PSO-Stacking improved ensemble model for campus building energy consumption forecasting based on priority feature selection

Journal of Building Engineering, Journal Year: 2023, Volume and Issue: 72, P. 106589 - 106589

Published: April 20, 2023

Language: Английский

---

Adaptive Forecasting in Energy Consumption: A Bibliometric Analysis and Review

Data, Journal Year: 2024, Volume and Issue: 9(1), P. 13 - 13

Published: Jan. 11, 2024

This paper addresses the challenges in forecasting electrical energy current era of renewable integration. It reviews advanced adaptive methodologies while also analyzing evolution research this field through bibliometric analysis. The review highlights key contributions and limitations models with an emphasis on traditional methods. analysis reveals that Long Short-Term Memory (LSTM) networks, optimization techniques, deep learning have potential to model dynamic nature consumption, but they higher computational demands data requirements. aims offer a balanced view advancements methods, guiding researchers, policymakers, industry experts. advocates for collaborative innovation enhance accuracy support development resilient, sustainable systems.

Language: Английский

---

New formulation for predicting total dissolved gas supersaturation in dam reservoir: application of hybrid artificial intelligence models based on multiple signal decomposition

Artificial Intelligence Review, Journal Year: 2024, Volume and Issue: 57(4)

Published: March 9, 2024

Abstract Total dissolved gas (TDG) concentration plays an important role in the control of aquatic life. Elevated TDG can cause gas-bubble trauma fish (GBT). Therefore, controlling fluctuation has become great importance for different disciplines surface water environmental engineering.. Nowadays, direct estimation is expensive and time-consuming. Hence, this work proposes a new modelling framework predicting based on integration machine learning (ML) models multiresolution signal decomposition. The proposed ML were trained validated using hourly data obtained from four stations at United States Geological Survey. dataset are composed from: ( i ) temperature T w ), ii barometric pressure BP iii discharge Q which used as input variables prediction. strategy conducted two steps. First, six singles model namely: multilayer perceptron neural network, Gaussian process regression, random forest iv vector functional link, v adaptive boosting, vi Bootstrap aggregating (Bagging), developed , their performances compared. Second, was introduced combination empirical mode decomposition (EMD), variational (VMD), wavelet transform (EWT) preprocessing algorithms with building hybrid models. signals decomposed to extract intrinsic functions (IMFs) by EMD VMD methods analysis (MRA) components EWT method. Then after, IMFs MRA selected regraded integral part thereof. single prediction compared several statistical metrics namely, root mean square error, absolute coefficient determination R 2 Nash–Sutcliffe efficiency (NSE). times high number repetitions, depending kind modeling process. results gave good agreement between predicted situ measured dataset. Overall, Bagging performed better than other five NSE values 0.906 0.902, respectively. However, extracted EMD, have contributed improvement models’ performances, significantly increased reaching 0.996 0.995. Experimental showed superiority more importantly improving predictive accuracy TDG. Graphical abstract

Language: Английский

---

Day-ahead load forecast based on Conv2D-GRU_SC aimed to adapt to steep changes in load

Energy, Journal Year: 2024, Volume and Issue: 302, P. 131814 - 131814

Published: May 26, 2024

Language: Английский

---

Online decoupling feature framework for optimal probabilistic load forecasting in concept drift environments

Applied Energy, Journal Year: 2025, Volume and Issue: 392, P. 125952 - 125952

Published: April 25, 2025

Language: Английский

---

Modeling heat capacity of liquid siloxanes using artificial intelligence methods

Fluid Phase Equilibria, Journal Year: 2025, Volume and Issue: unknown, P. 114423 - 114423

