Move the item, and both sides take the same logarithm.

log(V-V(t))= log(V-V 0)-t/τ

The above v, V(t) and t are known, and V0 and τ are unknown.

Let y=log(V-V(t))

x=t

Then kx+b=y

Where k =- 1/τ and b = log (v-v0).

The above process is to turn nonlinearity into linearity.

Substituting each data into the membership degree, the binary linear equations are obtained.

Equation 6 and variable 2 can be solved by least square method.

The following is the Matlab code:

v = 14；

t=[0.3，0.5， 1.0，2.0，4.0，7.0]；

Vt=[5.6873，6. 1434，7. 1633，8.8626， 1 1.0328， 12.6962]；

y = log(V-Vt)；

A=[t(:)，ones(size(t(:)))]；

kb = A \ y(:)；

τ=- 1/kb( 1)

V0=V-exp(kb(2))

Below%% is the result of fitting.

Vt_fit=V-(V-V0)*exp(-t/tau)

plot(Vt，Vt_fit，' * '，Vt，Vt，' r ')

Xlabel ("vt raw data")

Ylabel ("Ventricular tachycardia fitting data")

%%%%%%%%%%%%%%%

Fill in the blanks:

Methods: The nonlinear problem was transformed into a linear problem.

V0=5.000 1

τ= 3.6 165