Abstract
The gearbox is a critical component of modern MW wind turbines. An accurate model of the gearbox dynamics is needed to improve gearbox design, develop advanced control algorithms, and more effective fault diagnosis tools which could lead to lower the cost of energy from wind. The objective of this paper is to investigate how torque measurements can be used in a data-driven framework to build dynamic models of wind turbine gearboxes.An initial torsional model has been derived from first principles considering the stiffness of the gears, shafts, and structural components in the gearbox together with the mechanical components of the test bench. This model has been used to create simulated data of the experiments performed on gearboxes and to apply system identification techniques to the simulated signals, with a focus on predictor based subspace identification methods. System identification has been applied to torque and speed data measured on physical tests of two 3.4MW gearboxes. Gearbox excitation frequencies and their harmonics dominate the measured signals and disturb the system identification algorithms. Several techniques have been investigated to remove the shaft rotation and gear mesh frequency harmonics of the torque and rotational speed signals based on time synchronous averaging.
Highlights
The cost of energy has become the critical driver for wind energy research
Several techniques have been investigated to remove the shaft rotation and gear mesh frequency harmonics of the torque and rotational speed signals based on time synchronous averaging
This paper explores what information about the gearbox dynamics can be extracted from the torque signals in a back-toback gearbox test bench, to achieve data-driven models of the gearbox
Summary
The cost of energy has become the critical driver for wind energy research. Nowadays, 75% of onshore wind turbines have a gearbox [12], and the gearbox is one of the main contributors to capital expenditure of onshore wind energy projects. Three main applications are sought from data-driven models: improve gearbox design, develop advanced control algorithms, and more effective fault diagnosis tools, which would, in turn, lower the cost of energy. Complex physically derived models have been created to study gearbox dynamics but have not been used together with whole turbine models, and traditional wind turbine design codes lack insight into the dynamic behaviour of the internal drive train components [7]. A data-driven modelling methodology is proposed to improve the assessment of the remaining useful life of the gearbox components and develop novel operating concepts. When these models are used for control purposes, control performance is tied directly to the accuracy of the identified model [9].
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
