ODE in matlab -differential equations

64 Views Asked by At

I need help with this excercise : Find $k$(speed constant ) $$\frac{dx}{dt}=k(a-x)^2)$$ with data , $$t[min]=[0, 2.5, 5.6 ,9.6 ,14.6, 21.5, 32.5, 52.2]$$ ;

$(a-x)$ (mol/l) = [$0.0050$ (if x=0), $0.0045$ , $0.0040,0.0035,0.0030,0.0025,0.0020,0.0015]$

I´ve tried next steps:

Command W. :

$$t=[0 , 2.5, 5.6 , 9.6 , 14.6, 21.5 , 32.5 , 52.2];$$

$$x=[0:0.0005:0.0035];$$

$$a=0.0050;$$

$$c=a-x;$$

Editor ( my skript):

function $$dx=f(t,x)$$

$$dx=k*.((a-x).^2)$$

$$end + Run$$

Command W. :

$$tspan=0:0.0005:0.0035;$$

$$[x,t]=ode45(f,[0:0.0005:0.0035],0)$$

I don´t know how to find k

1

There are 1 best solutions below

2
On

Your ODE is separable and has as solution

$$ x(t) = \frac{a k t -1 + a c_0}{k t + c_0} $$

then one way to obtain the best fit to $k$ is by solving the following minimization problem

$$ \min_{k,c_0}\sum_{j=1}^n(a-x(t_j)-\delta_j)^2 $$

with

$$ \cases{ a = 0.0050\\ n = 8\\ t_j = \{0,2.5,5.6,9.6,14.6,21.5,32.5,52.2\}\\ \delta_j=\{0.0050,0.0045,0.0040,0.0035,0.0030,0.0025,0.0020,0.0015\} } $$

NOTE

If $x(0) = 0$ then

$$ x(t) = \frac{a^2k t}{a k t+1} $$

and now the minimization reads

$$ \min_{k}\sum_{j=1}^n\left(a-\frac{a^2k t_j}{a k t_j+1}-\delta_j\right)^2 $$