M.Sc Department of Mechanical and Materials Engineering Universiti Kebangsaan Malaysia, Selangor, Malaysia 2008-2010
B.Eng Department of Mechanical Engineering Universitas Syiah Kuala, Banda Aceh, Indonesia 1999-2005 Phone: +6285275155307 Address: Department of Mechanical Engineering Faculty of Engineering Universitas Syiah Kuala Jalan Syech Abdul Rauf no.7 Darussalam Banda Aceh 23111 Aceh - Indonesia
This study is performed to develop fatigue durability feature extraction and clustering algorithm... more This study is performed to develop fatigue durability feature extraction and clustering algorithms using the wavelet transform, in order to correlate the fatigue damage and the signal internal energy. Strain signals measured from a car lower suspension arm were used as input data for the simulation purposes. The data were obtained from a fatigue data acquisition experiment by involving car movements on different road surfaces, i.e. pavé, highway and UKM campus, at constant speeds of 20 to 70 km/h. The data were recorded using strain gauges and a data logger instrument. This study also used a random strain signal from Society of Automotive Engineers-owned fatigue strain data set, i.e. SAESUS. All data were analysed using the wavelet transform. The mother wavelet used was the Morlet since it resulted in an internal energy plot that was similar to the fatigue damage, so it could be applied to detect the locations of the signals that have high fatigue damage. The energy was etermined based on the Morlet wavelet coefficient. Furthermore, the highly fatigue damaging segments in the signals were extracted based on the high energy distribution in order to result segments which contribute to fatigue damage. In the extraction process, energy gate value was used as the parameter to determine the minimum magnitude value to be retained. These retained segments were attached together to form a shorter edited signal for the validation purpose. From the results obtained, it was found that the algorithm could shorten the signal by 74.3% and maintained the original statistical parameter and cumulative fatigue damage values by more than 90%. It ensured that the produced edited signals could maintain the original characteristics of the signals. In the later part of this paper, the segments were clustered into three groups in order to identify the segment scattering pattern. According to the findings of clustering using the k-means method, scatter plot was produced based on the relationship between the fatigue damage and its corresponding signal internal energy. Segments with higher energy presented higher fatigue damage, while segments with smaller energy represented smaller fatigue damage. It indicated that the relationship between the fatigue damage and the signal internal energy was strong and parallel, i.e. about 86.18%, and it was depended on the road types. This correlation could be used to predict fatigue failures. This algorithm could produce a more suitable method of segment-by-segment analysis for fatigue strain signal extraction and clustering. Finally, this algorithm is recommended as an alternative method in durability fatigue studies especially in the field of automotive engineering
Gaya luar yang bekerja pada Truss Tribun utama stadion Harapan Bangsa Banda Aceh terdiri dari ber... more Gaya luar yang bekerja pada Truss Tribun utama stadion Harapan Bangsa Banda Aceh terdiri dari berat atap, berat kuda-kuda, berat talang, berat ventilasi udara, berat rangka, beban angin serta reaksi tumpuan. Untuk mengetahui kemampuan Truss dalam menumpu semua berat yang bekerja, maka perlu dilakukan analisa kekuatan dari Truss itu sendiri. Dari hasil perhitungan Truss tiga dimensi dengan menggunakan software SAP 2000 versi 7.42, maka diperoleh gaya batang tarik maksimum Ft = 63.005,26 N dan gaya batang tekan maksimum Fc = -75.532,95 N, dengan perpindahan maksimum U = -0,0136 m. Selanjutnya diperoleh tegangan tarik yang terjadi σt = 26,753 MPa dan tegangan tekan yang terjadi σc = -32,073 MPa, dimana σt < σta dan σc < σcr dan σca, sehingga konstruksi Truss tersebut baik
This paper presents scattering of fatigue time series data to develop a relationship between the ... more This paper presents scattering of fatigue time series data to develop a relationship between the fatigue damage and the Morlet wavelet coefficient. In order to achieve the purpose, the SAESUS strain signal was inputted into a segmentation algorithm based on the wavelet coefficient amplitude level which produces 49 segments that each has an overall fatigue damage. Observation of the fatigue damage and the wavelet coefficient was made on each segment. In the end of process, this algorithm produced a more reliable and suitable method of fatigue stain signal segmentation and segment by segment analysis to identify any improvements in the data scattering for fatigue data clustering prospects. Scatter plot was produced based on the relationship between segmental fatigue damage and its corresponding wavelet coefficient value. As the result, the relationship between the fatigue damage and the Morlet wavelet coefficient is strong and parallel, since the higher wavelet coefficient presents the higher fatigue damage, otherwise, it is the lower fatigue damage.
