Wind speed short-term prediction using recurrent neural network GRU model and stationary wavelet transform GRU hybrid model

Energy Conversion and Management, Год журнала: 2024, Номер 308, С. 118333 - 118333

Опубликована: Апрель 11, 2024

Язык: Английский

---

Data driven net load uncertainty quantification for cloud energy storage management in residential microgrid

Electric Power Systems Research, Год журнала: 2023, Номер 226, С. 109920 - 109920

Опубликована: Окт. 14, 2023

Язык: Английский

---

Artificial Intelligence aided pharmaceutical engineering: Development of hybrid machine learning models for prediction of nanomedicine solubility in supercritical solvent

Journal of Molecular Liquids, Год журнала: 2024, Номер 397, С. 124127 - 124127

Опубликована: Янв. 26, 2024

Язык: Английский

---

A fuzzy time series forecasting model with both accuracy and interpretability is used to forecast wind power

Applied Energy, Год журнала: 2023, Номер 353, С. 122015 - 122015

Опубликована: Окт. 4, 2023

Язык: Английский

---

Short-term wind power forecasting based on multi-scale receptive field-mixer and conditional mixture copula

Applied Soft Computing, Год журнала: 2024, Номер 164, С. 112007 - 112007

Опубликована: Июль 17, 2024

Язык: Английский

---

Implementing and tuning machine learning-based models for description of solubility variations of nanomedicine in supercritical solvent for development of green processing

Case Studies in Thermal Engineering, Год журнала: 2023, Номер 49, С. 103200 - 103200

Опубликована: Июнь 17, 2023

Determination of drug solubility in supercritical solvents such as CO2 has been great importance for preparation nanomedicines. This study implements and tunes several machine learning models to describe the medicine density solvent at various pressure temperature. The dataset used this consisted input variables, temperature, pressure. methods AdaBoost algorithm boost performance base regression predicting mole fractions rivaroxaban SC-CO2 were developed. include Theil-Sen Regression (TSR), Gaussian Process (GPR), Automatic Relevance (ARD), Linear (LR). We employ Hunter-Prey Optimization technique tune hyper-parameters these models. results indicated that boosted outperform their counterparts. For fraction predictions, with ARD achieves an R2 value 0.95986, while GPR obtains score 0.99817. impressive 0.99906. Accordingly, is best model both outputs. These demonstrate strength enhancing predictive accuracy chemical properties.

Язык: Английский

---

Futuristic Streamflow Prediction Based on CMIP6 Scenarios Using Machine Learning Models

Water Resources Management, Год журнала: 2023, Номер 37(15), С. 6089 - 6106

Опубликована: Окт. 26, 2023

Язык: Английский

---

An integrated development environment based situational awareness for operational reliability evaluation in wind energy systems incorporating uncertainties

Electric Power Systems Research, Год журнала: 2024, Номер 233, С. 110467 - 110467

Опубликована: Май 17, 2024

Язык: Английский

---

Wind power data cleaning using RANSAC-based polynomial and linear regression with adaptive threshold

Scientific Reports, Год журнала: 2025, Номер 15(1)

Опубликована: Фев. 11, 2025

As the global demand for clean energy continues to rise, wind power has become one of most important renewable sources. However, data often contains a high proportion dense anomalies, which not only significantly affect accuracy forecasting models but may also mislead grid scheduling decisions, thereby jeopardizing security. To address this issue, paper proposes an adaptive threshold robust regression model (RPR model) based on combination Random Sample Consensus (RANSAC) algorithm and polynomial linear cleaning. The successfully captures nonlinear relationship between speed by extending features power, enabling handle nonlinearity. By combining RANSAC regression, is constructed tackle anomalous enhance During cleaning process, first fits raw randomly selecting minimal sample set, then dynamically adjusts decision thresholds median residuals absolute deviation (MAD), ensuring effective identification data. model's robustness allows it maintain efficient performance even with data, addressing limitations existing methods when handling densely distributed anomalies. effectiveness innovation proposed method were validated applying real from farm operated Longyuan Power. Compared other commonly used methods, such as Bidirectional Change Point Grouping Quartile Statistical Model, Principal Contour Image Processing DBSCAN Clustering Support Vector Machine (SVM) experimental results showed that delivered best in improving quality. Specifically, reduced average error (MAE) 72.1%, higher than reductions observed (ranging 37.3 52.7%). Moreover, effectively prediction Convolutional Neural Network (CNN) + Gated Recurrent Unit (GRU) model, accuracy. study innovative significant application potential. It provides new approach cleaning, applicable conventional scenarios low proportions complex datasets

Язык: Английский

---

Optimization of solar and wind power plants production through a parallel fusion approach with modified hybrid machine and deep learning models

Intelligent Data Analysis, Год журнала: 2025, Номер unknown

Опубликована: Фев. 16, 2025

Artificial Intelligence (AI) is becoming increasingly indispensable across diverse domains as technology rapidly advances. As traditional energy sources dwindle, there's a noticeable pivot towards renewable (RES). However, to effectively meet demands, integrating these RES into smart grids bolster efficiency imperative. Despite the transition, ongoing technical challenges persist, specifically in accurately predicting and optimizing grid parameters. To tackle hurdles enhance efficiency, various AI techniques are being harnessed. This study leverages real-time generation data (MWh) from solar wind plants over year, dependent on parameters such POA speed, respectively. Prediction outcomes derived using three machine learning (ML) models (XGBoost, CatBoost, LightGBM) deep (DL) (LSTM, BiLSTM, GRU). From individual models, two hybrid ML DL developed, yielding promising results. Subsequently, further refined through parallel fusion approach (PFA), resulting heightened accuracy reliability. The implementation of this technique notably reduces error rates 15.05% for ML, 19.18% DL, 8.1432% PFA. methodology holds substantial potential future research endeavors, supplementing existing enhanced efficiency.

Язык: Английский

---