Enhancing Power System Security: Neural Network Approaches for Quick and Robust Static Evaluation

In the modern era, sophisticated techniques for ranking and screening contingencies are necessary for the efficient operation of power systems. The Multi-Layer FeedForward Neural Network (MFNN) and the Radial Basis Function Network (RBFN) are the two artificial neural networks used by the Power System Static Security Evaluation (PSSSE) module. Two indices—the line stability index and the fast voltage stability index - that are derived from NewtonRaphson Load Flow (NRLF) analysis under different load scenarios during $\mathrm{N}-1$ line outage contingency is used to determine the system’s severity. The proposed MFNN and RBFN models use load scenarios, N-1 line outage scenarios, and power system operational states as input features to train the neural network models. Predicting weighted composite indices for undetermined network environments and ranking them for security evaluation are the primary objectives. Through simulation results, the suggested PSSSE module’s robustness and performance are verified on a typical IEEE 30bus test system. The proposed model is rapid, precise, and dependable for evaluating power system static security in the event of unforeseen network conditions, according to a comparison of accuracy and execution time with NRLF analysis. The outcomes show that the MFNN and RBFN model-based designed PSSSE module is suitable for online implementation and offers a trustworthy way to evaluate the security of power systems in real-time.