Read-First LSTM model: A new variant of long short term memory neural network for predicting solar radiation data

Energy Conversion and Management, Journal Year: 2024, Volume and Issue: 305, P. 118267 - 118267

Published: March 7, 2024

Language: Английский

---

Improving Solar PV Prediction Performance with RF-CatBoost Ensemble: A Robust and Complementary Approach

Renewable energy focus, Journal Year: 2023, Volume and Issue: 46, P. 207 - 221

Published: June 28, 2023

Language: Английский

---

A Performance Comparison Study on Climate Prediction in Weifang City Using Different Deep Learning Models

Water, Journal Year: 2024, Volume and Issue: 16(19), P. 2870 - 2870

Published: Oct. 9, 2024

Climate change affects the water cycle, resource management, and sustainable socio-economic development. In order to accurately predict climate in Weifang City, China, this study utilizes multiple data-driven deep learning models. The data for 73 years include monthly average air temperature (MAAT), minimum (MAMINAT), maximum (MAMAXAT), total precipitation (MP). different models artificial neural network (ANN), recurrent NN (RNN), gate unit (GRU), long short-term memory (LSTM), convolutional (CNN), hybrid CNN-GRU, CNN-LSTM, CNN-LSTM-GRU. CNN-LSTM-GRU MAAT prediction is best-performing model compared other with highest correlation coefficient (R = 0.9879) lowest root mean square error (RMSE 1.5347) absolute (MAE 1.1830). These results indicate that method a suitable model. This can also be used surface modeling. will help flood control management.

Language: Английский

---

Enhancing solar photovoltaic energy production prediction using diverse machine learning models tuned with the chimp optimization algorithm

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

Published: Aug. 10, 2024

Solar photovoltaic (PV) systems, integral for sustainable energy, face challenges in forecasting due to the unpredictable nature of environmental factors influencing energy output. This study explores five distinct machine learning (ML) models which are built and compared predict production based on four independent weather variables: wind speed, relative humidity, ambient temperature, solar irradiation. The evaluated include multiple linear regression (MLR), decision tree (DTR), random forest (RFR), support vector (SVR), multi-layer perceptron (MLP). These were hyperparameter tuned using chimp optimization algorithm (ChOA) a performance appraisal. subsequently validated data from 264 kWp PV system, installed at Applied Science University (ASU) Amman, Jordan. Of all 5 models, MLP shows best root mean square error (RMSE), with corresponding value 0.503, followed by absolute (MAE) 0.397 coefficient determination (R

Language: Английский

---

An innovative machine learning based on feed-forward artificial neural network and equilibrium optimization for predicting solar irradiance

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

Published: Jan. 25, 2024

Abstract As is known, having a reliable analysis of energy sources an important task toward sustainable development. Solar one the most advantageous types renewable energy. Compared to fossil fuels, it cleaner, freely available, and can be directly exploited for electricity. Therefore, this study concerned with suggesting novel hybrid models improving forecast Irradiance (I S ). First, predictive model, namely Feed-Forward Artificial Neural Network (FFANN) forms non-linear contribution between I dominant meteorological temporal parameters (including humidity, temperature, pressure, cloud coverage, speed direction wind, month, day, hour). Then, framework optimized using several metaheuristic algorithms create predicting . According accuracy assessments, attained satisfying training FFANN by 80% data. Moreover, applying trained remaining 20% proved their high proficiency in forecasting unseen environmental circumstances. A comparison among optimizers revealed that Equilibrium Optimization (EO) could achieve higher than Wind-Driven (WDO), Optics Inspired (OIO), Social Spider Algorithm (SOSA). In another phase study, Principal Component Analysis (PCA) applied identify contributive factors. The PCA results used optimize problem dimension, as well suggest effective real-world measures solar production. Lastly, EO-based solution yielded form explicit formula more convenient estimation

Language: Английский

---

Long-Short Term Memory Technique for Monthly Rainfall Prediction in Thale Sap Songkhla River Basin, Thailand

Symmetry, Journal Year: 2022, Volume and Issue: 14(8), P. 1599 - 1599

Published: Aug. 3, 2022

Rainfall is a primary factor for agricultural production, especially in rainfed region. Its accurate prediction therefore vital planning and managing farmers’ plantations. plays an important role the symmetry of water cycle, many hydrological models use rainfall as one their components. This paper aimed to investigate applicability six machine learning (ML) techniques (i.e., M5 model tree: (M5), random forest: (RF), support vector regression with polynomial (SVR-poly) RBF kernels (SVR- RBF), multilayer perceptron (MLP), long-short-term memory (LSTM) predicting multiple-month ahead monthly rainfall. The experiment was set up two weather gauged stations located Thale Sap Songkhla basin. development carried out by (1) selecting input variables, (2) tuning hyperparameters, (3) investigating influence climate variables on prediction, (4) multi-step-ahead prediction. Four statistical indicators including correlation coefficient (r), mean absolute error (MAE), root square (RMSE), overall index (OI) were used assess model’s effectiveness. results revealed that large-scale particularly sea surface temperature, significant tropical For projections basin whole, LSTM provided highest performance both stations. developed predictive rain acceptable performance: r (0.74), MAE (86.31 mm), RMSE (129.11 OI (0.70) 1 month ahead, (0.72), (91.39 (133.66 (0.68) 2 months (0.70), (94.17 (137.22 (0.66) 3 ahead.

