ENGINE COMPONENTS

BACKGROUND

[0001] 1. Field

[0002] Disclosed embodiments are generally related to turbine engines, and in particular to determining and applying a life evaluation for the gas turbine components.

[0003] 2. Description of the Related Art

[0004] Components of gas turbine engines need to be serviced from time to time in order to prevent failures that can result in down engine down time. Currently there are standards set in place that periodically inspect and determine life cycle fatigue life and periodic failure based on a deterministic approach. These methods are conservative and do not take into consideration the scatter in quality of the material properties due to manufacturing variances, site and project specific operating conditions. These methods result in rigid acceptance requirements for evaluating gas turbine components during overhaul and non-flexible mean time between overhaul intervals for gas turbine components.

[0005] Therefore being able to include more data in determining the times for inspection and overhaul can improve inspection times and increase efficiencies.

SUMMARY

[0006] Briefly described, aspects of the present disclosure relate to a system and method for establishing the probabilistic life evaluation of a gas turbine component.

[0007] An aspect of the present disclosure may be a method of determining the probabilistic life evaluation of a gas turbine component. The method may comprise receiving deterministic life data of the gas turbine component at a deterministic life data database, wherein the deterministic life data comprises physical properties of the gas turbine component; storing the deterministic life data in the deterministic life data database, receiving component life response data at a component life response database; wherein component life response data comprises operational and environmental data affecting the gas turbine component; storing the component life response data in the component life response database; and combining the component life response data and the deterministic life data in order to obtain a probabilistic life evaluation result for the gas turbine component.

[0008] Another aspect of the present invention may be a system for the determination of the life cycle of a gas turbine component. The system may comprise a processor adapted to process data; a compiled data database for storing data related to components of a gas turbine engine; sensors connected to the gas turbine component for determining properties of the gas turbine component, deterministic life data stored in a deterministic life database, wherein the deterministic life data is based on strength properties of the gas turbine engine; component life response data obtained from the sensors, wherein the component life response data is transmitted to a component life response data database; and wherein the processor combines the deterministic life data and component life data to establish a probabilistic life evaluation result of the gas turbine component. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Fig. 1 is a diagram of the system for performing the method for evaluating the life cycle of a component.

[0010] Fig. 2 is a flow chart of the steps for evaluating the life cycle of a component.

[0011] Fig. 3 shows two graphs that illustrate the determination of the predictive life result.

DETAILED DESCRIPTION

[0012] To facilitate an understanding of embodiments, principles, and features of the present disclosure, they are disclosed hereinafter with reference to implementation in illustrative embodiments. Embodiments of the present disclosure, however, are not limited to use in the described systems or methods and may be utilized in other systems and methods as will be understood by those skilled in the art.

[0013] The components described hereinafter as making up the various embodiments are intended to be illustrative and not restrictive. Many suitable components that would perform the same or a similar function as the components described herein are intended to be embraced within the scope of embodiments of the present disclosure.

[0014] The system and method of the present invention employs a probabilistic life evaluation algorithm that performs predictive analytics to estimate the risk of failure of gas turbine components. The system and method employs data from sensors and data from analysis of the physical properties of the gas turbine component. The data is defined as the deterministic life data and the component life response data.

[0015] “Deterministic life data” is data that is predicted deterministic life of the gas turbine components assuming several pre-set operating conditions, materials data scatter and inspection data plotted and stored as a response surface. Additionally the deterministic life database includes data on the materials properties of the gas turbine component as manufactured. This is the data pertaining to the strength properties of the gas turbine component and how that component will likely perform based upon its implementation in a system. This data may be supplied by the manufacturer of the gas turbine component. This data includes physical parameters of the component, such as creep, fatigue, tensile properties, pre-installation testing data, such as dimensional and weight variations.

[0016] “Component life response data” is data that is obtained from the gas turbine component while it is installed or located on the gas turbine engine. Component life response data may be ascertained from sensors located on the gas turbine component. The component life response data may also be compiled from inspection data. This data may include engine operational parameters, gas turbine component entry temperature, speeds, time spent at baseload conditions, time to ramp up and time to shutdown, location of damage on the component, the manner in which the damage occurred to the component, environmental conditions at the site where the component is located, such as ambient temperatures, conditions related to the unit operating in corrosive marine environments etc.

[0017] Disclosed is a method and system for determining the probabilistic life evaluation of a gas turbine component. The method and system comprise receiving predicted deterministic life data of the gas turbine components assuming several pre- set several operating conditions, material data scatter, and inspection data plotted and stored as a response surface in a deterministic life database. Additionally, the variation in components material properties to the specification requirements for a new component manufacture may be stored in a materials quality management database, wherein the materials quality data comprises of strength properties of the as manufactured gas turbine component taken from the in situ inspections and from inspections performed during overhaul stored in the component inspection database. The operational and environmental data may be maintained in a component health monitoring database; wherein component life response data comprises operational and environmental data affecting the gas turbine component; storing the component life response data in the component life response database; and combining the component life response data with the materials quality database,, inspection database and the deterministic life data base in order to obtain a probabilistic life evaluation result for the gas turbine component.

