ADAPTIVE AIR CHARGE ESTIMATION BASED ON SUPPORT VECTOR REGRESSION

• US 20120232772A1
• Filed: 03/07/2011
• Published: 09/13/2012
• Est. Priority Date: 03/07/2011
• Status: Active Grant

• Current Assignee: Toyota Jidosha Kabushiki Kaisha (Toyota Motor Corporation), Board of Regents of the University of Michigan (University of Michigan)
• Original Assignee: Toyota Motor Engineering & Manufacturing North America Incorporated (Toyota Motor Corporation)
• Inventors: Sun, Jing, Butts, Kenneth R., Faust, Jacob Edward
• Granted Patent: US 8,880,321 B2
• Time in Patent Office: Days
• US Class Current: 701/103
• CPC Class Codes: F02D 2041/001 for engines with variable valve actuation
• CPC Class Codes: F02D 2200/0402 the parameter being determined by using a model of the engine intake or its components
• CPC Class Codes: F02D 2200/0404 Throttle position
• CPC Class Codes: F02D 2200/0406 Intake manifold pressure
• CPC Class Codes: F02D 2200/0408 Estimation of intake manifold pressure
• CPC Class Codes: F02D 2200/101 Engine speed
• CPC Class Codes: F02D 41/18 by measuring intake air flow
• CPC Class Codes: F02D 41/2432 Methods of calibration
• CPC Class Codes: F02D 41/2451 characterised by what is learned or calibrated
• CPC Class Codes: G05B 13/042 in which a parameter or coefficient is automatically adjusted to optimise the performance
• CPC Class Codes: G05B 17/02 electric

Abstract

Examples of the present invention include air charge estimation models using linear programming support vector regression (LP-SVR). Air model parameters may be updated during operation of a vehicle, allowing the air model performance to be improved in the presence of part-to-part variation and part aging.

Claims

1. A method of estimating an air charge for an engine, the method comprising:

providing an air model for the engine, the air model including a throttle model and a cylinder flow model,

the cylinder flow model being a linear programming support vector regression (LP-SVR) model including support vectors having associated weighting parameters;

determining the support vectors and the weighting parameters for the LP-SVR model using steady-state engine operation data;

adjusting the weighting parameters using transient driving cycle data; and

estimating the air charge during engine operation using the air model.

2. The method of claim 1, further including validating the weighting parameters using a second set of transient driving cycle data.

3. The method of claim 1, the throttle flow model being a linear programming support vector regression (LP-SVR) throttle flow model having a second plurality of support vectors.

4. The method of claim 1, the air model receiving engine data including variable valve timing data, engine speed data, and throttle position data,

the air model providing an estimated throttle flow and an estimated manifold pressure.

5. The method of claim 4, wherein adjusting the weighting parameters includes comparing the estimated manifold pressure determined by the air model with a measured manifold pressure.

6. The method of claim 1, the air charge model being derived from an intake manifold dynamic model.

7. The method of claim 1, further including adjusting the weighting parameters during engine operation.

8. The method of claim 7, the weighting parameters being adjusted using a projection method during engine operation.

9. The method of claim 8, the weighting parameters being adjusted during periods of well excited and consistent engine operations.

10. The method of claim 1, the method being performed by an engine controller including a processor.

11. An apparatus, the apparatus being an engine controller for an engine,

the engine controller being operable to receive engine data including engine speed data, valve data, and throttle data,

the engine controller being operable to determine an air charge for the engine using an air charge model including a throttle model and a cylinder flow model,

the cylinder flow model being a linear programming support vector regression (LP-SVR) model including a plurality of support vectors having associated weighting parameters.

12. The method of claim 11, the engine controller being further operable to modify the air charge model during engine operation by adjustment of the weighting parameters of the LP-SVR model.

13. The apparatus of claim 12, the weighting parameters being adjusted using a projection method.

14. The apparatus of claim 12, the weighting parameters being adjusted during engine operation by comparing a measured manifold pressure and an estimated manifold pressure determined by the air model.

15. The apparatus of claim 11, the throttle model being a linear programming support vector regression (LP-SVR) throttle model.

16. The apparatus of claim 15, the throttle model being adjusted during engine operation by comparing a measured throttle flow and an estimated throttle flow determined by the throttle model.