Patent Application
20190056291
Embodiments of the present specification relate generally to prognostic of reciprocating machine components, and more particularly to systems and methods for inspection and prognostic of piston rings.
Piston rings are widely used in internal combustion (IC) engine cylinders, primarily for sealing purposes. Piston rings also perform other functions, such as, but not limited to, lubrication and heat transfer. Piston rings are made from special materials having capacity to withstand extreme operating conditions and are especially suitable for applications where high pressure, high temperature and high corrosive operating conditions are encountered. In automotive industry, performance evaluation of engine components, especially tribological performance of the piston ring and cylinder bore system, is central for achieving optimum engine efficiency and durability. Optimum condition of engine components, especially of piston rings, is required to manage power loss, fuel consumption, oil consumption, blow by, and harmful exhaust emissions.
A wear rate of a piston ring and a bore system is a significant parameter required to assess performance of an engine as wear results in poor performance, decreased oil economy and eventually requires an engine overhaul. Engine tests for evaluation of performance assessments are costly and time consuming. Mathematical models for various wear mechanisms are used for estimation of deterioration of piston ring performance.
The current operating life of a piston ring having a coating thickness of 200 microns is estimated at 30000 hours. Typically, about 50 microns of wear is observed during the current operating life of the piston. Presently, the piston ring is replaced in a scheduled maintenance after an estimated average life span of the piston ring. However, the estimated average life span is conservative and adds to overhead of maintenance.
In accordance with one aspect of the present invention, a method is disclosed. The method includes obtaining a reference profile of a piston ring. The reference profile comprises a plurality of reference fiducials. The method further includes receiving a measured profile of the piston ring. The measured profile comprises a plurality of measured fiducials. The method further includes aligning the measured profile with the reference profile based on one or more of the plurality of reference fiducials and one or more of the plurality of measured fiducials. The method also includes determining one or more wear parameters based on the aligned measured profile and the reference profile. The one or more wear parameters includes a wear depth value and a wear volume value. The method also includes generating a wear model based on the one or more wear parameters. The method also includes estimating a life of the piston ring based on the wear model.
In accordance with another aspect of the present invention, a system is disclosed. The system includes at least one processor unit and a memory unit communicatively coupled to a communications bus. The system further includes a data acquisition unit communicatively coupled to the communications bus and configured to obtain a reference profile of a piston ring. The reference profile includes a plurality of reference fiducials. The data acquisition unit is further configured to receive a measured profile of the piston ring. The measured profile includes a plurality of measured fiducials. The system also includes a profile registration unit communicatively coupled to the data acquisition unit and configured to align the measured profile with the reference profile based on one or more of the plurality of reference fiducials and one or more of the plurality of measured fiducials. The system further includes a model generator communicatively coupled to the profile registration unit and configured to determine one or more wear parameters based on the aligned measured profile and the reference profile. The one or more wear parameters comprises a wear depth value and a wear volume value. The model generator is further configured to generate a wear model based on the one or more wear parameters. The system also includes a life prediction unit communicatively coupled to the model generator and configured to estimate a life of the piston ring based on the wear model.
These and other features and aspects of embodiments of the present invention will become better understood when the following detailed description is read with reference to the accompanying drawings in which like characters represent like parts throughout the drawings, wherein:
Embodiments of systems and methods for determining prognostic of reciprocating machine components are described in detail hereinafter. More particularly, the systems and methods configured for determining prognostic of piston ring are disclosed in the subsequent paragraphs.
In one embodiment, the prognostic system 132 includes a data acquisition unit 116, a profile registration unit 118, a model generator 120, a life prediction unit 122, and a database unit 124. The plurality of units 116, 118, 120, 122, 124 is communicatively coupled to each other by a communications bus 126. The prognostic system 132 includes at least one processor unit 130 and a memory unit 128 communicatively coupled to a communications bus 126.
The data acquisition unit 116 is communicatively coupled to the communications bus 126 and configured to obtain a reference profile of the piston ring 106 from the memory unit 128. The reference profile includes a plurality of reference fiducials. In one embodiment, the plurality of reference fiducials includes a first forty-five degrees slope, a three-degrees slope, a second forty-five-degrees slope and an intersecting point of the three-degrees slope and the second forty-five-degrees slope. The data acquisition unit 116 is also communicatively coupled to the profilometer 110 and configured to receive the measurements 112. In one embodiment, the measurements 112 include a measured profile of the piston ring 106. The measured profile includes a plurality of measured fiducials corresponding to the plurality of reference fiducials.