[0018] For purposes of explaining the method and system, power turbine disks are discussed herein. It should be understood that while power turbine disks are discussed herein, the system and method may be applicable to other gas turbine components. The system and method discussed herein will provide the probability of failure of a gas turbine component, such as a power turbine disk, through the operational life cycle. The deterministic life data and the component life response data are combined. Once the sets of data are combined the probability of failure will be determined by comparing the results of the combined data to a threshold value. The threshold value will be based on allowable risk levels designated for the gas turbine engine component. This information can be used to determine when a product should be serviced or inspected. This can avoid unnecessary shut downs of the gas turbine engine, which enables more continuous energy output. Furthermore, the determination of the potential failure can also be used to order replacement gas turbine components prior to the anticipated loss of or serving of the component. This can also be used to prevent or reduce the unnecessary stoppage of power production.

[0019] Fig. 1 is a diagram of a probabilistic life evaluation system 100 in which the method is employed. Fig. 2 is a flow chart of the method employed in the system. The following discussion is made in reference to Fig. 1 and Fig. 2. It should be understood that while the diagram of the system shown in Fig. 2 is directed to a particular gas turbine components, it is contemplated that other gas turbine components may benefit from the predictive analysis.

[0020] Fig. 1 is a diagram of the probabilistic life evaluation (PLE) system 100. The gas turbine component 10 is part of a gas turbine engine. The gas turbine component 10 may be any part of a gas turbine engine, such as an airfoil, blade, combustor, power turbine disk, etc. In the embodiments discussed herein the gas turbine component 10 is a power turbine disk.

[0021] Attached to the gas turbine component 10 may be a sensor 12. The sensor 12 may be adapted to transmit data regarding conditions related to the gas turbine engine. For example, the sensor 12 may sense temperatures, pressures, material stress, operational parameters such as shaft speed, time spent at baseload, rate of ramp-up and down from baseload power, etc.

[0022] Data detected by the sensor 12 may be transmitted to the component life response data (CLRD) database 14. The CLRD database 14 may be located at a location proximate to the gas turbine engine, may be part of a cloud storage system or may be located in the same location as the other modules and components that make up the PLE system 100. The CLRD database 14 maintains and stores the component life response data such as discussed above. The CLRD database 14 may include environmental information, operations parameters and data determined from the sensor 12 and other material inspections.

[0023] The PLE system 100 also comprises the deterministic life data (DLD) database 16. The DLD database 16 may be located at a location proximate to the gas turbine engine, may be part of a cloud storage system or may be located in the same location as the other modules and components that make up the PLE system 100. The DLD database 16 maintains and stores the deterministic life data response such as discussed above. This may include data provided by suppliers directed to the material properties of the gas turbine component 10.

[0024] Both the CLRD database 14 and the DLD database 16 may transmit the data to or have the respective data accessed by the PLE processor 20 and the compiled data database 18. The PLE processor 20 is operably connected to the compiled data database 18 and accesses and compiles the data from the DLD database 16 and the CLRD database 14. The data is then stored and processed by the PLE processor 20 at the compiled data database 18. The compiled data database 18 may be located at a location proximate to the gas turbine engine, may be part of a cloud storage system or may be located in the same location as the other modules and components that make up the PLE system 100.

[0025] Reference will now be made to Fig. 2 that shows a flow chart setting out the method of performing the probabilistic life evaluation. In step 102, deterministic life data of the gas turbine component is received by the PLE system 100. The deterministic life data may be from a supplier of the gas turbine component 10. Alternatively, the deterministic life data may be independently determined. The deterministic life data sets forth the material properties of the gas turbine component 10 and the predicted life span of the product when placed within a gas turbine engine.

[0026] In step 104, the deterministic life data is stored in a DLD database 16. As discussed above the DLD database 16 may be located at a location proximate to the gas turbine engine, may be part of a cloud storage system or may be located in the same location as the other modules and components that make up the PLE system 100. The DLD database 16 maintains and stores the deterministic life data response such as discussed above. The deterministic life data may be an estimation of variation in the mechanical properties of a gas turbine component based on supplier quality information for a given manufactured gas turbine component 10 being installed into the engine.

[0027] In step 106, CLRD data from sensors 12 located on the gas turbine component 10 is received at the CLRD database 16. In step 108, the CLRD data is stored in the CLRD database 14. The CLRD database 14 may be located at a location proximate to the gas turbine engine, may be part of a cloud storage system or may be located in the same location as the other modules and components that make up the PLE system 100. Other component life response data may be received from operator or field service personnel identifying the location and size of damage as observed during inspection and entering the information into the PLE system 100.

[0028] In step 110, the component life response data and the deterministic life data are combined by the PLE processor 20 with the probabilistic life evaluation result stored in the compiled data database 18. The PLE result may then be compared to a threshold value. [0029] The comparison to the threshold value can be used by the PLE system 100 to institute a number of steps. For example, if a certain threshold value is exceeded, then the PLE system 100 may order a new gas turbine component 10. In yet another example, if the threshold value is exceeded then the PLE system 100 may institute an order to repair the gas turbine component 10. In another example, if the threshold value is exceeded the PLE system 100 may institute an inspection of the gas turbine component 10. Furthermore, the PLE system 100 may use different threshold values in order to institute different tasks.

