| written 9.4 years ago by | modified 3.8 years ago by |
Mumbai University > Electronics and Telecommunication > Sem 4 > Applied Maths 4
Marks: 8M
Year: Dec 2015
| written 9.4 years ago by | modified 3.8 years ago by |
Mumbai University > Electronics and Telecommunication > Sem 4 > Applied Maths 4
Marks: 8M
Year: Dec 2015
| written 9.4 years ago by |
Suppose we have a closed curve of fixed length $L = \int\sqrt{\dot{x}^2 + \dot{y}^2}dt$ and we swish to enclose the maximum amount of area Green's theorem allow us to express area as a line integral
$\int\frac{1}{2} (-y\dot{x} - x\dot{y})dt$ so that the variation problem becomes.
Maximize $J = \int\frac{1}{2} (-y\dot{x} - x\dot{y})dt - \lambda \sqrt{\dot{x}^2 + \dot{y}^2}dt$
The two variable Euless - Lagrange equations are
$\frac{\dot{y}}{2} - \frac{d}{dt} \bigg(-\frac{y}{2} - \frac{\lambda \dot{x}}{\sqrt{\dot{x}^2 + \dot{y}^2}}\bigg) = 0 \\ \frac{-\dot{x}}{2} - \frac{d}{dt} \bigg(\frac{y}{2} - \frac{\lambda \dot{y}}{\sqrt{\dot{x}^2 + \dot{y}^2}}\bigg) = 0$
Which may be written as
$\frac{d}{dt}\bigg(y + \frac{\lambda \dot{x}}{\sqrt{\dot{x}^2 + \dot{y}^2}}\bigg) = 0$
We then have
$y + \frac{\lambda \dot{x}}{\sqrt{\dot{x}^2 + \dot{y}^2}} = \cdot C$ and $-x + \frac{\lambda \dot{y}}{\sqrt{\dot{x}^2 + \dot{y}^2}} = \cdot D$
Simplifying both equations we get $(x - D)^2 + (y - C)^2 = \lambda^2$
Which is an equation of circle centered at (D, C) with radius $\lambda$.