Transient stability assessment and improvement are critical for power grid operation. It deals with the assessment of transient behavior of the power grid (especially the generators) when subjected to large disturbances. State-of-the-art approaches for transient assessment are classified into two: a) numerical methods and b) direct energy functions methods. Numerical methods are computationally expensive and current energy function methods require extensive system knowledge in advance. In this dissertation, two new approaches for transient stability assessment is investigated. First, a new method for predicting the behavior of power system generators is presented. The main advantage of this method is that it helps to find the critical generators, their critical clearing times and angles. Consequently, the system transient stability prediction can be performed. Also, using the Lyapunov theory and energy concept, the prediction can be used to find the unstable equilibrium point of the system. Second, an approach for assessing the potential energy capacity of the power system to prevent and control the transient instability of the power system is proposed. The approach can be used to find the appropriate control strategy so that system instability can be prevented.The proposed methods are tested on IEEE 9 bus, IEEE 39 bus, and North Carolina - South Carolina 500 bus systems. The results and discussions are provided.