This paper presents an algorithm for wavelet coefficient extraction for fatigue strain signal edi... more This paper presents an algorithm for wavelet coefficient extraction for fatigue strain signal editing based on the Morlet wavelet parameter. In order to achieve the objective of this paper, three mission signals produced at three gate values were simulated for the purpose of the verification of the effectiveness of the fatigue features extraction. The effectiveness was determined by comparing the global signal statistical parameter and total fatigue damaging values between the original and mission shortened signals. Ideally, all signals have approximately the same values. For the related analysis of this paper, the value of 37,000 me/Hz was selected as the optimum gate value since the level did not change the signal behaviour. When in use, this gate value gave 101 seconds shortened signal, containing at least 90.8% of the original fatigue damage. In conclusion, this study found that the use of the Morlet wavelet has led to the development of a method to summarise a fatigue strain signal, whilst preserving the originality of the signal behaviour and the fatigue damage.
This paper presents the comparative study on three types of fatigue data editing technique for su... more This paper presents the comparative study on three types of fatigue data editing technique for summarising long records of fatigue data. Two of techniques were developed based on timefrequency domain (continous wavelet and discrete wavelet) and another one technique was developed based on time domain. These three techniques are used to extract fatigue damaging events in the record that cause the majority of fatigue damage, whilst preserving the load cycle sequence in the data. The objective of this study is to observe the capability of each technique in summarising long records of fatigue data. For the purpose of this study, two set of nonstationary data that exhibits random behaviour was used. This random data was measured in microstrain unit on the SAE1045 material that were used as a lower suspension arm of a car. Experimentally, the data was collected for 60 seconds at sampling rate of 500Hz, which gave 30, 000 discrete data points. The result of the study indicates that all techniques are applicable in detecting and extracts fatigue damaging events that exist in the fatigue data. However, it was found that continous wavelet can extract the data better than the other technique based the shorten signals, retention of damage and statistical parameter.
The fatigue feature extraction using the Short-Time Fourier Transform (STFT) and wavelet transfor... more The fatigue feature extraction using the Short-Time Fourier Transform (STFT) and wavelet transform approaches are presented in this paper. The transformation of the time domain signal into time-frequency domain computationally implemented using the STFT and Morlet wavelet methods provided the signal energy distribution display with respect to the particular time and frequency information. In this study, cycles with lower energy content were eliminated, and these selections were based on the signal energy distribution in the time representation. The simulation results showed that the Morlet wavelet was found to be a better approach for fatigue feature extraction. The wavelet-based analysis obtained a 59 second edited signal with the retention of at least 94 % of the original fatigue damage. The edited signal was 65 seconds (52 %) shorter than length of the edited signal that was found using the STFT approach. Hence, this fatigue data summarising algorithm can be used for accelerating the simulation works related to fatigue durability testing.
This paper presents clustering of fatigue features resulted from the segmentation of SAESUS time ... more This paper presents clustering of fatigue features resulted from the segmentation of SAESUS time series data. The segmentation process was based on the Morlet wavelet coefficient amplitude level which produced 49 segments that each has overall fatigue damage. Observation of the fatigue damage and the wavelet coefficients was made on each segment. At the end of the process, the segments were clustered into three in order to identify any improvements in the data scattering for fatigue data clustering prospects. This algorithm produced a more reliable and suitable method of segment by segment analysis for fatigue strain signal segmentation. According to the findings, the higher Morlet wavelet coefficient presented damaging segment, otherwise, it was non-damaging segment. This indicated that the relationship between the Morlet wavelet coefficient and the fatigue damage was strong and parallel.
