Method and apparatus for estimating data for engine control

• US 6,405,122 B1
• Filed: 06/02/1999
• Issued: 06/11/2002
• Est. Priority Date: 10/14/1997
• Status: Expired due to Fees

• Current Assignee: Yamaha Hatsudoki Kabushiki Kaisha (Yamaha Corporation)
• Original Assignee: Yamaha Hatsudoki Kabushiki Kaisha (Yamaha Corporation)
• Inventors: Yamaguchi, Masashi
• Primary Examiner(s): Wolfe, Willis R.
• Assistant Examiner(s): Vo, Hien T.
• Application Number:
  US09/324,554
• Time in Patent Office: 1,105 Days
• Field of Search: 701/106, 701/104, 701/109, 701/102, 706/19, 706/25, 706/31, 706/26, 706/27, 123/478, 123/480, 123/674
• US Class Current: 701/106
• CPC Class Codes: F02D 2041/1433 using a model or simulation of the system
• 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 41/1404 Fuzzy logic control
• CPC Class Codes: F02D 41/1405 Neural network control
• CPC Class Codes: F02D 41/1458 with determination means using an estimation
• CPC Class Codes: F02D 41/1462 with determination means using an estimation
• CPC Class Codes: G05B 13/0285 using neural networks and fuzzy logic
• CPC Class Codes: G06N 3/043 based on fuzzy logic, fuzzy membership or fuzzy inference, e.g. adaptive neuro-fuzzy inference systems [ANFIS]

Abstract

A data estimation capability using a FNN to estimate engine state data for an engine control system is described. The data estimation capability provides for making data relating to the engine state available as control parameters in a simple, inexpensive manner. The data estimation includes using data from one or more sensors as inputs to a FNN to estimate unmeasured engine operating states. The data estimates are provided as control parameters to an engine control system. The FNN can be used to provide data estimates for engine state values (e.g. the exhaust air fuel ratio, the exhaust NO_x. value, the combustion chamber temperature, etc.) that are too difficult or too expensive to measure directly. Each FNN can be configured using a genetic optimizer to select the input data used by the FNN and the coupling weights in the FNN. For example, the air/fuel ratio in the engine exhaust is estimated from the crank-angle acceleration, the crank angle, the engine speed, the air intake volume, the intake air vacuum, the ignition timing, the valve timing angle, and/or the EGR valve control; thereby eliminating the need for an O₂ sensor in the exhaust.

Claims

1. An engine controller comprising:

a control parameter output;

a plurality of inputs configured to receive input information; and

a fuzzy neural network configured to estimate said control parameter output from said input information to produce an estimated control parameter, said fuzzy neural network comprising a neural network and a fuzzy inference system, said fuzzy neural network trained by comparing said estimated control parameter with said control parameter to produce an error signal and training said fuzzy neural network to produce said estimated control signal in a manner that reduces said error signal.

2. The engine controller of claim 1, wherein said control parameter output comprises intake air volume, said controller further comprising a vacuum sensor.

3. The engine controller of claim 1, wherein said control parameter output comprises a fuel deposition rate in an intake manifold, said controller further comprising at least one of a tachometer, an intake air volume sensor and an intake vacuum sensor.

4. The controller of claim 1, wherein said control parameter output comprises an evaporation time constant in an intake manifold, said controller further comprising at least one of an intake manifold wall temperature sensor, an engine speed sensor, an intake air volume sensor and an intake vacuum pressure sensor.

5. The controller of claim 1, wherein said control parameter output comprises a torque fluctuation amount, and said input information comprises at least one of a fluctuation of an engine revolution speed, the engine revolution speed, an intake air volume and a time data for a combustion chamber pressure.

6. The controller of claim 1, wherein said control parameter output comprises a combustion chamber temperature, and said input information comprises at least one of an intake manifold wall temperature, an ambient air temperature and an elapsed time from the engine start.

7. The controller of claim 1, wherein said control parameter output comprises an intake manifold wall temperature, and said input information comprises at least one of a coolant temperature, a cylinder block temperature and an oil temperature.

8. The controller of claim 1, wherein said control parameter output comprises an air/fuel ratio, and said input information comprises at least one of a time data for a crank angle acceleration speed, an engine speed, an intake air volume, an intake manifold vacuum pressure, an ignition timing, an intake valve timing, and exhaust valve timing, and an EGR valve timing.

9. The controller of claim 1, wherein said control parameter output refers to an air/fuel ratio, and said input information comprises at least one of an exhaust gas temperature, an atmospheric temperature, an engine speed, an intake air volume, an intake vacuum pressure, an ignition timing and a throttle valve opening angle.

10. The controller of claim 1, wherein said control parameter output comprises an air/fuel ratio near a spark plug, said controller further comprising at least one of an engine speed sensor, and a throttle position sensor, said input data comprising at least one of a fuel injection volume, a fuel injection timing, an ignition timing, and an intake/exhaust valve timing.

11. The controller of claim 1, wherein said control parameter output comprises an amount of NO_x components in an exhaust gas inside a catalyst system, and said input information comprises data from at least one of an engine speed, a throttle opening angle, a coolant temperature, a gas temperature at an input of said catalyst system, a gas temperature at an output of said catalyst system, a traveling distance, and an air/fuel ratio.

12. The controller of claim 1, wherein said control parameter output comprises user feedback for the engine operating condition, and said input information comprises at least one type of physiologic data such as a pulse, a heartbeat, a blood pressure, a breathing interval and an electroencephalogram (EEG).

13. The controller of claim 1, further comprising a genetic optimizer configured to optimize at least the number and types of input data for said fuzzy neural network by evolving an elite chromosome using a genetic algorithm based on an estimation accuracy.

14. An engine controller comprising a controller output;

at least one input configured to receive input information; and

estimator means for estimating said controller output from said input information by computing one or more membership functions, applying fuzzy rules to one or more results of said one or more membership functions, and combining one or more results of said fuzzy rules to produce a value for said controller output.

15. The engine controller of claim 14, further comprising optimizer means for optimizing said estimator means.

16. A data estimation method for engine control in which data for at least one engine condition is identified as a control parameter, the method comprising the steps of:

measuring sensor data;

providing said sensor data to a fuzzy neural network as input information to said fuzzy neural network; and

estimating said control parameter for the engine condition by using said fuzzy neural network to produce an estimated control parameter, said fuzzy neural network comprising a neural network and a fuzzy inference system, said fuzzy neural network trained by comparing said estimated control parameter with said control parameter to produce an error signal and training said fuzzy neural network to produce said estimated control signal in a manner that reduces said error signal, thereby allowing said engine to be controlled by using said estimated control parameter in lieu of said control parameter.

17. The data estimation method of claim 16, wherein said control parameter comprises intake air volume and said input information comprises intake vacuum pressure of the engine.

18. The data estimation method of claim 16, wherein said control parameter comprises a fuel deposition rate in an intake manifold, and said input information comprises at least one of an engine rotational speed, an intake air volume and an intake vacuum.

19. The data estimation method of claim 16, wherein said control parameter comprises an evaporation time constant in said intake manifold, and said input information comprises at least one of an intake manifold wall temperature, an engine speed, an intake air volume and an intake vacuum pressure.

20. The data estimation method of claim 16, wherein said control parameter comprises a torque fluctuation amount, and said input information comprises at least one of a fluctuation of an engine revolution speed, the engine revolution speed, an intake air volume and a time data for a combustion chamber pressure.

21. The data estimation method of claim 16, wherein said control parameter comprises a combustion chamber temperature, and said input information comprises at least one of an intake manifold wall temperature, an ambient air temperature and an elapsed time from the engine start.

22. The data estimation method of claim 16, wherein said control parameter comprises an intake manifold wall temperature, and said input information comprises at least one of a coolant temperature, a cylinder block temperature and an oil temperature.

23. The data estimation method of claim 16, wherein said control parameter comprises an air/fuel ratio, and said input information comprises at least one of a time data for a crank angle acceleration speed, an engine speed, an intake air volume, an intake manifold vacuum pressure, an ignition timing, an intake valve timing, and an exhaust valve timing, and an EGR valve timing.

24. The data estimation method of claim 16, wherein said control parameter comprises an air/fuel ratio, and said input information comprises at least one of an exhaust gas temperature, an atmospheric temperature, an engine speed, an intake air volume, an intake vacuum pressure, an ignition timing and a throttle valve opening angle.

25. The data estimation method of claim 16, wherein said control parameter comprises an air/fuel ratio around an ignition location of a spark plug, and said input information comprises at least one of an engine rotational speed, a throttle opening angle, fuel an injection volume, a fuel injection timing, an ignition timing, and an intake/exhaust valve timing.

26. The data estimation method of claim 16, wherein said control parameter comprises an amount of NO_x components in an exhaust gas inside a catalyst system, and said input information comprises at least one of an engine speed, a throttle opening angle, a coolant temperature, a gas temperature at an input of said catalyst system, a gas temperature at an output of said catalyst system, a traveling distance, and an air/fuel ratio.

27. The data estimation method of claim 16, wherein said control parameter comprises user feedback for the engine operating condition, and said input information comprises at least one type of physiologic data such as a pulse, a heart beat, a blood pressure, a breathing interval and an electroencephalogram (EEG).

28. The data estimation method of claim 16, further comprising the steps of:

identifying said input information as an input candidate data to be used as input data for said fuzzy neural network;

coding information regarding use of said candidate input data into chromosomes for use by a genetic algorithm;

generating a plurality of individual chromosomes; and

, optimizing at least the number and types of input data for said fuzzy neural network by evolving elite chromosomes using said genetic algorithm.

