Mi 180235
Mi 180235
Mi 180235
REGULAR ARTICLE
Abstract. This paper presents the optimization of spring fatigue life associated with suspension system
parameters using the design of experiment approach. The effects of suspension parameters on spring fatigue life
were analyzed because this process can improve spring fatigue life from a distinct perspective. A quarter car
model simulation was performed to obtain the force time histories for fatigue life prediction where the suspension
parameters were adjusted. Multiple input regression and interaction plots were conducted to identify the
interaction between these parameters. A full factorial experiment was performed to determine the optimal
suspension settings that would maximize the spring fatigue life. For the regression, a high R2 value of 0.9078 was
obtained, indicating good fitting. The established regression showed normality and homoscedasticity for
consistent prediction outcome. Reducing the spring stiffness and sprung mass while enhancing the damping
coefficient is therefore suggested to enhance fatigue life.
Keywords: Fatigue life / multiple input regression / design of experiment / automotive suspension /
finite element analysis
® ®
Fig. 4. Setup of fatigue life prediction using nCode DesignLife .
considered mean stress effects [29]. Hence, the SWT model process determined the optimal suspension parameters
was selected to perform the fatigue life analysis. when the objective function to maximize the spring fatigue
To expand the datasets, the suspension parameters life was assigned the constraints, as listed in Table 1.
were adjusted according to the proposed frequency range of
1–1.5 Hz [30]. After the parameters were adjusted, the
simulation was re-run to obtain the force time histories 3 Results and discussions
because the dynamic behavior of the model was changed.
The force time histories were then used to predict the The preliminary analysis of this research aimed to generate
spring fatigue life. After all the required data were the spring fatigue life prediction according to different
obtained, a regression analysis was conducted to study suspension parameters. Hence, the force loading signals
the effects and interaction of the parameters toward the under the influence of the suspension parameters were
response. Subsequently, eight axial points and three center simulated from the quarter car model, as shown in Figure 5.
points were added, for a total of 49 runs. The optimization These force signals were used to predict fatigue life together
Y.S. Kong et al.: Mechanics & Industry 20, 621 (2019) 5
Fig. 6. Fatigue life contour of original spring. Fig. 7. Pareto chart of standardized effects.
with the FEA analysis of the spring and material cyclic Table 3. Percentage variations of suspension parameters
properties. An example of spring fatigue life contour where and interactions.
the localized fatigue life was obtained is shown in Figure 6.
A spring sensitivity analysis was performed to determine Factor Percentage variation (%)
the effects of different spring designs in terms of fatigue life.
The spring sensitivity analysis was performed through B 33.68
adjusting the spring bar diameter. C 15.46
A total of 49 datasets were used in this analysis to D 11.80
determine the optimal solution of the suspension param- BD 7.45
eters for spring fatigue life. All effects of suspension BB 5.96
parameters were analyzed using a Pareto chart because the
BC 4.00
goal was to maximize the spring fatigue life, as shown in
Figure 7. The parameters, namely, spring stiffness CD 2.34
(K_stiff), tire stiffness (T_stiff), damping coefficient
(C_damp), sprung mass (Mass), and interactions between
parameters, were then analyzed. According to Figure 6, the spring fatigue life. Hence, tire stiffness was excluded from
parameters B, C, D, BD, BB, BC, and CD were related to the regression analysis.
spring fatigue life at the 0.1 level of significance. The significance level of each parameter, explained by
Meanwhile, the parameters of AC and below (red dot variation percentage after the less significant parameter
line) did not show significant effects, indicating that the was removed, is shown in Table 3. The table shows that the
tire stiffness had minimal effect on spring fatigue life. A flat designated spring stiffness had the highest variation
line was also obtained for the spring fatigue and tire (33.68%) on the spring fatigue life because the fatigue life
stiffness plots [31], indicating no direct relationship of a spring depends on the geometry, which is derived on
between these two parameters as shown in Figure 8a. the basis of the spring stiffness. The relationship between
The changes in tire vertical stiffness were not linear to the spring stiffness and fatigue life, where the fatigue life of the
6 Y.S. Kong et al.: Mechanics & Industry 20, 621 (2019)
Fig. 9. Response surface plot of fatigue life for different parameters: (a) spring stiffness and sprung mass, (b) spring stiffness and
damping coefficients, and (c) sprung mass and damping coefficients.
