Gradient-based optimization with ranking mechanisms for parameter identification of photovoltaic systems

Energy Reports, Journal Year: 2021, Volume and Issue: 7, P. 3979 - 3997

Published: July 3, 2021

Deriving optimal photovoltaic (PV) models' parameters have tremendous significance in simulating, evaluating, and controlling the systems. Determining unknown of these PV models is a multimodal, nonlinear, complex optimization problem. Hence, developing robust model to achieve effectively essential. This paper proposes an enhanced metaphor-free gradient-based optimizer (EGBO) for extracting quickly, precisely, reliably. In EGBO, rank-based mechanism employed update its efficiently. Also, logistic map (LC) implemented better use local escaping operator (LEO) original GBO algorithm. The proposed EGBO optimally identifies various model, such as single diodes, double modules. relevant results indicate that compared with most advanced methods, algorithm competitive reliability, accuracy, convergence speed. Moreover, from experimental data drawn manufacturer's datasheet demonstrate developed approach can offer highly accurate solutions at irradiances temperatures. Consequently, achieved confirm novel be presented utility tool deriving parameters, it helpful modeling

Language: Английский

---

Systematic Review of Deep Learning and Machine Learning for Building Energy

Frontiers in Energy Research, Journal Year: 2022, Volume and Issue: 10

Published: March 18, 2022

The building energy (BE) management has an essential role in urban sustainability and smart cities. Recently, the novel data science data-driven technologies have shown significant progress analyzing consumption demand sets for a smarter management. machine learning (ML) deep (DL) methods applications, particular, been promising advancement of accurate high-performance models. present study provides comprehensive review ML DL-based techniques applied handling BE systems, it further evaluates performance these techniques. Through systematic taxonomy, advances are carefully investigated, models introduced. According to results obtained forecasting, hybrid ensemble located high robustness range, SVM-based good limitation, ANN-based medium limitation linear regression low limitations. On other hand, DL-based, hybrid, ensemble-based provided highest score. ANN, SVM, single LR-based lower In addition, load higher score

Language: Английский

---

Boosting quantum rotation gate embedded slime mould algorithm

Expert Systems with Applications, Journal Year: 2021, Volume and Issue: 181, P. 115082 - 115082

Published: April 21, 2021

Language: Английский

---

Hybridizing of Whale and Moth-Flame Optimization Algorithms to Solve Diverse Scales of Optimal Power Flow Problem

Electronics, Journal Year: 2022, Volume and Issue: 11(5), P. 831 - 831

Published: March 7, 2022

The optimal power flow (OPF) is a practical problem in system with complex characteristics such as large number of control parameters and also multi-modal non-convex objective functions inequality nonlinear constraints. Thus, tackling the OPF becoming major priority for engineers researchers. Many metaheuristic algorithms different search strategies have been developed to solve problem. Although, majority them suffer from stagnation, premature convergence, local optima trapping during optimization process, which results producing low solution qualities, especially real-world problems. This study devoted proposing an effective hybridizing whale algorithm (WOA) modified moth-flame (MFO) named WMFO In proposed WMFO, WOA MFO cooperate effectively discover promising areas provide high-quality solutions. A randomized boundary handling used return solutions that violated permissible boundaries space. Moreover, greedy selection operator defined assess acceptance criteria new Ultimately, performance scrutinized on single multi-objective cases problems including standard IEEE 14-bus, 30-bus, 39-bus, 57-bus, IEEE118-bus test systems. obtained corroborate outperforms contender solving

Language: Английский

---

An improved moth flame optimization algorithm based on modified dynamic opposite learning strategy

Artificial Intelligence Review, Journal Year: 2022, Volume and Issue: 56(4), P. 2811 - 2869

Published: Aug. 16, 2022

Language: Английский

---