State-of-the-art review on energy and load forecasting in microgrids using artificial neural networks, machine learning, and deep learning techniques

Electric Power Systems Research, Год журнала: 2023, Номер 225, С. 109792 - 109792

Опубликована: Сен. 8, 2023

Язык: Английский

---

Machine Learning Based PV Power Generation Forecasting in Alice Springs

IEEE Access, Год журнала: 2021, Номер 9, С. 46117 - 46128

Опубликована: Янв. 1, 2021

The generation volatility of photovoltaics (PVs) has created several control and operation challenges for grid operators. For a secure reliable day or hour-ahead electricity dispatch, the operators need visibility their synchronous asynchronous generators' capacity. It helps them to manage spinning reserve, inertia frequency response during any contingency events. This study attempts provide machine learning-based PV power forecasting both short long-term. chosen Alice Springs, one geographically solar energy-rich areas in Australia, considered various environmental parameters. Different learning algorithms, including Linear Regression, Polynomial Decision Tree Support Vector Random Forest Long Short-Term Memory, Multilayer Perceptron are study. Various comparative performance analysis is conducted normal uncertain cases found that Regression performed better our dataset. impact data normalization on also analyzed using multiple metrics. may help choose an appropriate algorithm plan time-ahead volatility.

Язык: Английский

---

SolarNet: A hybrid reliable model based on convolutional neural network and variational mode decomposition for hourly photovoltaic power forecasting

Applied Energy, Год журнала: 2021, Номер 300, С. 117410 - 117410

Опубликована: Июль 15, 2021

Язык: Английский

---

Implementation of artificial intelligence techniques in microgrid control environment: Current progress and future scopes

Energy and AI, Год журнала: 2022, Номер 8, С. 100147 - 100147

Опубликована: Март 2, 2022

Microgrids are gaining popularity by facilitating distributed energy resources (DERs) and forming essential consumer/prosumer centric integrated systems. Integration, coordination control of multiple DERs managing the transition in this environment is a strenuous task. Classical techniques not enough to support dynamic microgrid environments. Implementation Artificial Intelligence (AI) seems be promising solution enhance operation microgrids future smart grid networks. Therefore, paper briefly reviews architectures, existing conventional controlling techniques, their drawbacks, need for intelligent controllers then extensively possibility AI implementation different structures with special focus on hierarchical layers. This also investigates AI-based strategies networked/interconnected/multi-microgrids It concludes summary scopes layers including single networked

Язык: Английский

---

A novel long term solar photovoltaic power forecasting approach using LSTM with Nadam optimizer: A case study of India

Energy Science & Engineering, Год журнала: 2022, Номер 10(8), С. 2909 - 2929

Опубликована: Май 11, 2022

Abstract Solar photovoltaic (PV) power is emerging as one of the most viable renewable energy sources. The recent enhancements in integration sources into grid create a dire need for reliable solar forecasting techniques. In this paper, new long‐term PV approach using long short‐term memory (LSTM) model with Nadam optimizer presented. LSTM performs better time‐series data it persists information more time steps. experimental models are realized on 250.25 kW installed capacity system located at MANIT Bhopal, Madhya Pradesh, India. proposed compared two and eight neural network different optimizers. obtained results present significant improvement accuracy 30.56% over autoregressive integrated moving average, 47.48% seasonal 1.35%, 1.43%, 3.51%, 4.88%, 11.84%, 50.69%, 58.29% RMSprop, Adam, Adamax, SGD, Adagrad, Adadelta, Ftrl optimizer, respectively. prove that methodology conclusive can be employed enhanced planning management.

