Detecting Shaft Misallignment in Gearbox of Helicopter Using Average Synchronous Analysis

International Journal of Engineering & Technology IJET-IJENS Vol: No: 5 Detecting Shaft Misallignment in Gearbox of Helicopter Using Average Synchronous Analysis Leila.nacib., Komi Midzodzi.pekpe, Saadi.sakhara. Leila.nacib@yahoo.fr LAGIS : : Laboratory of Computer Engineering and signal, university Lille.France. LSELM : laboratory of electromechanic system, university of Annaba, Algeria. Abstract-- This paper deals with signal processing techniques for non-destructive fault detection on gearbox of helicopter by using time synchronous averaging. Time synchronous averaging is a signal processing technique that is used to extract repetitive signals from additive noise. This process requires an accurate measurement of the repetitive frequency of the desired signal or a signal that is synchronous with the desired signal. The raw data is then divided up into segments of equal length blocks related to the synchronous signal and averaged together. When sufficient averages are taken, the random noise is canceled, leaving an improved estimate of the desired signal. Time synchronous averaging is a feature extraction technique that have been used successively to gearbox condition monitoring. rotating machinery. These techniques are applied for the early detection of misalignment in helicopters. In this article we have used several methods for detection of faults in rotating machines and from the cepstral analysis we detected a fault tree of misalignment at the direction of ASL in gearbox of helicopter. (Fig. ) Index Term-- time synchronous averaging, misalignment, diagnosis, cepstrum, spectrum. I. INTRODUCTION The gearboxes are widely used in mechanical and are found in all types of industries, processes. They are very busy mechanical elements, complex size and realize that failure may have limited their life. Given their importance, it is not surprising that they have been the subject of numerous studies on the analysis, design, study materials, lubrication, failure analysis, techniques monitoring and fault diagnosis, to avoid sudden breaks (helicopters) to reduce maintenance costs in industrial maintenance by practicing so-called conditional and finally, studies on the quality control of manufacturing gears has a direct influence on their life, but also the noise. Current techniques for monitoring machines are based on Fourier analysis. They give good results for defects that could be described as "simple" as the unbalance, some bearing defects. They are often ineffective when defects are characterized by complex variations of the spectrum, especially by non-stationary in the vibration signal. This is the case for gear vibration signatures that generate rich, which is often difficult to monitor developments and, in the case of certain faults makes early detection difficult. For other types of mechanical elements, several studies have shown the contribution of parametric methods for monitoring and fault diagnosis. In this article, we present recent technical applications of signal processing for early diagnosis of misalignment motor shaft in a gearbox of a helicopter. We first present techniques for high Resolution spectral analysis, which at present have not seen applications of vibration signals of Fig.. description of part in gearbox of helicopter A) VIBRATION GEAR AND IDENTIFICATION TECHNIQUES Techniques for detecting damage to the gears are based on the analysis of vibration signals. It is well known that the most important elements in the vibration spectra of gearbox are the teeth meshing frequency and harmonics, and sidebands. Amplitude modulations are present when a gear train mounted on a shaft bent or misaligned. If there is a default speed local angular velocity speed may change depending on the rotation. after the change of the speed modulation frequencies may occur. amplitude and frequency modulation are present in case of presence of default. severe defect causes the increase in the number and amplitude of the sidebands. modulation are caused by defects in the components of the machine, such as gear, bearing and shaft [9]. sidebands around harmonics engagement can also be the result of bearing wear accompanied by the transmission shaft [5]. The cepstrum is adapted to the detection of the sidebands in the vibrational spectra. In addition, since the cepstrum estimated average lateral spacing over a wide frequency range, it can measure very precisely the lateral periodicity. It is therefore applicable to the detection and diagnosis of gear faults [5-7]. Time synchronous averaging is a signal processing technique that is used to extract repetitive signals from additive noise. This process requires 7-55- IJET-IJENS @ December IJENS I J E N S

