Vortrag: Professor Ir. Dr. Shahrum Abdullah

Signal Processing Techniques For Fatigue Life Analysis, Monday, November 25, 2019, 14:00 Room 116, Institute for Risk and Reliability, LUH

Summary

This work focused on the capability of time-frequency signal processing techniques in determining the important features of fatigue life assessment using the variable amplitude strain loadings. Generally, the fatigue damage phenomenon can be studied in the condition of before and after the component or structural failure. Towards the implementation of integrity assessment to the needs of avoiding structural failure, therefore, it is vital to put more efforts on the signal-based life assessment before failure occurrence. It motivates to the exploration  of possible signal processing approach that can deeply analyse the fatigue features. Three domains of signal analysis can be highlighted for the interest of fatigue life assessment, i.e. the time domain, the frequency domain and the time-frequency domain. The traditional time domain technique for fatigue life analysis are restricted to only time information, hence, the complexity of analysis is also limited. Similar trend can also be seen in the frequency domain, where the main difference lies in the frequency distribution instead of the time range. For complexity of analysis, the time-frequency techniques can reveal valuable information in the scope of the simultaneous time-frequency localisation that is related to durability indicators. Three types of time-frequency localisation techniques were used, i.e. the continuous wavelet transform (CWT), discrete wavelet transform (DWT) and Hilbert-Huang transform (HHT), associated to the strain data analysis, hence, to assess the fatigue life of automotive component. The random data or signals were measured on different types of road surface, such as bumpy, highway, rural and campus road.

For example, the CWT was used for extracting singularities that represent a low frequency energy of signal while eliminating high frequency noises. In addition, the use of 4th order of Daubechies family in the DWT technique managed to decompose the respective signals into 13 levels based on their own frequency bands from the previous levels.  Meanwhile, the HHT produced eight signals of intrinsic mode function (IMF) and a residue by eliminating the high frequency signal behaviour. In conclusion, these time-frequency domain techniques revealed extra information related to fatigue damage from loading histories and offered other alternative for fatigue life assessment. The proposed techniques based on the time-frequency domain can offer alternative solution to the traditional time- and frequency- domain methods at more effective fatigue life prediction for automotive component.

 

 

Brief Biodata

In 2011, Shahrum Abdullah was appointed as the Professor in the Programme of Mechanical Engineering, Faculty of Engineering & Built Environment, Universiti Kebangsaan Malaysia, Bangi, Selangor. Back to his educational background, he received his B.Eng. in Mechanical Engineering (Hons, First Class) (1995) from Universiti Kebangsaan Malaysia (Malaysia), MSc in Engineering Design (1997) from Loughborough University (United Kingdom) and PhD in Mechanical Engineering (2005) from The University of Sheffield (United Kingdom). The research interests of Professor Abdullah are fatigue life assessment, durability, signal analysis and engineering design. He does more effort in evaluating the effectiveness of the signal processing parameters for determining fatigue life characteristic of materials and structures.

Since 2006, he obtained almost the total of RM6 million (Euro 1.3 million) funding as the project leader under various research grants in Malaysia. In addition, he has published more than 400 papers in highly indexed journals (SCOPUS, WoS) and conference proceedings, was ranked at #8 among the UKM researchers with highest numbers of publications in Web of Science (WoS) database for the year 2018.

Professor Abdullah was appointed as the Guest Professor in Universitaet Duisburg-Essen (Germany) for the winter semester 2012/2013 (October 2012 until March 2013), as well as the visiting professor in three universities in Japan. Within the period of July-September 2018, he was seconded as the research fellow at Liverpool John Moores University (United Kingdom). On another note, he committed in the involvement in several international conferences as the conference chairs, track chairs, organizing committee member, advisory committee member, etc., particularly for the national and international conferences related to fatigue and durability research. Finally, he is the founder of two established mechanical engineering conferences in Malaysia, i.e. International Conference on Recent Advances in Automotive Engineering (ReCARTM) and Symposium on Damage Mechanisms in Materials and Structures (SDMMS).

 

Professor Ir. Dr. Shahrum Abdullah

Program of Mechanical Engineering

Faculty of Engineering and Built Environment

Universiti Kebangsaan Malaysia

43600 UKM Bangi, Selangor

MALAYSIA

 

(shahrum@ukm.edu.my)