Field Experiences of Gas Turbines Vibrations - A

0 downloads 0 Views 709KB Size Report
Key words: Blade failures, rub, cracks, cracked shaft, gas turbines. 1. Introduction .... Beebe (1998) in a similar work for steam turbines reported that the breaking off of the ..... At this stage, high harmonics of x2, x3 and x4 rpm were noted.
APVC2007

APVC2007, Sapporo, Japan

Field Experiences of Gas Turbines Vibrations A Review and Case Studies M Salman LEONG Institute of Noise & Vibration, Universiti Teknologi Malaysia, Jalan Semarak 54100 Kuala Lumpur, Malaysia. E-mail:

[email protected]

Abstract Blade failures represent the highest percentage of failures in gas turbines. This paper presents some typical examples of blade related failures. A literature review of common types of blade faults and research on detection methods are presented. Some methods are however less feasible under practical operating conditions in the plant. Three case studies of gas turbines vibrations are presented. The first case reports on stator blades and labyrinth glands rubs resulting in recurring shaft seizure during the coast down after load removal. Comparison of vibration spectrum undertaken a day before failure with prior data showed increase in specific blade passing frequencies (BPF) with increased side bands at intervals of the synchronous rpm. These increases were from 13 to 28 times above standard deviations of baseline datum. The stator blades rub was suspected to originate from a distorted casing. Another case relates experiences of a cracked shaft which resulted in severe rubbing during a run up. The unit experienced steady increase in vibration levels on the compressor non-drive end bearing several months prior to the incident. A full rub occurred as the unit passed through the second critical, with further development of a thermal bent shaft aggravating the problem with instantaneous severe vibration excursion. An approximate 200o phase shift was also noted. A 100mm longitudinal crack on the shaft was found together with signs of severe rubs on the compressor blades. The third case involved time varying vibrations where vibration amplitudes and phase angles increased over a time period which dropped off after each time period; and repeated in regular cycles. This was due to oil leaks carbonization at the glands resulting in a rub. The paper concludes with a brief discussion on issues relating to relatively poor detection of blade faults in the plant. Key words: Blade failures, rub, cracks, cracked shaft, gas turbines.

1. Introduction

*Received XX Xxx, 200X (No. XX-XXXX) [DOI: 00.0000/ABCDE.2006.000000]

Vibration failures in gas turbines can be rather catastrophic with serious economic consequences. The turbines operate under relatively high vibratory and thermal cyclic stresses, especially when used in situations requiring daily start-stops (hot and cold starts) which are typical in peak load applications in power generation for example. The industry has seen its fair share of dramatic failures in vibration and other mechanical related failures in gas turbines. This paper presents a review of common failures in gas turbines, with several case studies of vibration related failures in gas turbines based on the experience of the author in the power generation industry.

1

APVC2007

APVC2007, Sapporo, Japan

2. Examples of Blade Failures in Gas Turbines The world’s largest insurance company Allianz Insurance (as cited by Meher-Homji, 1995d) reported that blade failures had accounted for as much as 42% of failures in gas turbines. Some typical failures are presented below. Fig. 1 shows a case from a 171 MW gas turbine which had 2 catastrophic failures totaling more than US$ 25million in production downtime, Barnard (2006). Hot section disk cracks lead to blade liberation (breakage) resulting in hot section blades and discs destroyed. Regular dental mold taken on machine showed no excessive erosion. Subsequent root cause analysis showed that the failure mechanism was fatigue driven by blade vibrations: either from short duration rotating stall during start up and shut down exciting the first tangential (sub-synchronous) mode of the blade or from synchronous vibration exciting the second bending mode of the blade. Photos in Fig. 2 shows lost of part in compressor blades, and rub marks on the rotor from stator blades from 100 MW gas turbines used in combined and open cycle generation in Malaysian power plants. Lost of parts and foreign object damage (FOD) are common failures. The photo also shows rub marks on the rotor from stator blades.