This paper presents comparison of the filtering methods of strain loadings or fatigue time series... more This paper presents comparison of the filtering methods of strain loadings or fatigue time series using the frequency spectrum and the wavelet transform (WT). Thus, a raw loading or signal can be simplified for the simulation purposes. For this reason, the Fast Fourier Transform (FFT) and the Morlet wavelet algorithms were used in order to transform the vibrational fatigue time series into the frequency domain signal, leading to the observation of the frequency characteristics of the signal. To retain high amplitude cycles in the FFT algorithm, a low pass filter technique was applied to remove the high frequency signals with small amplitude that are not damaging. The departure of a high frequency information was smoothening the low amplitude cycles at high frequency events in the fatigue signal. The Butterworth filter was selected as the most efficient filter design as it retained most of the fatigue damage and also its capability to remove 30 % of the original low amplitude cycles. On the other hand, the Morlet wavelet was managed to remove 64 % of the original 59 second signal. This wavelet filtering method removed 34 % more than the similar procedure applied through the FFT approach. Hence, this fatigue data summarising algorithm can be used for studying the durability characteristics of automotive components.
A Morlet wavelet algorithm was developed for eliminating lower wavelet coefficient amplitude henc... more A Morlet wavelet algorithm was developed for eliminating lower wavelet coefficient amplitude hence to retain higher wavelet coefficient amplitude. The 44.8 second original fatigue strain signal was edited based on Cut Off Level. Furthermore, the retained segments were joined in order to gain a new output signal having shorter time length. The COL gave the 39.5 second edited signal containing at least 93.2 % of the original fatigue damage. In conclusion, with the application of the Morlet wavelet, the fatigue damaging features can be identified and be extracted successfully in order to generate a new shortened edited signal.
An extraction computational algorithm for fatigue feature editing is presented in this paper. The... more An extraction computational algorithm for fatigue feature editing is presented in this paper. The magnitude of the time domain Morlet wavelet coefficient level was used as the parameter to set gate value for the eliminating process of the 60 second original signal. It was important to maintain the signal statistical parameters and the total fatigue damage of the mission signal as close as to the original signal, with the retention of the original load sequences. At the end of the process, by using this approach, segments containing the higher Morlet wavelet coefficients that contribute to the more fatigue damaging events were retained and were then joined so produce the optimum mission signal length of 13.8 seconds. This short signal gave a 77 % reduction in length with only 8.7 % reduction in the fatigue damage. In conclusion, the extraction of the fatigue features using the Morlet wavelet successfully created a new mission signal which retains the majority of the higher fatigue damaging events in the time history
This paper presents clustering of fatigue features resulted from the segmentation of the SAESUS t... more This paper presents clustering of fatigue features resulted from the segmentation of the SAESUS time series data. The segmentation process was based on the Morlet wavelet coefficient amplitude level which produces 49 segments that each has an overall fatigue damage. Observation of the fatigue damage and the wavelet coefficients was made on each segment. In the end of the process, the segments were clustered into three clusters in order to identify any improvements in the data scattering for fatigue data clustering prospects. This algorithm produced a more reliable and suitable method of segment by segment analysis for fatigue strain signal segmentation. According to the findings, the higher Morlet wavelet coefficient presented damaging segment, otherwise, it was undamaging segment. It indicated that the relationship between the Morlet wavelet coefficient and the fatigue damage was strong and parallel.
The fatigue feature extraction using the Short-Time Fourier Transform (STFT) and wavelet transfor... more The fatigue feature extraction using the Short-Time Fourier Transform (STFT) and wavelet transform approaches is presented in this paper. The transformation of the time domain signal into time-frequency domain computationally implemented using the STFT and Morlet wavelet methods provided the signal energy distribution display with respect to the particular time and frequency information. In this study, cycles with lower energy content were eliminated, and this selection was based on the signal energy distribution in the time representation. The simulation results showed that the Morlet wavelet was found to be a better approach for fatigue feature extraction. The wavelet - based analysis obtained a 59 second edited signal with the retention of at least 94 % of the original fatigue damage. The signal was 65 seconds (52 %) shorter than length of the edited signal that was found using the STFT approach. Hence, this fatigue data summarizing computational algorithm can be used for accelerating the simulation works related to fatigue durability testing.