amplitude amplitude amplitude International Journal of Engineering & Technology IJET-IJENS Vol: No: an accurate measurement of the repetitive frequency of the desired signal or a signal that is synchronous with the desired signal. The raw data is then divided up into segments of equal length blocks related to the synchronous signal and averaged together. When sufficient averages are taken, the random noise is canceled, leaving an improved estimate of the desired signal [9]. Synchronous time averaging is a technique for extracting features which have been successively used to control the state of the gearbox. The residual signal is obtained by removing the first shot and shaft components with their harmonic signal synchronous time average []. In this paper, the techniques mentioned above are applied to real data analysis record for stealing a helicopter. Capabilities of fault detection techniques discussed are compared in particular. Sensitivity to the seriousness of the breach specific technique is evaluated.[9]. A) TIME SIGNAL The time signal of a lack of misalignment at the pace of Fig. 3 Found a periodic phenomenon in the rotation frequency (period = turn), but also phenomena repeated every / turn and third turn. This will result in most cases by the presence of elements of order, 3 or even times the rotation frequency with amplitudes greater than the component of order. emps - - - B) RESULTATS AND DISCUSSIONS THE MESHING SIGNAL The importance of the shock depends on the teeth shape during the well operation and the fault nature during degraded operation. [] The meshing signal S e (t) is amplitude and frequency modulated by the signals S r (t) and S r (t) emitted from the pinion and the wheel whose frequencies are respectively f r and f r. These modulations, being multiplicative and convolved with meshing harmonics [], are defined by: - - - tour / tour..5.5.55..5.7 temps Fig.. time Signal faultless With: x c (t)the signal produced by the shock between two gear teeth. Te = / f e With: T the meshing period and f the e e meshing frequency. T =/f =N.T With T : the rotation r r e r period of wheel. f r : the rotation frequency of wheel. N : the number of teeth of wheel. T r =/ f =N.T r e With: T r : the rotation period of wheel. f : the rotation frequency of wheel. r N : the teeth number of wheel. T ABLE I GEAR MESH FREQUENCIES CALCULATIONS Shaft Speed Gear teeth S Fe= 7 Hz S Fe= 7 Hz 35 RPM N =3 F r = Hz 5 RPM N =33 F r = Hz 7 9 temps Fig. 3. Time signal with misalignment defect B) SPECTRAL ANALYSIS Spectral analysis is the major tool for the study of vibration signals of rotating machinery in the maintenance of conditional mean machines. Many problems related to the detection of defects on bodies rotating machines can be solved by Fourier analysis. However, there are cases where simple Fourier analysis is ineffective; we mainly refer to the case of signals with local non-stationary. The Fourier transform is not adapted to such situations because it is difficult to highlight local non-stationary are generally very short on this type of signals. Moreover, it should be noted that the vibration signals are very rich in harmonics, hence the difficulty of assessing the variations due to nonstationary It is legitimate to hope that more precise methods to bring more Fourier analysis. x -3 7-55- IJET-IJENS @ December IJENS I J E N S

International Journal of Engineering & Technology IJET-IJENS Vol: No: 7 x X: Y:.97e+5 X:.7 Y: 577 5 5 3 35 5 Fig.. Vibration spectrum faultless events of wheel gear shaft rotating with Hz and pinion rotating with Hz. Fig. 7. Vibration log spectrum x le spectre du signal rawvib 93, composante ASL, code défaut, fréquence d'engrennement 73 Hz 3.5.5 X: 5. Y: 3.e+ C) TIME SYNCHROUNOUS AVERAGE Time synchronous averaging is method of background noise reduction in spectra of complex signals. For instance, it can be used to sort out the contribution from one individual shaft and its associated gears from the complex vibration signature of a multistage gearbox. X:.5 Y: 5 X:. Y: 53.5.5 X: Y: 59 5 3 35 5 Fig. 5. Vibration spectrum events of wheel gear shaft rotating with Hz and pinion rotating with Hz..5 -.5 - - -.5.5.5 Fig.. Time synchronous average waveform circular diagram of wheel shaft (faultless). - 3 5 7. Fig.. Vibration log spectrum faultless.. -. -. -. -. - -. -. -. -...... Fig. 9. Time synchronous average waveform circular diagram of wheel shaft 7-55- IJET-IJENS @ December IJENS I J E N S

