Predicción del Margen de Estabilidad de Corredores de Transmisión Aplicando Criterios de Minería de datos y Algoritmos de Machine Learning

Voltage Stability refers to the ability of the system to maintain acceptable voltages in all busbars, considering normal operating conditions and after being subjected to a disturbance. The present work predicts the critical parameters of the system based on the P-V Curve determined by the Thevenin's Equivalent in a transmission corridor. A dataset is obtained via Monte Carlo simulations performed on the 39-bus test system model in PowerFactory controlled by Python. Given an operating condition, N simulations are performed to establish different system operating conditions under variations in the values of each of the system loads. From the obtained dataset, Data Mining is applied to train regression models based on artificial neural networks and support vector machines to predict the maximum power transfer condition. Afterwards, the MSE (Mean-squared error) is used to analyze the performance of the regression models. The proposed methodology can be applied in control centers to predict the maximum transfer power point of a congested transmission corridor. This prediction offers early warning signs in operations and might allow structuring criteria for security constrained dispatch in planning.