spring reduced with every increment of spring stiffness, is spring stiffness and sprung mass, a response surface plot for
plotted in Figure 8b. The second important parameter was fatigue life is shown in Figure 9. This condition could be
the damping coefficient, which also affected the spring explained using an interaction plot, which described how
fatigue life [32]. The reason was that the damper transforms the mean of fatigue life changed when the two other
shock from road excitations into heat and dissipates it. relevant factors were altered. In Figure 10, the interaction
Therefore, the damping properties play a critical role in plots for spring stiffness and sprung mass indicate that an
determining the fatigue behavior of automotive compo- increment in spring stiffness reduced the fatigue life. When
nents, including the spring. the sprung mass was increased, the fatigue life was further
The damping coefficients showed a positive increasing reduced, but the magnitude was small. The interaction of
relationship toward the spring fatigue life, as illustrated in spring stiffness and damping coefficient was then consid-
Figure 8c. Apart from spring stiffness and damping ered, as shown in Figure 9b. According to the interaction
coefficients, the sprung mass was an important parameter plot, increasing the damping could enhance the spring
which affected the spring fatigue life. Kamal and Rahman fatigue life where an apparent increment of spring fatigue
[33] proposed that the sprung mass changed the equilibri- life was illustrated. According to the damping coefficient
um position where the spring deformed to counter the road and sprung mass interaction plot, the enhancement of
disturbances. The effects of sprung mass were clear because fatigue life through increasing damping coefficient could be
it had direct effects on the spring which was designed to further improved using reduced sprung mass. Regarding
support heavy loads. When the sprung mass was high, the the design aspect, the interaction plots between parameters
load exerted on the spring was also high, thereby reducing provided important information when the combination
the spring fatigue life. Hence, a negative relationship for effects of parameters toward response were required.
spring fatigue life and sprung mass was obtained, as shown Subsequently, multiple regression modeling was per-
in Figure 8d. Subsequently, the interactions between formed for the parameters, with spring stiffness, damping
parameters toward spring fatigue life were also investigat- coefficients, and vehicle sprung mass as the independent
ed because the combination of every parameters led to variables and spring fatigue life as the dependent variable.
varying effects on spring fatigue life. The fatigue model was obtained as follows:
Table 3 shows that the interactions between spring
N f ¼ 17:25 0:00063 K _stiff þ 0:00438 C _damp
stiffness and sprung mass had only a small percentage of
variation. To further investigate the relationship between 0:04048 Mass þ 0:00002C _damp Mass: ð2Þ
Y.S. Kong et al.: Mechanics & Industry 20, 621 (2019) 7
References
Fig. 15. Optimized fatigue life of the spring. [1] E.S. Palma, E.S. dos Santos, Fatigue damage analysis in an
automobile stabilizer bar, Proc. Inst. Mech. Eng. D 216,
865–871 (2002)
[2] S. Zhang, A. Chen, A practical design process to optimize
spring fatigue life could be maximized. The predicted fatigue performance for chassis components, SAE Technical
optimized fatigue life was 7,012,938 blocks to failure, and Paper 2009-01-1227, 2009
the fatigue contour of the spring is shown in Figure 15. The [3] A.S. Sener, Determination of vehicle components fatigue life
lowest fatigue occurred at the inner surface, where the based on FEA method and experimental analysis, Int. J.
crack was initiated, and propagated toward the outer Electr. Mech. Mechatr. Eng. 2, 133–145 (2011)
[4] C.K. Clarke, G.E. Borowski, Evaluation of a leaf spring
surface [40]. To estimate and reduce the uncertainties,
failure, J. Fail. Anal. Prevent. 5, 54–63 (2005)
2 replicates were used to increase accuracy and variability
[5] C. Iyidiker, I. Cokal, A. Yilmaz, N. Kilic, Fatigue life
of estimates. The 95% confidence interval is the range of prediction of a bus body structure using CAE tools, FISITA
likely values for the mean fatigue life which between 2010 World Automotive Congress, Italy, 319–329, 2010
3,162,278 and 12,589,254 blocks to failure. [6] Y.S. Kong, S. Abdullah, D. Schramm, M.Z. Omar, S.M.