29. A data estimation method for engine control in which data for at least one engine condition is identified as an estimated parameter, the method comprising the steps of:

obtaining input information from one or more sensors;

providing said input information to one or more inputs of a fuzzy neural network;

using said fuzzy neural network to estimate said estimated parameter; and

obtaining said estimated parameter as an output of said fuzzy neural network, wherein said estimated parameter comprises a fuel deposition rate in an intake manifold, and said input information comprises at least one of an engine rotational speed, an intake air volume and an intake vacuum.

30. A data estimation method for engine control in which data for at least one engine condition is identified as an estimated parameter, the method comprising the steps of:

obtaining input information from one or more sensors;

providing said input information to one or more inputs of a fuzzy neural network;

using said fuzzy neural network to estimate said estimated parameter; and

obtaining said estimated parameter as an output of said fuzzy neural network, wherein said estimated parameter comprises an evaporation time constant in said intake manifold, and said input information comprises at least one of an intake manifold wall temperature, an engine speed, an intake air volume and an intake vacuum pressure.

31. A data estimation method for engine control in which data for at least one engine condition is identified as an estimated parameter, the method comprising the steps of:

obtaining input information from one or more sensors;

providing said input information to one or more inputs of a fuzzy neural network;

using said fuzzy neural network to estimate said estimated parameter; and

obtaining said estimated parameter as an output of said fuzzy neural network, wherein said estimated parameter comprises a torque fluctuation amount, and said input information comprises at least one of a fluctuation of an engine revolution speed, the engine revolution speed, an intake air volume and a time data for a combustion chamber pressure.

32. A data estimation method for engine control in which data for at least one engine condition is identified as an estimated parameter, the method comprising the steps of:

obtaining input information from one or more sensors;

providing said input information to one or more inputs of a fuzzy neural network;

using said fuzzy neural network to estimate said estimated parameter; and

obtaining said estimated parameter as an output of said fuzzy neural network, wherein said estimated parameter comprises a combustion chamber temperature, and said input information comprises at least one of an intake manifold wall temperature, an ambient air temperature and an elapsed time from the engine start.

33. A data estimation method for engine control in which data for at least one engine condition is identified as an estimated parameter, the method comprising the steps of:

obtaining input information from one or more sensors;

providing said input information to one or more inputs of a fuzzy neural network;

using said fuzzy neural network to estimate said estimated parameter; and

obtaining said estimated parameter as an output of said fuzzy neural network, wherein said estimated parameter comprises an intake manifold wall temperature, and said input information comprises at least one of a coolant temperature, a cylinder block temperature and an oil temperature.

34. A data estimation method for engine control in which data for at least one engine condition is identified as an estimated parameter, the method comprising the steps of obtaining input information from one or more sensors;

providing said input information to one or more inputs of a fuzzy neural network;

using said fuzzy neural network to estimate said estimated parameter; and

obtaining said estimated parameter as an output of said fuzzy neural network, wherein said estimated parameter comprises an air/fuel ratio, and said input information comprises at least one of a time data for a crank angle acceleration speed, an engine speed, an intake air volume, an intake manifold vacuum pressure, an ignition timing, an intake valve timing, and an exhaust valve timing, and an EGR valve timing.

35. A data estimation method for engine control in which data for at least one engine condition is identified as an estimated parameter, the method comprising the steps of:

obtaining input information from one or more sensors;

providing said input information to one or more inputs of a fuzzy neural network;

using said fuzzy neural network to estimate said estimated parameter; and

obtaining said estimated parameter as an output of said fuzzy neural network, wherein said estimated parameter comprises an air/fuel ratio, and said input information comprises at least one of an exhaust gas temperature, an atmospheric temperature, an engine speed, an intake air volume, an intake vacuum pressure, an ignition timing and a throttle valve opening angle.

36. A data estimation method for engine control in which data for at least one engine condition is identified as an estimated parameter, the method comprising the steps of:

obtaining input information from one or more sensors;

providing said input information to one or more inputs of a fuzzy neural network;

using said fuzzy neural network to estimate said estimated parameter; and

obtaining said estimated parameter as an output of said fuzzy neural network, wherein said estimated parameter comprises an air/fuel ratio around an ignition location of a spark plug, and said input information comprises at least one of an engine rotational speed, a throttle opening angle, fuel an injection volume, a fuel injection timing, an ignition timing, and an intake/exhaust valve timing.

37. A data estimation method for engine control in which data for at least one engine condition is identified as an estimated parameter, the method comprising the steps of:

obtaining input information from one or more sensors;

providing said input information to one or more inputs of a fuzzy neural network;

using said fuzzy neural network to estimate said estimated parameter; and

obtaining said estimated parameter as an output of said fuzzy neural network, wherein said estimated parameter comprises an amount of NO_x components in an exhaust gas inside a catalyst system, and said input information comprises at least one of an engine speed, a throttle opening angle, a coolant temperature, a gas temperature at an input of said catalyst system, a gas temperature at an output of said catalyst system, a traveling distance, and an air/fuel ratio.