In this letter, we study the performance of a wireless power transfer system under energy harvesting distribution uncertainty. The uncertainty captures practical nonidealities of the rectification process and is modelled as the maximum Kullback-Leibler distance of the actual distribution from a nominal distribution. The case where symmetrized divergence is considered for the statistical distance between distributions is also considered. By formulating a convex optimization problem, we investigate a mathematical framework that provides closed form expressions for the minimum average harvested energy (the worst performance) and the associated statistical distribution. Theoretical results show that the energy harvesting performance is significantly degraded as the level of the uncertainty increases.