Proper setting of process conditions in the melt spinning setup is one way to yield uniform quality of spinline tension. The optimum setting to give uniform spinline tension is determined using experiment plans in the Taguchi method and significant process parameters are also identified. When the setting shifts from the optimum, the spinline tension becomes non-uniform and downgrades product quality. This study aims to diagnose single or double fault conditions of those significant process parameters based on the spinline tension signal. The critical procedures of fault diagnosis are feature extraction and classification. The tension signal is decomposed into a wavelet packet tree of four resolution levels. Four entropies from the best-basis wavelet packet tree and the lowest entropy at level four are selected as features. The back-propagation neural network acts as a classifier. The experimental results demonstrate that the features and classifier actually work well to identify the single and double fault conditions with high accuracy in melt spinning.
LeeJSJungHWKimSH. Effect of fluid viscoelasticity on the draw resonance dynamics of melt spinning. J Non-Newtonian Fluid Mech2001; 99: 159–166.
2.
JungHWSongHSHyunJC. Analysis of the stabilizing effect of spinline cooling in melt spinning. J Non-Newtonian Fluid Mech1999; 87: 165–174.
3.
KaseSArakiM. Studies on melt spinning. VIII. Transfer function approach. J Appl Polym Sci1982; 27: 4439–4465.
4.
DevereuxBMDennMM. Frequency response analysis of polymer melt spinning. Ind Eng Chem Res1994; 33: 2384–2390.
5.
ZieminskiKFSpruiellJ. On-line studies and computer simulation of the melt spinning of nylon-66 filaments. J Appl Polym Sci1988; 35: 2223–2245.
6.
WilczynskiK. A computer model for single-screw plasticating extrusion. Polym Plast Technol Eng1996; 35: 449–477.
7.
ZiabickiA. Fundamentals of fiber formation, London: John Wiley, 1976.
8.
DuttaANadkaniVM. Identifying critical process variable in poly(ethylene terephthalate) melt spinning. Text Res J1984; 54: 35–42.
9.
GuptaVBMondalSABhuvaneshYC. Spinning speed–throughput rate relationships for polyester, nylon, and polypropylene fibers. J Appl Polym Sci1997; 65: 1773–1788.
10.
ChenRSLeeHHYuCY. Application of Taguchi’s method on the optimal process design of an injection model PC/PBT automobile bumper. Compos Struct1997; 39: 209–214.
11.
KimJDChoiMS. Stochastic approach to experimental analysis of cylindrical lapping process. Int J Mach Tools Manuf1995; 35: 51–59.
12.
LauJHChangC. Taguchi design of experiment for wafer bumping by stencil printing. IEEE Trans Electron Packag Manuf2000; 23: 219–225.
13.
JaisinghANarasimhanKDatePP. Sensitivity analysis of a deep drawing process for miniaturized products. J Mater Process Technol2004; 147: 321–327.
14.
YangWHTarngYS. Design optimization of cutting parameters for turning operations based on the Taguchi method. J Mater Process Technol1998; 84: 122–129.
15.
LeeSHLeeSH. Optimisation of cutting parameters for burr minimization in face-milling operations. Int J Prod Res2003; 41: 497–511.
16.
HuangCCTangTT. Optimizing multiple qualities in as-spun polypropylene yarn by neural networks and genetic algorithms. J Appl Polym Sci2006; 100: 2532–2541.
17.
WangSZhuGTangYY. Feature extraction of radar multiple-target echoes using wavelet packet transform with the best bases. Int J Pattern Recognit Artif Intell2003; 17: 127–150.
18.
LaineAFanJ. Texture classification by wavelet packet signatures. IEEE Trans Pattern Anal Mach Intell1993; 15: 1186–1191.
19.
PunCMLeeMC. Rotation-invariant texture classification using a two-stage wavelet packet feature approach. IEE Proc Vision Image Signal Process2001; 184: 422–428.
20.
Su LeeCChoiCHChoiJY. Surface defect inspection of cold rolled strips with features based on adaptive wavelet packet. IEICE Trans Inf Syst1997; E80-D: 594–604.
21.
SantosoSPowersEJGradyWM. Power quality assessment via wavelet transform analysis. IEEE Trans Power Deliv1996; 11: 924–930.
22.
ElmitwallyAFarghalSKandilM. Proposed wavelet-neurofuzzy combined system for power quality violations detection and diagnosis. IEE Proc Gener Transm Distrib2001; 148: 15–20.
23.
AngrisaniLDapontePD’ApuzzoM. A measurement method based on the wavelet transform for power quality analysis. IEEE Trans Power Deliv1998; 13: 990–998.
24.
LouXLoparoKA. Bearing fault diagnosis based on wavelet transform and fuzzy inference. Mech Syst Signal Process2004; 18: 1077–1095.
25.
Peng YH, Xu XG and Zhao HX. Application of wavelet packet analysis in turbine fault diagnosis. In: Proceedings of the Fifth International Conference on Machine Learning and Cybernetics, 2006, 4028556, pp.2897–2900.
26.
LeeDSSLithgowBJMorrisonRE. New fault diagnosis of circuit breakers. IEEE Trans Power Deliv2003; 18: 454–459.
27.
WangZYLiuYLGriffinPJ. Neural net and expert system diagnose transformer faults. IEEE Comput Appl Power2000; 13: 50–55.
28.
MallatSG. A theory for multi-resolution signal decomposition: the wavelet representation. IEEE Trans Pattern Anal Mach Intell1989; 11: 674–693.
29.
HouXLiuG. Wavelet packet image coding algorithm combined with the actual coding costs. Opt Eng2003; 42: 1640–1648.
30.
ChangTKuoCCJ. Texture analysis and classification with tree-structured wavelet transform. IEEE Trans Image Process1993; 2: 429–441.
31.
GurneyK. An introduction to neural networks, London: UCL Press, 1997.
