打的源头，当压气机中压缸打开时发现大部分的动叶和静叶由于巨大的变形从叶台根部断裂，断裂的叶片表面已经完全变形（顺时针方向）。因此，尽管一系列动叶和静叶已损坏，问题的关键是哪一片是第一片断裂的叶片，不论是这些断裂叶片中的一片还是被击打导致叶片断裂，但这些击打物都是来自压气机。因此，击打物通常是已断裂叶片中的任何一片或者以下的一些组合：
1. Something left during last inspection
2. Failure of fixing material and hitting the other components
3. Dislodging of metallic piece from stationary blade and hitting the other components. 4. Dislodging of metallic piece from moving blades and hitting the other components.
5. Dislodging of metallic piece from moving and stationary blades and hitting the other components. In order to find the root cause, the factors mentioned above, theory of elimination have been applied to come to the final cause of failure. 1、上一次检查时遗留在压气机内部的物品。 2、压气机的固定部件脱落击中了其他部件。
3、压气机静叶中含金属物质的部件掉块击打了其他部件。 4、压气机动叶中含金属物质的部件掉块击打了其他部件。。
5、压气机动叶和静叶中含金属物质的部件掉块击打了其他部件。为了寻找断裂的根源，综合如上所述的各项因素，采用排除法推测最终的叶片断裂源头。
Failure due Something left during last inspection:-:-Inspection was carried out under the supervision of GE & BHEL hence chances of leaving some object inside the compressor is very remote, besides if some thing was left then the damages would have occurred during re commissioning itself when the machine was started & stopped for a number of times. Therefore the probability of damage of compressor blades due to left over material is very remote and can be ruled out.
若上一次检查时有物品遗留在压气机内，检查是在GE专家和印度巴拉特重型电气有限公司联合指导之下完成的，因此检查时有物品遗留在压气机内部基本是不可能的，假设确实有物品遗留，那么，压气机在试车时就会发生损坏事故，何况机组已经启停了许多次。因此，有物品遗留在压气机内部基本确定是不可能的，而且完全可以将其排除。
Failure of fixing material: -If the source of DOD is due to the failure or dislodging of fixing material, then missing of fixing material would have observed but on physical inspection of compressor no fixing material was found to be missing. Hence the failure of compressor blades due to Failure of fixing material is ruled out.
若叶片断裂的源头是压气机的固定部件脱落击中了其他部件，那么，在压气机开缸进行检查时肯定会有固定部件脱落的痕迹，但是，实际上刚好相反。因此，压气机的固定部件脱落击中了其他部件也完全可以排除了。
Failure due to dislodging of metallic piece from moving/ stationary blade: A number of moving as well as Stationary blades have been found to be uprooted from the platform. And this failure is sufficient to cause extensive damage. Now the question is why this component has failed. The component will fail due to the following reason :
若压气机动叶或静叶中含金属物质的部件掉块击打了其他部件：动叶和静叶从叶台根部连根拔起的数量一样多。因此，这些大量的断裂叶片足够产生巨大的力量将压气机损坏。现在问题的关键是为什么这些叶片会断裂，叶片断裂无外乎如下原因： a) Fatigue A、疲劳
. HCF
1、高周疲劳
i. Resonance during critical Speed, ①过临界转速时产生的共振。 ii. Flow induced vibration-Flutter. ② 流量变化引起的振动---颤振 iii. Rotating Stall. .③旋转失速
LCF not applicable for compressor 2、低周疲劳不适用于压气机
Thermo mechanical Fatigue crack (TMF) is very rare for initial 5 stages of compressor blades. 3、热机械疲劳裂纹基本不会发生在压气机叶片的前5级 b) Metallurgical Non Conformity. B、冶金工艺不合格
c) Bending over load (Impact). C、叶片的挠曲故障
d) Machining Defect (Notch, Tool mark) D、制造因素（刻痕、刀痕） e) Forging Defect E、锻造工艺的影响 f) Over Load F、负荷过载
g) Corrosion/Erosion. G、叶片腐蚀
By doing the fractographic analysis of the failure surface of the failed component it is possible to know the exact cause of failure. But the fracture surface of the damaged uprooted stationary as well as moving blades are so badly deformed that fractographic analysis will not give any useful information.
通过断裂叶片部件表面的断口分析是可能发现导致断裂原因的。但是动叶和静叶叶片根部连根拔起产生的断裂面已经严重的变形，因此，断口分析很难得到有用的信息。
In all cases the failed surface, the failed surface is having different colour (dark as well as bright)
这三起叶片断裂事故的叶片表面有不同的颜色（黑色和白色）
相似的叶片断裂表面
From the preliminary study and the failure pattern, cause of failure of
GT#2B,2A & 3A appears to be same.
经过初步研究故障形式表明，导致GT#2B,2A & 3A叶片断裂的原因相同。
压气机叶片颤振导致高周疲劳断裂
Gas turbine hot path components are very prone to high temperature Creep, Thermo mechanical fatigue and Low Cycle fatigue failure. Designers, depending upon duty condition (Cyclic Load), design hot path components, considering Creep life/ fatigue life of almost 90% of the life of material after which development of these type of defect i.e. Creep Crack / fatigue crack is expected.
燃机透平的热通道部件非常容易遭受高温蠕变，热机械疲劳以及低周疲劳。从设计者角度说，这些取决于负载的条件（启停的负荷），设计热通道部件，必须考虑蠕变寿命、在材料寿命范围内承受90%的疲劳寿命，因此，在日后机组的运行中这些设计缺陷也即蠕变裂纹、疲劳裂纹在可预见期是会产生的。
Depending upon the amplitude and frequency of cyclic load, cyclic fatigue is again classified in to Low cycle fatigue failure (LCF) where the amplitude of cyclic load is such that it reaches up to yield stress but the frequency of cyclic load is less. Where as in case of High cycle fatigue failure (HCF), the amplitude of cyclic load is much below the yield stress but the frequency of cyclic load is more.
机组日常启停中由于振动的幅度和频率，循环疲劳不停的重复出现各级的低周疲劳，当机组负荷变动时产生的振幅会达到屈服应力，但是负荷变动时的产生的频率较少发生。如果发生高周疲劳，那么，负荷变动产生的振幅低于屈服应力，但是负荷变动时的产生的频率较多发生。
In any fatigue failure, the failed surface will have the following indications: . Beach mark in some cases Beach Marks is visible even with naked eye. . Thump Nail Shape, this is also visible with naked eye. . Striations this requires high resolution Microscope.
When the cleaned failed surface is seen with high resolution microscope e.g. with Scanning Electron Microscope (SEM), then the presence of those indications are evidenced as shown below. 任何疲劳故障导致断裂的表面有如下特征：
1、海滩状特征甚至在很多类似案例中用肉眼即可看见。 2、重击形成的钉子形状，同样可以用肉眼看见。 3、条纹状特征需要高分辨率的显微镜。
当叶片断裂表面清理干净后，使用高分辨率的显微镜（例如电子显微镜），那么，就可看见如下图所示的特征：
Presence of Thump Nail Shape, Beach mark with naked eye indicates that the failure mode is due to fatigue, obviously this is to be confirmed with Fractographic analysis and the cause of fatigue most likely will be due to flow induced vibrations coupled with rotating stall /resonance/ Flutter because of modulation of IGV with frequency. As a result of IGV modulation, change in GT load with change in frequency is more compare to change in GT load without IGV modulation.
肉眼可看见重击形成的钉子形状特征、海滩状特征表明叶片断裂的形式为疲劳断裂，断口分析可以确切地证明，叶片疲劳的原因最大可能是由于IGV频繁的调整导致流量变化引起自激振动，再加上旋转失速、共振、颤振。因为IGV的调整，燃机的负荷也跟着频繁的调整，结果燃机负荷超过了IGV的调整范围。
This additional fluctuation of load due to modulation of IGV is creating more flow induced vibration which is pulsating in nature due to aerodynamic flow instabilities. This aerodynamic flow instabilities (separation of flow on both leading and trailing edges), tend to formation of vortex. The vortex Shedding frequency is determined by STROUHAL NUMBER(St).The Strouhal number is named after Vincenc Strouhal & is an integral part of fundamentals of fluid mechanics.
