A CNN-LSTM based deep learning model with high accuracy and robustness for carbon price forecasting: A case of Shenzhen's carbon market in China

Journal of Environmental Management, Год журнала: 2024, Номер 352, С. 120131 - 120131

Опубликована: Янв. 23, 2024

Язык: Английский

---

An Adaptive Strategy for Wind Speed Forecasting Under Functional Data Horizon: A Way Toward Enhancing Clean Energy

IEEE Access, Год журнала: 2024, Номер 12, С. 68730 - 68746

Опубликована: Янв. 1, 2024

An important issue in competitive energy markets is the accurate and efficient wind speed forecasting for power production. However, models developed one location usually do not match other site various reasons like changes terrain, different patterns, atmospheric factors such as temperature, pressure, humidity, etc. Thus, introducing a flexible model that captures all features challenging task. This paper proposes functional data analysis (FDA) approach to forecast variant daily profiles with higher accuracy. Unlike traditional methods, FDA more attractive it forecasts complete profile, thus, can be obtained ultra-short period. To this end, first filtered extreme values. The series then divided into deterministic (Component-I) stochastic (Component-II) components. Component-I modeled forecasted based on generalized additive modeling technique. On hand, Component-II using autoregressive (FAR) FAR explanatory variables (FARX). For comparison purposes, from univariate integrated moving average (ARIMA), seasonal ARIMA (SARIMA), SARIMA exogenous information (SARIMAX), neural network (NNAR) are also obtained. empirical analysis, NASA project Canada located Durham, England, one-day-ahead out-of-sample year. performance of assessed through accuracy measures, namely mean error, root squared absolute standard error. results indicate outperform classical ARIMA, SARIMA, SARIMAX, deep learning model, NNAR. Within models, ability FARX superior FAR.

Язык: Английский

---

Evaluating LSTM and NARX neural networks for wind speed forecasting and energy optimization in Tetouan, Northern Morocco

Energy Exploration & Exploitation, Год журнала: 2025, Номер unknown

Опубликована: Янв. 7, 2025

Generating electricity from renewable sources is crucial for advancing toward a low-carbon economy, with wind power playing significant role. Effective energy management essential meeting societal needs and protecting the environment. This study aims to optimize production by improving accuracy of speed predictions. Building on previous research comparing MLP, NARX, Elman models Tetouan City, we introduce novel comparison between nonlinear autoregressive exogenous inputs (NARX) model long short-term memory (LSTM) network. Utilizing MATLAB, analyzed 12 years meteorological data City determine which provides most accurate Our results reveal that LSTM significantly outperforms NARX model, achieving lower values mean absolute error (MAE = 0.18855), squared (MSE 0.0666), root (RMSE 0.25808). demonstrates network's superior capability handle complex, long-term data. These findings offer valuable insights enhancing in similar regions, highlighting model's potential optimization efficiency.

Язык: Английский

---

Machine learning-based wind speed prediction using random forest: a cross-validated analysis for renewable energy applications

Turkish Journal of Engineering, Год журнала: 2025, Номер 9(3), С. 508 - 518

Опубликована: Март 8, 2025

Wind speed prediction plays a crucial role in renewable energy planning and optimization. This study presents comprehensive analysis of wind forecasting using Random Forest (RF) models. The research utilized high-resolution data collected throughout 2023 at the Bowen Abbot facility. Our methodology employed RF with cross-validation techniques to ensure model stability reliability. demonstrated robust performance across multiple evaluation metrics, achieving an average R² score 0.9155 (±0.0035) through 5-fold cross-validation. Error revealed consistent training, testing, validation sets, root mean square errors (RMSE) 0.6624 (±0.0098) m/s. Feature importance that 3-hour rolling was most influential predictor, accounting for 89.84% model's predictive power, followed by 1-hour (2.59%) (2.57%) lagged speeds. hierarchical temporal features suggests recent patterns are accurate predictions. error distribution showed approximately normal distributions slight deviations tails, particularly set (kurtosis: 5.2146). Key findings indicate maintains high accuracy different scales, absolute (MAE) averaging 0.4998 partitions its reliability operational deployment. These results demonstrate potential algorithms applications, providing valuable tool power generation management. study's contribute growing body on machine learning applications energy, offering insights into methodologies systems.

