METHOD FOR DETERMINATION OF FATIGUE LIFETIME CONSUMPTION OF AN ENGINE COMPONENT

Abstract

A method for determining fatigue lifetime consumption of an engine component, by defining a reference thermal load cycle, the reference thermal load cycle being characterized by a reference load cycle amplitude and a reference load cycle time, and determining a reference load cycle lifetime consumption. The method includes measuring a temperature of the engine component, determining a thermal load cycle based upon the temperature measurement, determining a load cycle amplitude, determining a load cycle time, relating the load cycle time to the reference load cycle time, thereby determining a load cycle time factor, relating the load cycle amplitude to the reference load cycle amplitude, thereby determining a load cycle amplitude factor, combining the load cycle time factor and the load cycle amplitude factor into a combined load cycle factor for determining a load cycle lifetime consumption.

Claims

1. A method for the determination of fatigue lifetime consumption of an engine component, the method comprising: defining a reference thermal load cycle, the reference thermal load cycle being characterized by a reference load cycle amplitude and a reference load cycle time; determining a reference load cycle lifetime consumption; measuring a temperature of the engine component; determining a thermal load cycle based upon said temperature measurement; determining a load cycle amplitude, defined as a difference between a maximum measured temperature during the thermal load cycle and a minimum measured temperature during the thermal load cycle; determining a load cycle time; relating the load cycle time to the reference load cycle time, thereby determining a load cycle time factor; relating the load cycle amplitude to the reference load cycle amplitude, thereby determining a load cycle amplitude factor; combining the load cycle time factor and the load cycle amplitude factor into a combined load cycle factor; and determining a load cycle lifetime consumption dependent upon the combined load cycle factor.

2. The method according to the claim 1, wherein the reference load cycle time is defined as a time in which an absolute temperature change of the reference thermal load cycle amounts to a specific percentage of the reference load cycle amplitude, and the load cycle time is defined as a time in which the measured temperature changes by a same percentage of the load cycle amplitude of the thermal load cycle.

3. The method according to claim 1, wherein the combined load cycle factor further takes a load cycle mechanical stress factor into account.

4. The method according to claim 1, wherein the combined load cycle factor is determined as:
X=R.sub.T.Math.R.sub.Δt.sup.a.Math.R.sup.T.sup.−b.Math.R.sub.F, wherein X represents a load factor, R.sub.T represents a load cycle amplitude factor, a and b are constants, R.sub.Δt represents the load cycle time factor, and R.sub.F represents the load cycle mechanical stress factor.

5. The method according to the preceding claim, wherein a equals 0.5 and b equals 0.25

6. The method according to claim 1, comprising: determining a reference number of thermal load cycles which equals a number of reference thermal load cycles until a crack is initiated in the engine component; determining a load cycle count factor based upon the combined load cycle factor, wherein the load cycle count factor is defined as the load cycle lifetime consumption divided by the reference load cycle lifetime consumption; and summing up the load cycle count factors of all load cycles during an engine component lifetime, thereby determining a cumulated load cycle consumption.

7. The method according to claim 6, comprising: determining a residual load cycle lifetime expectation as a difference between the reference number of thermal load cycles and the cumulated load cycle consumption.

8. The method according to claim 6, comprising: determining a relative lifetime consumption as a ratio of the cumulated load cycle consumption divided by the reference number of thermal load cycles.

9. The method according to claim 6, wherein the load cycle count factor is determined as:
R.sub.N=N.sub.R/N=R.sub.s.sup.−1/c, wherein R.sub.N represents a load cycle count factor, N represents a number of thermal load cycles with a specific load cycle amplitude and cycle time to crack initiation, N.sub.R represents the number of reference thermal load cycles to crack initiation, R.sub.s is a strain ratio which is determined as a function of the combined load cycle factor, and c is a material parameter.

10. The method according to claim 9, wherein the load cycle strain ratio is expressed in a polynomial form of the combined load cycle factor.

11. The method according to claim 10, wherein the load cycle strain ratio is expressed as a first order polynomial of the combined load cycle factor.

12. The method according to claim 1, wherein the engine component is a component of a gas turbine engine, and the reference thermal load cycle is a load cycle from cold start to full load of the gas turbine engine at a maximum load gradient.

13. The method according to claim 1, comprising: applying the method on-line.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0032] The subject matter of the present disclosure is now to be explained in more detail by means of selected exemplary embodiments shown in the accompanying drawings. The figures show

[0033] FIG. 1 a flow diagram illustrating an exemplary mode of performing the method as disclosed above;

[0034] FIG. 2 a diagram illustrating the correlation between the combined load cycle factor and the strain ratio for a first exemplary engine component; and

[0035] FIG. 3 a diagram illustrating the correlation between the combined load cycle factor and the strain ratio for a second exemplary engine component.

