In this paper, we consider an unmanned aerial vehicle (UAV)-enabled wireless power transfer (WPT) network, where the UAV is responsible for sustaining the network with all devices on the ground. The sustainable charge process operates periodically, i.e., in each working period, the UAV flies from and back to a landing position. The initial remaining energy in the batteries and battery storage limits are considered in this study. We assume only the initial remaining network lifetime (RNL) known and no further lifetime updates available, and provide a sustainable design aiming at minimizing the average UAV consumed power while sustaining the network including all devices. We formulate the problem via determining the UAV trajectory and lengths of working period and charging phase, and introduce an iterative algorithm for an efficient solution. The repeats of designed working period are proved to guarantee the sustentation of the whole network. Moreover, when UAV behaves the same in all working periods, the obtained solution is strongly influenced by the initial RNL. To get rid of this influence, we further propose a second approach, namely a transition-based sustainable design, in which a transition state is introduced for improving the RNL, and the UAV updates its behaviour in each working period. The convergence and limit of this transition approach are addressed. After transitions, the network shifts to a steady state, and operates periodically as in the first design. Finally, through simulations, we validate the sustaining performance of both designs and the performance improvement via the transition approach.