Dynamic Analysis of a Hybrid Squeeze Film Damper Mounted Rub-Impact Rotor-Stator System

An investigation is carried out on the systematic analysis of the dynamic behavior of the hybrid squeeze-film damper HSFD mounted a rotor-bearing system with strongly nonlinear oil-film force and nonlinear rub-impact force in the present study. The dynamic orbits of the system are observed using bifurcation diagrams plotted using the dimensionless rotating speed ratio as control parameters. The onset of chaotic motion is identified from the phase diagrams, power spectra, Poincaré maps, bifurcation diagrams, maximum Lyapunov exponents, and fractal dimension of the rotor-bearing system. The dynamic behaviors are unlike the usual ways into chaos 1T ⇒ 2T ⇒ 4T ⇒ 8T ⇒ 16T ⇒ 32T · · · ⇒ chaos or periodic ⇒ quasi-periodic ⇒ chaotic , it suddenly gets in chaos from the periodic motion without any transition. The results presented in this study provide some useful insights into the design and development of a rotor-bearing system for rotating machinery that operates in highly rotating speed and highly nonlinear regimes.


Introduction
Squeeze-film damper SFD bearing is actually a special type of journal bearing with its journal mechanically prevented from rotating but free to vibrate within the clearance space.The hybrid squeeze-film damper HSFD and the porous squeeze-film damper PSFD are the well-known applications of SFD and also useful for industry.Some literatures discussed dynamic behaviors in SFD bearings and also found many interesting and useful results.Holmes

Mathematical Modeling
Figure 1 shows a rotor supported on HSFDs in parallel with retaining springs.The bearing consists of four hydrostatic chambers and four hydrodynamic regions.The oil film supporting force is dependent on the integrated action of hydrodynamic pressure and hydrostatic pressure of HSFD. Figure 2 a represents the cross-section of HSFD and rubimpact rotor-stator model.The structure of this kind bearing should be popularized to consist of 2N N 2, 3, 4 . . .hydrostatic chambers and 2N hydrodynamic regions.In this study, oil pressure distribution model in the HSFD is proposed to integrate the pressure distribution of dynamic pressure region and static pressure region.

The Instant Oil Film Supporting Force for HSFD
To analyze the pressure distribution, the Reynolds equation for constant lubricant properties and noncompressibility should be assumed, then the Reynolds equation is introduced as follows 12 : The supporting region of HSFD should be divided into three regions: static pressure region, rotating direction dynamic pressure region, and axial direction dynamic pressure region, as shown in Figure 2. In the part of HSFD with −a ≤ z ≤ a, the long bearing theory is assumed and Reynolds equation is solved with the boundary condition of static pressure region p c,i acquiring the pressure distribution p 0 θ .In the part of HSFD with a ≤ |z| ≤ L/2, the short bearing theory is assumed and solves the Reynolds equation with the boundary condition of p z, θ | z ±a p 0 θ and p z, θ | z ±L/2 0, yielding the pressure distribution in axis direction dynamic pressure region p z, θ .Finally, a formula of pressure distribution in whole supporting region is obtained.According to the above conditions, the instant oil film pressure distribution is as follows.The instant pressure in rotating direction within the range of −a ≤ z ≤ a is where 3

2.4
The instant pressure in the axis direction within the range of where

2.6
The instant oil film forces of the different elements are determined by integrating 2.2 and 2.5 over the area of the journal sleeve.In the static pressure region, the forces are

2.7
In the rotating direction dynamics pressure region, the forces are p i θ R2a sin θdθ.

2.8
In the axial direction dynamic pressure region, the forces are

2.9
The resulting damper forces in the radial and tangential directions are determined by summing the above supporting forces.It is as follows: 2.10

Rub-Impact Force
Figure 2 b shows the radial impact force f 1 and the tangential rub force f 2 .f 1 and f 2 could be expressed as 17

2.11
Then we could get the rub-impact forces in the horizontal and vertical directions as follows: 2.12

Dynamics Equation
The equations of rotor motion in the Cartesian coordinates can be written as m ÿ d ẏ ky mρω 2 sin ωt f y ky 0 R y .

2.13
The origin of the o-xyz-coordinate system is taken to be the bearing center O b .Dividing these two equations by mcω 2 and defining a nondimensional time φ ωt and a speed parameter s ω/ω n , one obtains the following nondimensionalized equations of motion:

2.15
Equations 2.14 ∼ 2.15 describe a nonlinear dynamic system.In the current study, the approximate solutions of these coupled nonlinear differential equations are obtained using the fourth-order Runge-Kutta numerical scheme.