International Journal of Engineering & Technology IJET-IJENS Vol: No: D) CEPSTRUM ANALYSIS Cepstrum technique appears to be efficient for detecting changes not easily notable in the spectrum. Major benefit of using cepstrum technique would be earlier damage identification because it is clear and easier to see changes as many authors pointed out. Time synchronous averaging is very efficient in damage localization as many other authors refer. For complete gearbox damage identification it is necessary to repeat the analysis with reference to each gear in the gearbox. There is need to access gear shaft with a reason to collect tachometer signal. Many authors do not point out this significant limitation of time synchronous averaging technique, especially its ability to implement in practice..9. PIGION ROUE (WEEL) synchronous cepstre method and the expertise report. The application of this technique to the vibration signal emitted by the gear reductor system permits to conclude that it can play an important role in the study of gear vibrations. In fact, the use state of a reductor is strongly related to modulation phenomena that present the vibrations relative to the meshing signal. We have shown that by Time signal and Cepstrum energy of the vibration signal averaged according the LCM of the first gear we can detect the fault which is misalignment of drive shaft in gearbox of helicopter. FUTURE WORK In future work we will offer a new technique for fault detection based on the geometric mean calculates and computes statistics to extract the vibration severity in the cepstrum, the technique is the automatic detection of defects not require a great analysis.7..5..3.. 5 3 35 5 5 55 5 Fig.. Cepstrum calculated energy of the vibration signal averaged according the LCM of the first gear (faultless)..9..7..5..3.. X: 5 Y:. X: Y:.3 X: Y:. 5 3 35 5 5 55 5 Fig.. Cepstrum calculated energy of the vibration signal averaged according the LCM of the first gear. Misalignment is indicated by peak amplitude preponderant generally times the frequency of rotation (sometimes 3 or times). It is a vibration component of order frequency rotation (rarely of order 3, or exceptionally order) with amplitudes greater than the components of order [Fig. ]. CONCLUSION In this paper, a gear box diagnosis technique based on cepstrum analysis. The performances of this technique in the gear system diagnosis have been compared to those of the REFERENCES [] A. kraker 9, "cepstrum analysis as a usefulsupplement to spectrum analysis for monitoring gear-box, "experimental stress analysis: proceedings of the th international conference,amsterdam, netherlands, may -, p. -9. [] B. k. n., "handbook of condition monitoring", elseiver advanced technology, oxford 99. [3] Robert Randall bond, wiley, "vibration-based condition monitoring: industrial, automotive and aerospace applications ",. [] Randall, r. b., 9, "a new method of modeling gear faults", asme journal of mechanical design, april 9, vol., p. 59-7. [5] Fan, x., and j. zuo.,, "gearbox fault detection using hilbert and wavelet packet transform ", mechanical systems and signal processing (), p. 9-9. [] Timothy a dunton, "an introduction to time waveform analysis", universal technologies. [7] Randall r.b. 9, "cepstrum analysis", brüel & kjær technical review, no. 3, 3-. [] komgon, n.c. mureithi, n. lakis, a. thomas, m., 7, "on the use synchronous averaging of time, independent component analysis and support vector machines for bearing fault diagnosis ", st international conference on industrial risk engineering, montreal, canada, p.-. [9] P. d. mc fadden, "detecting fatigue cracks in gears by amplitude and phase demodulation of the meshing vibration ", journal of vibration, acoustics, stress and reliability in design (9) p. 5-7. [] B d forrester, "advanced vibration analysis techniques for fault detection and diagnosis in geared transmission systems ", ph.d. thesis, swinburne university of technology, australia, 99. [] Leon cohen, hunter college, "time-frequency analysis", prentice hall 995. [] Forrester, bd, "analysis of gear vibration in the timefrequency domain ", in current practices and trends in mechanical failure prevention, edited by hc and sc 7-55- IJET-IJENS @ December IJENS I J E N S

International Journal of Engineering & Technology IJET-IJENS Vol: No: 9 pussy pussy (vibration institute, willow brook), pp. 5 to 3, 99. [3] J. hong suh, s. arcep kumara, s. p. mysore, "machinery fault diagnosis and prognosis: application of advanced signal processing techniques " cirp annals-manufacturing technology, volume, issue, 999, p. 37-3. [] S. n. engin, k. gülez and mr. badi, "advanced signal processing techniques for faults diagnostics-a review ", mathematical and computational applications, vol. no., p. -3, 999. [5] Alexej v. barkov, dr. natalia a. barkova "diagnostics of gearing and geared couplings using envelope spectrum methods " vibroacoustical systems and technologies inc., st. petersburg, russia. [] Alain b. et pachaud c. (99) surveillance des machines par analyse des vibrations e tirage afnor, pp -5. [7] H. zheng, z. li and x. chen, ()gear fault diagnosis based on continuous wavelet transform mechanical systems and signal processing (-3),7. [] Hamzaoui n., rapport sur la maintenance conditionnelle des machines tournantes. Consulting laboratoire vibratoire acoustique I.N.S.A Lyon. [9] Rusmir bajrić, denijal sprečić, ninoslav zuber () "review of vibration signal processing techniques towards gear pairs damage identification " [] Robert b. randall and nader sawalhi () a new method for separating discrete components from a signal, university of new south wales, sydney, australia. [] Noureddine haloui, djamel chikouche, messaoud benidir (7) diagnosis of gear systems by spectral analysis of vibration signals. IJCSNS international journal of computer science and network security, vol.7 no.. [] Landolsi fouad :cours de technique de surveillance. [3] Amit aherwar, md. saifullah khalid () vibration analysis techniques for gearbox diagnostic: a review [] Sidahmed m. 99 bulletin s.f.m. revue française de mécanique, 3-5. Détection précoce de défauts dans les engrenages par analyse vibratoire. [5] G. dalpiaz, a. rivola and r. rubini gear fault monitoring: comparison of vibration analysis techniques [] Ebersbach et al, (5), the investigation of the condition and faults of a spur gearbox using vibration and wear debris analysis techniques, international conference on wear of materials, wear, pp. - [7] Decker Harry. J (), Gear Crack detection using tooth analysis at NASA Research Center. [] Polyshchuk V.V et al (), Gear fault detection with time frequency based parameter NP, International Journal of Rotating Machinery (), pp 57-3. [9] Lin J. and Zuo M. J (3), Gear box fault diagnosis using adaptive wavelet filter, Mechanical Systems and Signal Processing, Vol. 7(), pp.59-9. [3] Ghemari. Z and Saad. S, Development of measurement precision of sensor vibration Journal of Vibration and Control, Published online before print May 3,. 7-55- IJET-IJENS @ December IJENS I J E N S