Midspan Deflection Prediction of Long-Span Cable-Stayed Bridge Based on DIWPSO-SVM Algorithm DOI Creative Commons
Lilin Li, Qing He, Hua Wang

и другие.

Applied Sciences, Год журнала: 2025, Номер 15(10), С. 5581 - 5581

Опубликована: Май 16, 2025

With the increasing emphasis on safety and longevity of large-span cable-stayed bridges, accurate prediction midspan deflection has become a critical aspect structural health monitoring (SHM). This study proposes novel hybrid model, DIWPSO-SVM, which integrates dynamic inertia weight particle swarm optimization (DIWPSO) with support vector machines (SVMs) to enhance accuracy deflection. The model incorporates wavelet transform decompose signals into temperature vehicle load effects, allowing for more detailed analysis their individual impacts. DIWPSO algorithm dynamically adjusts balance global exploration local exploitation, optimizing SVM parameters improved performance. proposed was validated using real-world data from long-span bridge, demonstrating superior compared traditional PSO-SVM models. DIWPSO-SVM achieved an average error 1.43 mm root-mean-square (RMSE) 2.05, significantly outperforming original had 5.29 RMSE 5.62. These results highlight effectiveness in providing reliable predictions, offering robust tool bridge maintenance decision-making.

Язык: Английский

A robust control scheme for optimized pitch angle estimation of offshore wind turbine under varied climatic conditions using Osprey algorithm DOI
Prince Kumar, Nabanita Adhikary

Measurement, Год журнала: 2025, Номер unknown, С. 117122 - 117122

Опубликована: Фев. 1, 2025

Язык: Английский

Процитировано

1
Fractional order ANFIS controllers for LFC in RES integrated three-area power system DOI Creative Commons
Yaw Opoku Mensah Sekyere, Francis Boafo Effah, Philip Yaw Okyere

и другие.

Journal of Electrical Systems and Information Technology, Год журнала: 2025, Номер 12(1)

Опубликована: Март 26, 2025

Abstract The existing adaptive neuro-fuzzy inference system (ANFIS) for load frequency control in multi-area power systems has two inputs consisting of area error (ACE) and its integer order derivative. A recently proposed ANFIS added another input integral the ACE. In this paper, controllers with fractional derivative ACE ACE, referred to as (FO-ANFIS) are used instead improve performance controllers. FO-ANFIS training dataset is obtained from a cascaded PI-fractional PID filters (FOPI-FOPIDN) tuned by particle swarm optimization variant called dynamic inertia weight acceleration coefficient algorithm. 2-input 3-input FO-ANFIS, tested on three-area integrated renewable energy sources. results compared those their (IO-AFIS) counterparts FOPI-FOPIDN which data was using overshoot, undershoot, settling time, steady-state tie-line responses well time absolute values. Their real-world applicability validated incorporating communication delay governor dead band one four experimental scenarios evaluation. Robustness parameter uncertainty further assessed through variation ± 25%. From results, controller emerges best performing between followed controller.

Язык: Английский

Процитировано

0
Midspan Deflection Prediction of Long-Span Cable-Stayed Bridge Based on DIWPSO-SVM Algorithm DOI Creative Commons
Lilin Li, Qing He, Hua Wang

и другие.

Applied Sciences, Год журнала: 2025, Номер 15(10), С. 5581 - 5581

Опубликована: Май 16, 2025

With the increasing emphasis on safety and longevity of large-span cable-stayed bridges, accurate prediction midspan deflection has become a critical aspect structural health monitoring (SHM). This study proposes novel hybrid model, DIWPSO-SVM, which integrates dynamic inertia weight particle swarm optimization (DIWPSO) with support vector machines (SVMs) to enhance accuracy deflection. The model incorporates wavelet transform decompose signals into temperature vehicle load effects, allowing for more detailed analysis their individual impacts. DIWPSO algorithm dynamically adjusts balance global exploration local exploitation, optimizing SVM parameters improved performance. proposed was validated using real-world data from long-span bridge, demonstrating superior compared traditional PSO-SVM models. DIWPSO-SVM achieved an average error 1.43 mm root-mean-square (RMSE) 2.05, significantly outperforming original had 5.29 RMSE 5.62. These results highlight effectiveness in providing reliable predictions, offering robust tool bridge maintenance decision-making.

Язык: Английский

Процитировано

0