Aeroelastic Stability Analysis of Large Scale Horizontal Axis Wind Turbines Using Reduced Order System Identification Based on Flexible Nonlinear Multi-body Dynamics

2008 ◽  
Author(s):  
Fanzhong Meng ◽  
Marilena Pavel ◽  
Michel van Tooren
Keyword(s):  
Stability Analysis ◽  
System Identification ◽  
Large Scale ◽  
Horizontal Axis ◽  
Order System ◽  
Aeroelastic Stability ◽  
Reduced Order ◽  
Body Dynamics ◽  
Horizontal Axis Wind Turbines ◽  
Multi Body

scholarly journals Reduced order system identification for UAVs

2015 ◽  
Vol 119 (1218) ◽  
pp. 961-980 ◽  
Author(s):  
P-D. Jameson ◽  
A. K. Cooke
Keyword(s):  
System Identification ◽  
Equations Of Motion ◽  
Real Data ◽  
Flight Test ◽  
Rate Of Change ◽  
Order System ◽  
Parameter Estimates ◽  
Reduced Order Models ◽  
Test Scenario ◽  
Reduced Order

Abstract Reduced order models representing the dynamic behaviour of symmetric aircraft are well known and can be easily derived from the standard equations of motion. In flight testing, accurate measurements of the dependent variables which describe the linearised reduced order models for a particular flight condition are vital for successful system identification. However, not all the desired measurements such as the rate of change in vertical velocity (Ẇ) can be accurately measured in practice. In order to determine such variables two possible solutions exist: reconstruction or differentiation. This paper addresses the effect of both methods on the reliability of the parameter estimates. The methods are used in the estimation of the aerodynamic derivatives for the Aerosonde UAV from a recreated flight test scenario in Simulink. Subsequently, the methods are then applied and compared using real data obtained from flight tests of the Cranfield University Jetstream 31 (G-NFLA) research aircraft.

Projective Jacobi and Gauss-Seidel on the GPU for Non-Smooth Multi-Body Systems

2014 ◽  
Author(s):  
Gabriel Nützi ◽  
Adrian Schweizer ◽  
Michael Möller ◽  
Christoph Glocker
Keyword(s):  
Granular Media ◽  
Large Scale ◽  
Contact Problems ◽  
Rigid Body Dynamics ◽  
Contact Dynamics ◽  
Problem Size ◽  
Body Dynamics ◽  
Smooth Contact ◽  
Multi Body ◽  
Speedup Factor

Large-scale contact problems with impacts and Coulomb friction arise in the simulation of rigid body dynamics treated within the non-smooth contact dynamics approach using set-valued force and impact laws. In this paper the parallelization of two popular numerical methods for solving such contact problems on the GPU, being the projected over-relaxed Jacobi (JOR Prox) and projected Gauss-Seidel iteration (SOR Prox), is studied in detail. Performance tests for the parallel JOR and SOR Prox iterations are conducted and a speedup factor of up to 16, depending on the problem size, can be achieved compared to a sequential implementation. This work forms the stepping stone to the simulation of granular media on a computer cluster.

Order Reduction and Control of Large-Scale Linear Time Periodic Dynamical Systems

1999 ◽  
Author(s):  
Venkatesh Deshmukh ◽  
S. C. Sinha
Keyword(s):  
Large Scale ◽  
Linear Time ◽  
Parametric Instability ◽  
Order Reduction ◽  
System Stability ◽  
Order System ◽  
Control Laws ◽  
Reduced Order ◽  
Time Invariant ◽  
Time Periodic

Abstract This paper provides methodology for designing reduced order controllers for large-scale, linear systems represented by differential equations having time periodic coefficients. The linear time periodic system is first converted into a form in which the system stability matrix is time invariant. This is achieved by the application of Liapunov-Floquet transformation. Then a system called an auxiliary system is constructed which is a completely time invariant. Order reduction algorithms are applied to this system to obtain a reduced order system. The control laws are calculated for the reduced order system by minimizing the least square error between the auxiliary and the transformed system. These control laws when transformed back to time varying domain provide the desired control action. The schemes formulated are illustrated by designing full state feedback and output feedback controllers for a five mass inverted pendulum exhibiting parametric instability.

Torsional Vibration Analysis for Large-Scale Reciprocating Compressor Crankshaft

2013 ◽  
Vol 457-458 ◽  
pp. 428-432
Author(s):  
Ke Zhan ◽  
Xiao Ling Yu ◽  
Bin Yan Yu ◽  
Jia Xie
Keyword(s):  
Finite Element ◽  
Large Scale ◽  
New Method ◽  
Reciprocating Compressor ◽  
Connecting Rod ◽  
Transient Stress ◽  
Finite Element Technology ◽  
Body Dynamics ◽  
Vibration Stress ◽  
Multi Body

This paper presents a new method which combined multi-body dynamics theory and finite element technology to calculate transient stress of the crankshaft of the large-scale reciprocating compressor. On the basis of multi-body dynamics theory, the kinematical simulation of the crankshaft, the connecting rod, the piston and other components were performed, and thus to get the vibration modal of the crankshaft. So we can judge whether the crankshafts torsional resonance will happen, as well as get the real loads on the crankshaft when it worked. Then the transient stress of the crankshaft can be calculated using finite element technology. Comparing to traditional stress calculating methods, this new method not only considers the variable inertia which caused by reciprocating masss movement, but also can calculate the integrated vibration stress of crankshaft in three directions, including torsion, lateral and axial. Therefore, this method can describe dynamic characteristics of the crankshaft more accurately and more entirely.

Sign in / Sign up

Export Citation Format

Share Document