Fig. 1:

Photos of damaged turbine blades (left), and missing blade (source Barnard)

Fig. 2:

Photos of lost part of blades, and stator blade rubs

Less dramatic but more frequent failures are cracks and looseness in the intermediate packing piece and blade roots which holds the blades to the shaft. Photos in Fig. 3 shows a gas turbine with 21 stages axial flow compressor section on a single, composite welded shaft. All the blades in the compressor stator and rotor were made of high-tensile, ferritic chrome steel; and manufactured partly by milling and partly by precision forging. They are fitted into radial grooves machined in the shaft and compressor casing, and separated from one another by intermediate pieces. The rotor blades of the first ten rows were fitted directly in the shaft by specially shaped roots, while the remaining blades were rooted in intermediate pieces. There had been frequent experiences of cracks at the roots. These cracks would cause the intermediate pieces and the blades to become loose, and potentially

2

APVC2007

APVC2007, Sapporo, Japan

detach from the rotor and consequently causing foreign object damage (FOD) to other components downstream.

Fig. 3: Roots of compressor disk, and cracks at intermediate piece. While failures in blades are on its own are already an issue of grave concern, the consequential damages could be more catastrophic. Photo (a) given in Figure 4 shows a fire started in a transformer that was hit by a blade that went through a wall which then hit the transformer. The photo (b) shows a completely damaged compressor section (as a result of damaged turbine blades for the same machine cited in Fig. 1).

(a) Transformer fire (b) Compressor section Fig. 4: Consequential damages from blades failures: (a) fire after blade went through wall and hit transformer; (b) damage in compressor section (source Barnard).

3. Literature Review of Blade Related Failures Gas turbines often operate at the design limit of blades, bearings and combustion components under extreme conditions. This meant that these components are life limited and are more likely to experience failures than other less stressed parts. Hot section blades typically fail because of creep, oxidation, low-cycle fatigue (LCF), and high-cycle fatigue (HCF). Contributing factors include cyclic loads, over firing, inadequate refurbishment cyclic loads, corrosion and environmental attack. Hot section blades are life-limited items and require refurbishment or replacement at intervals dependent upon thermal exposure. High-cycle fatigue (HCF) is avoided in rotating turbine blades by controlling vibratory stresses. Long blades are often designed by tuning the lower natural frequencies to avoid operating near strong excitations and high vibratory stresses. When the natural frequency variance is greater than an acceptable band, the blades must be designed with sufficient stiffness to limit stresses when blades operate at resonance. HCF failures could occur if these conditions are not met or if excessive excitations occur because of vane damage, rotating stall, flutter, or blade tip rubbing. The variability of blade fabrication can influence natural