The main tribune of the Harapan Bangsa stadium in Banda Aceh is a three-dimensional structure. Th... more The main tribune of the Harapan Bangsa stadium in Banda Aceh is a three-dimensional structure. The dead loads from the roofs, the gutters and the ventilations cause the frames to experience direct stress, not only tensile stress σt but also compression stress σc. The objective of this paper is to recalculate the maximum live load which can be supported by the frames. In this study, Finite Element Method (FEM) was used to analyze the strength of the frames using the program SAP2000 version 8.3.3. The program calculated the joint reaction forces, the shear force diagrams, the moment diagrams, and the capacity ratio caused by the dead loads. Analyzing the results, the maximum capacity ratio was equal to 0.787, with axial force of 8,201.394 N. Therefore, the frames accept the live load equal to 2,219.691 N.
This paper discussed on the effectiveness of the Morlet wavelet to generate new edited signal. Th... more This paper discussed on the effectiveness of the Morlet wavelet to generate new edited signal. The 122.4 second SAESUS strain signal was edited based on the Morlet wavelet coefficient amplitude level. Segments with wavelet coefficient amplitude level lower than Cut Off Level (COL) were removed. Furthermore, extracted fatigue damaged segments were retained and then were joined in order to gain new edited signal. The signal statistical parameter and fatigue damaging values should be as accurate as possible for all signals. From the analysis, the 25,000 !" was selected to be the optimum COL value since the level did not change the signal behaviour. This value gave a 14 % reduction in length with only 6.1 % reduction in the fatigue damage. This indicated that the Morlet wavelet can be successfully applied to compress the original signal without changing the main history as well.
This paper presents a new approach to identify fatigue damaging potential locations using the Mor... more This paper presents a new approach to identify fatigue damaging potential locations using the Morlet wavelet coefficients. For solving the subject matter, the 122.4 second SAESUS strain signal was selected for the simulation purpose. As the result, the Morlet wavelet coefficients predicted that the maximum fatigue damage occurs at 40.4 - 42.6 seconds and 67.4 – 70 seconds. For the validation purpose, the Morrow’s fatigue damaging value was calculated and was obtained that the maximum fatigue damage occurs at 0 seconds and 99.7 seconds. The fatigue damaging value at the points was 0.0047 cycles to failure. Since both the plots had similar pattern, the Morlet wavelet coefficients could be used as the early warning of the fatigue damaging potential locations, although the locations were not entirely correct.
This paper is aimed to develop a correlation between the fatigue damage and the Morlet wavelet. I... more This paper is aimed to develop a correlation between the fatigue damage and the Morlet wavelet. In order to achieve the purpose, the SAESUS strain signal was edited based on the wavelet coefficients. The process resulted 49 damaged segments that each has an overall fatigue damage. Furthermore, each segment was analysed to obtain the fatigue damaging and wavelet coefficient values. Based on the plot, it was found that the fatigue damage and the Morlet wavelet had a strong correlation. The higher wavelet coefficient presented the higher fatigue damage, otherwise, it was the lower fatigue damage.
This paper presents the convenient wavelet family for fatigue strain signal analysis based on the... more This paper presents the convenient wavelet family for fatigue strain signal analysis based on the wavelet coefficients. This study involved the Morlet and Daubechies wavelet coefficients using both the Continuous Wavelet Transform (CWT) and the Discrete Wavelet Transform (DWT), respectively. The original fatigue strain signals were collected from a front lower suspension arm of a sedan by placing strain gauges on highest stress locations. The car ran over public road surfaces, i. e. pavé, highway, and UKM campus. In conclusion, the Daubechies wavelet is the convenient wavelet family for fatigue strain signal analysis. It is because the wavelet family gives higher coefficient values indicating that the resemblance between the wavelet and the signals is strong, more close, and similar.
