Application of Sensitivity Analysis to the Development of High Performance Adaptive Hydraulic Engine Mounts

Abstract

In this paper, the sensitivity analysis is applied to the development of high performance adaptive hydraulic mounts. The analysis allows us to select the most effective design parameters for tuning an adaptive mount to different road and engine conditions. It is shown that in the low frequency road excitation, the upper chamber compliance and inertia of the fluid column in the inertia track are the most influential properties in changing the dynamic stiffness of the hydraulic mount. These properties for the high frequency engine excitations are the upper compliance and the inertia of the fluid column of the decoupler. For tuning the adaptive mount to different road and engine excitation, a global optimization technique is used to find the magnitude of the adjusting parameters to minimize objective functions in low and high frequency excitations. The results indicate significant improvement over conventional hydraulic mounts. It is further shown that when the upper compliance is used as the adjusting parameter, a simple on-off control which is triggered by the engine revolution and vehicle speed is sufficient for tuning the adaptive mount.

Transient Response of Hydraulic Engine Mount to a Realistic Excitation: Improved Non-Linear Models and Validation

“…Jazar and Golnaraghi [7] proposed a simple non-linear mathematical model to characterize the decoupler resistance in terms of the Duffing's equation (continuous non-linearity). Foumani et al [8] conducted a sensitivity analysis and concluded that C 1 and I i are the most influential parameters in the dynamic stiffness model over the lower frequency range. Tiwari et al [9] refined the bench experiments [1] and further quantified C 1 and C 2 under several F m .…”

“…2(a) is commonly applied in dynamic stiffness measurement [1][2][3][4][5][6][7][8], where the constant x m (corresponding to a specific F m ) is usually neglected in the analysis, thus leaving only the sinusoidal component x(t) = X⋅sin(2πft + φ). In steady state elastomer tests, the dynamic stiffness K(f,X) is evaluated only at the frequency of excitation f and superharmonics are ignored [1][2][3][4][5][6][7][8]. A quasi-linear model could also be estimated from the K(f,X) data [11].…”

“…By using p 1 (t) as an indicator of the operating condition, the multi-staged C 1 (p 1 ) non-linearity is mathematically formulation by Eq. (8). For mount D, the empirical coefficients of Eq.…”

“…Constructed using convoluted thin rubber membranes [9], the bottom chamber is intentionally designed to yield a large C 2 to accommodate fluid displaced from the top chamber. Due to the difficulty of analytically calculating C 2 , most researchers [1][2][3][4][5][8][9][10] assumed a linearized (constant) C 2 value, which could be curve-fitted from measurements at a certain operating point. For instance, a nominal value C 20 of 2.4×10 -9 m 5 /N was measured for mount D and it seems to work well for triangular excitations under a preload F m of 1200 N. Note that F m (or x m ) plays a pivot role on the chamber compliances [2,[9][10]: First, F m determines the operating point about which the non-linear compliances are estimated; second, F m dictates the mean fluid pressure (under the static equilibrium).…”

“…Hydraulic engine mount is designed to be highly nonlinear as its parameters, such as stiffness and damping parameters, significantly vary with the amplitude (X) and frequency (f) of sinusoidal excitation [1][2][3][4][5][6][7][8][9][10][11][12][13][14][15]. Over the past two decades, extensive experimental and analytical studies have been conducted on the non-linear characterization under steady state condition [1][2][3][4][5][6][7][8], but their non-linear behaviors under transient or arbitrary loading excitations (say with multiple frequency components) are not well understood. To better illustrate the full extent of the underlying research issues, consider the fluid system of Fig.…”