Feature extraction and quality grading method by virtue of acoustic signals generated from hardwood logs based on AR-MED

PDF(2171 KB)

Journal of Vibration and Shock ›› 2019, Vol. 38 ›› Issue (14) : 181-188.

QU Yuying,YANG Yang,XU Feng

Author information +

History +

Abstract

Internal defect detection and quality grading for hardwood logs could improve its utilization rate and economic benefit. Aiming at the insufficient acoustic parameters in quality assessment due to the non-stationary features of acoustic signals and the overlapping of defect features, a method for feature extraction and quality grading was proposed based on the autoregressive model (AR) and minimum entropy deconvolution (MED).An AR-based linear filter was applied to filter periodic deterministic components from original signals according to the Akaike Information Criterion (AIC).Then, the residual signal containing defect informations was enhanced by using the MED technique, and the kurtosis value of the enhanced signal was used as the characteristic parameter.The quality of sample logs was graded by the kurtosis, and the grading result was compared with that graded by velocity. Numerical simulations and measured results of the hardwood logs show that the proposed method can significantly increase the kurtosis values of defect signals and efficiently classify the hardwood logs in quality.

Key words

hardwood logs / quality assessment / kurtosis / Minimum entropy deconvolution / Autoregressive model

Cite this article

EndNote

Ris (Procite)

Bibtex

Download Citations

QU Yuying,YANG Yang,XU Feng. Feature extraction and quality grading method by virtue of acoustic signals generated from hardwood logs based on AR-MED[J]. Journal of Vibration and Shock, 2019, 38(14): 181-188

References

[1] Carpenter RD, Sonderman DL, Rast ED, et al. Defects in hardwood timber [M]. U.S. Department of Agriculture, Forest Service, Washington, DC, 1989: 88.
[2] Huang CL, Lindstrom H, Nakada R, et al. Cell wall structure and wood properties determined by acoustics-A selective review [J]. Holz als Roh-und Werkstoff, 2003,61(5):321-335.
[3] Wiedenbeck J, Wiemann M, Alderman D, et al. Defining hardwood veneer log quality attributes [R]. Gen. Tech. Rep. NE-313. Newtown Square, PA: U.S. Department of Agriculture, Forest Service, Northeastern Research Station, 2003, 36.
[4] Aratake S, Arima T. Estimation of modulus of rupture (MOR) and modulus of elasticity (MOE) of lumber using higher natural frequency of log in pile of logs II-Possibility of application for Sugi square lumber with pith [J]. Mokuzai Gakkaishi, 1994, 40(9):1003-1007.
[5] Ross RJ, McDonald KA, Green DW, et al. Relationship between log and lumber modulus of elasticity [J]. Forest products Journal, 1997, 47(2):89-92.
[6] Wang X. Acoustic measurements on trees and logs: a review and analysis [J]. Wood Science and Technology, 2013, 47(5): 965-975.
[7] Carter P, Briggs D, Ross RJ, et al. Acoustic testing to enhance western forest values and meet customer wood quality needs [R]. PNW-GTR-642. USDA Forest Service, Pacific Northwest Research Station, Porland, OR, 121-129.
[8] Gaunt D. A revolution in structural timber grading [C]//Quenneville P. World Conference on Timber Engineering. Auckland, New Zealand: Curran Associates, Inc., 2012. 276-283.
[9] Murphy G, Cown D. Sand, stem and log segregation based on wood properties: a review [J]. Scandinavian Journal of Forest Research, 2015, 30(8):757-770.
[10] Wang X, Ross RJ, Green DW, et al. Stress wave sorting of red maple logs for structural quality [J]. Wood Science and Technology, 2004, 37(6):531-537.
[11] Wang X, Verrill S, Lowell E, et al. Acoustic sorting models for improved log segregation [J]. Wood and Fiber Science, 2013, 45(4):343-352.
[12] Farrell R, Nolan G. Sorting plantation Eucalyptus nitens logs with acoustic wave velocity [R]. Project No.PN07.3018, Victoria, Australia: Forest & Wood Products Australia Limited, 2008.
[13] Xu F, Wang X, Thomas E, et al. Defect detection and quality assessment of hardwood logs: part 1-acoustic impact test and wavelet analysis [J]. Wood and Fiber Science, 2018, 50(3): 291-309.
[14] Kohei H, Masato N, Takanobu N, et al. Close/distant talker discrimination based on kurtosis of linear prediction residual signals[C]//2014 IEEE International Conference on Acoustic, Speech and Signal Processing (ICASSP), 2014: 2327-2331.
[15] Al-Bugharbee H, Trendafilova I. A fault diagnosis methodology for rolling element bearings based on advanced signal pretreatment and autoregressive modeling [J]. Journal of Sound and Vibration, 2016,369:246-265.
[16] 曹展, 王细洋. 基于AR模型的齿轮故障诊断 [J]. 失效分析与预防, 2016, 11(5): 270-275.
Cao Z, Wang XY. Gear fault diagnosis based on AR model [J]. Failure Analysis and Prevention, 2016, 11(5): 270-275.
[17] Wang W, Wong A K. Autoregressive model-based gear fault diagnosis [J]. Journal of Vibration and Acoustics, 2002, 124(2): 172-179.
[18] 田艳, 王胜春, 王月婷, 等. 基于AR模型的塔机结构损伤检测研究 [J]. 工程机械, 2013, 44(12):22-26.
Tian Y, Wang SC, Wang YT, et al. Research of structure damage detection of tower crane based on AR model [J]. Construction Machinery and Equipment, 2013, 44(12):22-26.
[19] Wiggins RA. Minimum entropy deconvolution [J]. Geoexploration, 1978, 16(1-2):21-35.
[20] Donoho D L. On minimum entropy deconvolution. In: Applied Time Series Analysis II [M], New York: Academic Press, 1981:565-608.
[21] Wang W, Wong A K. A model-based gear diagnostic techniques [R], DSTO, TR-1079, Airframes and Engine Division, Aeronautical and Maritime Research Laboratory, 2000.
[22] National Hardwood Lumber Association. Rules for the measurement and inspection of hardwood and cypress [M]. Memphis, Tennessee, U.S.A. 2015:104.
[23] Mucciardi A N, Luley C J, Gormally K H. Preliminary evidence for using statistical classification of vibration waveforms as an initial decay detection tool [J]. Arboriculture & Urban Forestry, 2011, 37(5):191-199.
[24] Wang X, Stelzer H E, Wiedenbeck J, et al. Assessing wood quality of borer-infested red oak logs with a resonance acoustic technique [J]. Wood and Fiber Science, 2009, 41(2):180-193.