Spark advance controller and production machine of spark advance control logic in internal combustion engine

• US 20060157029A1
• Filed: 01/18/2006
• Published: 07/20/2006
• Est. Priority Date: 01/19/2005
• Status: Active Grant

• Current Assignee: Hitachi Astemo Ltd. (Hitachi, Ltd.)
• Original Assignee: Hitachi, Ltd.
• Inventors: Nemoto, Mamoru, Shiraishi, Takuya, Suzuki, Kunihiko, Miyazaki, Taizo
• Granted Patent: US 7,290,529 B2
• Time in Patent Office: Days
• US Class Current: 123/406.330
• CPC Class Codes: F02D 2041/1437 Simulation
• CPC Class Codes: F02D 41/1406 with use of a optimisation method, e.g. iteration
• CPC Class Codes: F02P 5/1514 with means for optimising the use of registers or of memories, e.g. interpolation
• CPC Class Codes: F02P 5/152 dependent on pinking detecting or indicating knocks in internal-combustion engines G01L23/22
• CPC Class Codes: Y02T 10/40 Engine management systems

Abstract

A spark advance controller capable of controlling spark advance in an internal combustion engine with high accuracy without requiring a larger number of data maps and performing laborious optimization tests using an actual engine. In the spark advance controller, map data for controlling the spark advance is computed by a numerical simulator simulating an internal combustion engine, and the map data for spark advance control is approximated with a regression model. The spark advance is computed by using the regression model depending on the detected operating conditions.

Claims

1. A spark advance controller in an internal combustion engine, said controller including means for detecting operating conditions of said internal combustion engine,

wherein said controller stores a regression model approximating map data for spark advance control, and decides a spark advance with computation using the regression model depending on the detected operating conditions, thereby performing the spark advance control.

2. The spark advance controller in the internal combustion engine according to claim 1, wherein the regression model is a polynominal made up of at least one of a term including only an engine speed, a term including only a load, a term including only a modification variable, an interaction term including the engine speed and the load, an interaction term including the modification variable and the engine speed, an interaction term including the modification variable and the load, and an interaction term including the engine speed, the load and the modification variable.

3. The spark advance controller in the internal combustion engine according to claim 2, wherein the regression model is divided into a reference function and a modification function, and the spark advance is decided as the sum of a spark advance reference value computed by the reference function and a spark advance modification value computed by the modification function.

4. The spark advance controller in the internal combustion engine according to claim 3, wherein the reference function is a polynominal made up of at least one of the term including only the engine speed, the term including only the load, the interaction term including the engine speed and the load, and the modification function is a polynominal made up of at least one of the term including only the modification variable, the interaction term including the modification variable and the engine speed, the interaction term including the modification variable and the load, and the interaction term including the engine speed, the load and the modification variable.

5. The spark advance controller in the internal combustion engine according to claim 3, wherein said modification variable includes an operating characteristic value of an intake valve, and the spark advance modification value is computed based on the operating characteristic of the intake valve by using the modification function with the operating characteristic value of the intake valve set as the modification variable.

6. The spark advance controller in the internal combustion engine according to claim 3, wherein said modification variable includes an intake temperature, and the spark advance modification value is computed based on the intake temperature by using the modification function with the intake temperature set as said modification variable.

7. The spark advance controller in the internal combustion engine according to claim 3, wherein said modification variable includes an equivalence ratio, and the spark advance modification value is computed based on the equivalence ratio by using the modification function with the equivalence ratio set as the modification variable.

8. The spark advance controller in the internal combustion engine according to claim 3, wherein said modification variable includes an EGR (Exhaust Gas Recirculation) rate, and the spark advance modification value is computed based on the EGR rate by using the modification function with the EGR rate set as the modification variable.

9. The spark advance controller in the internal combustion engine according to claim 3, wherein when there are a plurality of modification variables, the spark advance modification value is computed based on each of the modification variables, and the total sum of the respective spark advance modification values computed for all the modification variables is set as a total spark advance modification value.

10. The spark advance controller in the internal combustion engine according to claim 3, wherein said reference function is defined as a dual and third-degree polynominal in which the engine speed and the load are independent variables, and said modification function is defined as a triple and third-degree polynominal in which the engine speed, the load and the modification variable are independent variables.

11. The spark advance controller in the internal combustion engine according to claim 1, wherein the spark advance is decided as the sum of an MBT (Minimum spark advance for Best Torque) computed by using an approximated MBT regression model and a knock retard computed by using an approximated knock-retard regression model.

12. A production machine of spark advance control logic, wherein map data for spark advance control is computed by a numerical simulator simulating an internal combustion engine, and the map data is approximated with a regression model to produce control logic for spark advance.

13. The production machine of spark advance control logic according to claim 12, wherein each term of said regression model is multiplied by a partial regression coefficient decided to more accurately approximate the map data, and a degree of influence of the term upon the spark advance, which is multiplied by the partial regression coefficient, is evaluated based on magnitudes of respective absolute values of the partial regression coefficients, and if the degree of influence upon the spark advance is not larger than a predetermined value, the relevant term multiplied by the relevant partial regression coefficient is omitted from the regression model.

14. The production machine of spark advance control logic according to claim 12, wherein the map data for spark advance control is constituted by an MBT map and a knock retard map, said MBT map is approximated with one regression model, and said knock retard map is approximated with another regression model.

15. The production machine of spark advance control logic according to claim 12, wherein the map data for spark advance control is subdivided into plural groups depending on individual operating ranges, and each subdivided group of the map data is approximated with a regression model.

16. The production machine of spark advance control logic according to claim 12, wherein the number of degrees of the regression model is set depending on demanded control accuracy.

17. The production machine of spark advance control logic according to claim 12, wherein said numerical simulator computes at least a cylinder volume, computes change of gas mass in a cylinder during a gas exchange process, computes heat generated in a burning process, computes a heat loss, determines the occurrence of knock, computes a relationship between the cylinder pressure and the cylinder volume per cycle based on said numerical simulation, computes indicated torque based on the relationship between the cylinder pressure and the cylinder volume, computes an MBT based on the relationship between the indicated torque and the spark advance, and computes a knock limit based on a result of the determination as to the occurrence of knock.