3

APVC2007

APVC2007, Sapporo, Japan

frequencies on assembly; wear and erosion can reduce resonant frequency margins over time. Either of these conditions can result in high-cycle fatigue damage. In principle, blade faults diagnosis can be undertaken from measurements and monitoring of gas turbine operating parameters such as the pressure, vibration, strain and stress, and acoustic signals in an attempt to obtain information to assess the blades’ condition. This is however easier said than done under practical operating situations in the plant. Common blade faults could be broadly classified into the following categories: rubbing, cracking and foreign object damage (FOD) or lost part, blade deformation (twisting, creeping, corrosion and erosion), blade fouling and rotating stall, blade fatigue failure, and blade root attachment problems (root crack and loose blade). A review of each of such faults and diagnostic methods proposed in the literature is presented. 3.1 Blade Rubbing Meher-Homji (1995b) reported that blade rubbing in gas turbine could be detected from increased rotor sub-harmonic, sub-synchronous vibration at the rotor natural frequency observed in the vibration spectrum. It was also suggested that the acoustic diagnostic method is viable to detect blade rubbing using stethoscopes during gas turbine start up and shut down, Meher-Homji (1995b). Simmons et. al (1992) reported that the measurement of the blade tip clearance using optical probe could provide early indication for blade rubbing problem if the measured blade tip clearance has reduced. Fisher (2000) commented that techniques such as the performance and vibration monitoring are dependent on a secondary effect being produced by the fault. For instance, blade rubbing or other blade faults could only be monitored by these techniques provided sufficient material loss had occurred and results in an imbalance or loss of performance. Kubiak et. al (2001) reported that blade rubbing could also be detected if the blade passing frequencies (BPF) amplitude is found to be extremely high in the vibration spectrum. 3.2 Blade Cracking, FOD and Lost Part Beebe (1998) in a similar work for steam turbines reported that the breaking off of the blade (lost part) could be detected using the vibration analysis and performance monitoring. Yuan et. al (1999) used the short time Fourier Transform (STFT) and Mallat wavelet decomposition method to analyze the vibration signal of a cracked blade. It was shown that this method was viable in detecting cracked blades based on the presence of higher-frequency impact signals in the STFT maps and wavelet decomposition map. These impact signals were absent in these vibration maps if the blade is in good condition. Doftman et.al (1999) presented in their paper that the turbine blade root cracks was induced by the fatigue initiated by the rotor torsional vibration. Torsional vibration is generally a sporadic, transient phenomenon resulting from a sudden load changes due to changes in load. The measurement and monitoring of the torsional vibration could then be used to provide early warning for the turbine blade root crack if the torsional vibration detected differed from the baseline data. This method was also reported by Maynard (2001). Damage of turbine blades, including cracks, tends to shift the blade resonant frequencies. The monitoring of the blade’s resonance could therefore be adopted as a blade damage indicator. Based on this principle, Mercadal et. al (2001) proposed a method using a non-contacting stress monitoring system to monitor the blade’s resonance in an attempt to diagnose the cracked blade. In addition, turbine blades crack susceptibility could also be predicted based on the life evaluation method using fracture mechanics theory in computer simulation and modeling. This method has been studied by Bernstein et.al (1992), Rao et. al (2001) amongst others.

4

APVC2007

APVC2007, Sapporo, Japan

3.3 Blade Deformation (Twisting, Creeping, Erosion, and Corrosion) Blade deformation faults in principle could be detected by measuring the pressure field around the blades since the deformed blade would alter the pressure field around its body. This method had been proposed by Mathioudakis et. al (1991). Beebe (1998) commented that the erosion of the blade due to the solid particles could also be detected using the performance monitoring method when turbine efficiency drops. 3.4 Blade Fouling and Rotating Stall Meher-Homji (1990) presented a case study on blade fouling in axial compressor. He found that compressor blade fouling could be detected based on performance monitoring. A drop in the overall power output and thermal efficiency of the gas turbine could potentially indicate the presence of fouling. Meher-Homji (1995a) commented that rotating stall in the compressor section could also be detected by using performance monitoring. When stall occurs, the overall performance of compressor decreased thus indicating a likely fault. Mathioudakis et. al (1991) reported that measurement of the pressure field around rotating blades in turbomachinery is more sensitive in detection of fouled blades than conventional performance monitoring. 3.5 Blade Fatigue Flow-induced resonant vibration is usually the cause for the blade fatigue failure in gas turbine. Blade fatigue could be detected using the strain gauge method. This method was studied by Mercadal et. al (2001) amongst others. 3.6 Blade Root Attachment Problems (Loose Blade) Kuo (1995) studied the diagnosis of the loose blade with Fourier analysis of the vibration signals. He used neural networks and fuzzy logic methods to develop a pattern recognition algorithm to enhance the vibration analysis of the loose blade. Lim et.al (2005) reported on laboratory studies on the use of wavelet analysis in the detection of loose blades. Changes in the wavelet decomposition maps during run up coast down conditions were demonstrated to show up loose blades which were not apparent in the FFT spectrum. 3.7 Use of Vibration Measurements in Blade Faults Diagnosis Under practical field conditions, vibration inevitably represents the most readily and widely used method for blade fault diagnosis. Vibration signal is conventionally analyzed with Fourier analysis to obtain the vibration spectrum. Mitchell (1975) presented that the relative changes in the blade passing frequency (BPF) and its harmonics could provide useful information. BPF amplitude was reported to increase at both low flow (near surge) and at high (approaching choke) conditions. Baines (1987) however commented that vibration analysis could only detect the blade faults if severe damage had occurred at the blade, while a minor shape modification of a few blades would not be evident in the spectrum. The use of vibration analysis for blade faults diagnosis was also studied by Simmons et. al (1986, 1987) and Parge (1990). It was reported that blade faults could be detected from observations of relative changes in the BPF and its harmonics amplitude. Mathioudakis et. al (1990) conducted an experimental investigation of the compressor casing vibration correlated to the pressure field around the compressor blade for different blade faults condition. In the investigation, he found that the unsteady pressure field measurements provided a clearer picture of the blade faults than the conventional vibration signals. Using non-parametric identification methods, acceleration outputs were correlated to unsteady pressure measurements taken by fast response transducers at the inner surface of the compressor casing. The transfer functions allow reconstruction of unsteady pressure signal features from the accelerometers readings. It was demonstrated that information