[0030] The probabilistic life evaluation algorithm used by the PLE system 100 is a statistical response surface failure function that can map the impact of variation of random variables taken from the DLD database 14 and the CLRD database 16. These random variables may be things such as operating PT entry temperature, operating speed, disk cavity temperatures as a function of bleed air, material properties, field inspection data that inputs to the material models for deterministic lifing models that quantify life at several predetermined operating conditions based on thermochemical fatigue and creep life damage.

[0031] The PLE system 100 uses the data on actual operating history using the engine parameters and provides real time risk levels to disk burst, cyclic and creep lives of the damaged disks. The PLE system 100 provides guidance on acceptance/rejection of observed rotor damage as well as for the subsequent inspection and overhaul of the power turbine rotor. The PLE system 100 can provide field service/fleet management personnel a capability to track the risk profile of the operating gas turbine components such as disks in real time utilizing the PLE system 100 rather than stopping an engine, stripping a PT rotor and scrapping the parts each time damage is observed. Thus the PLE system 100 improves reliability, availability and extended mean time between the next overhaul while reducing the overall scrap rate of the components. The PLE system will enable overall reduction in the currently defined inspection intervals thus increasing the engine availability and reducing costs.

[0032] FIGS. 3 and 4 show two graphs that illustrate the determination of the threshold value. Ni stands for the cycles to crack initiation. Pf stands for product failure. Fig. 3 shows the frequency vs. crack initiation. The frequency is the number of times a given range of Ni is achieved. It is to show the variation in the range of life cycle fatigue life due to variations in parameters such as temperature, speed, power, inspection data, material properties etc.

[0033] Fig. 4 shows the probability of failure. Equation 1 is shown below. Ni is the cycles to crack initiation. The Ni is a function of different variables which represent operating conditions such as temperature, speed, power, inspection data, material properties etc. The subscript i in equation 1 represents the life cycle fatigue computed for one set of combination of variables“a”. The deterministic life data feeds into the modeled response surface that is computed for multiple conditions using equation 2 shown below.

Lg (Ni) creates a linear regression model of Ni on logarithmic scale and forms the mathematical basis for the response surface of deterministic lives. Variable“b” represents the fitting constants that are used to fit Ni vs ai data as shown in equation (1). Various physical properties of the gas turbine component 10 are utilized in order to ascertain a probability of failure, which is the probabilistic life expectancy result. The probability of failure will increase over time as the parameters that impact the probability of failure increase. The determination of the probability of failure and thus the PLE result is discussed below. The determined PLE result is then compared to a threshold value in order to take action on the gas turbine engine.

[0034] In the following example the PLE result it determined for a gas turbine component 10, which in this instance is a power turbine disk. It should be understood that for other gas turbine components different variables may be used.

[0035] The deterministic life data used in this example is the result of the following evaluations; crushing stress evaluation, unzipping assessment and disk burst and yield assessment. These various evaluations are the application of algorithms and calculations know to one of ordinary skill in the art for determining the properties. While the individual combinations are known the combination and application within the manner set forth herein is not known.

[0036] The component life response data used in this example is the result of life cycle fatigue, creep life evaluation, half cycle evaluation, tie bolt mechanical integrity and life assessment evaluation. This data can be taken from properties measured in the environment by sensors 12 located on or near the gas turbine component 10.

[0037] Life cycle fatigue (LCF) may be performed by taking the strain vs component life curves from strain controlled smooth specimen data for estimating initiation life. Additionally fracture mechanics may be based on the estimation for propagation life. The typical lives of components may be converted to minimum lives by using scaling factors.

[0038] A stress analysis of the gas turbine component 10 provides the strain range and maximum von-Mises stress required for the calculation of the LCF damage.

[0039] For crack propagation, effective initial flaw size is based on the capability for detecting cracks on power turbine discs. Crack propagation life is computed only for those power turbine disc features that had predicted initiation life less than the acceptable levels.

[0040] Some of the variables used for evaluation can be LCF -Data, strain range to number of cycles to crack initiation. Stress-strain curve, Fracture toughness, Fatigue crack growth rate curves, PT entry temperature, centrifugal load, rated PT speed, blade mass, constraint and contact boundary conditions etc. These calculations are used to determine the PLE result for the power turbine disc.

[0041] The threshold value to which the PLE results may be compared can be established by having a life assessment performed for the worst engine condition operating at max continuous speed (i.e. 105% of the rated speed). Thus the calculations and variables used for establishing the PLE result are the same that are used in establishing the threshold variable; however the data that is used for the basis of the calculations is different. In other words, a hypothetical situation is presented that would initiate failure and provides a scenario that would likely result in failure.

[0042] While embodiments of the present disclosure have been disclosed in exemplary forms, it will be apparent to those skilled in the art that many modifications, additions, and deletions can be made therein without departing from the spirit and scope of the invention and its equivalents, as set forth in the following claims.