Power System Transient Voltage Stability Region Boundary Approximation and Stability Margin Estimation Based On Residual Neural Network

The complexity and diversity of new power system make the traditional methods of mechanism stability assessment methods have shortcomings such as poor applicability and low computational efficiency. The intelligent assessment methods represented by neural network have outstanding advantages in terms of efficiency, but its black box model is difficult to be understood by power system operators, which seriously affects its practicability. Therefore, taking the transient voltage stability problem as an example, this paper proposes a method to derive the analytical expression of the boundary of the transient voltage stability region (TVS) described by the neural network evaluation model. On the one hand, the proposed method establishes the connection between the mechanical transient voltage stability assessment and the intelligent assessment based on neural network. On the other hand, the proposed method constructs the stability margin estimation based on boundary approximation, which further improves the practicability of the intelligent assessment method. Test results on dual induction motor 4-bus system and IEEE 39-system verify the effectiveness of the proposed method.