5

APVC2007

APVC2007, Sapporo, Japan

could be deduced from casing measurements without intrusive measurements in the inner working sections. Meher-Homji (1995c) commented that a considerable amount of experimental work still needs to be done to correlate blade problems with the vibration signatures, but there is little doubt that useful information does exist in the vibration signatures as demonstrated from the case studies presented below.

4. Shaft Seizure as From Stator Blade Rubs Four gas turbines each rated 100MW were used in a combined cycle plant. The turbines had been in service for several years, but had no baseline vibration spectrum taken during new conditions. Baseline spectra blade faults diagnosis was nevertheless obtained under existing operating conditions. Periodic monitoring was undertaken over several months. Detailed vibration spectra at base load condition were obtained on a particular unit GT4 before a scheduled overhaul. The measurements were then correlated to the blades’ condition during overhaul inspections. During the overhaul, minor maintenance works on the compressor blades were conducted. No major defects were found on rotor and stator blades. Measurements immediately after the overhaul were undertaken to be used as baseline information for blade faults diagnosis based on the assumption that all blades in GT4 were in good condition immediately after the overhaul. Comparison of vibration spectra of GT4 to the three other operating units was also undertaken. Assessment of BPF and sidebands with statistical analysis to determine standard deviations under normal conditions were undertaken to establish alarm limits for blade faults diagnosis. Three months of operation after the overhaul, rubbing like noise was audible from the compressor section during the coast down of GT4. Under steady state operating conditions (with load and full speed no load), this noise was inaudible. Plant personnel also noted that the coast down time of the unit from load to stand still conditions reduced significantly. Under prior normal conditions, the coast down time was approximately four hours which had now reduced to approximately 20 minutes instead. A vibration investigation was initiated. Typical vibration spectrum of the measurement is given in Fig. 5(b) & (d). Vibration spectrum at the same measurement location and direction undertaken two months earlier are given in Fig. 5(a) and (c). Comparison of the spectrum showed distinct increase in specific blade passing frequencies (BPF), and with increased side bands of the BPF at intervals of the synchronous rpm. BPF at 3050 Hz, 3250 Hz, 3850 Hz and 3900 Hz were exceptionally high in vertical axis. Measurements in axial axis showed that the BPF of 3050 Hz was extraordinary high. The BPF of 3050 Hz (corresponding to compressor blades row 11, 12 and 14) in the vertical and axial directions showed increase with standard deviations as high as 13 times. For BPF of 3250 Hz (row 13), an equally high amplitude exceeding 28x standard deviations was noted. Other BPF of 3850 Hz (row 15-21) and 3900 Hz (stator blades row 15-20) also increased above 7x standard deviation. Typical trends of the BPFs are shown in Fig. 6. Multiple harmonics of 3400 Hz, 3500 Hz, and 3600 Hz were also noticeable. Correlated with the physical evidence of reduced coast down time (which was obviously a strong indicator of increased friction i.e. rub) and with an audible rubbing sound, it was suspected that there were blade rubs. It could in fact be concluded with sufficient confidence that blade faults had occurred in the latter rows of compressor section of GT4. Despite these concerns on the unit, the turbine was to remain in operations since the overall vibration levels as recorded in the DCS was substantially below alarm levels. The rotor of GT4 seized up the following day after this measurement during its coast down. The unit could not be turned (barring) until the subsequent day after cooling down to cold conditions. Due to load demands from the electricity grid, the unit was brought back to service. This resulted in a recurring shaft seizure during coast down.