This study is performed to develop fatigue durability feature extraction and clustering algorithm... more This study is performed to develop fatigue durability feature extraction and clustering algorithms using the wavelet transform, in order to correlate the fatigue damage and the signal internal energy. Strain signals measured from a car lower suspension arm were used as input data for the simulation purposes. The data were obtained from a fatigue data acquisition experiment by involving car movements on different road surfaces, i.e. pavé, highway and UKM campus, at constant speeds of 20 to 70 km/h. The data were recorded using strain gauges and a data logger instrument. This study also used a random strain signal from Society of Automotive Engineers-owned fatigue strain data set, i.e. SAESUS. All data were analysed using the wavelet transform. The mother wavelet used was the Morlet since it resulted in an internal energy plot that was similar to the fatigue damage, so it could be applied to detect the locations of the signals that have high fatigue damage. The energy was etermined based on the Morlet wavelet coefficient. Furthermore, the highly fatigue damaging segments in the signals were extracted based on the high energy distribution in order to result segments which contribute to fatigue damage. In the extraction process, energy gate value was used as the parameter to determine the minimum magnitude value to be retained. These retained segments were attached together to form a shorter edited signal for the validation purpose. From the results obtained, it was found that the algorithm could shorten the signal by 74.3% and maintained the original statistical parameter and cumulative fatigue damage values by more than 90%. It ensured that the produced edited signals could maintain the original characteristics of the signals. In the later part of this paper, the segments were clustered into three groups in order to identify the segment scattering pattern. According to the findings of clustering using the k-means method, scatter plot was produced based on the relationship between the fatigue damage and its corresponding signal internal energy. Segments with higher energy presented higher fatigue damage, while segments with smaller energy represented smaller fatigue damage. It indicated that the relationship between the fatigue damage and the signal internal energy was strong and parallel, i.e. about 86.18%, and it was depended on the road types. This correlation could be used to predict fatigue failures. This algorithm could produce a more suitable method of segment-by-segment analysis for fatigue strain signal extraction and clustering. Finally, this algorithm is recommended as an alternative method in durability fatigue studies especially in the field of automotive engineering
Gaya luar yang bekerja pada Truss Tribun utama stadion Harapan Bangsa Banda Aceh terdiri dari ber... more Gaya luar yang bekerja pada Truss Tribun utama stadion Harapan Bangsa Banda Aceh terdiri dari berat atap, berat kuda-kuda, berat talang, berat ventilasi udara, berat rangka, beban angin serta reaksi tumpuan. Untuk mengetahui kemampuan Truss dalam menumpu semua berat yang bekerja, maka perlu dilakukan analisa kekuatan dari Truss itu sendiri. Dari hasil perhitungan Truss tiga dimensi dengan menggunakan software SAP 2000 versi 7.42, maka diperoleh gaya batang tarik maksimum Ft = 63.005,26 N dan gaya batang tekan maksimum Fc = -75.532,95 N, dengan perpindahan maksimum U = -0,0136 m. Selanjutnya diperoleh tegangan tarik yang terjadi σt = 26,753 MPa dan tegangan tekan yang terjadi σc = -32,073 MPa, dimana σt < σta dan σc < σcr dan σca, sehingga konstruksi Truss tersebut baik
This paper presents scattering of fatigue time series data to develop a relationship between the ... more This paper presents scattering of fatigue time series data to develop a relationship between the fatigue damage and the Morlet wavelet coefficient. In order to achieve the purpose, the SAESUS strain signal was inputted into a segmentation algorithm based on the wavelet coefficient amplitude level which produces 49 segments that each has an overall fatigue damage. Observation of the fatigue damage and the wavelet coefficient was made on each segment. In the end of process, this algorithm produced a more reliable and suitable method of fatigue stain signal segmentation and segment by segment analysis to identify any improvements in the data scattering for fatigue data clustering prospects. Scatter plot was produced based on the relationship between segmental fatigue damage and its corresponding wavelet coefficient value. As the result, the relationship between the fatigue damage and the Morlet wavelet coefficient is strong and parallel, since the higher wavelet coefficient presents the higher fatigue damage, otherwise, it is the lower fatigue damage.
This paper presents an algorithm for wavelet coefficient extraction for fatigue strain signal edi... more This paper presents an algorithm for wavelet coefficient extraction for fatigue strain signal editing based on the Morlet wavelet parameter. In order to achieve the objective of this paper, three mission signals produced at three gate values were simulated for the purpose of the verification of the effectiveness of the fatigue features extraction. The effectiveness was determined by comparing the global signal statistical parameter and total fatigue damaging values between the original and mission shortened signals. Ideally, all signals have approximately the same values. For the related analysis of this paper, the value of 37,000 me/Hz was selected as the optimum gate value since the level did not change the signal behaviour. When in use, this gate value gave 101 seconds shortened signal, containing at least 90.8% of the original fatigue damage. In conclusion, this study found that the use of the Morlet wavelet has led to the development of a method to summarise a fatigue strain signal, whilst preserving the originality of the signal behaviour and the fatigue damage.