因为IGV的调整引起额外的负荷波动导致更多的自激振动，这种有规律的振动从空气动力学本质来看就是流动的不稳定。这些空气流动的不稳定（主导流量和跟随的流量是脱离的），经常容易形成涡流脱落，涡流脱落频率由斯特罗哈数所决定。斯特罗哈数在特劳哈尔数之后，他们都是流体力学不可或缺的组成部分。
Vortex Shedding Frequency (Fv) = Strouhal No.(St) X Flow Velocity (Vf) / Vortex Shedder Width(Wv) which is a Hydraulic parameter and depends upon the angle of inlet and exit.
Fv =St XVf /Wv =(StX Q)/(AXWv) Where Q is the flow &A is the area of flow 涡流脱落频率（Fv）=斯特罗哈数（St）×流速(Vf)/涡流特征的宽度(Wv)，涡流特征的宽度(Wv)取决于涡流进入和离开时的角度
Fv =St ×Vf /Wv =(St×Q)/(A×Wv)，其中Q为涡流流量，A为涡流区域的面积
Since for a particular, compressor St , Q, A, Wv are assumed to be constant, but in actual working condition St , Q, A remain constant but Vortex Shedder Width (Wv) varies with IGV position & fouling on compressor blades, as a result, Vortex Shedding Frequency changes . Since maximum fouling takes place in the initial stages of stationary & moving blades of compressor, hence even for a very less change in IGV angle, separation of flow takes place resulting in formation of vortex shedding. For Reynolds number in the range of 800 – 200,000 there exist two values of Strouhal number. The lower frequency is attributed to the large scale instability of the wake and is independent of Reynolds number. The higher frequency Strouhal number is caused by small scale instabilities from the separation of shear layer. If this Vortex Shedding Frequency coincides with natural frequency of Blade, the Blade will oscillate in harmony with the Vortex Shedding and begin to FLUTTER. FLUTTER imposes significant aerodynamic lateral and torsional forces on the blade, resulting in more than expected stress concentration just above the platform of the blade, at subsynchoronus frequency that can have a detrimental effect on blade life.
具体到本文案例来说，压气机的斯特罗哈数St、涡流流量Q、涡流区域面积A、涡流特征的宽度Wv假设是不变的，但是在真实的工作条件，压气机的斯特罗哈数St、涡流流量Q、涡流区域面积A保持不变，涡流特征的宽度Wv是随着IGV的角度和压气机叶片的结垢程度而变化，因此，涡流脱落频率也跟着改变。因为结垢最大的地方发生在压气机静叶和动叶的前面几级，因此，即使IGV改变一点点，发生空气气流分离就会产生涡流脱落。雷诺数在800-200，000范围之内有两种不同特罗哈数，大规模的不稳定气流会产生低频，而且不受雷诺数的影响。更高频率的斯特罗哈数由小规模的剪切层不稳定气流产生。如果涡流脱落的频率和叶片本身固有的频率一致，那么，叶片就会产生与涡流脱落频率同样的振动，然后开始发生颤振。颤振强加于气流的气动力侧面，同时扭矩力作用于叶片，结果是更多的应力集中在叶台上部就如推测的那样，这种频率影响叶片的寿命。
As already mentioned above, fouling is more in the initial stages of compressor, therefore, initial stages are subjected to flutter induced vibration more compare to other stages because of flow separation. This is why the initial stages of compressor blades and vanes fail due to flutter. In case of RGPPL-Dabhol, the failures took place in R3 & R1; therefore, the following can be concluded:
综上所述，压气机前几级叶片很容易遭受污染、结垢，因此，因为气流脱离导致压气机前几级相比其他级要承受自激振动。这就解释了为什么压气机的前几级动叶和静叶因为颤振而发生叶片断裂。在RGPPL-Dabhol的案例中，叶片断裂的地方发生在压气机R3 & R1，因此，可考虑采取如下措施：
1. Presence of primary & secondary failed surfaces indicate that the crack developed much earlier and propagated gradually with loading cycle and finally the component has failed due is Tensile over load. Fractographic Analysis will confirm this observation
1、从目前的第一级和第二级叶片断裂表面表明，裂纹扩展更容易，随着负荷的变化逐步地传播，最终导致力矩过载断裂。断口分析可证实此推测。
2. Presence of Thump nail shape & Beach mark on the failed surface of the failed component, indicate that the failure mode of GT#2B , 2A&3A are due to High Cycle Fatigue.
2、通过分析叶片断裂表面表明重击形成的钉子状特征和条纹状特征是导致GT#2B , 2A&3A叶片断裂的主要原因是高周疲劳。