Abstract: With the penetration of distributed generation (DG), especially photovoltaic, into distribution networks, the issue of node voltage violation has become increasingly prominent. The stochastic nature of distributed generation and load uncertainty may result in uncertain power flow in distribution feeders, leading to node voltage collapse or overvoltage. In addition, multiple voltage regulators usually exist in complex distribution networks, while DG-based voltage regulation measures may conflict with traditional voltage regulators. Moreover, the traditional local control method has the disadvantages of high cost, redundant control operations, and limited support for the maximum utilization of new energy sources. Considering the aforementioned factors, in this paper, we propose a voltage violation mitigation approach for low-voltage distribution networks based on optimal control input selection. Through the synergy between distributed energy storage and DG, the supply-demand balance and flexibility level of active control within the feeder are improved, achieving simultaneous control of local node voltages. To evaluate the impact of any control signal on the voltages at all the nodes, the concepts of the comprehensive influence factor and the generalized electrical distance are proposed. Based on the comprehensive influence factor, a two-stage control input selection algorithm is introduced. In the first stage, the algorithm screens out the total control set that is effective for all the violated node voltages using node voltage iteration. On this basis, in the second stage, the algorithm screens out the redundant control signals therein, and ultimately realizes mitigation of voltage violations in the whole network with the smallest cost and the simplest control operation. The effectiveness of the proposed method is verified through simulations conducted on the modified IEEE 123-node distribution network.