Язык: Английский

---

An Hour-Ahead PV Power Forecasting Method Based on an RNN-LSTM Model for Three Different PV Plants

Energies, Год журнала: 2022, Номер 15(6), С. 2243 - 2243

Опубликована: Март 18, 2022

Incorporating solar energy into a grid necessitates an accurate power production forecast for photovoltaic (PV) facilities. In this research, output PV was predicted at hour ahead on yearly basis three different plants based polycrystalline (p-si), monocrystalline (m-si), and thin-film (a-si) technologies over four-year period. Wind speed, module temperature, ambiance, irradiation were among the input characteristics taken account. Each plant parameter. A deep learning method (RNN-LSTM) developed evaluated against existing techniques to of selected plant. The proposed technique compared with regression (GPR, GPR (PCA)), hybrid ANFIS (grid partitioning, subtractive clustering FCM) machine (ANN, SVR, SVR (PCA)) methods. Furthermore, LSTM structures also investigated, recurrent neural networks (RNN) 2019 data determine best structure. following parameters prediction accuracy measure considered: RMSE, MSE, MAE, correlation (r) determination (R2) coefficients. comparison all other approaches, RNN-LSTM had higher minimum (RMSE MSE) maximum (r R2). p-si, m-si a-si showed lowest RMSE values 26.85 W/m2, 19.78 W/m2 39.2 respectively. Moreover, found be robust flexible in forecasting considered plants.

Язык: Английский

---

Short-term photovoltaic power forecasting using meta-learning and numerical weather prediction independent Long Short-Term Memory models

Renewable Energy, Год журнала: 2023, Номер 216, С. 118997 - 118997

Опубликована: Июль 13, 2023

Short-term photovoltaic (PV) power forecasting is essential for integrating renewable energy sources into the grid as it provides accurate and timely information on expected output of PV systems. Deep learning (DL) networks have shown promising results in this area, but depending weather conditions particularities each system, different DL architectures may perform best. This paper proposes a meta-learning method to improve one-hour-ahead deterministic forecasts systems by dynamically blending base multiple models learn under what model performs Four long short-term memory are used produce production without using numerical predictions, with objective enhance generalizability proposed solution. The accuracy meta-learner evaluated three rooftop Lisbon, Portugal. Results indicate that best at plants, can up 5% over most per plant 4.5% equal-weighted combination forecasts. These improvements statistically significant even larger during peak hours.

Язык: Английский

---

A bibliometric analysis of machine learning techniques in photovoltaic cells and solar energy (2014–2022)

Energy Reports, Год журнала: 2024, Номер 11, С. 2768 - 2779

Опубликована: Фев. 22, 2024

Solar energy presents a promising solution to replace fossil-based sources, mitigating global warming and climate change. However, solar faces socio-economic, environmental, technical challenges. Computational tools like machine learning offer solutions these Despite numerous studies, there's lack of comprehensive research on ML applications in Photovoltaics Energy. This study conducts critical analysis Energy using publication trends bibliometric analysis, employing the PRISMA approach Scopus database. Results reveal high output, citations, international collaboration. Notable researchers include G. E. Georghiou Haibo Ma, with Ministry Education (China) being prolific affiliation. China emerges as most active nation due funding programs National Natural Science Foundation Key Research Development Program. contributes terms providing an patterns from 2014 2022, including topic categories important metrics, at levels country, institution, organisation. Analysing author-keyword data aggregate publishing themes identify influential journals. Enhancing comprehension hotspots focal points research. also aims discuss role Cognitive Computing cancer/tumor oncological research, emphasising potential for significant advancements obstacles that need be overcome order fully utilise its advantages. Future studies could extensive into cybersecurity systems particularly wake malware, phishing, other intrusion attacks grid infrastructure worldwide.

Язык: Английский

---

Energy 4.0: AI-enabled digital transformation for sustainable power networks

Computers & Industrial Engineering, Год журнала: 2024, Номер 193, С. 110253 - 110253

Опубликована: Май 24, 2024

Язык: Английский

---

A cohesive structure of Bi-directional long-short-term memory (BiLSTM) -GRU for predicting hourly solar radiation

Renewable Energy, Год журнала: 2024, Номер 222, С. 119943 - 119943

Опубликована: Янв. 2, 2024

Язык: Английский

---