[0036] It is understood that the drawings are highly schematic, and details not required for instruction purposes may have been omitted for the ease of understanding and depiction. It is further understood that the drawings show only selected, illustrative embodiments, and embodiments not shown may still be well within the scope of the herein disclosed and/or claimed subject matter.

EXEMPLARY MODES OF CARRYING OUT THE TEACHING OF THE PRESENT DISCLOSURE

[0037] With reference to FIG. 1, an exemplary mode of carrying out the method as described above is lined out in more detail. In a first step 100, calculations for an engine component and a reference load cycle are performed. The reference load cycle is for instance a cold start of a gas turbine engine, and loading up the gas turbine engine to rated full power with the maximum load gradient for the gas turbine engine, or for the gas turbine engine when operated in a combined cycle power plant and the combined cycle power plant is loaded up at a maximum load gradient. These calculations may include computing the stresses in the engine component which arise due to the temperature differences within the material of the engine component. With this knowledge, a reference number of load cycles until a crack is initiated in the engine component can be derived. In a next step 200, said calculation may be repeated for a variety of load cycles in which the load cycle amplitude and load cycle time are varied. The number of load cycles until crack initiation may be computed for each of the load cycles. Generally, the results of said computations may be represented dependent on the combined load cycle factor. In particular, the strain ratio for each load cycle, wherein the strain ratio is given by

R.sub.s=(N/N.sub.R).sup.c,

wherein N represents the number of thermal load cycles to crack initiation with a specific load cycle amplitude and load cycle time, N.sub.R represents the number of reference thermal load cycles to crack initiation, R.sub.s is a strain ratio which is determined as a function of the combined load cycle factor, and c is the fatigue ductility exponent or Coffin-Manson exponent, which is a material specific parameter, which may also depend on the absolute temperature of the material. In a following step 300 a correlation function between, for instance, the strain ratio and the combined load cycle factor is developed. Such, the strain factor may be represented as R.sub.s=R.sub.s (X), wherein X is the combined load cycle factor.

[0038] It is noted that steps 100 through 300 are generally performed off-line. The following steps will be performed on-line, while an engine is running. Thus, the exemplarily shown method may be considered as comprising two modules, an off-line module 1000 and an on-line module 2000. In the on-line module, in a first step 400 of the on-line module 2000, a temperature of the engine component is measured. Based upon consecutive temperature measurements of the component material temperature, as lined out above, in step 500 a decision is taken whether a thermal load cycle was initiated, and whether said thermal load cycle is completed, or not. If no thermal load cycle was completed, steps 400 and 500 will be repeated. If a thermal load cycle was completed, the combined load cycle factor will be calculated in step 600 based upon the measured load cycle amplitude and load cycle time. Further, in step 600 the strain factor and subsequently the load cycle count factor for the load cycle are computed. In step 700, all load cycle count factors during the lifetime of the engine component are summed up and constitute a cumulated load cycle consumption. In step 800, it is verified whether the cumulated load cycle consumption is larger than the reference number of thermal load cycles, which equals the number of reference thermal load cycles until a crack is initiated. If this is not the case, the on-line module 2000 starts again at step 400. If the cumulated load cycle consumption is larger than the reference number of thermal load cycles, an event or warning may be generated in step 900. The on-line module starts again at step 400. It is appreciated that step 900 may be initiated before the cumulative load cycle consumption actually reaches the reference number of thermal load cycles, such that a power plant operator might schedule a plant outage well before an actual damage occurs.

[0039] FIGS. 2 and 3 depict the correlation between the strain ratio R.sub.s and the combined load cycle factor X for two exemplary engine components. The data points show the results from numerical simulations, while the lines represent the function applied as a correlation in step 600 above. As is seen, an excellent fit of the simulation points and the correlation function results with a third order polynomial. As is furthermore seen, even with a first order polynomial a fairly good fit would be achieved. As is apparent, the correlation varies with different engine component geometries, but fits excellent for a given engine component.

[0040] While the subject matter of the disclosure has been explained by means of exemplary embodiments, it is understood that these are in no way intended to limit the scope of the claimed invention. It will be appreciated that the claims cover embodiments not explicitly shown or disclosed herein, and embodiments deviating from those disclosed in the exemplary modes of carrying out the teaching of the present disclosure will still be covered by the claims.