Язык: Английский

---

Tensile behavior evaluation of two-stage concrete using an innovative model optimization approach

REVIEWS ON ADVANCED MATERIALS SCIENCE, Год журнала: 2025, Номер 64(1)

Опубликована: Янв. 1, 2025

Abstract Two-stage concrete (TSC) is a sustainable material produced by incorporating coarse aggregates into formwork and filling the voids with specially formulated grout mix. The significance of this study to improve predictive accuracy TSC’s tensile strength, which essential for optimizing its use in construction applications. To achieve objective, novel reliable models were developed using advanced machine learning algorithms, including random forest (RF) gene expression programming (GEP). performance these was evaluated important evaluation metrics, coefficient determination ( R 2 ), mean absolute error (MAE), squared error, root square (RMSE), after they trained on comprehensive dataset. results suggest that RF model outperforms GEP model, as evidenced higher value 0.94 relative 0.91 reduced MAE RMSE values. This suggests has superior capability. Additionally, sensitivity analyses SHapley Additive ExPlanation analysis revealed water-to-binder (W/B) ratio most influential input parameter, accounting 51.01% outcomes presented model. research emphasizes TSC design, enhancing performance, promoting sustainable, cost-effective construction.

Язык: Английский

---

Interpretable machine learning for coastal wind prediction: Integrating SHAP analysis and seasonal trends

Journal of Coastal Conservation, Год журнала: 2025, Номер 29(3)

Опубликована: Май 6, 2025

Язык: Английский

---

Performance analysis of ARIMA Model for wind speed forecasting in Jerusalem, Palestine

Energy Exploration & Exploitation, Год журнала: 2024, Номер 42(5), С. 1727 - 1746

Опубликована: Апрель 24, 2024

Palestine lacks sufficient conventional energy sources that meet the daily needs of Palestinian people, and consequently, it heavily relies on neighboring countries for its supply with compensations. Wind is recognized as an abundant, effective, eco-friendly power source, but poses several challenges in harnessing due to inherent variability wind characteristics. The main objective this research study delve into landscape Palestine, offer some insights feasibility speed forecasting implementing sustainable solutions, a special focus ARIMA; widely used statistical method time series forecasting. It specifically explores potential using ARIMA models forecast data captured from meteorological station located east Jerusalem, duration 2 years—January 1, 2021 December 31, 2022. To find optimal values parameters (p, d, q) considered site, set experiments were conducted model's accuracy was evaluated three metrics: RMSE, MAE, coefficient determination (R ). results have shown (21,2) emerges most accurate structure input period demonstrates superior estimation minimal RMSE (1.74), MAE (1.58) higher R (0.76) values. This means achieved when autoregressive process based previous two lagged observations moving average incorporates dependency between observation residual error second-order applied observations. These findings give valuable precision emphasize region clarified by accuracy.

Язык: Английский

---

Seasonal ARIMA and LSTM Models for Wind Speed Prediction at El-Oued Region, Algeria

Опубликована: Апрель 24, 2024

The Saharan state of El-Oued becomes a hub for agricultural activity, and leads the production potatoes dates in Algeria. However, region faces challenges due to limited access rural electricity wind erosion. Given cube relationship between power drift potentials with velocity, it is imperative obtain accurate speed predictions. These forecasts are essential optimize feasibility off-grid powered by mitigate erosion risks. One major challenge predicting its stochastic, intermittent, non-dispatchable characteristics. This study focuses on short-term prediction using statistical methods like seasonal ARIMA deep learning techniques such as LSTM. Hourly data measured at 10 m AGL was utilized compare models. While SARIMA's performance improved significantly rolling prediction, LSTM exhibited superior performance, demonstrated achieving lowest RMSE MAE values deployed dataset.

Язык: Английский

---

Comparison of Artificial Intelligence Approaches for Estimating Wind Energy Production: A Real-World Case Study

Results in Engineering, Год журнала: 2024, Номер unknown, С. 103626 - 103626

Опубликована: Дек. 1, 2024

Язык: Английский

---

Enhanced Short-Term Photovoltaic Power Prediction using a Hybrid Improved Whale Optimization Algorithm with XGBoost

Опубликована: Апрель 2, 2024

Grid-connected photovoltaic (PV) systems are becoming more and popular in the renewable energy space, but their volatility unpredictability present difficulties, especially event of variable weather. In order to address these issues, accurate PV power forecasting becomes essential for efficient grid management. this, we a model called IMWOA-XGB, which combines XGBoost Improved Whale Optimization Algorithm (IMWOA). With help our model, projections should be accurate, allowing creation well-informed generation plans that reduce effect systems. The root mean square error (RMSE) is 18.867 absolute (MAE) 8.639 suggested model.

Язык: Английский

---