In one embodiment, the data acquisition unit 116 is configured to obtain a first plurality of geometrical features of the piston ring from a computer aided design (CAD) model of the piston ring. The first plurality of geometrical features is representative of the reference fiducials of the reference profile. The data acquisition unit 116 is further configured to measure a second plurality of geometrical features of the piston ring using a co-ordinate measuring machine (CMM). The second plurality of geometrical features is representative of the measured fiducials of the reference profile. In an alternate embodiment, the data acquisition unit 116 is further configured to acquire a three-dimensional (3D) scan of the piston ring 106. In such an embodiment, the data acquisition unit is further configured to obtain a plurality of two-dimensional (2D) images at a plurality of cross-sections based on the 3D scan. The second plurality of geometrical features of the piston ring corresponding to the measured fiducials, is determined based on the plurality of 2D images. In one embodiment, a measurement system, such as but not limited to, the profilometer 110 and a camera based measurement system may acquire the measurements 112 and store them in the memory unit 128 which is accessed by the data acquisition unit 116 at a later time instant. Further, a CAD design may also be stored in the memory unit 128. In such an embodiment, the data acquisition unit 116 is configured to receive the first plurality of geometrical features and the second plurality of geometrical features from the memory unit 128. The first plurality of geometrical features and the second plurality of geometrical features are selected from a non-wearable region.
The profile registration unit 118 is communicatively coupled to the data acquisition unit 116 and configured to align the measured profile with the reference profile based on one or more of the plurality of reference fiducials and one or more of the plurality of measured fiducials. The profile registration unit 118 is further configured to register the measured profile with the reference profile based on the first plurality of geometrical features and the second plurality of geometrical features using a registration technique. The profile registration unit 118 is configured to transform the first plurality of geometrical features and generate a modified first plurality of geometrical features representative of a transformed reference profile aligned with the measured profile.
The model generator 120 is communicatively coupled to the profile registration unit 118 and configured to determine one or more wear parameters based on the measured profile and the transformed reference profile. In one embodiment, model generator 120 is configured to perform data analysis of the modified first plurality of geometrical features and the second plurality of geometrical features. Alternatively, in another embodiment, the model generator 120 is configured to perform data analysis of the first plurality of geometrical features and the modified second plurality of geometrical features.
In the embodiment where the modified first plurality of geometrical features is considered, the data analysis includes modelling the transformed reference profile and the measured profile by polynomial functions. Specifically, modified first plurality of geometrical features, representative of the aligned reference profile, is modelled by a first polynomial function. Further, the second plurality of geometrical features, representative of the measured profile, is modelled by a second polynomial function. Additional features between successive modified first plurality of geometrical features are computed by interpolating the first polynomial function. Similarly, additional features between successive second plurality of geometrical features are computed by interpolating the second polynomial function. The model generator 120 is further configured to compute a difference between the interpolated first polynomial and the interpolated second polynomial to determine one or more wear parameters. In one embodiment, the first polynomial and the second polynomial are cubic polynomials. In alternate embodiments, other types of mathematical functions may be used to model the first plurality of geometrical features and the second plurality of geometrical features. The one or more wear parameters includes a wear depth value, a wear volume value and a wear contact area value. The model generator 120 is further configured to generate a wear model based on the one or more wear parameters.
Further, the model generator 120 is configured to generate a machine learning model corresponding to one of the wear parameters. The machine learning model is determined using a machine learning technique. In one embodiment, the machine learning technique trains a machine learning model to learn progression of the wear parameter based on historical data acquired during previous maintenance schedules. The machine learning model is used for predicting the wear parameter at a future time instant.
The life prediction unit 122 is communicatively coupled to the model generator 120 and configured to estimate the prognostic parameter 114 of the piston ring based on the wear model. The prognostic parameter 114 is determined based on a predicted wear parameter estimate at a future time instant. In one embodiment, the prognostic parameter 114 is representative of an estimate of remaining useful life of the piston ring based on the wear model. In another embodiment, the life prediction unit 122 is further configured to estimate a future time instant at which the predicted wear parameter estimate exceeds a pre-determined wear limit value.
The database unit 124 is communicatively coupled to the other units 116, 118, 120, 122, 130, 128 via the communications bus 126 and configured to store the data in a structured way using a plurality of tables. In one embodiment, the database unit 124 is part of the memory unit 128 and may employ at least one of the relational database management system (RDBMS), a structured query language (SQL) database and a not only structured query language (NoSQL) database. The database unit 124 includes historical data corresponding a plurality of piston rings collected over an extended time period spanning entire life cycle of each of the plurality of piston rings. Specifically, the database unit 124 includes reference profile for each of the plurality of piston rings, a plurality of measured profiles for each of the plurality of piston rings, and remaining useful life estimates corresponding to each of the plurality of measured profiles. Each of the plurality of measured profiles are acquired during scheduled maintenance of the engine in which the corresponding piston ring is assembled.
Moreover, the processor unit 130 may include at least one of a general purpose processor, a controller, a digital signal processor or a customized processing element such as, but not limited to, an application specific integrated circuit (ASIC) and field programmable gate array (FPGA). The processor unit 130 may receive additional inputs from a user through a user interface 162 or any other input device such as a keyboard. The processor unit 130 may also be configured to provide one or more outputs to an output device such as a display 160. The processor unit 130 may also be configured to perform the functionality provided by the data acquisition unit 116, profile registration unit 118, the model generator 120, and the life prediction unit 122. The processor unit 130 is also configured to store data into and retrieve data from the memory unit 128 or the database 124.
In one embodiment, the memory unit 128 is a random access memory (RAM), read only memory (ROM), flash memory or any other type of computer readable memory medium accessible by at least one of the data acquisition unit 116, profile registration unit 118, the model generator 120 and the life prediction unit 122. Also, in certain embodiments, the computer readable memory medium may be encoded with a program to instruct the processor unit 130 to enable a sequence of steps to generate prognostic of piston rings.
In one embodiment, the memory unit 128 may include a non-transitory computer readable medium having instructions to enable at least one processor unit to obtain a reference profile of a piston ring. The reference profile includes a plurality of reference fiducials. Further, the instructions enable the at least one processor unit to receive a measured profile of the piston ring. The measured profile includes a plurality of measured fiducials. The instructions enable the at least one processor unit to align the measured profile with the reference profile based on one or more of the plurality of reference fiducials and one or more of the plurality measured fiducials. Further, the instructions enable the at least one processor unit to determine one or more wear parameters based on the aligned measured profile and the reference profile. The one or more wear parameters includes a wear depth value, a wear contact area value and/or a wear volume value. The instructions enable the processor unit 130 to generate a wear model based on the one or more wear parameters. Also the instructions enable the at least one processor unit to predict a life of the piston ring based on the wear model.
In step 508, the wear depth value is compared with a wear limit value obtained at step 510. In one embodiment, the wear limit value is specified by the user. In other embodiment, the wear limit value is obtained from a memory location. In this embodiment, the wear limit value 510 is representative of a maximum allowable wear depth value beyond which the piston ring is to be replaced. If the wear depth value determined at step 506 is greater than the wear limit value obtained at step 510, the piston ring is replaced at step 512. If the wear depth value determined at step 506 is not greater than the wear limit value obtained from step 510, the piston ring is not replaced and the maintenance work flow is completed. Further, a wear model is generated in step 516 based on the wear depth value determined at step 506. Although the wear model may be a mathematical or a physical model, in one embodiment, a machine learning technique is used in step 520 to generate the wear model of step 516. The machine learning technique is based on historical data that includes one or more wear parameters and life parameters corresponding to the one or more wear parameters. The historical data is recorded over an extended period of time spanning life period of a plurality of piston rings having same or similar specifications as that of the piston ring in current use.
In step 518, the wear model is used to generate an estimate of remaining useful life of the piston ring. The estimate of the remaining useful life is used by personnel who are managing the work flow to ensure that the piston ring is usable till next scheduled maintenance inspection.
The maintenance work flow disclosed by the flow chart 500 is explained with help of a use case for a piston ring with an average life duration of thirty thousand hours as specified by the piston manufacturer. After using the piston ring for about twenty-five thousand hours, and before completion of thirty thousand hours, wear of the piston ring coating is measured while performing maintenance procedures. For acquiring piston ring measurements, cylinder of the engine is to be opened and the piston ring may have to be disassembled. Further, an estimate of remaining useful life duration of the piston ring is estimated as seventeen thousand hours at step 518. The piston ring is continued to be used after twenty five thousand hours for the next seventeen thousand hours without replacement. The cylinder is not opened during subsequent scheduled maintenance, and advantageously, the piston ring is used for additional twelve thousand hours beyond average life duration of thirty thousand hours as specified by the manufacturer.
The method further includes receiving a measured profile of the piston ring in step 604. The measured profile includes a plurality of measured fiducials. It may be noted that one or more of the plurality of measured fiducials corresponds to one or more of the reference fiducials. The receiving of the measured profile includes measuring a second plurality of geometrical features of the piston ring using a co-ordinate measuring machine (CMM).
In another embodiment, the step 604 of receiving the measured profile includes acquiring a three-dimensional (3D) scan of the piston ring. Further, a plurality of two-dimensional (2D) images at a plurality of cross-sections of the piston ring is obtained based on the 3D scan of the piston ring. The step 604 further includes determining the second plurality of geometrical features of the piston ring based on the plurality of 2D images.
The method further includes, in step 606, aligning the measured profile with the reference profile based on one or more of the plurality of reference fiducials and one or more of the plurality of measured fiducials. In one embodiment, aligning the measured profile includes registering the measured profile with the reference profile based on the plurality of reference fiducials and the plurality of measured fiducials.
Further, at step 608, the method also includes determining one or more wear parameters based on the aligned measured profile and the reference profile. The one or more wear parameters includes at least one of a wear depth value, a wear contact area value and a wear volume value. In one embodiment, determining the one or more wear parameters includes interpolating the first plurality of geometrical features by a first polynomial and the second plurality of geometrical features using a second polynomial. Further, the step of determining includes computing a difference between the first polynomial and the second polynomial to determine one or more of the wear parameter.
Further, the method further includes determining a wear model based on the one or more wear parameters as illustrated in step 610. In another embodiment, the step 610 of determining the wear model includes generating a machine learning model for predicting the wear parameter at a future time instant. The machine learning model is determined using a machine learning technique using historical data corresponding to a plurality of piston rings acquired during their entire life span.
In step 612, the method also includes estimating a life of the piston ring based on the wear model. In another embodiment, estimating the life of the piston ring includes estimating a future time instant at which the predicted wear parameter exceeds a pre-determined wear limit value.
Disclosed embodiments measure wear in a plurality of locations of a piston ring and thereby enable accurate estimation of a residual life period of the piston ring. Measuring wear in the piston ring using the present technique is more accurate compared to measuring cylinder bore wear or end gap measurements using conventional techniques. Advantageously, the present technique is able to extend the life of piston rings from an average of thirty thousand operating hours to about forty-five thousand operating hours. Further, the disclosed technique is able to provide prognosis that helps to decide whether it is required to remove/replace the piston ring to avoid imminent road failure. Although the disclosed embodiments are directed towards piston rings of internal combustion (IC) engines, the present technique is not limiting and may be applied to assess life of components exposed to wearing in other industrial applications.
It is to be understood that not necessarily all such objects or advantages described above may be achieved in accordance with any particular embodiment. Thus, for example, those skilled in the art will recognize that the systems and techniques described herein may be embodied or carried out in a manner that achieves or improves one advantage or group of advantages as taught herein without necessarily achieving other objects or advantages as may be taught or suggested herein.
While the technology has been described in detail in connection with only a limited number of embodiments, it should be readily understood that the specification is not limited to such disclosed embodiments. Rather, the technology can be modified to incorporate any number of variations, alterations, substitutions or equivalent arrangements not heretofore described, but which are commensurate with the spirit and scope of the claims. Additionally, while various embodiments of the technology have been described, it is to be understood that aspects of the specification may include only some of the described embodiments. Accordingly, the specification is not to be seen as limited by the foregoing description, but is only limited by the scope of the appended claims.
| Number | Date | Country | Kind |
|---|---|---|---|
| 201741029279 | Aug 2017 | IN | national |
