This question is likely asking you to recall Parseval's identity, which is stated as
$$\sum_{n=-\infty}^\infty |c_n|^2 = \dfrac{1}{2\pi}\int_{-\pi}^\pi |f(x)|^2 \, dx$$
where \$c_n\$ are the coefficients of the Fourier series of the function \$f(x)\$.
You can see that the right-hand side of the identity is proportional to the average power of the signal. So, with some manipulation of the scaling factors, you can use this to assign a portion of the power of the single to each of the fourier components.
Also note, I assume that your signal repeats on every interval of time T (e.g, \$f(t+T)=f(t)\$. The way you've drawn it,it looks like the signal is 0 outside the region \$0 < t < T\$, but that would make it a non-periodic signal and it wouldn't make sense to talk about its Fourier series representation or average power.