Analytical Tools for Observing Nonlinear Dynamics of Rotor-Bearing System
In the present study, the nonlinear dynamics of the rotor-bearing system equipped with HSFD shown in Figure 1 are analyzed using Poincaré maps, bifurcation diagrams, the Lyapunov exponent and the fractal dimension.The basic principles of each analytical method are reviewed in the following subsections.

Dynamic Trajectories and Poincar é Maps
The dynamic trajectories of the rotor-bearing system provide a basic indication as to whether the system behavior is periodic or nonperiodic.However, they are unable to identify the onset of chaotic motion.Accordingly, some other form of analytical method is required.In the current study, the dynamics of the rotor-bearing system are analyzed using Poincaré maps derived from the Poincaré section of the rotor system.A Poincaré section is a hypersurface in the state-space transverse to the flow of the system of interest.In nonautonomous systems, points on the Poincaré section represent the return points of a time series corresponding to a constant interval T , where T is the driving period of the excitation force.The projection of the Poincaré section on the y nT plane is referred to as the Poincaré map of the dynamic system.When the system performs quasiperiodic motion, the return points in the Poincaré map form a closed curve.For chaotic motion, the return points form a fractal structure comprising many irregularly distributed points.Finally, for nT-periodic motion, the return points have the form of n discrete points.

Power Spectrum
In this study, the spectrum components of the motion performed by the rotor-bearing system are analyzed by using the Fast Fourier Transformation method to derive the power spectrum of the displacement of the dimensionless dynamic transmission error.In the analysis, the frequency axis of the power spectrum plot is normalized using the rotating speed, ω.

Bifurcation Diagram
A bifurcation diagram summarizes the essential dynamics of a rotor-train system and is therefore a useful means of observing its nonlinear dynamic response.In the present analysis, the bifurcation diagrams are generated using two different control parameters, namely the dimensionless unbalance coefficient, β, and the dimensionless rotating speed ratio, s, respectively.In each case, the bifurcation control parameter is varied with a constant step, and the state variables at the end of one integration step are taken as the initial values for the next step.The corresponding variations of the y nT coordinates of the return points in the Poincaré map are then plotted to form the bifurcation diagram.

Lyapunov Exponent
The Lyapunov exponent of a dynamic system characterizes the rate of separation of infinitesimally close trajectories and provides a useful test for the presence of chaos.In a chaotic system, the points of nearby trajectories starting initially within a sphere of radius ε 0 form after time t an approximately ellipsoidal distribution with semiaxes of length ε j t .The Lyapunov exponents of a dynamic system are defined by λ j lim t → ∞ 1/t log ε j t /ε 0 , where λ j denotes the rate of divergence of the nearby trajectories.The exponents of a system are usually ordered into a Lyapunov spectrum, that is, A positive value of the maximum Lyapunov exponent λ 1 is generally taken as an indication of chaotic motion 16 .