When observing the parameter sensitivity, the fatigue Haris, T. Bruckmann, Mission profiling of road data
life could be enhanced through reducing the spring stiffness measurement for coil spring fatigue life, Measurement
and sprung mass. This analysis was significant in 107, 99–110 (2017)
determining the suspension parameter configuration that [7] T.E. Putra, Husaini, Investigating the road surface effect to
led to the optimal spring fatigue life. A good spring design the fatigue life of an automotive coil spring, Mater. Sci. Eng.
results in sufficiently long fatigue life, but current coil 352, 012016 (2018)
spring design always depends solely on the spring geometry [8] M. Buciumeanu, A.S. Miranda, A.C.M. Pinho, F.S. Silva,
and materials. This analysis proposes the effects of Design improvement of an automotive-formed suspension
suspension parameter toward the spring fatigue life and component subjected to fretting fatigue, Eng. Fail. Anal. 14,
thus serves as guidance for suspension designers. In terms 810–821 (2017)
of vehicle dynamics, engineers have attempted to maximize [9] Z.M. Zhang, C. Sun, R. Bridgelall, M. Sun, Application of
vehicle ride comfort through reducing the spring stiffness machine learning method to evaluate road roughness
and increasing sprung mass and damping coefficients. from connected vehicles, J. Transp. Eng. B 144, 04018043
However, the effects of changing the spring stiffness and (2018)
damping coefficients toward spring’s fatigue life are [10] M.A. Aly, A.O. Nassef, K. Hamza, Multi-objective design of
significant but have not been conducted. Through utilizing vehicle suspension systems via a local diffusion genetic
algorithm for disjoint pareto frontiers, Eng. Optim. 47,
the developed multiple regression, spring fatigue life can be
706–717 (2015)
predicted under constrained suspension parameters.
[11] K.R. Kashyzadeh, M.J. Ostad-Ahmad-Ghorabi, A. Arghavan,
Fatigue life prediction of package of suspension automotive
under random vibration based on road roughness, Mediter.
4 Conclusion J. Model. Simul. 4, 37–50 (2015)
[12] A. Saoudi, M. Bouazara, D. Marceau, Study of the fatigue life
In this study, design of experiments was developed on the and weight optimization of an automobile aluminium alloy
basis of SWT strain–life models to predict the fatigue life of part under random road excitation, Shock Vib. 17, 107–121
an automotive coil spring using suspension parameters, (2010)
namely, spring stiffness, damping coefficients, and sprung [13] M. Kamal, M. Mustafizur Rahman, M.S.M. Sani, Applica-
mass. The interactions between parameters were exam- tion of multibody simulation for fatigue life estimation, Int.
ined, and the spring stiffness showed the largest influence J. Autom. Mech. Eng. 7, 912–923 (2013)
10 Y.S. Kong et al.: Mechanics & Industry 20, 621 (2019)
[14] M.C. Levesley, S.A. Kember, D.C. Barton, P.C. Brooks, air-ride suspension-based weight estimation, 2014 ASABE
O.M. Querin, Dynamic simulation of vehicle suspension and CSBE/SCG AB Annual International Meeting, Canada,
systems for durability analysis, Mater. Sci. Forum 440–441, 141897468, 2014
103–110 (2003) [28] B. He, S. Wang, F. Gao, Failure analysis of an automobile
[15] B. Li, H. Yuan, T. Zhao, G. Wang, Research on dynamic damper spring tower, Eng. Fail. Anal. 17, 498–505 (2010)
characteristics and fatigue robust optimization of integrated [29] A. Ince, G. Glinka, A modification of Morrow and Smith-
vehicle model, Proc. Inst. Mech. Eng. D 1–18 (2018) Watson-Topper stress correction models, Fatig. Fract. Eng.
[16] S.J. Heo, D.O. Kang, J.H. Lee, I.H. Kim, S.M.H. Darwish, Mater. Struct. 34, 854–867 (2011)
Shape optimization of lower control arm considering multi- [30] B. Lacroix, P. Seers, Z. Liu, A passive nonlinear damping
disciplinary constraint condition by using progress meta- design for a road race car application, SAE technical paper
model method, Int. J. Autom. Technol. 14, 499–505 series 2006-01-1984, 2006
(2013) [31] F.B. Grosselle, G. Timelli, F. Bonollo, Doe applied to
[17] J. Fang, Y. Gao, G. Sun, C. Xu, Q. Li, Multiobjective robust microstructural and mechanical properties of Al-Si-Cu-Mg
design optimization of fatigue life for a truck cab, Reliab. casting alloys for automotive application, Mater. Sci. Eng. A
Eng. Syst. Saf. 135, 1–8 (2015) 527, 3536–3545 (2010)
[18] X.G. Song, J.H. Jung, H.J. Son, J.H. Park, K.H. Lee, Y.C. [32] B. Breytenbach, P.S. Els, Optimal vehicle suspension
Park, Metamodel-based optimization of a control arm characteristics for increased structural fatigue life, J.
