Amplitude of wave Does amplitude of wave changes during refraction of wave
If it will changes,why?
I have been searched for internet but still cannot find a complete answer
 A: No material is an ideally refracting or reflecting medium. There will always be a finite  amount of absorption. Correspondingly the amplitude of the wave would decrease.
If you treat the light as a wave you have to attribute the decrease to the dissipative term of the refractive index. If you want a physical mechanism you have to attribute this to absorption of photons by atoms and molecules in the medium and subsequent emission in a different direction.
A: Assuming you are talking about light, you need to define what you mean when you talk about the amplitude of the wave. Is it amplitude of electric field, magnetic field, or the square-root of intensity?
A plane wave in a non-magnetic loss-less medium with refractive index $n$  is given by:
electric field: $\mathbf{E}=\mathbf{E_0}\exp\left(i\omega\left( t - n\mathbf{\hat{k}}.\mathbf{r}/c\right)\right)+\mathbf{E_0^\dagger}\exp\left(-i\omega\left( t - n\mathbf{\hat{k}}.\mathbf{r}/c\right)\right)$
where $\omega$ is angular frequency, $c$ is the speed of light in vacuum, $t$ is time, $\mathbf{r}$ is position, $\mathbf{\hat{k}}$ is unit vector in the direction of wave propagation, and $\mathbf{E_0}$ is the complex 3d vector giving the polarization of the wave
magnetic field: $\mathbf{H}=\frac{n}{\mu_0 c}\,\mathbf{\hat{k}}\times\left[\mathbf{E_0}\exp\left(i\omega\left( t - n\mathbf{\hat{k}}.\mathbf{r}/c\right)\right)+\mathbf{E_0^\dagger}\exp\left(-i\omega\left( t - n\mathbf{\hat{k}}.\mathbf{r}/c\right)\right)\right]$
the time-averaged Poynting vector (proportional to intensity) is: $\left<\mathbf{S}\right>=\left<\mathbf{E} \times \mathbf{H}\right>=\frac{2n}{\mu_0 c}\Re\left(\mathbf{E_0}^\dagger.\mathbf{E_0}\right) \mathbf{\hat{k}}$
So, provided there are no reflections, if you talk about electric field  - the amplitude stays the same (when refracting from one medium into another one with different $n$), if about magnetic field - the amplitude changes (since $n$ appears in the expression).
If there are reflections, then the intensity of light must change, which would lead to changes to the left-handside of my last fomula. The right-hand side, with $\mathbf{E_0}^\dagger.\mathbf{E_0}$ must then change too.