6

APVC2007

APVC2007, Sapporo, Japan

(a)

Axial direction: 2mths before seizure

(c) Vertical direction: 2mths before seizure Fig. 5:

(b) Axial direction: 1 day before seizure

(d) Vertical direction: 1 day before seizure

Acceleration spectrum as measured on bearing 2 months and 1 day before shaft seizure

The extremely high BPF of rotor rows 11-21 and stator rows 15-20 implied that the clearance between the rotor with stator and casing on these rows had been reduced. This was consistent with published works of Kubiak et.al (2001) that reported that if the BPF had increased significantly in the vibration spectrum, blade rubbing could probably be present. The multiple harmonics noticed around 3400 Hz to 3600 Hz (which is not the BPF) in the vibration spectrums provided further evidence of rubbing. The rubbing between the rotor and stationary components could generate excessive localized heat and potentially causing the rotor to bow. A bowed rotor could further increase the rubbing. This therefore explained why the rotor seized up. On the next day, with cooling the bowed rotor in all likelihood straighten which made it possible for the rotor to turn - only to be brought back to service with obvious consequences of a repeat shaft seizure that obviously confirmed the vibration assessment.

7

APVC2007

APVC2007, Sapporo, Japan

Amplitute, Gs

An emergency outage was initiated on the unit for a physical 1.6 inspection. It was found that some 1.4 (Row 11,12,14) 1.2 compressor stator blades particularly 1 0.8 at row 20 and row 21 at the upper (Stator 15-20) 0.6 part of compressor casing showed 0.4 0.2 traces of rub marks at the tip of the (Row 15-21) 0 blades. This finding was in good Date agreement with the assessment that the higher BPF seen on the vibration Fig. 6 Trending of selected BPFs spectrum was due to the rubbing in gas turbine. Further inspections confirmed more severe and underlying faults on labyrinth glands. The labyrinth glands immediately after the last row of compressor blades experienced a severe rubbing against the rotor shaft. The rubbing was approximately 3 mm in depth which could have reduced the efficiency of the gas turbine due to the pressure leakage in the compressor. Root cause analysis of the rubbing in the unit suggested that rotor eccentricity was the likely reason of the rub. Radial blade rubs are a more common type of blade tip rubbing caused by excessive differential expansion of the rotor and casing components during transient event (shut down, emergency trip, start up). Neither of these anticipated conditions was found after the inspection. There was a suggestion by the OEM that the root cause could be due to the compressor casing distortion. Casing distortion in turn could be due to the non-uniform insulation on the compressor casing. Major correction works on the unit was carried out to improve the insulation on the compressor casing. The unit was put into service again although the root cause of the rubbing was not conclusively identified. Freq 3050 Freq 3850 Freq 3900

5. Cracked Shaft and Rotor Rub This case was for a 100 MW in a different power generation plant in Malaysia. Several months before the crack was detected, steady and significant increase in vibration levels were noted at the control panel. 12