This paper presents the comparative study on three types of fatigue data editing technique for su... more This paper presents the comparative study on three types of fatigue data editing technique for summarising long records of fatigue data. Two of techniques were developed based on timefrequency domain (continous wavelet and discrete wavelet) and another one technique was developed based on time domain. These three techniques are used to extract fatigue damaging events in the record that cause the majority of fatigue damage, whilst preserving the load cycle sequence in the data. The objective of this study is to observe the capability of each technique in summarising long records of fatigue data. For the purpose of this study, two set of nonstationary data that exhibits random behaviour was used. This random data was measured in microstrain unit on the SAE1045 material that were used as a lower suspension arm of a car. Experimentally, the data was collected for 60 seconds at sampling rate of 500Hz, which gave 30, 000 discrete data points. The result of the study indicates that all techniques are applicable in detecting and extracts fatigue damaging events that exist in the fatigue data. However, it was found that continous wavelet can extract the data better than the other technique based the shorten signals, retention of damage and statistical parameter.
The fatigue feature extraction using the Short-Time Fourier Transform (STFT) and wavelet transfor... more The fatigue feature extraction using the Short-Time Fourier Transform (STFT) and wavelet transform approaches are presented in this paper. The transformation of the time domain signal into time-frequency domain computationally implemented using the STFT and Morlet wavelet methods provided the signal energy distribution display with respect to the particular time and frequency information. In this study, cycles with lower energy content were eliminated, and these selections were based on the signal energy distribution in the time representation. The simulation results showed that the Morlet wavelet was found to be a better approach for fatigue feature extraction. The wavelet-based analysis obtained a 59 second edited signal with the retention of at least 94 % of the original fatigue damage. The edited signal was 65 seconds (52 %) shorter than length of the edited signal that was found using the STFT approach. Hence, this fatigue data summarising algorithm can be used for accelerating the simulation works related to fatigue durability testing.
This paper presents clustering of fatigue features resulted from the segmentation of SAESUS time ... more This paper presents clustering of fatigue features resulted from the segmentation of SAESUS time series data. The segmentation process was based on the Morlet wavelet coefficient amplitude level which produced 49 segments that each has overall fatigue damage. Observation of the fatigue damage and the wavelet coefficients was made on each segment. At the end of the process, the segments were clustered into three in order to identify any improvements in the data scattering for fatigue data clustering prospects. This algorithm produced a more reliable and suitable method of segment by segment analysis for fatigue strain signal segmentation. According to the findings, the higher Morlet wavelet coefficient presented damaging segment, otherwise, it was non-damaging segment. This indicated that the relationship between the Morlet wavelet coefficient and the fatigue damage was strong and parallel.
This paper presents comparison of the filtering methods of strain loadings or fatigue time series... more This paper presents comparison of the filtering methods of strain loadings or fatigue time series using the frequency spectrum and the wavelet transform (WT). Thus, a raw loading or signal can be simplified for the simulation purposes. For this reason, the Fast Fourier Transform (FFT) and the Morlet wavelet algorithms were used in order to transform the vibrational fatigue time series into the frequency domain signal, leading to the observation of the frequency characteristics of the signal. To retain high amplitude cycles in the FFT algorithm, a low pass filter technique was applied to remove the high frequency signals with small amplitude that are not damaging. The departure of a high frequency information was smoothening the low amplitude cycles at high frequency events in the fatigue signal. The Butterworth filter was selected as the most efficient filter design as it retained most of the fatigue damage and also its capability to remove 30 % of the original low amplitude cycles. On the other hand, the Morlet wavelet was managed to remove 64 % of the original 59 second signal. This wavelet filtering method removed 34 % more than the similar procedure applied through the FFT approach. Hence, this fatigue data summarising algorithm can be used for studying the durability characteristics of automotive components.
