Method of learning torque-stroke curve of electric motor controlled dry clutch system

1. A method for learning a torque-stroke (T-S) curve of an electric motor controlled dry clutch system, the method comprising:

calculating, by a control unit, a position change value A for allowing a position change point P 3 corresponding to an arbitrary torque y 3 on a previous T-S curve C 1 to follow-up and be moved to an expectation T-S curve C 3 ;

calculating, by the control unit, a probability Pr_X 3 to allow a consideration of various environmental factors of a clutch based on a maximum value of a valid range in which the position change value A varies, and multiplying the probability Pr_X 3 to the position change value A to calculate a final position change value A;

calculating, by the control unit, a new point P 3 by applying the final position change value A to the position change point P 3 of the previous T-S curve C 1 , and generating a final T-S curve connecting the new point P 3 and a touch point; and

controlling, by the control unit, a stroke of an electric motor based on the final T-S curve so that a dry clutch transmits a target torque from an engine, the electric motor changing a position of the dry clutch.

2. The method of claim 1 ,

wherein the step of calculating a position change value A comprises:

moving, by the control unit, the position change point P 3 corresponding to the arbitrary torque y 3 on the previous T-S curve C 1 by a predetermined unit bias;

generating, by the control unit, a virtual T-S curve C 2 connecting the point P 3 of which position is moved by the unit bias on the previous T-S curve C 1 and the touch point;

calculating, by the control unit, a difference value X between a position at an adaptation point of the virtual T-S curve C 2 and a position at an adaptation point of the previous T-S curve C 1 corresponding to a torque of the adaptation point of the virtual T-S curve C 2 ;

calculating, by the control unit, a position change amount Δpos at the adaptation point of an expectation T-S curve C 3 corresponding to a torque of the previous T-S curve C 1 ; and

calculating, by the control unit, the position change value A based on a position difference value X at the adaptation points of the previous T-S curve C 1 and the virtual T-S curve C 2 , and a position variation amount Δpos at the adaptation points of the previous T-S curve C 1 and the expectation T-S curve C 3 .

3. The method of claim 2 ,

wherein the previous T-S curve C 1 , the virtual T-S curve C 2 , and the expectation T-S curve C 3 are generated by connecting the touch point and position change points of each of the T-S curves C 1 , C 2 and C 3 in a spline curve shape using a spline function.

4. The method of claim 3 ,

wherein the previous T-S curve C 1 is a T-S curve stored initially or a T-S curve learned and stored previously,

the expectation T-S curve C 3 is a T-S curve generated by connecting the position points corresponding to respective torque at entire regions while a vehicle actually drives, and

the virtual T-S curve C 2 is a T-S curve generated by connecting a point P 3 of which position is moved on the previous T-S curve C 1 by the unit bias and the touch point.

5. The method of claim 2 , wherein the unit bias is set to 1 mm.

6. The method of claim 1 , wherein the probability Pr_X 3 corresponds to a resulting value of a probability density function that is calculated on a basis of a normal distribution curve of the position change value A.

7. The method of claim 6 , wherein the probability Pr_X 3 is expressed as an equation of:

1

σ

⁢

⁢

1

⁢

2

⁢

⁢

π

⁢

e

-

a

2

2

⁢

⁢

σ

⁢

⁢

1

2

wherein σ1 is a standard deviation of aMax/2, the aMax is the maximum value, and a is the position change value at P 3 .