An energy function based unified power flow controller (UPFC) is developed for improving transient stability of network-preserving power systems. In order to consider model uncertainties, we also propose a forward neural networks controller to deal with such model uncertainties. This controller can be treated as neural network approximations of energy function control actions and provides online learning ability. Simulations on two power systems demonstrate that the proposed control strategy is very effective for suppressing power swing even under severe system conditions.