Please use this identifier to cite or link to this item:
https://ptsldigital.ukm.my/jspui/handle/123456789/519705| Title: | Establishment of mathematical model for fatigue life assessment of an automotive coil spring under random strain loads |
| Authors: | Reza Manouchehrynia (P84483) |
| Supervisor: | Shahrum Abdullah, Prof. Ir. Dr. |
| Keywords: | Universiti Kebangsaan Malaysia -- Dissertations Dissertations, Academic -- Malaysia Fatigue life assessment Automotive coil spring Strain loads |
| Issue Date: | 8-Dec-2021 |
| Description: | This research aims at establishing a mathematical model based on fatigue life assessment using a probabilistic technique for the strain-time history signals of automobile coil spring. A probabilistic model for fatigue data must be used to evaluate automotive suspension systems which fail due to unforeseen loads and different road conditions, since most fatigue data is deterministic. The objective of this study is to formulate a mathematical model using fatigue reliability assessment based on probabilistic modelling relationships for strain-life models. The aim is to establish the relationship between strain data and fatigue life prediction for vehicle coil spring durability assessment through the proposed probabilistic modelling due to non- stationary and non-Gaussian distributions of fatigue data. The hypothesis establishes a probabilistic mathematical model that is appropriate for any strain-life models. Strain gauge was mounted on coil surface to collect strain signals using a data acquisition system to evaluate the durability of spring relative to different road surfaces. Fatigue data is edited based on 100%, 95% and 90% retained damage in order to obtain shorter signals in an attempt to accelerate the fatigue life assessment that has no adverse effect on the original features by removing low-amplitude cyclic loads. Finite element analysis was performed to determine hotspots at the critical area of coil spring by mounting strain gauge properly. Simulation based modal analysis was carried out to evaluate strain amplitude of spring in each mode. Maximum likelihood estimation was used to determine location and scale parameters of the Gumbel model to constitute the proposed model. The Akaike information criterion method was considered to determine which model should estimate the most suitable fit parameters. Based on the fatigue data editing results, the highest fatigue life was obtained at 4.22×107 blocks to failure for the original highway road data by employing the Coffin-Manson model. Meanwhile, the lowest fatigue life was predicted for 90% retained damage rural road data using the SmithWatson-Topper model at 2.28×104 blocks to failure. Furthermore, the highest life was obtained for mode 1 at 2.55×108 blocks to failure for the Dirlik method. Whereas, mode 3 was shown the lowest life at 1.00×106 blocks to failure for narrowband method. The proposed mathematical model displayed a safe life for the Smith-Watson-Topper life model. Furthermore, the Coffin-Manson and Morrow models demonstrated a conservative probabilistic for rural, campus and highway roads. Therefore, for the highest correlation curve, R 2 values of 0.9998, 0.9997 and 0.9993, with the lowest rootmean-square error values of 0.0233, 0.0267 and 0.0417 for highway, campus and rural roads according to the Smith-Watson-Topper life model were calculated. Moreover, all data points were safe with the 100% survivability for these three road profiles. All lifetime data was almost fitted within the boundary of 1:2 and 2:1 lines for these three strain-life models. It implies that the proposed model and measured strain signals provide very good fatigue life prediction. A novel model is established to evaluate strain-life fatigue curves based on the proposed probabilistic mathematical model where fatigue results were non-Gaussian and non-stationary distribution. The generated prediction relationships are the novel and significant contribution to fatigue reliability assessment. Ph.D |
| Pages: | 260 |
| Publisher: | UKM, Bangi |
| Appears in Collections: | Faculty of Engineering and Built Environment / Fakulti Kejuruteraan dan Alam Bina |
Files in This Item:
| File | Description | Size | Format | |
|---|---|---|---|---|
| Establishment of mathematical model for fatigue life.pdf Restricted Access | 13.53 MB | Adobe PDF |  View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.