Method and apparatus to diagnose intake airflow

• US 20040230368A1
• Filed: 05/14/2003
• Published: 11/18/2004
• Est. Priority Date: 05/14/2003
• Status: Active Grant

• Current Assignee: GM Global Technology Operations LLC (General Motors Company)
• Original Assignee: GM Global Technology Operations Incorporated (General Motors Company)
• Inventors: Kropinski, Michael A., Miotke, Mark D.
• Granted Patent: US 7,181,334 B2
• Time in Patent Office: Days
• US Class Current: 701/114
• CPC Class Codes: F02D 2200/0408 Estimation of intake manifold pressure
• CPC Class Codes: F02D 41/1405 Neural network control
• CPC Class Codes: F02D 41/222 relating to the failure of sensors or parameter detection devices
• CPC Class Codes: G05B 23/024 Quantitative history assessment, e.g. mathematical relationships between available data; Functions therefor; Principal component analysis [PCA]; Partial least square [PLS]; Statistical classifiers, e.g. Bayesian networks, linear regression or correlation analysis; Neural networks
• CPC Class Codes: G05B 9/02 electric
• CPC Class Codes: Y02T 10/40 Engine management systems

Abstract

An engine control system including an engine controller, a plurality of sensors coupled to the engine controller, and a neural network operating in the engine controller, and where upon failure of at least one of the plurality sensors, the neural network generates a representative value of the failed sensor.

Claims

1. An engine control system comprising:

an engine controller;

a plurality of sensor coupled to said engine controller;

a neural network operating in said engine controller; and

wherein upon failure of at least one of said plurality of sensors said neural network generates a representative value of said failed sensor.

2. The engine control system of claim 1 wherein said plurality of sensors includes a manifold pressure sensor, an air flow sensor, and a throttle position sensor.

3. The engine control system of claim 1 wherein said neural network continually generates representative values for said plurality of sensors.

4. The engine control system of claim 1 wherein said neural network has a transfer function of unity.

5. The engine control system of claim 1 wherein said neural network is an auto associative neural network.

6. The engine control system of claim 1 wherein said plurality of sensors includes, a transmission speed sensor, an engine speed sensor, and wheel speed sensors.

7. The engine control system of claim 1 further including a compression ignition engine.

9. The method of claim 1 wherein said step of processing said sensor measurement to determine representative values of said sensor measurements comprises processing said sensor measurements with a neural network.

8. A method of engine control comprising:

determining a set of sensor measurements;

processing said sensor measurement to determine representative values of said sensor measurements; and

determining a sensor fault based on a comparison of said set of sensor measurements and said representative values of said sensor measurements.

10. A control system for a vehicle comprising:

an engine;

a controller controlling said engine;

a plurality of sensors coupled to said controller;

a neural network operating in said controller generating sensor signals for said controller; and

wherein said plurality of sensors generates a virtual sensor value.

11. The control system of claim 10 wherein said plurality of sensors comprise a pressure sensor, an air flow sensor, and a throttle position sensor.

12. The control system of claim 10 wherein said neural network has a transfer function of unity.

13. The control system of claim 10 wherein said neural network examines the plurality of sensors for faults and generates a representative signal for a faulted sensor.

14. A method of engine control comprising:

providing an internal combustion engine;

providing a controller to control the internal combustion engine;

providing a neural network operating in said controller;

determining a set of sensor measurements;

processing said sensor measurements with the neural network to determine representative values of said sensor measurements; and

wherein at least one of said representative values of said sensor measurements is a virtual sensor element where said virtual sensor measurement replaces a traditional physical sensor measurement.