considering strength and durability performance, Comput. Terramech. 48, 397–408 (2011)
Math. Appl. 60, 976–980 (2010) [33] M. Kamal, M.M. Rahman, Study on dynamic behaviour of
[19] J.K. Kim, Y.J. Kim, W.H. Yang, Y.C. Park, K.H. Lee, Wishbone suspension system, Mater. Sci. Eng. 36, 012019
Structural design of an outer tie rod for a passenger car, Int. (2012)
J. Autom. Technol. 12, 375 (2011) [34] J. Gerretzen, E. Szymanska, J.J. Jansen, J. Bart, H.J. van
[20] T.H. Lee, J.J. Jung, Metamodel-based shape optimization of Manen, E.R. van de Heuvel, L.M.C. Buydens, Simple and
connecting rod considering fatigue life, Key Eng. Mater. effective way for data processing selection based on design of
306–308, 211–216 (2006) experiments, Anal. Chem. 87, 12096–12103 (2015)
[21] C. Schäfer, E. Finke, Shape optimization by design of [35] M.K. Assadi, S.M. Zahraee, J. Taghdisi, Integration of
experiments and finite element methods an application of computer simulation, design of experiments and particle
steel wheels, Struct. Multidiscipl. Optim. 36, 477–491 (2008) swarm optimisation to optimise the production line efficiency,
[22] A.C. Mitra, T. Soni, G.R. Kiranchand, Optimization of Int. J. Swarm Intell. Evol. Comput. 5, 1000136 (2016)
automotive suspension by design of experiments: a nonde- [36] P. Sivák, E. Ostertagová, Evaluation of fatigue tests by
rivative method, Adv. Acoust. Vib. 2016, 3259026 (2016) means of mathematical statistics, Proc. Eng. 48, 636–642
[23] A.F. Neto, A.F.B. Costa, M.F. Lima, Use of factorial designs (2012)
and the response surface methodology to optimize a heat [37] N.A. Wahir, M.E. Nor, M.S. Rusiman, K. Gopal, Treatment
staking process, Exp. Tech. 42, 319 (2018) of outliers via interpolation method with neural network
[24] D.C. Montgomery, Design and analysis of experiments, John forecast performances, J. Phys. 995, 012025 (2018)
Wiley & Sons, U.S.A, 2009 [38] C. Leys, O. Klein, Y. Dominicy, C. Ley, Detecting
[25] P.Y. Mhaske, P.A. Narwade, M.P. Nagarkar, Stiffness multivariate outliers: Use a robust variant of the Mahala-
analysis of passenger car tire using nitrogen, Int. J. Sci. Res. nobis distance, J. Exp. Soc. Psychol. 74, 150–156 (2018)
Publ. 6, 140–148 (2016) [39] P. Li, J. Lam, K.C. Cheung, Control of vehicle suspension
[26] T.E. Putra, S. Abdullah, D. Schramm, M.Z. Nuawi, T. using an adaptive inerter, Proc. Inst. Mech. Eng. D 229,
Bruckmann, Generating strain signals under consideration 1934–1943 (2015)
of road surface profile, Mech. Syst. Signal Proces. 60–61, [40] L. Del Llano-Vizcaya, C. Rubio-González, G. Mesmacque, T.
485–497 (2015) Cervantes-Hernández, Multiaxial fatigue and failure
[27] A.D. Balmos, A.D. Layton, A. Aaron, J.V. Krogmeier, D.R. analysis of helical compression springs, Eng. Fail. Anal.
Buckmaster, Toward understanding the errors in online 13, 1303–1313 (2006)
Cite this article as: Y.S. Kong, S. Abdullah, D. Schramm, S.S.K. Singh, Determining optimal suspension system parameters for
spring fatigue life using design of experiment, Mechanics & Industry 20, 621 (2019)