We can also test linearity and time invariance formally, using the operator that the equation defines, rather than just inspecting the terms.
Define the operator $L[y]=t^2y''+7ty'+8ty$. To test linearity, check whether $L[c_1y_1+c_2y_2]=c_1L[y_1]+c_2L[y_2]$ for any constants $c_1,c_2$ and any two functions $y_1(t),y_2(t)$.
Since differentiation itself is linear:
\[ L[c_1y_1+c_2y_2]=t^2(c_1y_1''+c_2y_2'')+7t(c_1y_1'+c_2y_2')+8t(c_1y_1+c_2y_2) \]Regrouping the terms:
\[ =c_1(t^2y_1''+7ty_1'+8ty_1)+c_2(t^2y_2''+7ty_2'+8ty_2)=c_1L[y_1]+c_2L[y_2] \]The superposition property holds exactly, so $L$ is a linear operator, confirming the equation $L[y]=10\sin t$ is linear.
Now test time invariance by shifting time. If $y(t)$ solves the equation and we define $z(t)=y(t-t_0)$, a genuinely time-invariant system would have $z(t)$ solve the same equation with the input shifted to $10\sin(t-t_0)$. But $y(t-t_0)$ actually satisfies
\[ (t-t_0)^2y''(t-t_0)+7(t-t_0)y'(t-t_0)+8(t-t_0)y(t-t_0)=10\sin(t-t_0) \]which uses coefficients $(t-t_0)^2,7(t-t_0),8(t-t_0)$, not the original coefficients $t^2,7t,8t$, unless $t_0=0$. Since the shifted function does not satisfy the equation with the original, unshifted coefficients, the system is not time-invariant, only linear.
Combining both checks: the equation is an ordinary (not partial) equation, it is linear, and it is time-varying rather than time-invariant.
\[ \boxed{\text{Linear differential equation}} \]Let \( y = f(x) \) be the solution of the differential equation\[\frac{dy}{dx} + \frac{xy}{x^2 - 1} = \frac{x^6 + 4x}{\sqrt{1 - x^2}}, \quad -1 < x < 1\] such that \( f(0) = 0 \). If \[6 \int_{-1/2}^{1/2} f(x)dx = 2\pi - \alpha\] then \( \alpha^2 \) is equal to ______.
If \[ \frac{dy}{dx} + 2y \sec^2 x = 2 \sec^2 x + 3 \tan x \cdot \sec^2 x \] and
and \( f(0) = \frac{5}{4} \), then the value of \[ 12 \left( y \left( \frac{\pi}{4} \right) - \frac{1}{e^2} \right) \] equals to: