Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
SYSTEM AND METHOD FOR VALVE CHARACTERISTIC DETERMINATION AND MODELLING
Document Type and Number:
WIPO Patent Application WO/2019/185158
Kind Code:
A1
Abstract:
A method for computer based modelling of an optimized intake and exhaust valve characteristic for an internal combustion engine. The method includes defining a first collection of valve characteristic values for an intake valve and a second collection of valve characteristic values for an exhaust valve, simulating, via a computer implemented engine model, operation at an operating point of the internal combustion engine, and for each value combination of the first collection and the second collection, to determine a simulated indicated mean effective pressure (IMEP) and simulated exhaust mass flow (EMF) at each value combination, calculating a deviation of each simulated IMEP and EMF from experimentally determined IMEP and EMF for the internal combustion engine, statistically modelling the deviations in IMEP and EMF determined at each value combination, and identifying an optimized intake valve characteristic value and an optimized exhaust valve characteristic value based on the statistically modelled IMEP and EMF deviations, wherein the optimized intake valve characteristic value and the optimized exhaust valve characteristic value correspond to a point at which deviation in IMEP and EMF is minimized over the first and second collection of values.

Inventors:
BOUILLY, Julien (European Technical Administration DivisionAvenue du Bourget 60, 1140 BRUSSELS, 1140, BE)
LAFOSSAS, Francois (European Technical Administration DivisionAvenue du Bourget 60, 1140 BRUSSELS, 1140, BE)
RIVAS, Manuel (European Technical Administration DivisionAvenue du Bourget 60, 1140 BRUSSELS, 1140, BE)
Application Number:
EP2018/058229
Publication Date:
October 03, 2019
Filing Date:
March 29, 2018
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
TOYOTA MOTOR EUROPE (Avenue du Bourget 60, 1140 BRUSSELS, 1140, BE)
International Classes:
F02D41/24; G05B17/02; F02D35/02; F02D41/14
Domestic Patent References:
WO2016209522A12016-12-29
Foreign References:
US20070265805A12007-11-15
DE102007039691A12009-02-26
US20150275711A12015-10-01
Other References:
FATHI-ELAMIN: "Computing and Engineering Researchers' Conference", December 2009, UNIVERSITY OF HUDDERSFIELD, article "Detection Of Diesel Engine Valve Clearance By Acoustic Emission"
S. M. JAFARI: "School of Mechanical Engineering", February 2014, SHARIF UNIVERSITY OF TECHNOLOGY, article "Valve Fault Diagnosis in Internal Combustion Engines Using Acoustic Emission and Artificial Neural Network"
Attorney, Agent or Firm:
BELL, James et al. (CABINET BEAU DE LOMENIE, 158 RUE DE L'UNIVERSITE, PARIS CEDEX 07, 75340, FR)
Download PDF:
Claims:
CLAIMS

1. A method for computer based modelling of an optimized intake and exhaust valve characteristic for an internal combustion engine, the method comprising:

defining a first collection of valve characteristic values for an intake valve and a second collection of valve characteristic values for an exhaust valve;

simulating, via a computer implemented engine model, operation at an operating point of the internal combustion engine, and for each value combination of the first collection and the second collection, to determine a simulated indicated mean effective pressure (IMEP) and simulated exhaust mass flow (EMF) at each value combination;

calculating a deviation of each simulated IMEP and EMF from experimentally determined IMEP and EMF for the internal combustion engine;

statistically modelling the deviations in IMEP and EMF determined at each value combination; and

identifying an optimized intake valve characteristic value and an optimized exhaust valve characteristic value based on the statistically modelled IMEP and EMF deviations, wherein the optimized intake valve characteristic value and the optimized exhaust valve characteristic value correspond to a point at which deviation in IMEP and EMF is minimized over the first and second collection of values. 2. The method according to claim 1, wherein the defining is performed using a design of experiments (DoE).

3. The method according to any of claims 1-2, wherein the simulating is performed using a ID engine model.

4. The method according to claim 3, wherein the ID engine model is implemented such that the outputs of the ID model are provided to a Gaussian regression model.

5. The method according to any of claims 1-4, wherein the modelled intake and exhaust valve characteristic is selected from a valve lash and a valve timing.

6. The method according to any of claims 1-5, wherein the simulating is carried out for at least 2 engine operating points, and preferably carried out for at least 3 engine operating points, the statistical modelling being carried out for all value combinations at each simulated engine operating point.

7. The method according to claim 6, wherein an engine operating point is defined by at least an engine rotational speed and an engine load.

8. The method according to claim 7, wherein the engine load is modelled based on a brake mean effective pressure.

9. The method according to any of claims 1-8, wherein each valve characteristic value in the first and second collection is based on possible physical variations of an associated valve characteristic value of an intake valve and an exhaust valve, respectively, of the modelled internal combustion engine.

10. The method according to any of claims 1-9, comprising:

bench testing an actual internal combustion engine corresponding to the internal combustion engine to be modelled;

obtaining high frequency intake, exhaust, and in-cylinder pressure traces during the bench test; and

storing IMEP and EMF values derived from the traces as the experimentally determined IMEP and EMF values.

11. The method according to any of claims 1-10, comprising: prior to the simulating, calibrating the computer implemented engine model

corresponding to the internal combustion engine, for each unique combination of values from the first collection and the second collection.

12. The method according to claim 11, wherein the calibrating comprises modifying at least one parameter associated with an engine wall heat transfer, and/or filtering an apparent heat release rate to determine an actual end of combustion.

13. The method according to claim 12, wherein the actual end of combustion is found when the filtered apparent heat release rate is equal to 0.

14. The method according to any of claims 12-13, wherein the at least one parameter associated with an engine wall heat transfer is a heat transfer multiplier associated with an in-cylinder pressure.

15. The method according to any of claims 12-13, wherein the at least one parameter associated with an engine wall heat transfer is a cylinder wall temperature.

Description:
SYSTEM AND METHOD FOR VALVE CHARACTERISTIC

DETERMINATION AND MODELLING

FIELD OF THE DISCLOSURE

[0001] The present disclosure is related to determination via simulation, of valve characteristics in an engine, and more particularly, determination of valve lash and valve timing based on a dual-model analysis.

BACKGROUND OF THE DISCLOSURE

[0002] Path of an airflow into a cylinder and exhaust flow out the cylinder, of an internal combustion engine can be highly sensitive to valve timing and valve lash (i.e., tappet/rocker arm clearance), among others. Such sensitivity can lead to dramatic impacts on engine performance such as power and fuel economy, as well as other engine characteristics. This is due, at least in part to the overall effect of valve lash on volumetric efficiency, which in turn affects air mass trapped in the cylinder and pumping losses, an consequently the indicated mean effective pressure.

[0003] Because more and more emphasis is being placed on increasing fuel economy, and thereby, reduction of vehicle emissions, while maintaining other performance characteristics of the engine, the desire to determine an optimum valve lash and timing for an engine design has become a focus.

[0004] Experimentally determining and adjusting valve lash typically involves removing a valve cover from cylinder head of an engine, and manually measuring (e.g., with a feeler gauge) a clearance between the actuator of the valve (e.g., tappet, rocker arm, etc.) and the contact surface of the valve stem. Further, determining and adjusting valve timing requires an even more complex set of tools and sensors.

[0005] Therefore, continuous and iterative determination and adjustment of valve lash and timing via intrusive methods is difficult if not impossible, and requires much time to determine and/or derive optimal settings.

[0006] Because modelling of engine characteristics, both performance based animations based have become more important during the design phase, and because valve characteristics play an important role in such modeling, focus has been put on the ability to accurately determine and/or model valve

characteristics during the design phase.

[0007] Previous attempts have been made to determine valve lash and timing via non-intrusive methods. For example, some researchers have attempted to use vibration analysis for determining a corresponding valve defect.

[0008] For example, Fathi-Elamin, "Detection Of Diesel Engine Valve

Clearance By Acoustic Emission," Computing and Engineering Researchers' Conference, University of Huddersfield, Dec 2009, describes a technique whereby acoustic emissions (AE) of a diesel engine may be analyzed for approximation of a valve lash. In this document the AE events associated with exhaust valve opening was clearly observed and the difference between healthy and faulty valves was shown.

[0009] In another study, S. M. Jafari, "Valve Fault Diagnosis in Internal Combustion Engines Using Acoustic Emission and Artificial Neural Network," School of Mechanical Engineering, Sharif University of Technology, February 2014, AE were coupled with neural network concepts in order to identify valve defects in a spark ignited engine.

SUMMARY OF THE DISCLOSURE

[0010] The present inventors have recognized that improvement in valve lash and timing determination, particularly during design phases of a vehicle engine is desirable. They have further determined that a model based analysis can be implemented, particularly based on a dual-model approach, which can result in highly accurate correlations of valve lash and timing to actual and optimized values.

[0011] According to embodiments of the present disclosure, a method for computer based modelling of an optimized intake and exhaust valve

characteristics for an internal combustion engine is provided. The method includes defining a first collection of valve characteristic values for an intake valve and a second collection of valve characteristic values for an exhaust valve, simulating, via a computer implemented engine model, operation at an operating point of the internal combustion engine, and for each value combination of the first collection and the second collection, to determine a simulated indicated mean effective pressure (IMEP) and simulated exhaust mass flow (EMF) at each value combination, calculating a deviation of each simulated IMEP and EMF from experimentally determined IMEP and EMF for the internal combustion engine. The method also includes statistically modelling the deviations in IMEP and EMF determined at each value combination, and identifying an optimized intake valve characteristic value and an optimized exhaust valve characteristic value based on the statistically modelled IMEP and EMF deviations, wherein the optimized intake valve characteristic value and the optimized exhaust valve characteristic value correspond to a point at which deviation in IMEP and EMF is minimized over the first and second collection of values.

[0012] By providing such a method, the inventors have enabled

determination of optimized valve characteristic values, e.g., lash and/or timing, that may be used to increase accuracy during modelling of other engine operation characteristics, for example, emission and performance modelling. In addition, based on the techniques described herein, time savings due to automation can be gained during the modelling and optimization process.

[0013] The defining may be performed using a design of experiments (DoE).

[0014] The simulating may be performed using a ID engine model.

[0015] The outputs of the ID engine model may be configured for providing to a Gaussian regression model.

[0016] The modelled intake and exhaust valve characteristic may be selected from a valve lash and a valve timing.

[0017] The simulating may be carried out for at least 2 engine operating points, and preferably carried out for at least 3 engine operating points, the statistical modelling being carried out for all value combinations at each simulated engine operating point.

[0018] An engine operating point may be defined by at least an engine rotational speed and an engine load.

[0019] The engine load may be modelled based on a brake mean effective pressure.

[0020] Each valve characteristic value in the first and second collection may be based on possible physical variations of an associated valve characteristic value of an intake valve and an exhaust valve, respectively, of the modelled internal combustion engine. [0021] The method may include bench testing an actual internal combustion engine corresponding to the internal combustion engine to be modelled,

obtaining high frequency intake, exhaust, and in-cylinder pressure traces during the bench test, storing IMEP and EMF values derived from the traces as the experimentally determined IMEP and EMF values.

[0022] The method may comprise, prior to the simulating, calibrating the computer implemented engine model corresponding to the internal combustion engine, for each unique combination of values from the first collection and the second collection.

[0023] The calibrating may include modifying at least one parameter associated with an engine wall heat transfer, and/or filtering an apparent heat release rate to determine an actual end of combustion.

[0024] The actual end of combustion may be determined when the filtered apparent heat release rate is equal to 0 following an initial spike in the apparent heat release rate.

[0025] The at least one parameter associated with an engine wall heat transfer may be a heat transfer multiplier associated with an in-cylinder pressure.

[0026] The at least one parameter associated with an engine wall heat transfer may be a cylinder wall temperature.

[0027] It is intended that combinations of the above-described elements and those within the specification may be made, except where otherwise contradictory.

[0028] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments of the disclosure and together with the description, and serve to explain the principles thereof.

BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Fig. 1 shows a schematic representation of an exemplary

configuration for bench testing of an one cylinder of an actual internal combustion engine;

[0030] Fig. 2 shows a flowchart of an exemplary method according to embodiments of the disclosure;

[0031] [CANCELLED] [0032] Fig. 3 shows a schematic representation of an exemplary computer implemented ID engine model configuration;

[0033] Figs. 4A and 4B show graphs highlighting an exemplary technique for calibrating end of combustion calculations according to the present disclosure;

[0034] Figs. 5A and 5B show graphs highlighting a technique for heat transfer calibration; and

[0035] Figs. 6 shows a table and chart highlighting deviation results for an exemplary collection of valve characteristics.

DESCRIPTION OF THE EMBODIMENTS

[0036] Reference will now be made in detail to exemplary embodiments of the disclosure, examples of which are illustrated in the accompanying drawings. Wherever possible, the same reference numbers will be used throughout the drawings to refer to the same or like parts.

[0037] Fig. 1 shows a schematic representation of an exemplary

configuration for bench testing of an one cylinder of an actual internal combustion engine. While an exemplary single cylinder is shown at Fig. 1, one of ordinary skill understands that the system may be implemented across the multi-cylinder internal combustion engine, e.g. 4 cylinders, 5 cylinder, 6 cylinders, etc.

[0038] An internal combustion prepared for such bench testing, may include an intake camshaft 5, and exhaust camshaft 10, an intake valve 15, an exhaust valve 20, and a piston 100 configured to reciprocate resulting in rotation of a crank (not shown). Intake valve 15 may be configured to open and close an intake port 45 in fluid communication with an intake manifold of the internal combustion engine, while the exhaust valve 20 may be configured to open and close and exhaust port 50 in fluid communication with exhaust manifold of the internal combustion engine.

[0039] In order to accomplish opening and closing of the intake and exhaust ports 45, 50, intake camshaft 5 and exhaust camshaft 10 may be rotated in conjunction with the crank shaft of the internal combustion engine such that lobes of the camshafts contact stems of each valve 15, 20 to apply force thereto resulting in reciprocating motion of the valves. The correlation between rotation of the camshafts and rotation of the crank relates to what is referred to herein as a valve timing. In other words, the position of the camshaft relative to the position of the crankshaft during their respective rotations defines the valve timing of the internal combustion engine.

[0040] Similarly, during the period where a lobe of the camshaft is not in contact with the stem of the valve (i.e. while the intake port 45 or the exhaust port 50 is closed) a gap between the cam lobe and the valve stem is

maintained, and this gap will be referred to herein as valve lash. In other words, valve lash may be defined as the distance between the cam lobe and the valve stem during a period in which the cam lobe does not apply force to the valve stem.

[0041] For purposes of the present disclosure, valve timing and valve lash shall be understood to correspond to "valve characteristics," and values associated there with some be understood to correspond with "valve

characteristic values."

[0042] The test configuration of Fig. 1 may further include an intake temperature sensor 35A, configured to measure the temperature of gases passing through the intake port 45, and exhaust temperature sensor 35B, configured to measure the temperature of gases passing through the exhaust port 50, instantaneous pressure transducer 30, and average pressure

transducer 25, among others. Instantaneous pressure transducer 30 may comprise one or more pressure transducers, for example an intake pressure transducer 30A and exhaust pressure transducer 30B, while average pressure transducer 25 may also comprise one or more pressure transducers, for example, an intake pressure transducer 25A and exhaust pressure transducer 25B. These transducers may be linked to a processing means (not shown), e.g., a computer, for purposes of determining and storing an instantaneous pressure, instantaneous pressure differential, an average pressure, and an average pressure differential, between the intake port 45 and the exhaust port 50. The transducers may be configured to operate at a relatively high sampling frequency (e.g. assuming a max engine speed of 7000rpm and a resolution of 0.1 CAD, a frequency of 420kHz may be chosen) to provide high frequency output traces.

[0043] Experimental data may be obtained by operating the internal combustion engine configured according to Fig. 1, such experimental data including for example, indicated mean effective pressure, exhaust mass flow, intake mass flow, brake mean effective pressure, end of combustion, heat transfer coefficient, etc. One of skill in the art understands how these values may be obtained based on the configuration of Fig. 1, and a detailed

explanation will not be provided herein. Notably, the experimental data obtained via the bench testing of the internal combustion engine may be stored and referenced for purposes of carrying out embodiments of the present disclosure.

[0044] According to some embodiments, experimental data may be obtained at a plurality of operating points of the internal combustion engine. For example, a first set of data may be obtained for an engine operating point where the engine rotates at a speed of 3000 RPM at a brake mean effective pressure of 6.0 bar, a second set of data may be obtained for an engine operating point where the engine rotates at a speed of 5000 RPM at a brake mean effective pressure of 9.0 bar, and a third set of data may be obtained where the engine rotates at 6000 RPM at a brake mean effective pressure of 15.6 bar. Each of these sets of data may be stored and compiled into various data maps relative to engine speed and load, for example. One of skill in the art will recognize that the above exemplary operating points are not intended to be limiting, and that any operating points for particular engine design may be used.

[0045] As an exemplary selection scheme, 3 operating points (OP9) by increasing speed and load may be selected. For example :

[0046] OP1 : 1500rpm 25% load

[0047] OP2 : 3500rpm 50% load

[0048] OP3 : 5500rpm 75% load

[0049] Once the experimental data has been compiled and stored, e.g. on storage medium associated with a computer on which simulations are to be run, modeling of the engine to find optimal valve characteristics may be undertaken.

[0050] Fig. 2A shows a flowchart of an exemplary method according to embodiments of the disclosure. In order to provide a desirable number of relevant valve characteristic values, e.g. intake and exhaust valve lash and/or intake and exhaust valve timing, a design of experiments (DoE) may be created to define a first collection of valve characteristic values for an intake valve and a second collection of valve characteristic values for an exhaust valve (step 205). [0051] For example, criteria for designating a valve characteristic value may be input such that the valve characteristic values sweep over ranges that are defined according to the possible physical variations of the valve lash and/or valve timing of the real engine. According to such an example, an exemplary valve lash of an actual internal combustion engine may range between about 0 and 0.3 millimeters, and this range may form an input to a design of

experiment generator (e.g. ETAS ASCMO) such that a collection (e.g. between 70 and 120 points, and according to some embodiments, 90 points) of intake valve lash values and a collection of exhaust valve lash values may be created over the range specified.

[0052] Similarly with regard to valve timing, a range for example ± 5 degrees of crank angle rotation may be specified, and a collection of the intake valve timing and exhaust valve timing values may be created over the range by the design of experiments. One of skill in the art will recognize that the presently recited ranges are exemplary only and not intended to be limiting.

Any range corresponding to possible actual variation in an actual internal combustion engine may be implemented for creating a desirable number of points, as may any other suitable design of experiment software (e.g., IBM SPSS, CAMO DOE, etc.)

[0053] An engine model, e.g., one-dimensional (ID), may be created using any desired modeling software an exemplary one-dimensional model is shown at Fig. 3, along with its primary inputs, main boundary conditions, and primary outputs. For example as inputs the ID model may take instantaneous in cylinder pressure, injected fuel mass, intake valve lash and timing, exhaust valve lash and timing, spark advance, and geometrical parameters.

[0054] Boundary conditions may be set for example, instantaneous inlet port pressure, instantaneous outlet port pressure, time average intake temperature, and surface temperatures of for example, pistons, valves, cylinder walls, etc. One of skill in the art is aware as to how such boundary conditions may be determined, and a detailed explanation will not be provided herein.

[0055] The outputs of the ID model include at least an indicated mean effective pressure and an exhaust mass flow, among others. Notably the primary inputs to the ID model other than intake valve lash and/or timing, and exhaust valve lash and/or timing, may be set as constants based on an operating point of the modeled engine. For example, the modeled engine as discussed earlier operating at RPM of 3000 at a brake mean effective pressure of 6.0 bar will typically have a known and/or preferred injected fuel mass, instantaneous in-cylinder pressure, spark advance, and geometrical parameters which the modeling software has received from an operator during the modeling process. Likewise for each selected engine operation point.

[0056] Similarly the boundary conditions may be predefined by the operator to known or approximate values based on a particular engine design. For example surface temperatures of pistons, cylinder walls, valves, etc. may be set as min and max, and for an engine operating in a steady-state condition, it may be understood that the engine is maintained at a substantially constant temperature. Selection of certain of these parameters will be discussed further with regard to calibration below.

[0057] The outputs of the ID model will be discussed in greater detail below with regard to implementation of the statistical model (e.g., Gaussian

regression model) related to deviations and indicated mean effective pressure and exhaust mass flow from the experimental data determined during the bench testing.

[0058] In order to calibrate end of combustion and heat transfer coefficient characteristics, among others of the ID engine model, the model may be run at each engine operating point (e.g. three operating points) for each combination of valve characteristic values from each collection. In other words, where 90 intake valve lash points and 90 exhaust valve lash points exist in the intake valve collection and exhaust valve collection, respectively, each value pair, i.e. one intake valve lash value and one exhaust valve lash value, may be simulated in the ID engine (step 215) during a calibration phase, to permit calibration of end of combustion (EOC) determination and heat transfer coefficient for the engine model, among others. During the calibration phase the output values of the ID engine model, i.e., IMEP and EMF may be stored to a temporary or permanent location for utilization during the calibration phase.

[0059] Figs. 4A and 4B show graphs highlighting an exemplary technique for calibrating end of combustion calculations for the ID engine model according to embodiments of the disclosure (step 220). The inventors of the present disclosure have determined that ID engine models may have difficulty determining end of combustion during simulated engine operation. Therefore, they have determined that a signal corresponding to apparent heat release rate (AHRR) although typically noisy (see Fig. 4A), may be used by implementing a convolution function to filter noise therefrom (see Filtered AHRR of Fig. 4A). Once the filtering has been performed, end of combustion, as measured in crank angle degrees may be found at the first occurrence where AHRR equals 0 following an initial spike (corresponding to ignition) in AHRR.

[0060] The EOC value determined in this manner is then provided to the ID model as a fixed position for each given operating point of the engine. In other words, viewing Fig. 4B, the end of combustion condition for the internal combustion engine under consideration, can be seen to be approximately 71 crankshaft angle degrees, and this value may be fixed across the internal combustion engine at the given operating point. Each subsequent operating point of the internal combustion engine may then have its end of combustion crank angle value calibrated in the same manner.

[0061] Similarly, Figs. 5A and 5B show graphs highlighting a technique for heat transfer calibration. Each internal combustion engine may have differing heat transfer characteristics that depend on, for example, cylinder wall thickness, material from which the cylinder head and block are constructed, etc. Based on the measured information during bench testing of the internal combustion engine to be modeled, the cylinder pressure trace taken by the high-frequency pressure sensors 30A and 30B/25A and 25B for each engine operating point, may be used to determine a heat transfer coefficient for calibrating the simulated internal combustion engine.

[0062] In order to determine a heat transfer multiplier to be applied in the ID engine model for calibrating the heat transfer coefficient, one may consider that the heat transfer multiplier impacts the cylinder pressure curve and therefore may minimize the absolute value of the error between the simulated value of cylinder pressure (P) and the measured value in order to adjust the corresponding curve to find an appropriate heat transfer multiplier for a particular engine operating point. In order to iterate (i.e. sweep) through a range of heat transfer multiplier values to determine where the absolute value of the error is minimized, the equation below may be used. [0063] EVO corresponds to the exhaust valve opening point in degrees, P si m is the simulated cylinder pressure, and P eXp is the experimentally determined cylinder pressure, while S. A corresponds to the spark advance angle in degrees.

[0064] When the value is less than a threshold value (e.g., ~-0.25), the heat transfer multiplier may be decreased, and when the value is greater than the threshold value the transfer multiplier may be increased. The results of the evaluation may then be graphed as shown at Fig. 5A, and the minimum determined to be the heat transfer multiplier to be used for calibration, in the present example, 1.3.

[0065] Other calibrations may also be carried out using the high-frequency pressure curve obtained during the bench testing of the internal combustion engine, for example average exhaust pressure calibration, so as to calibrate DR between the intake and exhaust pressures.

[0066] The ID engine model may be implemented to output values of mass flow and IMEP to a Gaussian regression model having mass flow and IMEP as primary inputs. In other words, the primary outputs of the ID model simulated by operating at selected intake and exhaust valve characteristic points are provided to the Gaussian model for learning and optimization of valve characteristics.

[0067] Secondary inputs to the Gaussian regression model in which the ID engine model has been implemented may include, for example, spark advance, injected fuel mass, and certain geometrical parameters related to the internal combustion engine being modeled. Further boundary conditions may be set, including, instantaneous inlet port pressure, instantaneous outlet port pressure, time average intake temperature, and surface temperatures, e.g. of pistons, valves, cylinder walls, etc.

[0068] With the calibration information determined at step 220 configured for the model, the ID engine model may be run in a looped manner (step 230), to determine, i.e., simulate, indicated mean effective pressure and exhaust mass flow over the entire range of valve characteristic values for the intake and exhaust valves.

[0069] For each of the indicated mean effective pressure and exhaust mass flows determined corresponding to a particular combination of valve

characteristic values (i.e., intake characteristic value: exhaust characteristic value), a deviation from the experimentally determined values of indicated mean effective pressure and exhaust mass flow, DIMER and AEMF may be calculated (step 225) and the deviations stored for modeling in the Gaussian process regression model at step 235.

[0070] The looped calculations may be carried over each characteristic value combination, for each of the engine operating points selected for the design of experiments (DoE), and each set of resulting deviations in IMEP and EMF values corresponding to its respective engine operating point fed to the Gaussian process regression model for analysis (step 235).

[0071] Once all points have been modeled in the Gaussian process regression model, the corresponding valve lash values (i.e. those having a minimum deviation of indicated mean effective pressure and exhaust mass flow) may be determined (step 240). Of course, one of skill will recognize that other requirements/limits may be used for selecting the appropriate valve lash/timing values may be implemented in order to determine an "optimum" point (e.g., using a Pareto curve according to said requirements/limits).

[0072] Figs. 6 shows a table and chart highlighting deviation results for an exemplary collection of valve characteristics. As shown at Fig. 6, calculated deviations of indicated mean effective pressure and exhaust mass flow may be graphed against one another for each engine operating point. In Fig. 6, only three exemplary combinations of intake valve lash and exhaust valve lash are evaluated, however, one of skill in the art will recognize that these three exemplary combinations have been selected for purposes of demonstration, from the larger representation of the Gaussian regression model, based on their proximity to a threshold accuracy target area within the graph. For example, an initial proximity of ± 20 percent may be set and only values falling within this initial proximity range may be further considered. This initial proximity maybe modified in order to obtain a finer result set, particularly where a great number of points lie within the initial proximity. For example, an initial range of variation of valve lashes may be defined, and then the ID model may use such a range to determine a combination of lashes that satisfy an predetermined accuracy target value, based on the EMF and IMEP outputs. This allows for a poorly calibrated model to be identified, at which point it the target value may not be achieved. In this case, a combination of valve lashes that is the closest to the targeted accuracy may be selected. [0073] As shown at Fig. 6 intake valve/exhaust valve lash combinations of 0.10,0.15; 0.00, 0.00; and 0.30, 0.30 have been considered, all measurements being in millimeters, and plotted for two exemplary engine operating points.

[0074] For each of these engine operating points it can be seen that combination 1, 0.10, 0.15mm, renders the most accurate results in terms of percent deviation from the experimental data for indicated mean effective pressure and exhaust mass flow. Thus for purposes of the engine simulation, and for calculation of additional engine characteristics, e.g.

performance/emissions type modeling, this valve lash combination may be utilized with a high level of accuracy.

[0075] By implementing methods according to the present disclosure, it becomes possible to accurately determine a corresponding valve characteristic (i.e., lash and/or timing) based on corresponding deviations in IMEP and EMF of simulated values from experimentally determined data. The corresponding valve characteristic values thus determined may then be utilized in various other engine modelling procedures to improve accuracy of, for example, performance and emissions results, thereby leading to greater flexibility and improved engine design.

[0076] Throughout the description, including the claims, the term

"comprising a" should be understood as being synonymous with "comprising at least one" unless otherwise stated. In addition, any range set forth in the description, including the claims should be understood as including its end value(s) unless otherwise stated. Specific values for described elements should be understood to be within accepted manufacturing or industry tolerances known to one of skill in the art, and any use of the terms "substantially" and/or "approximately" and/or "generally" should be understood to mean falling within such accepted tolerances.

[0077] Although the present disclosure herein has been described with reference to particular embodiments, it is to be understood that these embodiments are merely illustrative of the principles and applications of the present disclosure.

[0078] It is intended that the specification and examples be considered as exemplary only, with a true scope of the disclosure being indicated by the following claims.