field IEEBs. Such mostly focus on specific subfields or a limited number papers. This paper presents systematic meta-survey, i.e., articles, that compares state art IEEBs using Prisma approach. more detail, our meta-survey aims give broader view, respect already published surveys, state-of-the-art IEEB field, investigating use supervised, unsupervised, semi-supervised, self-supervised variety IEEB-based scenarios. Moreover, compare surveys by answering five important research questions about definitions, architectures, methods/models used, datasets real implementations utilized, main challenges/research directions defined. provides insights useful both newcomers researchers who want learn methodologies IEEBs’ design implementation.

Язык: Английский

---