Sensors bias detection for electronic returnless fuel system

1. Method for isolating an actual sensor bias in a fuel delivery system having a fuel pump comprising:

monitoring a plurality of fuel pump parameters, comprising:

monitoring a first fuel pump parameter comprising a pump current measured by a current sensor,

monitoring a second fuel pump parameter comprising a pump pressure measured by a pressure sensor, and

monitoring a third fuel pump parameter comprising a pump voltage;

modeling a first fuel pump modeled parameter based on the monitored second fuel pump parameter and the monitored third fuel pump parameter, said modeled first fuel pump modeled parameter corresponding to a modeled pump current;

modeling a second fuel pump modeled parameter based on the monitored first fuel pump parameter and the monitored third fuel pump parameter, said modeled second fuel pump modeled parameter corresponding to a modeled pump pressure;

detecting first and second fuel pump sensor biases, comprising

detecting the first fuel pump sensor bias based on a comparison between the first fuel pump parameter to the modeled first fuel pump modeled parameter, said detected first fuel pump sensor bias comprises a current sensor bias, and

detecting the second fuel pump sensor bias based on a comparison between the modeled second fuel pump modeled parameter to the second fuel pump parameter, said detected second fuel pump sensor bias comprises a pressure sensor bias;

modeling a fourth fuel pump modeled parameter based on the monitored second and third fuel pump parameters; and

isolating the actual sensor bias in one of the detected first and second fuel pump biases based on the monitored third fuel pump parameter and the modeled fourth fuel pump modeled parameter.

2. The method of claim 1 wherein the modeled fourth fuel pump modeled parameter comprises a modeled angular pump speed.

3. The method of claim 1 wherein the pump voltage is monitored in response to a pulse width modulation voltage to the fuel pump.

4. The method of claim 1 wherein detecting first and second fuel pump sensor biases comprises:

determining a first fuel pump parameter difference based on the comparison between the modeled first fuel pump modeled parameter to the monitored first fuel pump parameter;

determining a second fuel pump parameter difference based on the comparison between the modeled second fuel pump modeled parameter to the monitored second fuel pump parameter;

detecting the first fuel pump sensor bias when the first fuel pump parameter difference deviates from a first detected bias threshold; and

detecting the second fuel pump sensor bias when the second fuel pump parameter difference deviates from a second detected bias threshold.

5. The method of claim 4 further comprising:

filtering the first and second fuel pump differences.

6. The method of claim 1 wherein isolating the actual sensor bias in one of the detected first and second fuel pump sensor biases comprises isolating a fictitious sensor bias in the other one of the detected first and second fuel pump sensor biases based on the monitored third fuel pump parameter and the modeled fourth fuel pump modeled parameter.

7. The method of claim 6 further comprising:

flagging the isolated actual sensor bias in one of the detected first and second fuel pump sensor biases; and

resetting the fictitious sensor bias as a non-detected fuel pump sensor bias in the other one of the detected first and second fuel pump sensor biases.

8. The method of claim 1 wherein isolating the actual sensor bias in one of the detected first and second fuel pump sensor biases comprises:

determining an unbiased fourth fuel pump parameter based on the monitored third fuel pump parameter;

comparing the modeled fourth fuel pump modeled parameter and the unbiased fourth fuel pump parameter; and

isolating the actual sensor bias in one of the detected first and second fuel pump sensor biases based on the comparing.

9. The method of claim 8 wherein the unbiased fourth fuel pump parameter comprises an unbiased angular pump speed.

10. The method of claim 8 wherein isolating the actual sensor bias in one of the detected first and second fuel pump sensor biases comprises:

determining a difference between the modeled fourth fuel pump modeled parameter and the unbiased fourth fuel pump parameter; and

isolating the actual sensor bias in the detected first fuel pump sensor bias when the determined difference is not greater than an actual sensor bias threshold.

11. The method of claim 8 wherein isolating the actual sensor bias in one of the detected first and second fuel pump sensor biases comprises:

determining a difference between the modeled fourth fuel pump modeled parameter and the unbiased fourth fuel pump parameter;

isolating the actual sensor bias in the detected second fuel pump sensor bias when the determined difference is at least the actual sensor bias threshold plus an absolute value of the detected second fuel pump sensor bias as a function of the third fuel pump parameter.

12. The method of claim 1 wherein the fuel delivery system is an electronic returnless fuel system.

13. The method of claim 12 wherein the electronic returnless fuel system maintains a desired fuel system pressure by applying closed-loop correction derived from the monitored first and second fuel pump parameters as feedback.

