Evaluating the Performance of Machine Learning Models for Energy Load Prediction in Residential HVAC Systems

Energy and Buildings, Год журнала: 2025, Номер unknown, С. 115517 - 115517

Опубликована: Фев. 1, 2025

Язык: Английский

---

Predictive Analytics for Sustainable Energy: An In-depth Assessment of Novel Stacking Regressor Model in the Off-Grid Hybrid Renewable Energy Systems

Energy, Год журнала: 2025, Номер unknown, С. 135916 - 135916

Опубликована: Апрель 1, 2025

Язык: Английский

---

Enhancing the safety of hydroelectric power generation systems: an intelligent identification of axis orbits based on a nonlinear dynamics method

Energy, Год журнала: 2025, Номер unknown, С. 135864 - 135864

Опубликована: Апрель 1, 2025

Язык: Английский

---

Building Energy Optimization Using an Improved Exponential Distribution Optimizer Based on Golden Sine Strategy Minimizing Energy Consumption Under Uncertainty

Results in Engineering, Год журнала: 2025, Номер unknown, С. 104840 - 104840

Опубликована: Апрель 1, 2025

Язык: Английский

---

Enhancing Smart Home Efficiency with Heuristic-Based Energy Optimization

Computers, Год журнала: 2025, Номер 14(4), С. 149 - 149

Опубликована: Апрель 16, 2025

In smart homes, heavy reliance on appliance automation has increased, along with the energy demand in developing urban areas, making efficient management an important factor. To address scheduling of appliances under Demand-Side Management, this article explores use heuristic-based optimization techniques (HOTs) homes (SHs) equipped renewable and sustainable resources (RSERs) storage systems (ESSs). The optimal model for minimization peak-to-average ratio (PAR), considering user comfort constraints, is validated by using different techniques, such as Genetic Algorithm (GA), Binary Particle Swarm Optimization (BPSO), Wind-Driven (WDO), Bacterial Foraging (BFO) Modified (GmPSO) algorithm, to minimize electricity costs, PAR, carbon emissions delay discomfort. This research investigates results three real-world scenarios. scenarios demonstrate benefits gradually assembling RSERs ESSs integrating them into SHs employing HOTs. simulation show substantial outcomes, scenario Condition 1, GmPSO decreased from 300 kg 69.23 kg, reducing 76.9%; bill prices were also cut unplanned value 400.00 cents 150 cents, a 62.5% reduction. PAR was unscheduled 4.5 2.2 which reduced 51.1%. 2 showed that 0.5 (unscheduled) 0.2, 60% reduction; costs 500.00 200.00 250.00 reduction GmPSO. 3, where batteries integrated, algorithm emission 158.3 208.3 24%. cost 500 GmPSO, decreasing overall 40%. achieved 57.1% 2.8 1.2.

Язык: Английский

---

Energy Use Intensity Analysis of Office Buildings Using Green BIM-Integrated Interpretable Machine Learning

Journal of Building Engineering, Год журнала: 2025, Номер unknown, С. 112760 - 112760

Опубликована: Апрель 1, 2025

Язык: Английский

---

Development and application of an intelligent nitrogen removal diagnosis and optimization framework for WWTPs: Low-carbon and stable operation

Water Research, Год журнала: 2024, Номер 266, С. 122337 - 122337

Опубликована: Авг. 30, 2024

Язык: Английский

---

Adaptive Difference Least Squares Support Vector Regression for Urban Road Collapse Timing Prediction

Symmetry, Год журнала: 2024, Номер 16(8), С. 977 - 977

Опубликована: Авг. 1, 2024

The accurate prediction of urban road collapses is paramount importance for public safety and infrastructure management. However, the complex variable nature subsidence mechanisms, coupled with inherent noise non-stationarity in data, poses significant challenges to development precise real-time models. To address these challenges, this paper develops an Adaptive Difference Least Squares Support Vector Regression (AD-LSSVR) model. AD-LSSVR model employs a difference transformation process input output effectively reducing enhancing stability. This extracts trends features from leveraging symmetrical characteristics within it. Additionally, parameters were optimized using grid search cross-validation techniques, which systematically explore parameter space evaluate performance multiple subsets ensuring both precision generalizability selected parameters. Moreover, sliding window method was employed data sparsity anomalies, robustness adaptability experimental results demonstrate superior predicting collapse timing, highlighting its effectiveness handling nonlinear data.

Язык: Английский

---

Experimental studies on the cooling and heating performance of a highly emissive coating

Heliyon, Год журнала: 2024, Номер 10(19), С. e38233 - e38233

Опубликована: Сен. 21, 2024

Язык: Английский

---

Developing an integrated prediction model for daylighting, thermal comfort, and energy consumption in residential buildings based on the stacking ensemble learning algorithm

Building Simulation, Год журнала: 2024, Номер unknown

Опубликована: Сен. 10, 2024

Язык: Английский

---