10

Brg Brg Brg Brg Brg

6

No.1 No.2 No.3 No.4 No.5

4

2

5-Feb

4-Feb

3-Feb

2-Feb

1-Feb

31-Jan

30-Jan

29-Jan

28-Jan

27-Jan

26-Jan

25-Jan

24-Jan

23-Jan

22-Jan

21-Jan

20-Jan

19-Jan

18-Jan

17-Jan

16-Jan

15-Jan

14-Jan

13-Jan

12-Jan

11-Jan

9-Jan

10-Jan

8-Jan

7-Jan

0 6-Jan

Velocity (mm/s Pk)

8

Date

Fig. 7:

Overall levels for gas turbine with (subsequent detected) cracked shaft

Fig. 7 shows trends of the x1 synchronous component for bearing casings in vertical direction over a month time span. Bearing No. 1 (compressor non-drive end), No. 3

8

APVC2007

APVC2007, Sapporo, Japan

(turbine drive end) showed marked increase over the latter two weeks period. Vibration levels in the four months period prior to this were less than 1 mm/s for bearings No. 1 and 3, and 3.3 mm/s for bearing No. 5 (generator non-drive end). The alarm limit was 12mm/s and a trip level at 25mm/s (0-Pk). A vibration investigation was initiated a month later. The synchronous amplitude and phase angle plots against rpm are shown in Fig. 8. The vibration spectra during the run up past the first critical speed and as the unit approached the second critical speed were dominated by the synchronous x1 RPM component. Vibration amplitudes increased substantially as the second critical of 2280 rpm was approached. At this stage, high harmonics of x2, x3 and x4 rpm were noted. This suggested a rotor rub at the second critical due to excessive amplification as the unit approached this critical. After passing the second critical, vibration levels dropped with less prominent harmonic components. This suggested the rub being relieved at this point in time. As the unit approached full speed no load (FSNL) conditions at 3000rpm vibration levels were below trip levels. While settling at full speed (with marginal over speed) an instantaneous increase of vibration levels most severe at the outboard (non-drive end) compressor bearing with peak velocity levels of up to 194 mm/s occurred. This was accompanied by a phase shift of approximately 200o. A manual trip was in progress (due to an apparent time delay in the automatic multiplexed protection system used on the unit).

Velocity mm/s

40 30 20 10 0 0

500

1000

1500

2000

2500

3000

3500

RPM

180 Phase (degree)

120 60 0

~200o phase shift

-60 -120 -180 0

500

1000

1500

2000

2500

3000

3500

RPM

Fig. 8:

Run up amplitude (x1 RPM) and phase angles

During the coast down, the dynamic amplification at the second critical was more severe as compared to the run up; Fig. 9. The vibration response in the vertical directions was significantly higher than response in the horizontal direction particularly in the compressor NDE (Brg. 1) and turbine DE bearings (Brg. 3). A cracked shaft and/or thermally bent shaft were suspected to the root cause of the increasing vibrations particularly at the compressor bearing. A full rub was suspected to occur as the unit passes through the second critical, with further development of a thermally bent shaft aggravating the problem and hence the severe vibration excursion.

9

APVC2007

APVC2007, Sapporo, Japan

(a) During run up at 2826 rpm Fig. 8:

(b) At FSNL 3008 rpm with rubs

(a) During run up (b) During coast down Fig. 9: Peak held spectra during run up and coast down after a rub on Brg. 1 (vertical) A subsequent run up investigation of the gas turbine was initiated by the plant with the generator uncoupled to isolate any fault from the generator. This run-up re-confirmed higher response in the vertical directions than the horizontal, with highest response at the compressor bearing. The unit was removed for an inspection where a 100mm longitudinal crack was found in the compressor section, together with signs of severe rubs on the compressor blades. This resulted in a downtime of more than 8 months awaiting a replacement shaft at a time when there was almost no spare capacity in the national grid.

