Research on Wind Turbine Fault Detection Based on CNN-LSTM

Energies, Journal Year: 2024, Volume and Issue: 17(17), P. 4497 - 4497

Published: Sept. 7, 2024

With the wide application of wind energy as a clean source, to cope with challenge increasing maintenance difficulty brought about by development large-scale power equipment, it is crucial monitor operating status turbines in real time and accurately identify specific location faults. In this study, CNN-LSTM-based motor fault detection model constructed for four types typical faults, namely gearbox electrical yaw pitch faults motors, combining CNN’s advantages excelling feature extraction LSTM’s dealing long-time sequence data, achieve simultaneous multiple types. The accuracy turbine reaches 90.06%, optimal results are achieved effective discovery system obtaining 94.09%, 96.46%, 97.39%, respectively. CNN-LSTM proposed study improves effect, avoids further deterioration provides direction preventive maintenance, reduces downtime loss due restorative essential sustainable use service life, which helps improve operation level farms.

Language: Английский

Impact contact mechanical characteristics assessment of piezoelectric material using the semi-analytical method

Mathematics and Mechanics of Solids, Journal Year: 2024, Volume and Issue: unknown

Published: Dec. 10, 2024

The objective of this paper is to study the impact mechanical performance piezoelectric materials. Therefore, three-dimensional frictionless contact between a half-space and conducting rigid impactor investigated. A semi-analytical model developed based on second-order Newmark method discrete convolution-fast Fourier transform algorithm. unknown reaction force impactors obtained by exploiting conjugate gradient dichotomy method. detailed parametric investigation conducted different types explore influences several parameters, including electric charge density, time step, velocity. results reveal that velocity density have significant effects response material. Higher densities lead smaller force, but larger deformation. insights gained from can be applied analyze charged particles arbitrary shape

Language: Английский