Power is the probability of correctly rejecting the null hypothesis ($H_0$) when the true parameter precisely equals $\mu_a$.
$$ \text{Power} = 1 - \beta $$
Factors Affecting Power
Effect Size: Greater distance between $\mu_a$ and $\mu_0$ dramatically increases Power.
Sample Size ($n$): Larger datasets compress the Standard Error, curving the probability exponentially closer to 1.0.
Significance ($\alpha$): Increasing $\alpha$ artificially raises Power at the expense of more Type I Errors.
The Minimum Trap
Notice how the lowest point on the curve (when $\mu_a = \mu_0$) is exactly exactly $\alpha$. We can never have zero probability of rejecting $H_0$ by definition of the test!