6. Time Varying Vibrations from Seal Rubs This study while less dramatic was initially a little puzzling. Scheduled maintenance work was undertaken on a 100 MW gas turbine. When the unit was brought back to service vibration of the turbine bearing was observed to increase gradually after start up; increasing from amplitudes of less than 2 mm/s initially to 7 mm/s. Trim balancing works were attempted by the OEM which did not reduce the vibrations. It was also noted that the turbine vibrations after some time (up to 8 hours operation) in fact dropped and increased with time over a cycle of 30 to 150 minutes. A typical chart recording showing bearing vibrations and exhaust temperature (confirming steady load conditions) are given in Fig. 10. It was also observed that phase angles were changing with time together with the amplitude time variations. This phenomenon was repeated with daily operations. Sacchi et al (1982) reported a similar case where an oil gland rub resulted in a vibration with continuous phase angles increase. The fault was identified as oil leaks where the seal was completely covered by a carbonized oil layer whose thickness was such that it reduced the diametrical clearance between the seal and shaft. The problem was related to insufficient seal pressurization, where oil leakage from the seal into a hot ambient was subjected to a carbonization process with a progressive formation of a carbonized oil layer. High vibration levels resulted in wearing of the carbonized oil layer; and the process was repeated. Inspection of the oil glands on this unit confirmed signs of leaks, and the problem

10

APVC2007

APVC2007, Sapporo, Japan

resolved with rectifications of the oil gland and restoration of proper pressurization. Time varying vibrations over ~30 to

Fig. 10 Station chart showing bearings vibrations and process parameters

7. Issues in Poor Detection of Blades Related Failures There ought to be concern that the single largest failure in gas turbines are often not readily detectable by conventional overall vibration monitoring as evident from the field experiences presented here. Protection systems as used in most installations in the field are traditionally based on overall vibration amplitudes. These overall amplitudes are more often than not dominated by the synchronous and possibly second and third harmonics of rpm. As such only x1, x2 and higher components associated with mass unbalance and misalignment would be readily detected. Blades related failures do not generate sufficient excitation that would show up as a substantial increase in the overall vibration levels. Plant operators are often concerned only if the vibration levels approaches the alarm or trip levels. This therefore often results in blade related failures not detected. The case studies also showed that blade related faults can indeed be detected before catastrophic failure provided adequate analysis was undertaken. This involved analysis of blade passing frequencies, sidebands of BPF, and wavelet analysis. Some faults such as looseness at blade roots are also not readily detected under steady state conditions, but only during transients with wavelet decomposition maps for example. There is still a gap before some of the diagnostic techniques that showed promise undertaken in research find its way into practice for operational machines.

References (1) (2) (3) (4) (5)

(6)

Baines, N., Modern Vibration Analysis in Condition Monitoring; Noise and Vibration Control Worldwide, May 1987, pp.148-151. Barnard, I., Engineering Asset Management: An Insurance Perspective; 1st World Congress on Engineering Asset Management, Australia, July 2006. Beebe, R., Condition Monitoring of Steam Turbines by Performance Analysis; 52nd Conference of the Machinery Failure Prevention Society, Virginia Beach, 1998. Bernstein, H.L., Allen, J.M., Analysis of Cracked Gas Turbine Blades, Journal of Engineering for Gas Turbines and Power, Vol. 114, April 1992, pp.293- 300. Dorfman, L.S., Miroslav, T., Torsional Monitoring of Turbine-Generators for Incipient Failure Detection; Sixth EPRI Steam Turbine/Generator Workshop. August 17-20, 1999. St. Louis, Missouri. Fisher, C., Gas Turbine Condition Monitoring Systems - an Integrated Approach; IEEE

11

APVC2007

APVC2007, Sapporo, Japan

(7)

(8) (9) (10)

(11) (12)

(13)

(14) (15) (16) (17) (18)

(19)

(20)

(21)

(22) (23) (24)

(25) (26)