Language: Английский

---

Solar irradiation prediction using empirical and artificial intelligence methods: A comparative review

Heliyon, Journal Year: 2023, Volume and Issue: 9(6), P. e17038 - e17038

Published: June 1, 2023

Solar irradiation data is essential for the feasibility of solar energy projects. Notably, intermittent nature influences use in all forms, whether or agriculture. Accurate prediction only solution to effectively different forms. The estimation most critical factor site selection and sizing projects selecting a suitable crop area. But physical measurement irradiation, due cost technology involved, not possible locations across globe. Numerous techniques have been implemented predict this purpose. two types approaches that are frequently employed empirical artificial intelligence (AI). Both demonstrated good accuracy various places world. To find out best method, thorough review research articles discussing has done compare methods prediction. In paper, predicting using AI published from 2017 2022 reviewed, both compared. showed more accurate than methods. models, modified sunshine-based models (MSSM) highest accuracy, followed by (SSM) non-sunshine-based (NSM). NSM little lower MSSM SSM, but can give results sunshine unavailability. Also, literature confirmed simple could accurately, increasing model's polynomial order cannot improve results. Artificial neural networks (ANN) Hybrid among methods, support vector machine (SVM) adaptive neuro-fuzzy inference system (ANFIS). increase efficiency hybrid minimal, complexity requires very sophisticated programming knowledge. ANN's important input factors maximum minimum temperatures, temperature differential, relative humidity, clearness index precipitation.

Language: Английский

---

Daily solar radiation estimation in Belleville station, Illinois, using ensemble artificial intelligence approaches

Engineering Applications of Artificial Intelligence, Journal Year: 2023, Volume and Issue: 120, P. 105839 - 105839

Published: Jan. 14, 2023

Language: Английский

---

Prediction of Pea (Pisum sativum L.) Seeds Yield Using Artificial Neural Networks

Agriculture, Journal Year: 2023, Volume and Issue: 13(3), P. 661 - 661

Published: March 12, 2023

A sufficiently early and accurate prediction can help to steer crop yields more consciously, resulting in food security, especially with an expanding world population. Additionally, related the possibility of reducing agricultural chemistry is very important era climate change. This study analyzes performance pea (Pisum sativum L.) seed yield by a linear (MLR) non-linear (ANN) model. The used meteorological, agronomic phytophysical data from 2016–2020. neural model (N2) generated highly predictions yield—the correlation coefficient was 0.936, RMS MAPE errors were 0.443 7.976, respectively. significantly outperformed multiple regression (RS2), which had error 6.401 148.585. sensitivity analysis carried out for network showed that characteristics greatest influence on seeds date onset maturity, harvest, total amount rainfall mean air temperature.

Language: Английский

---

Adaptive Stochastic Conjugate Gradient Optimization for Backpropagation Neural Networks

IEEE Access, Journal Year: 2024, Volume and Issue: 12, P. 33757 - 33768

Published: Jan. 1, 2024

Backpropagation neural networks are commonly utilized to solve complicated issues in various disciplines. However, optimizing their settings remains a significant task. Traditional gradient-based optimization methods, such as stochastic gradient descent (SGD), often exhibit slow convergence and hyperparameter sensitivity. An adaptive conjugate (ASCG) strategy for backpropagation is proposed this research. ASCG combines the advantages of techniques increase training efficiency speed. Based on observed gradients, algorithm adaptively calculates learning rate search direction at each iteration, allowing quicker greater generalization. Experimental findings benchmark datasets show that outperforms standard regarding time model performance. The provides viable method improving process networks, making them more successful tackling problems across several domains. As result, information initial seeds formed while being trained grows. coordinated efforts ASCG's Conjugate Gradient components improve achieve global minima. Our results indicate our achieves 21 percent higher accuracy HMT dataset performs better than existing methods other datasets(DIR-Lab dataset). experimentation revealed has an 95 when utilizing principal component analysis features, compared 94 using correlation heatmap features selection approach with MSE 0.0678.

Language: Английский

---