A Morlet wavelet algorithm was developed for eliminating lower wavelet coefficient amplitude henc... more A Morlet wavelet algorithm was developed for eliminating lower wavelet coefficient amplitude hence to retain higher wavelet coefficient amplitude. The 44.8 second original fatigue strain signal was edited based on Cut Off Level. Furthermore, the retained segments were joined in order to gain a new output signal having shorter time length. The COL gave the 39.5 second edited signal containing at least 93.2 % of the original fatigue damage. In conclusion, with the application of the Morlet wavelet, the fatigue damaging features can be identified and be extracted successfully in order to generate a new shortened edited signal.
An extraction computational algorithm for fatigue feature editing is presented in this paper. The... more An extraction computational algorithm for fatigue feature editing is presented in this paper. The magnitude of the time domain Morlet wavelet coefficient level was used as the parameter to set gate value for the eliminating process of the 60 second original signal. It was important to maintain the signal statistical parameters and the total fatigue damage of the mission signal as close as to the original signal, with the retention of the original load sequences. At the end of the process, by using this approach, segments containing the higher Morlet wavelet coefficients that contribute to the more fatigue damaging events were retained and were then joined so produce the optimum mission signal length of 13.8 seconds. This short signal gave a 77 % reduction in length with only 8.7 % reduction in the fatigue damage. In conclusion, the extraction of the fatigue features using the Morlet wavelet successfully created a new mission signal which retains the majority of the higher fatigue damaging events in the time history
This paper presents clustering of fatigue features resulted from the segmentation of the SAESUS t... more This paper presents clustering of fatigue features resulted from the segmentation of the SAESUS time series data. The segmentation process was based on the Morlet wavelet coefficient amplitude level which produces 49 segments that each has an overall fatigue damage. Observation of the fatigue damage and the wavelet coefficients was made on each segment. In the end of the process, the segments were clustered into three clusters in order to identify any improvements in the data scattering for fatigue data clustering prospects. This algorithm produced a more reliable and suitable method of segment by segment analysis for fatigue strain signal segmentation. According to the findings, the higher Morlet wavelet coefficient presented damaging segment, otherwise, it was undamaging segment. It indicated that the relationship between the Morlet wavelet coefficient and the fatigue damage was strong and parallel.
The fatigue feature extraction using the Short-Time Fourier Transform (STFT) and wavelet transfor... more The fatigue feature extraction using the Short-Time Fourier Transform (STFT) and wavelet transform approaches is presented in this paper. The transformation of the time domain signal into time-frequency domain computationally implemented using the STFT and Morlet wavelet methods provided the signal energy distribution display with respect to the particular time and frequency information. In this study, cycles with lower energy content were eliminated, and this selection was based on the signal energy distribution in the time representation. The simulation results showed that the Morlet wavelet was found to be a better approach for fatigue feature extraction. The wavelet - based analysis obtained a 59 second edited signal with the retention of at least 94 % of the original fatigue damage. The signal was 65 seconds (52 %) shorter than length of the edited signal that was found using the STFT approach. Hence, this fatigue data summarizing computational algorithm can be used for accelerating the simulation works related to fatigue durability testing.
The main tribune of the Harapan Bangsa stadium in Banda Aceh is a three-dimensional structure. Th... more The main tribune of the Harapan Bangsa stadium in Banda Aceh is a three-dimensional structure. The dead loads from the roofs, the gutters and the ventilations cause the frames to experience direct stress, not only tensile stress σt but also compression stress σc. The objective of this paper is to recalculate the maximum live load which can be supported by the frames. In this study, Finite Element Method (FEM) was used to analyze the strength of the frames using the program SAP2000 version 8.3.3. The program calculated the joint reaction forces, the shear force diagrams, the moment diagrams, and the capacity ratio caused by the dead loads. Analyzing the results, the maximum capacity ratio was equal to 0.787, with axial force of 8,201.394 N. Therefore, the frames accept the live load equal to 2,219.691 N.
This paper discussed on the effectiveness of the Morlet wavelet to generate new edited signal. Th... more This paper discussed on the effectiveness of the Morlet wavelet to generate new edited signal. The 122.4 second SAESUS strain signal was edited based on the Morlet wavelet coefficient amplitude level. Segments with wavelet coefficient amplitude level lower than Cut Off Level (COL) were removed. Furthermore, extracted fatigue damaged segments were retained and then were joined in order to gain new edited signal. The signal statistical parameter and fatigue damaging values should be as accurate as possible for all signals. From the analysis, the 25,000 !" was selected to be the optimum COL value since the level did not change the signal behaviour. This value gave a 14 % reduction in length with only 6.1 % reduction in the fatigue damage. This indicated that the Morlet wavelet can be successfully applied to compress the original signal without changing the main history as well.