Aerospace Conference 2000. Kubiak, J., Gonzalez, G., Garcia, G., Urquiza, B., Hybrid Fault Pattern for the Diagnosis of Gas Turbine Component Degradation, International Joint Power Generation Conference, New Orleans, Louisiana., June 4-7, 2001, Paper no: PWR-19112. Kuo, R. J., Intelligent Diagnosis for Turbine Blade Faults Using Artificial Neural Network and Fuzzy Logic; Engineering Application of Artificial Intelligence, Vol. 8, 1995, 25-34. Lim, M.H., Leong, M.S., Diagnosis for Loose Blades in Gas Turbines Using Wavelet Analysis, ASME Journal of Eng. for Gas Turbines and Power, Vol. 127, 2005, pp. 315-322 Maynard, K., Application of Torsional Vibration Measurement to Blade and Shaft Crack Detection in Operating Machinery; Maintenance and Reliability Conference, Gatlinburg, Tennessee, May 6-9, 2001. Mathioudakis,K., Loukis, E, Papailiou, K.D., Casing Vibration and Gas Turbine Operating Conditions; ASME Journal of Eng. for Gas Turbine and Power, Vol. 112, 1990, 478-485. Mathioudakis,K., Papathanasiou, A., Loukis, E., Papailiou, K.D.,Fast Response Wall Pressure as a Means of Gas Turbine Blade Fault Identification; ASME Journal of Engineering for Gas Turbine and Power, Vol. 113, 1991, 269-275. Meher-Homji, C.B., Gas Turbine Axial Compressor Fouling – a Unified Treatment of its Effects, Detection and Troubleshooting; ASME Cogen Turbo IV, New Orleans, August 27-29, 1990. Meher-Homji, C.B., Blading Vibration and Failures in Gas Turbines: Part A – Blading Dynamics & the Operating Environment, 1995, ASME paper no. 95-GT-418. Meher-Homji, C.B., Blading Vibration and Failures in Gas Turbines: Part B – Compressor and Turbine Airfoil Distress, 1995, ASME paper no. 95-GT-419. Meher-Homji, C.B., Blading Vibration and Failures in Gas Turbines: Part C – Detection and Troubleshooting, 1995, ASME paper no. 95-GT-420. Meher-Homji, C.B., Blading Vibration and Failures in Gas Turbines: Part D – Case Studies; 1995, ASME paper no. 95-GT-421. Mercadal, M., VonFlotow, A., Tappert, P., Damage Identification by NSMS Blade Resonance Tracking in Mistuned Rotors, IEEE Aerospace Conference Proceedings, Vol. 7, 2001. Mitchell, J., Examination of Pump Cavitation, Gear Mesh and Blade Performance Using External Vibration Characteristics, Proceedings of the 4th Turbomachinery Symposium, Texas A&M University, Oct 14-16, 1975, pp. 39-45. Parge, P., Non-Intrusive Vibration Monitoring for Turbine Blade Reliability, Proceedings of Second International Machinery Monitoring and Diagnostic Conference, Los Angeles, Oct 22-25, 1990, pp. 435-446. Rao, J.S., Pathak, A., Chwla, A, Blade Life: A Comparison by Cumulative Damage Theories, Journal of Engineering for Gas Turbine and Power, Vol. 123, No. 4, October 2001, pp. 886-892. Saachi G., Moro, B., Vibratory Problems on Gas Turbines Rotors, International Federation for Theory of Machine and Mechanisms Conference Proceedings, Italy, 1982, pp. 7-15. Simmons, H.R., Non-Intrusive Detection of Turbine Blade Resonance, Third EPRI Conference on Incipient Failure Detection in Power Plants, Philadelphia, 1987. Simmons, H.R., A.J. Smalley, C.E. Edlund, and R.W. Frischmuth,. Monitoring Compressor Blade Tip Clearance in Combustion Turbines, EPRI Steam and Combustion Turbine Blading Conference, Orlando, Florida, January 29-31, 1992. Simmons, H.R., A Non-Intrusive Method for Detecting HP Blade Resonance, 1996, ASME Paper No. 86-JPGC-Pwr-36. Yuan, Q., Xie, H., Meng, Q.J., Fault Diagnosis of Turbine Blade by Wavelet, A-PVC Conference, Singapore, 1999, Vol. 1, pp. 170-175.

12