Published: March 1, 2025

Language: Английский

---

AI Technologies and Their Applications in Small-Scale Electric Power Systems

IEEE Access, Journal Year: 2024, Volume and Issue: 12, P. 109984 - 110001

Published: Jan. 1, 2024

As the landscape of electric power systems is transforming towards decentralization, small-scale have garnered increased attention. Meanwhile, proliferation artificial intelligence (AI) technologies has provided new opportunities for system management. Thus, this review paper examines AI technology applications and their range uses in electrical systems. First, a brief overview evolution importance integration given. The background section explains principles systems, including stand-alone grid-interactive microgrids, hybrid virtual plants. A thorough analysis conducted on effects aspects such as energy consumption, demand response, grid management, operation, generation, storage. Based foundation, Acceleration Performance Indicators (AAPIs) are developed to establish standardized framework evaluating comparing different studies. AAPI considers binary scoring five quantitative Key (KPIs) qualitative KPIs examined through three-tiered scale – established, evolved, emerging.

Language: Английский

---

Studying the Thermodynamic Phase Stability of Organic–Inorganic Hybrid Perovskites Using Machine Learning

Molecules, Journal Year: 2024, Volume and Issue: 29(13), P. 2974 - 2974

Published: June 22, 2024

As an important photovoltaic material, organic–inorganic hybrid perovskites have attracted much attention in the field of solar cells, but their instability is one main challenges limiting commercial application. However, search for stable among thousands perovskite materials still faces great challenges. In this work, energy above convex hull values was predicted based on four different machine learning algorithms, namely random forest regression (RFR), support vector (SVR), XGBoost regression, and LightGBM to study thermodynamic phase stability perovskites. The results show that algorithm has a low prediction error can effectively capture key features related Meanwhile, Shapley Additive Explanation (SHAP) method used analyze algorithm. third ionization B element most critical feature stability, second electron affinity ions at X site, which are significantly negatively correlated with (Ehull). screening high site worthy priority. help us understand correlation between features, assist rapid discovery highly materials.

Language: Английский

---

Analysis and prediction of the changes in groundwater resources under heavy precipitation and ecological water replenishment

Journal of Water and Climate Change, Journal Year: 2023, Volume and Issue: 14(6), P. 1762 - 1778

Published: May 22, 2023

Abstract Identifying the influence of heavy precipitation and ecological water replenishment (EWR) on groundwater resources is essential for management risk prevention. This study innovatively developed a resource analysis prediction model integrated with level fluctuation method, correlation analysis, machine learning method under EWR. Water results showed that compared January 1, 2021, area increased 4.46 × 108 m3 August 28. Compared small flow EWR, was main contributor to rise in level. Correlation found elevation, specific yield, permeability coefficient show positive correlations recharge. Machine among models 35 monitoring wells, extreme gradient boosting (XGB) random forest (RF) performed best 30 wells five respectively. The increase storage predicted deviated from actual value by only 0.6 107 (prediction bias 1.3%), indicating performance good condition. can help better understand change trend conditions

Language: Английский

---

Evaluating the EEMD-LSTM model for short-term forecasting of industrial power load: A case study in Vietnam

International Journal of Renewable Energy Development, Journal Year: 2023, Volume and Issue: 12(5), P. 881 - 890

Published: Aug. 2, 2023

This paper presents the effectiveness of ensemble empirical mode decomposition-long short-term memory (EEMD-LSTM) model for short term load prediction. The prediction performance proposed is compared to that three other models (LR, ANN, LSTM). contribution this research lay in developing a novel approach combined EEMD-LSTM enhance capability industrial forecasting. was field where there had been limited proposals improvement, as these hybrid primarily developed industries such solar power, wind CO2 emissions, and not widely applied forecasting before. First, raw data preprocessed using IQR method, serving input all four models. Second, processed then used train each evaluated regression-based metrics mean absolute error (MAE) squared (MSE) assess their respective output. Seojin Vietnam, results showed it outperformed terms RMSE, n-RMSE, MAPE errors both 1-step 24-step highlighted model's capture intricate nonlinear patterns electricity data. study underscored significance selecting suitable concluded dependable precise predicting future assets. robust accurate forecasts showcased its potential assisting decision-making processes energy sector.

Language: Английский

---