Fractal Dimension
The presence of chaotic vibration in a system is generally detected using either the Lyapunov exponent or the fractal dimension property.The Lyapunov exponent test can be used for both dissipative systems and nondissipative i.e. conservative systems, but is not easily applied to the analysis of experimental data.Conversely, the fractal dimension test can only be used for dissipative systems but is easily applied to experimental data.In contrast to Fourier transform-based techniques and bifurcation diagrams, which provide only a general indication of the change from periodic motion to chaotic behavior, dimensional measures allow chaotic signals to be differentiated from random signals.Although many dimensional measures have been proposed, the most commonly applied measure is the correlation dimension d G defined by Grassberger and Procaccia due to its computational speed and the consistency of its results.However, before the correlation dimension of a dynamic system flow can be evaluated, it is first necessary to generate a time series of one of the system variables using a time-delayed pseudo-phase-plane method.Assume an original time series of x i {x iτ ; i 1, 2, 3, . . .N}, where τ is the time delay or sampling time .If the system is acted upon by an excitation force with a frequency ω, the sampling time, τ, is generally chosen such that it is much smaller than the driving period.The delay coordinates are then used to construct an n-dimensional vector X x jτ , x j 1 τ , x j 2 τ , . . ., x j n−1 τ , where j 1, 2, 3, . . .N − n 1 .The resulting vector comprises a total of N − n 1 vectors, which are then plotted in an n-dimensional embedding space.Importantly, the system flow in the reconstructed n-dimensional phase space retains the dynamic characteristics of the system in the original phase space.In other words, if the system flow has the form of a closed orbit in the original phase plane, it also forms a closed path in the n-dimensional embedding space.Similarly, if the system exhibits a chaotic behavior in the original phase plane, its path in the embedding space will also be chaotic.The characteristics of the attractor in the n-dimensional embedding space are generally tested using the function N i,j 1 H r − |x i − x j | to determine the number of pairs i, j lying within a distance , where H denotes the Heaviside step function, N represents the number of data points, and r is the radius of an n-dimensional hypersphere.For many attractors, this function exhibits a power law dependence on r as r → 0, that is c r ∝ r d G .Therefore, the correlation dimension, d G , can be determined from the slope of a plot of log c r versus log r .Chen and Yau 18 showed that the correlation dimension represents the lower bound to the capacity or fractal dimension d c and approaches its value asymptotically when the attracting set is distributed more uniformly in the embedding phase space.A set of points in the embedding space is said to be fractal if its dimension has a finite noninteger value.Otherwise, the attractor is referred to as a "strange attractor."To establish the nature of the attractor, the embedding dimension is progressively increased, causing the slope of the characteristic curve to approach a steadystate value.This value is then used to determine whether the system has a fractal structure or a strange attractor structure.If the dimension of the system flow is found to be fractal i.e. to have a noninteger value , the system is judged to be chaotic.
In the current study, the attractors in the embedding space were constructed using a total of 60000 data points taken from the time series corresponding to the displacement of the system.Via a process of trial and error, the optimum delay time when constructing the time series was found to correspond to one third of a revolution of the system.The reconstructed attractors were placed in embedding spaces with dimensions of n 2, 4, 6, 8, 10, 12, 14, 16, 18, and 20, respectively, yielding 10 different log c r versus log r plots for each attractor.The number of data points chosen for embedding purposes i.e., 60000 reflects the need for a compromise between the computation time and the accuracy of the results.In accordance with Grassberger and Procaccia 19 , the number of points used to estimate the intrinsic dimension of the attracting set in the current analysis is less than 42 M , where M is the greatest integer value less than the fractal dimension of the attracting set.

Numerical Results and Discussions
The nonlinear dynamic equations presented in 2.14 to 2.15 for the HSFD rotor-bearing system with strongly nonlinear oil-film force and nonlinear rub-impact force were solved using the fourth-order Runge-Kutta method.The time step in the iterative solution procedure was assigned a value of π/300, and the termination criterion was specified as an error tolerance of less than 0.0001.The time series data corresponding to the first 800 revolutions of the rotor was deliberately excluded from the dynamic analysis to ensure that the analyzed data related to steady-state conditions.The sampled data were used to generate the dynamic trajectories, Poincaré maps, and bifurcation diagrams of the spur rotor system in order to obtain a basic understanding of its dynamic behavior.The maximum Lyapunov exponent and the fractal dimension measure were then used to identify the onset of chaotic motion.The rotating speed ratio s is one of the most significant and commonly used as a control parameter in analyzing dynamic characteristics of bearing systems.Accordingly, the dynamic behavior of the current rotor-bearing system was examined using the dimensionless rotating speed ratio s as a bifurcation control parameter.The bifurcation diagram in Figure 3 shows the long-term values of the rotational angle, plotted with rotor displacement against the dimensionless speed s without rub-impact effect.Qualitatively different behavior was observed at values of s within the range of 0 < s < 5.It can be seen that the dynamic motion of rotor trajectory in low speed is T-periodic motion both in X and Y directions, and it drops to a lower spatial displacement mode at the speed s 2.3.As the speed is increased, the T-period motion loses its stability at s 2.52, and a 2Tperiodic motion starts to build up.The jump phenomenon is also occurred under 2T -periodic motion at s 2.7.As the speed is further increased, the 2T -periodic motion loses its stability at s 2.82, and a T -periodic motion suddenly appears.The rotor trajectory, the Poincaré map, and the displacement power spectrum in the X and Y directions at s 2.6 are given in Figure 4, from which the 0.5-subharmonic motion is shown by the double loops of the rotor trajectory, two discrete points in the Poincaré map and peaks at 0.5 in the power spectrum.The pressure distributions in the four oil chambers are shown in Figure 5.It can be seen that the variations of pressure distributions are periodic, and the period is the same with the rotor trajectory.Figures 6 a and 6 b show the bifurcation diagrams for the rotor displacement against the dimensionless rotating speed ratio with rub-impact effect.Compared with bifurcation results without rub-impact effect, bifurcation results with rub-impact effect show that dynamic trajectories perform strongly nonperiodic at low rotating speeds, but it would escape nonperiodic motions to periodic motions.The bifurcation diagrams show that the geometric centers of rotor in the horizontal and vertical directions perform nonperiodic motion or the so-called chaotic motion at low values of the rotating speed ratio, that is, s < 0.61.Figures 7, 8, and 9 represent phase diagrams, power spectra, Poincaré maps, Lyapunov exponents, and the fractal dimensions of pinion center with s 0.32, 0.36, and s 0.42, respectively.The simulation results show that phase diagrams show disordered dynamic behaviors with s 0.32, 0.36 and s 0.42; power spectra reveal numerous excitation frequencies; the return points in the Poincaré maps form some geometrically fractal structures, but the maximum Lyapunov exponent is positive with s 0.36 and maximum Lyapunov exponent is negative with s 0.32 and s 0.42.Thus, the results show that the dynamic trajectory performs chaotic motion with s 0.36, but they present no chaotic motions with s 0.32 and s 0.42.Figures 10 and 11 are phase diagrams and Poincaré maps for the route of subharmonic motion into chaos, out of chaos to periodic response at different rotating speed ratios of s with rub-impact effect .Unlike the usual ways into chaos 1T ⇒ 2T ⇒ 4T ⇒ 8T ⇒ 16T ⇒ 32T • • • ⇒ chaos or periodic ⇒ quasi-periodic ⇒ chaotic , it suddenly gets in chaos from the periodic motion without any transition or suddenly escape from irregular motions into periodic motions in accordance with phase diagrams and Poincaré maps.