This paper presents a new approach to identify fatigue damaging potential locations using the Mor... more This paper presents a new approach to identify fatigue damaging potential locations using the Morlet wavelet coefficients. For solving the subject matter, the 122.4 second SAESUS strain signal was selected for the simulation purpose. As the result, the Morlet wavelet coefficients predicted that the maximum fatigue damage occurs at 40.4 - 42.6 seconds and 67.4 – 70 seconds. For the validation purpose, the Morrow’s fatigue damaging value was calculated and was obtained that the maximum fatigue damage occurs at 0 seconds and 99.7 seconds. The fatigue damaging value at the points was 0.0047 cycles to failure. Since both the plots had similar pattern, the Morlet wavelet coefficients could be used as the early warning of the fatigue damaging potential locations, although the locations were not entirely correct.
This paper is aimed to develop a correlation between the fatigue damage and the Morlet wavelet. I... more This paper is aimed to develop a correlation between the fatigue damage and the Morlet wavelet. In order to achieve the purpose, the SAESUS strain signal was edited based on the wavelet coefficients. The process resulted 49 damaged segments that each has an overall fatigue damage. Furthermore, each segment was analysed to obtain the fatigue damaging and wavelet coefficient values. Based on the plot, it was found that the fatigue damage and the Morlet wavelet had a strong correlation. The higher wavelet coefficient presented the higher fatigue damage, otherwise, it was the lower fatigue damage.
This paper presents the convenient wavelet family for fatigue strain signal analysis based on the... more This paper presents the convenient wavelet family for fatigue strain signal analysis based on the wavelet coefficients. This study involved the Morlet and Daubechies wavelet coefficients using both the Continuous Wavelet Transform (CWT) and the Discrete Wavelet Transform (DWT), respectively. The original fatigue strain signals were collected from a front lower suspension arm of a sedan by placing strain gauges on highest stress locations. The car ran over public road surfaces, i. e. pavé, highway, and UKM campus. In conclusion, the Daubechies wavelet is the convenient wavelet family for fatigue strain signal analysis. It is because the wavelet family gives higher coefficient values indicating that the resemblance between the wavelet and the signals is strong, more close, and similar.
This paper presents the fatigue strain signal behaviour analysis. The signals were measured at a ... more This paper presents the fatigue strain signal behaviour analysis. The signals were measured at a front lower suspension arm of a road vehicle driven over public road surfaces. These signals consisted of 10 stationary and 14 nonstationary signals. Based on the road loadings, the brick-pavéd road represented noisy but mostly consistent load features. The highway road represented mostly consistent load features, and the university road represented load features that might include turning and braking, rough road surfaces, and speed bumps. In conclusion, the university signals exhibited a lower frequency background that contains occasional shocks. These features contributed to the higher fatigue damaging potential compared to the pavé and highway road surfaces.
This paper discussed on the fatigue feature extraction of a fatigue strain signal using the Morle... more This paper discussed on the fatigue feature extraction of a fatigue strain signal using the Morlet wavelet. The transformation of the time domain signal into time-frequency domain provided the signal energy distribution display with respect to the particular time and frequency information. In this process, lower amplitude level contained in the original signal was eliminated based on the gate value in order to produce a new shortened signal. The simulation results showed that the Morlet wavelet able to shorten the signal more than 63 % of the original time length retaining at least 94 % of the original fatigue damage. Hence, this fatigue data summarising computational algorithm can be used for accelerating the simulation works related to fatigue durability testing.
Uploads
level did not change the signal behaviour. This value gave a 14 % reduction in length with only 6.1 % reduction in the fatigue damage. This indicated that the Morlet wavelet can be successfully applied to compress the original signal without changing the main history as well.
level did not change the signal behaviour. This value gave a 14 % reduction in length with only 6.1 % reduction in the fatigue damage. This indicated that the Morlet wavelet can be successfully applied to compress the original signal without changing the main history as well.