14. Method for isolating an actual sensor bias in an electronic returnless fuel delivery system having a fuel pump including a pressure sensor and a current sensor comprising:

monitoring a pump pressure, a pump current and a pump voltage;

modeling pump current based on the monitored pump pressure and the monitored pump voltage;

modeling pump pressure based on the monitored pump current and the monitored pump voltage;

detecting biases in the current sensor and the pressure sensor, comprising:

detecting the bias in the current sensor based on a comparison between the monitored pump current to the modeled pump current, wherein the detected bias in the current sensor does not differentiate between an actual sensor bias in the current sensor and a fictitious sensor bias in the current sensor resulting from an actual sensor bias in the pressure sensor, and

detecting the bias in the pressure sensor based on a comparison between the monitored pump pressure to the modeled pump pressure, wherein the detected bias in the current sensor does not differentiate between the actual sensor bias in the pressure sensor and a fictitious bias in the pressure sensor resulting from the actual sensor bias in the current sensor;

modeling an angular pump speed based on the monitored pump pressure and the monitored pump voltage;

determining an unbiased angular pump speed based on the monitored pump voltage,

comparing the modeled angular pump speed and the unbiased angular pump speed;

determining a change in angular pump speed based on a difference between the modeled angular pump speed and the unbiased angular pump speed; and

isolating the corresponding actual sensor bias in one of the detected biases in the current sensor and the pressure sensor and isolating the corresponding fictitious sensor bias in the other one of the detected biases in the current sensor and the pressure sensor based on the determined change in the angular pump speed.

15. The method of claim 14 wherein isolating the corresponding actual sensor bias in one of the detected biases in the current sensor and the pressure sensor and isolating the corresponding fictitious sensor bias in the other one of the detected biases in the current sensor and the pressure sensor based on the determined change in angular pump speed, comprises:

isolating the corresponding actual sensor bias in the detected bias in the current sensor and isolating the corresponding fictitious sensor bias in the detected bias in the fuel pump pressure sensor when the determined change in angular pump speed is less than an actual bias sensor threshold.

16. The method of claim 14 wherein isolating the corresponding actual sensor bias in one of the detected biases in the current sensor and the pressure sensor and isolating the corresponding fictitious sensor bias in the other one of the detected biases in the current sensor and the pressure sensor, comprises:

isolating the corresponding actual sensor bias in the detected bias in the pressure sensor and isolating the corresponding fictitious sensor bias in the detected bias in the current sensor when the determined change in angular pump speed is at least an actual bias sensor threshold plus an absolute value of the detected bias in the fuel pump pressure sensor as a function of the pump voltage.

17. The method of claim 14 wherein the electronic returnless fuel delivery system maintains a desired fuel system pressure by applying closed-loop correction derived from the monitored pump pressure measured by the pressure sensor and the monitored pump current measured by the current sensor as feedback.

18. The method of claim 14 wherein isolating the corresponding actual sensor bias in one of the detected biases in current sensor and the pressure sensor and isolating the corresponding fictitious sensor bias in the other one of the detected biases in the current sensor and the pressure sensor comprises:

flagging the corresponding isolated actual sensor bias in one of the detected biases in the current sensor and the pressure sensor; and

removing the corresponding fictitious sensor bias from a detected bias in the other one of the detected biases in the current sensor and the pressure sensor.

19. An apparatus for isolating an actual sensor bias in an electronic returnless fuel delivery system including a first sensor and a second sensor comprising:

an internal combustion engine; and

an electronic returnless fuel delivery system comprising:

a fuel tank;

a fuel pump positioned within the fuel tank and supplying fuel from the fuel tank to the engine; and

a controller in communication with the fuel pump

monitoring a plurality of fuel pump parameters, comprising:

monitoring a first fuel pump parameter comprising a pump current measured by a current sensor,

monitoring a second fuel pump parameter comprising a pump pressure measured by a pressure sensor, and

monitoring a third fuel pump parameter comprising a pump voltage,

modeling a first fuel pump modeled parameter based on the monitored second fuel pump parameter and the monitored third fuel pump parameter, said modeled first fuel pump modeled parameter corresponding to a modeled pump current,

modeling a second fuel pump modeled parameter based on the monitored second fuel pump parameter and the monitored third fuel pump parameter, said modeled second fuel pump modeled parameter corresponding to a modeled pump current,

detecting first and second fuel pump sensor biases, comprising:

detecting the first fuel pump sensor bias based on a comparison between the first fuel pump parameter to the modeled first fuel pump modeled parameter, said detected first fuel pump sensor bias comprises a current sensor bias, and

detecting the second fuel pump sensor bias based on a comparison between the modeled second fuel pump modeled parameter to the second fuel pump parameter, said detected second fuel pump sensor bias comprises a pressure sensor bias,

modeling a fourth fuel pump modeled parameter based on the monitored second and third fuel pump parameters, and

isolating the actual sensor bias in one of the detected first and second fuel pump biases based on the monitored third fuel pump parameter and the modeled fourth fuel pump modeled parameter.