The increase in the number of distributed energy resources (DERs) in the low-voltage grid causes reverse active power flow, which induces voltage regulation issues across the feeder. We employ the parameter-free copula estimation of distribution algorithm (EDA) that optimally controls the reactive power of DERs to minimize voltage deviations. EDAs iteratively learn from data and sample an explicit probability distribution that models the dependencies between variables, allowing for a more effective exploration of the optimal solution space with fewer iterations. A copula offers additional flexibility, since the dependence structure between the decision variables and the marginal distributions can be modeled independently. The effectiveness of the proposed method is illustrated on a modified IEEE 123 node test feeder with 10 smart photovoltaic inverters. The results show that the proposed method achieves improved voltage profiles and offers many opportunities for further adaptability.