Tuning apparatus and tuning method for clutch temperature estimation model

Patent metadata

Field	Value
Publication number	US-11727172-B2
Application number	US-202017096022-A
Country	US
Kind code	B2
Filing date	Nov 12, 2020
Priority date	Apr 17, 2020
Publication date	Aug 15, 2023
Grant date	Aug 15, 2023

Abstract

A tuning method for a clutch temperature estimation model may include, generating n tuning genes, calculating a tuning value corresponding to a tuning variable by using information of each of the n tuning genes, calculating a temperature estimation accuracy by applying the calculated tuning value to the clutch temperature estimation model, extracting n tuning genes of highest calculated accuracies, and regenerating m tuning genes through recombination of the extracted n tuning genes.

First claim

What is claimed is:

1. An automatically tuning method without user interference for a clutch temperature estimation model, the tuning method comprising:
generating, by a tuning value generation module, a plurality of tuning genes;
calculating, by the tuning value generation module, a tuning value corresponding to a tuning variable by using information of each tuning gene of the plurality of tuning genes;
calculating, by an accuracy calculation module, a temperature estimation accuracy by applying the calculated tuning value to the clutch temperature estimation model;
extracting, by the accuracy calculation module, a first number of tuning genes having highest calculated accuracies among the plurality of tuning genes;
and regenerating, by a recombination module, a second number of tuning genes through recombination of the extracted first number of tuning genes with highest tuning accuracy,
wherein
regenerating the second number of tuning genes comprises:
randomly selecting a third number of tuning genes from the extracted first number of tuning genes;
extracting a tuning gene component for a randomly assigned region for each tuning gene of the third number of tuning genes; and
generating a new tuning gene by combining the extracted tuning gene components in corresponding regions.

2. The tuning method of claim 1 , wherein calculating a tuning value, calculating temperature estimation accuracies, and extracting the first number of tuning genes are further performed for the regenerated second number of tuning genes.

3. The tuning method of claim 2 , wherein extracting the first number of tuning genes comprises extracting tuning genes with highest accuracies among tuning genes with high accuracies in a previous generation and tuning genes in a subsequent generation.

4. The tuning method of claim 2 , further comprising:
determining whether a current generation is a final generation while increasing a number of generation by repeating the regenerating of the second number of tuning genes through recombination of the extracted first number of tuning genes, the calculating of the tuning value for the second number of tuning genes, the calculating of the accuracies for the second number of tuning genes, and the extracting of the first number of tuning genes for the second number of tuning genes; and
when a current generation is a final generation, finishing the repeating operation and tuning the clutch temperature estimation model by a tuning variable value with a highest accuracy.

5. The tuning method of claim 1 , wherein regenerating the second number of tuning genes further comprises regenerating a post-mutation tuning gene by setting a mutation probability for each chromosome in the new tuning gene.

6. The tuning method of claim 5 , wherein, in the regenerating of the second number of tuning genes, for each in a plurality of sub-tuning genes included in the first number of tuning genes, a mutation probability for a lower bit of a sub-tuning gene is set higher than a mutation probability for an upper bit of a sub-tuning gene, which is higher than the lower bit.

7. The tuning method of claim 1 , wherein in calculating the accuracies, the accuracies are calculated based on a pressure plate maximum temperature, a pressure plate cooling curve, a flywheel maximum temperature, and a flywheel cooling curve.

8. The tuning method of claim 7 , wherein calculating the accuracies comprises: calculating an accuracy of the clutch temperature estimation model by multiplying an accuracy of the pressure plate maximum temperature, an accuracy of the pressure plate cooling curve, an accuracy of the flywheel maximum temperature, and an accuracy of the flywheel cooling curve and by multiplying a mode weight value.

9. An automatically tuning apparatus without user interference for a clutch temperature estimation model, the automatically tuning apparatus comprising:
a computing system having a memory and a processor, the computing system having a set of modules stored on the memory and processed by the processor, the modules comprising:
a tuning value generation module configured to: randomly generate a plurality of first generation tuning genes and calculate a tuning value corresponding to a tuning variable by using information of each first generation tuning gene of the plurality of first generation tuning genes;
an accuracy calculation module configured to: apply the calculated tuning value to the clutch temperature estimation model, calculate an accuracy of temperature estimation, and extract tuning genes having highest calculated accuracies among the plurality of first generation tuning genes;
and a recombination module configured to regenerate second generation tuning genes by recombining the extracted first generation tuning genes with highest tuning accuracy,
wherein
the recombination module is further configured to: randomly select a number of tuning genes from the extracted first generation tuning genes, extract a tuning gene component for a randomly assigned region for each of the randomly selected number of tuning genes, and generate a new tuning gene by combining the extracted tuning gene components in corresponding regions.

10. The tuning apparatus of claim 9 , wherein:
the tuning value generation module is configured to calculate the tuning value for the second generation tuning genes; and
the accuracy calculation module is configured to apply the calculated tuning value for the second generation tuning genes to the clutch temperature estimation model, calculate the accuracy of temperature estimation, and extract third generation tuning genes with highest accuracies among the first generation tuning genes and the second generation tuning genes.

11. The tuning apparatus of claim 10 , wherein the recombination module is configured to recombine the extracted third generation tuning genes to regenerate fourth generation tuning genes.

12. The tuning apparatus of claim 9 , wherein the recombination module is configured to regenerate a post-mutation tuning gene by setting a mutation probability for each chromosome in the new tuning gene.

13. The tuning apparatus of claim 12 , wherein, for each in a plurality of sub-tuning genes included in the first generation tuning genes, a mutation probability for a lower bit of a sub-tuning gene is set higher than a mutation probability for an upper bit of a sub-tuning gene, which is higher than the lower bit.

14. The tuning apparatus of claim 9 , wherein the accuracy calculation module is configured to calculate the tuning accuracies based on accuracies of a pressure plate maximum temperature, a pressure plate cooling curve, a flywheel maximum temperature, and a flywheel cooling curve.

15. The tuning apparatus of claim 14 , wherein the accuracy calculation module is configured to calculate an accuracy of the clutch temperature estimation model by multiplying an accuracy of the pressure plate maximum temperature, an accuracy of the pressure plate cooling curve, an accuracy of the flywheel maximum temperature, and an accuracy of the flywheel cooling curve and by multiplying a mode weight value.

Assignees

• Hyundai Motor Co Ltd
• Kia Motors Corp

Inventors

• Lee Geontae
• Kim Min Ki
• Song Kyoung

Classifications

• G06F30/20 Primary

  Design optimisation, verification or simulation (optimisation, verification or simulation of circuit designs G06F30/30) · CPC title

• G01K13/00

  Thermometers specially adapted for specific purposes · CPC title

• G06F2119/08

  Thermal analysis or thermal optimisation · CPC title

• G06F30/15 Primary

  Vehicle, aircraft or watercraft design · CPC title

• F16D48/08 Primary

  Regulating clutch take-up on starting · CPC title