Vehicle fuel tank modeling systems and methods

Patent metadata

Field	Value
Publication number	US-11615658-B2
Application number	US-202017037829-A
Country	US
Kind code	B2
Filing date	Sep 30, 2020
Priority date	Sep 30, 2020
Publication date	Mar 28, 2023
Grant date	Mar 28, 2023

Abstract

Vehicle fuel tank modeling systems and methods are provided herein. An example method includes introducing known volumes of fuel into the fuel tank, determining correlations of fuel level percentage values relative to the known volumes of fuel placed into the fuel tank, offsetting the correlations to compensate for unusable fuel volume of the fuel tank, selecting a non-linear fuel tank model that fits the offset correlations, and applying the non-linear fuel tank model to a target vehicle having a fuel tank with a fuel tank part number.

First claim

What is claimed is:

1. A method, comprising:
introducing known volumes of fuel into a fuel tank of a first vehicle, wherein the fuel tank of the first vehicle comprises a fuel tank part number;
determining correlations of fuel level percentage values relative to the known volumes of fuel placed into the fuel tank of the first vehicle;
offsetting the correlations to compensate for unusable fuel volume of the fuel tank of the first vehicle;
selecting a non-linear fuel tank model that fits the offset correlations; and
applying the non-linear fuel tank model to a fuel tank comprising the fuel tank part number of a target vehicle,
wherein generating the fuel tank model comprises determining an estimate of unusable fuel volume left in the fuel tank, the offsetting of the correlations being based on the unusable fuel volume.

2. The method according to claim 1 , further comprising:
receiving a request to determine a fuel level volume for the target vehicle, the request comprising a fuel level percentage value of the target vehicle; and
converting the fuel level percentage value into the fuel level volume using the non-linear fuel tank model.

3. The method according to claim 2 , further comprising:
determining a vehicle identification number (VIN) for the target vehicle included in the request;
determining a fuel tank part number of the target vehicle based on the VIN; and
selecting the non-linear fuel tank model using the fuel tank part number.

4. The method according to claim 1 , wherein selecting the non-linear fuel tank model comprises:
determining non-linear model candidates for the fuel tank;
performing non-linear regression analysis on the fuel level percentage values and the known volumes of fuel using the non-linear model candidates; and
evaluating the non-linear model candidates by determining at least one of a cross-validation error, mean absolute deviation, and/or a model error tolerance.

5. The method according to claim 1 , further comprising:
draining the fuel tank to empty;
determining, based on the fuel tank being empty, a fuel level percentage value of the fuel level percentage values; and
wherein the known volumes of fuel vary in amount.

6. A method, comprising:
emptying a fuel tank of a vehicle;
determining a fuel level percentage value for the fuel tank when the fuel tank is empty to determine an offset;
incrementally filling the fuel tank with known volumes of fuel until the fuel tank is full;
determining correlations between fuel level percentage values and the known volumes of fuel;
applying the offset to the correlations;
selecting a non-linear model for the correlations;
determining a plurality of fuel tank models for a plurality of vehicles that comprise a similar or identical fuel tank to the fuel tank;
selecting a non-linear model candidate for the fuel tank;
performing non-linear regression analysis on the fuel level percentage values and the known volumes of fuel using the non-linear model candidate; and
evaluating the non-linear model candidate by determining at least one of a cross-validation error, mean absolute deviation, and/or a model error tolerance.

7. The method according to claim 6 , further comprising:
evaluating a plurality of non-linear model candidates that include the non-linear model candidate; and
determining a best-performing one of the plurality of non-linear model candidates, the best performing one of the plurality of non-linear model candidates having a cross-validation error value that is at or below a cross-validation error threshold, the best-performing one of the plurality of non-linear model candidates being the non-linear model.

8. The method according to claim 6 , further comprising determining an estimate of unusable fuel volume left in the fuel tank.

9. The method according to claim 6 , wherein an amount of each of the known volumes of fuel is varied.

10. A system, comprising:
a processor; and
a memory for storing instructions, the processor executing the instructions to:
determining an estimate of unusable fuel volume left in a fuel tank of a first vehicle when the fuel tank is empty;
incrementally filling the fuel tank with known volumes of fuel;
determining correlations between fuel level percentage values and the known volumes of fuel;
applying an offset to the fuel level percentage values to compensate for the estimate of the unusable fuel volume;
determining a best-performing non-linear model candidate from a plurality of non-linear model candidates that matches the correlations between the fuel level percentage values and the known volumes of fuel; and
determining a fuel level volume for a second vehicle having a similar or same fuel tank using the best-performing non-linear model candidate.

11. The system according to claim 10 , wherein the processor is configured to:
receive a request to determine a fuel level volume for second vehicle, the request comprising a fuel level percentage value of the second vehicle; and
convert the fuel level percentage value into the fuel level volume using a non-linear fuel tank model.

12. The system according to claim 11 , wherein the processor is configured to:
determine a vehicle identification number (VIN) for the second vehicle included in the request;
determine a fuel tank part number of the second vehicle based on the VIN; and
select the non-linear fuel tank model using the fuel tank part number.

13. The system according to claim 10 , wherein the processor is configured to generate the fuel tank model by determining an estimate of unusable fuel volume left in the fuel tank, the offsetting of the correlations being based on the unusable fuel volume.

14. The system according to claim 10 , wherein the processor is configured to select a non-linear fuel tank model by:
determining non-linear model candidates for the fuel tank;
performing non-linear regression analysis on the fuel level percentage values and the known volumes of fuel using the non-linear model candidates; and
evaluating the non-linear model candidates by determining at least one of a cross-validation error, mean absolute deviation, and/or a model error tolerance.

15. The system according to claim 10 , wherein the processor is configured to:
cause the fuel tank to drain to empty; and
determine, when the fuel tank is empty, a fuel level percentage value of the fuel level percentage values.

16. The system according to claim 10 , wherein the known volumes of fuel vary in amount.

17. The system according to claim 10 , wherein the processor is configured to stream the fuel level volume for the second vehicle to a recipient system.

18. A method, comprising:
introducing known volumes of fuel into a fuel tank of a first vehicle, wherein the fuel tank of the first vehicle comprises a fuel tank part number;
determining correlations of fuel level percentage values relative to the known volumes of fuel placed into the fuel tank of the first vehicle;
offsetting the correlations to compensate for unusable fuel volume of the fuel tank of the first vehicle;
selecting a non-linear fuel tank model that fits the offset correlations; and
applying the non-linear fuel tank model to a fuel tank comprising the fuel tank part number of a target vehicle,
wherein selecting the non-linear fuel tank model comprises:
determining non-linear model candidates for the fuel tank;
performing non-linear regression analysis on the fuel level percentage values and the known volumes of fuel using the non-linear model candidates; and
evaluating the n

Assignees

• Ford Global Tech Llc

Inventors

• Park Hyongju
• Bulat Ben
• Gordon Richard
• Babila Arnold

Classifications

• B60K15/03 Primary

  Fuel tanks · CPC title

• B60K2015/03473

  for draining or emptying a fuel tank · CPC title

• B60K2015/03203

  during refuelling · CPC title

• G07C5/0816 Primary

  Indicating performance data, e.g. occurrence of a malfunction · CPC title

• G07C5/008

  communicating information to a remotely located station (transmission systems for measured values G08C) · CPC title