Data Analysis Method
Finite Element Modeling in VibronRotor
The FEM approach in VibronRotor is used to discretize a rotating shaft into smaller, manageable beam
elements to analyze its lateral dynamic behavior. The modeling assumes the rotor is flexible and operating
under high-speed conditions, which necessitates consideration of gyroscopic effects and distributed inertia.
1. Element Definition
• Each finite element, as illustrated in Fig. 2, is a two-node beam element.
• Each node has four degrees of freedom (DOFs):
– Translational: x and y (displacements in horizontal and vertical directions)
– Rotational: θx and θy (rotations about the x and y axes)
• The system ignores axial and torsional motions to focus purely on lateral dynamics.
2. Generalized Displacement Vector
The nodal displacements for an element i are packed into a single displacement vector:
Qi = [xi , yi , θxi , θyi , xi+1 , yi+1 , θx(i+1) , θy(i+1) ]T
This vector captures all translational and rotational movements at both ends of the element.
3. Shape Functions and Matrix Formulation
Using standard Euler-Bernoulli beam theory, shape functions S1 , S2 , S3 , S4 are defined (Eq. 1B in the image)
based on the position along the beam’s length. These shape functions help describe how displacements and
rotations vary along the element.
From these, VibronRotor computes the following element-level matrices:
• Mass matrix (Me ): includes translational and rotational inertial terms
• Stiffness matrix (Ke ): models resistance due to bending stiffness
• Gyroscopic matrix (Ge ): accounts for dynamic effects due to rotation speed
• Damping matrix (when included): reflects energy loss mechanisms like bearing damping
An example of the translational mass matrix Mt is shown in the image (lower right), illustrating its
symmetric structure.
4. Global Assembly
• Element matrices are assembled into large global matrices using rotor geometry and mesh configuration
defined by the user.
• These global matrices are used in:
– Eigenanalysis (to find natural frequencies)
– Imbalance response
– Stability evaluation
1
Eigenanalysis
Purpose:
To compute the rotor system’s natural frequencies and assess damping behavior by solving the equations of
motion using eigenvalue decomposition.
Key Equation:
The equation of motion (as defined in Eq. 6) is converted to a state-space form:
0 I
A=
−M−1 (KB + KS ) −M−1 (CB + ΩG)
– A: State matrix
– M: Global mass matrix
– KS , KB : Structural and bearing stiffness matrices
– CB : Bearing damping matrix
– ΩG: Gyroscopic matrix scaled by rotational speed
Interpretation:
By solving for the eigenvalues of matrix A, one obtains the system’s damped natural frequencies and damping
ratios. The real parts of the eigenvalues indicate system stability. These results are used to generate Campbell
diagrams, which map natural frequencies versus rotor speed and identify critical speeds.
Steady-State Imbalance Response
Purpose:
To simulate the dynamic response of the rotor when subjected to harmonic excitation due to mass imbalance.
Assumed Solution:
The displacement of each degree of freedom is assumed to follow harmonic motion:
X(t) = x cos(ωt − β) = QC cos(ωt) + QS sin(ωt)
Where:
– x: Amplitude
– β: Phase lag
– QC , QS : Cosine and sine response components
Substituting into the motion equation yields:
K − Mω 2
ωC QC F
= C
−ωC K − Mω 2 QS FS
This matrix equation (Eq. 14) is solved to obtain the rotor’s steady-state response under the specified
imbalance conditions.
Execution:
– The user provides the imbalance forces, phase angles, and locations.
– The analysis is performed over a user-defined speed range.
2
– Bearing stiffness and damping coefficients are modeled as speed-dependent and interpolated using cubic
splines.
– The resulting amplitude and phase values are plotted as Bode diagrams.
– Orbit plots are generated to visualize the lateral motion of the rotor.
Instability Threshold Analysis
Purpose:
To determine the critical rotational speed at which the rotor system transitions from stable to unstable
behavior by analyzing the real parts of complex eigenvalues (damping exponents).
Approach:
– The state matrix is evaluated over a range of rotor speeds.
– For each speed, the eigenvalues are computed. The real parts of these eigenvalues correspond to the
damping exponents of the system’s vibrational modes.
– Instability is identified when a damping exponent transitions from negative to positive.
– The exact speed at which this transition occurs is referred to as the instability threshold.
Execution:
– The user provides speed-dependent bearing stiffness and damping coefficients.
– Because these coefficients are usually only known at discrete speeds, cubic spline interpolation is used
to estimate values at intermediate speeds.
– The damping exponents are plotted as functions of rotor speed to visualize stable and unstable regions.
– These plots help define safe operational limits for the rotor and avoid regions that could lead to self-
excited vibrations or failure.
Active Vibration Control
Purpose:
To stabilize rotor systems using feedback-based electromechanical control strategies, especially in systems
where passive damping is insufficient or contact-free operation is required.
Model Overview:
– A simplified 1-degree-of-freedom (1-DOF) model of a cantilever beam with electromagnetic actuation
is used.
– The system equation is derived from Newton’s law and includes force-current and force-displacement
relationships.
– The dynamic behavior of the system is represented in the Laplace domain.
Open-Loop System:
The open-loop transfer function is:
Ki
G(s) =
ms2 − Kx
Where:
3
– Ki : Force-current gain
– Kx : Stiffness coefficient
– m: Mass of the beam
– This open-loop system is inherently unstable (a pole in the right-half plane).
Control Strategy:
A Proportional-Derivative (PD) controller is applied:
Gc (s) = kp + kd s
– The resulting closed-loop transfer function is derived and rewritten into a state-space form for dynamic
analysis.
Execution:
– The closed-loop dynamics are analyzed via eigenvalue tracking of the augmented state matrix.
– This allows investigation of how control parameters (kp , kd ) influence stability and vibration attenua-
tion.
– This capability enables VibronRotor to model and evaluate real-world control systems like Active
Magnetic Bearings (AMB) and their interaction with rotor dynamics.