Testing the Wind Energy Data Based on Environmental Factors Predicted by Machine Learning with Analysis of Variance DOI Creative Commons
Yasemin Ayaz Atalan, Abdülkadir Atalan

Applied Sciences, Journal Year: 2024, Volume and Issue: 15(1), P. 241 - 241

Published: Dec. 30, 2024

This study proposes a two-stage methodology for predicting wind energy production using time, environmental, technical, and locational variables. In the first stage, machine learning algorithms, including random forest (RF), gradient boosting (GB), k-nearest neighbors (kNNs), linear regression (LR), decision trees (Tree), were employed to estimate output. Among these, RF exhibited best performance with lowest error metrics (MSE: 0.003, RMSE: 0.053) highest R2 value (0.988). second analysis of variance (ANOVA) was conducted evaluate statistical relationships between independent variables predicted dependent variable, identifying speed (p < 0.001) rotor as most influential factors. Furthermore, GB models produced predictions closely aligned actual data, achieving values 88.83% 89.30% in ANOVA validation phase. Integrating highlighted robustness methodology. These findings demonstrate integrating verification methods.

Language: Английский

The Evolution of AI Applications in the Energy System Transition: A Bibliometric Analysis of Research Development, the Current State and Future Challenges DOI Creative Commons
Daniel Icaza, Fernando González-Ladrón-de-Guevara, Jorge Rojas Espinoza

et al.

Energies, Journal Year: 2025, Volume and Issue: 18(6), P. 1523 - 1523

Published: March 19, 2025

The transformation of energy markets is at a crossroads in the search for how they must evolve to become ecologically friendly systems and meet growing demand. Currently, methodologies based on bibliographic data analysis are supported by information communication technologies have necessary. More sophisticated processes being used systems, including new digitalization models, particularly driven artificial intelligence (AI) technology. In present review, 342 documents indexed Scopus been identified that promote synergies between AI transition (ET), considering time range from 1990 2024. methodology includes an evaluation keywords related areas ET. analyses extend review authorship, co-authorship, AI’s influence system subareas. integration resources, supply demand, which renewable sources play leading role end-customer level, now conceived as both producer consumer, intensively studied. results has experienced notable growth last five years will undoubtedly future achieving decarbonization goals. Among applications it enable be design up execution start-up power plants with control optimization. This study aims baseline allows researchers, legislators, government decision-makers compare their benefits, ambitions, strategies, novel formulating policies field. developments scope sector were explored relation domain parts chain. While these involve complex analysis, techniques provide powerful solutions designing managing high penetration. represents fundamental shift market design, enabling more efficient sustainable transitions. Future lines research could focus demand forecasting, dynamic adjustments distribution different generation sources, storage, usage

Language: Английский

Citations

0
Cloud-IoT Framework for EV Charge Station Allocation and Scheduling: A Spotted Hyena Jellyfish Search Optimization Approach DOI

G. Saravanan,

Ramamani Tripathy, Raghuveer Rao

et al.

Sustainable Computing Informatics and Systems, Journal Year: 2025, Volume and Issue: unknown, P. 101118 - 101118

Published: March 1, 2025

Language: Английский

Citations

0
A Short-Term Load Forecasting Method Considering Multiple Factors Based on VAR and CEEMDAN-CNN-BILSTM DOI Creative Commons
Baoshan Wang, Li Wang,

Yanru Ma

et al.

Energies, Journal Year: 2025, Volume and Issue: 18(7), P. 1855 - 1855

Published: April 7, 2025

Short-term load is influenced by multiple external factors and shows strong nonlinearity volatility, which increases the forecasting difficulty. However, most of existing short-term methods rely solely on original data or take into account a single factor, results in significant errors. To improve accuracy, this paper proposes method considering contributing based VAR CEEMDAN-CNN- BILSTM. Firstly, strongly correlated with are selected Spearman correlation analysis, vector autoregressive (VAR) model multivariate input derived, Levenberg–Marquardt algorithm introduced to estimate parameters. Secondly, complete ensemble empirical mode decomposition adaptive noise (CEEMDAN) permutation entropy (PE) criterion combined decompose reconstruct relatively stationary components, respectively CNN-BILTSM network for forecasting. Finally, sine–cosine Cauchy mutation sparrow search (SCSSA) used optimize parameters combinative accuracy. The actual simulation utilizing Australian validate accuracy proposed model, achieving reduction root mean square error 31.21% 18.04% compared CEEMDAN-CNN-BILSTM, respectively.

Language: Английский

Citations

0
Electric Vehicle Shared Services: A Decade of Innovation, Challenges, and Transformative Impact on Sustainable Urban Mobility — A Systematic Literature Review DOI Open Access

Meis Musida,

Ivan Hanafi, Moch. Sukardjo

et al.

The Open Transportation Journal, Journal Year: 2025, Volume and Issue: 19(1)

Published: May 6, 2025

Introduction Research on Electric Vehicle Shared Services (EVSS) has significantly grown over the past decade, emerging as a transformative solution to urban mobility challenges while advancing sustainable transportation. Through innovation and scalable solutions, EVSS garnered attention for their potential address pressing environmental issues, including climate change air quality. Material Methods This Systematic Literature Review (SLR) examines evolution, challenges, impacts of from 2014 2023. A total 52 studies were analyzed using PRISMA methodology, ensuring comprehensive rigorous evaluation literature. Key themes identified synthesize trends, benefits associated with these services. Results Findings reveal significant growth in research driven by technological advancements, supportive policy frameworks, heightened global awareness issues. Studies highlight that can achieve reduction greenhouse gas emissions 14–65% compared traditional vehicles, alongside notable improvement local These are pivotal efforts mitigate enhance health. Moreover, provides affordable flexible transportation options, particularly underserved populations, contributing social equity. Integration public systems further reduces traffic congestion enhances efficiency. Discussion Despite promise, faces several challenges. Limited charging infrastructure necessitates investment networks. High upfront costs purchasing maintaining electric vehicle (EV) fleets remain financial obstacle operators. Furthermore, user perception such range anxiety, require targeted education campaigns acceptance. Collaborative among policymakers, community organizations, private operators crucial addressing barriers maximizing shared EV Conclusion represents approach achieving mobility. Their environmental, social, underscore role critical However, overcoming adoption will robust coordinated framework investments engagement strategies. Continued stakeholder collaboration essential unlocking full fostering equitable systems.

Language: Английский

Citations

0
Testing the Wind Energy Data Based on Environmental Factors Predicted by Machine Learning with Analysis of Variance DOI Creative Commons
Yasemin Ayaz Atalan, Abdülkadir Atalan

Applied Sciences, Journal Year: 2024, Volume and Issue: 15(1), P. 241 - 241

Published: Dec. 30, 2024

This study proposes a two-stage methodology for predicting wind energy production using time, environmental, technical, and locational variables. In the first stage, machine learning algorithms, including random forest (RF), gradient boosting (GB), k-nearest neighbors (kNNs), linear regression (LR), decision trees (Tree), were employed to estimate output. Among these, RF exhibited best performance with lowest error metrics (MSE: 0.003, RMSE: 0.053) highest R2 value (0.988). second analysis of variance (ANOVA) was conducted evaluate statistical relationships between independent variables predicted dependent variable, identifying speed (p < 0.001) rotor as most influential factors. Furthermore, GB models produced predictions closely aligned actual data, achieving values 88.83% 89.30% in ANOVA validation phase. Integrating highlighted robustness methodology. These findings demonstrate integrating verification methods.

Language: Английский

Citations

0