Conclusions
A hybrid squeeze-film damper mounted rotor-bearing system with nonlinear oil-film force and nonlinear rub-impact force has been presented and studied by a numerical analysis        Pressure in the static pressure chamber R: Inner radius of the bearing housing r: Radius of the journal.r, t: Radial and tangent coordinates s: Speed parameter ω/ω n U: ρ/δ x, y, z: Horizontal, vertical and axial coordinates x 0 , y 0 : Damper static displacements X, Y, X 0 , Y 0 : x/δ, y/δ, x 0 /δ, y 0 /δ ρ: Mass eccentricity of the rotor φ: Rotational angle φ ωt ω: Rotational Derivatives with respect to t and φ.
et al. 1 published a paper dealing with aperiodic behavior in journal bearings and what may very well have been the first paper about aperiodic behavior in journal bearing systems.Nikolajsent and Holmes 2 reported their observation of

Figure 3 :
Figure 3: Bifurcation diagram of X nT a and Y nT b versus rotor speed s without rub-impact effect .

Figure 4 :
Figure 4: Subharmonic motion at s 2.6 case 1 ; a Rotor trajectory; b Poincaré map; c and d Displacement power in X and Y directions without rub-impact effect .

Figure 5 :
Figure 5: Pressure distribution in the static pressure chamber at s 2.6 without rub-impact effect .

Figure 6 :
Figure 6: Bifurcation diagram of HSFD rotor-bearing system using dimensionless rotating speed coefficient, s, as bifurcation parameter with rub-impact effect .

Figure 7 :
Figure 7: Simulation results obtained for rotor-bearing system with s 0.32 with rub-impact effect .

Figure 8 :
Figure 8: Simulation results obtained for rotor-bearing system with s 0.36 with rub-impact effect .

Figure 9 :
Figure 9: Simulation results obtained for rotor-bearing system with s 0.42 with rub-impact effect .
Damper eccentricity εδ f x , f y : Components of the fluid film force in horizontal and vertical coordinates F r , F τ : Components of the fluid film force in radial and tangential directions h: Oil film thickness, h δ 1 ε cos θ k: S t i ffness of the retaining springs k d : Proportional gain of PD controller k p : Derivative gain of PD controller L: Bearing length m: Masses lumped at the rotor mid-point O m : Center of rotor gravity O b , O j : Geometric center of the bearing and journal p θ : Pressure distribution in the fluid film p s : Pressure of supplying oil p c,i : speed of the shaft ϕ b : Angle displacement of line O b O j from the x-coordinate see Figure 1 Ω: φb δ: Radial clearance R − r, θ: The angular position along the oil film from line O 1 O 3 see Figure 1 μ: Distribution angle of static pressure region • , :