Extending AC Security Constrained Optimal Power Flow for Low Inertia Systems with Artificial Neural Network-based Frequency Stability Constraints

Frequency stability in inverter-based renewable energy sources (RES)-dominated, low-inertia, power systems is a timely challenge. This paper employs a systematic approach, utilizing an artificial neural network (ANN) and dynamic simulation, to infer two key frequency stability indicators: nadir and rate of change of frequency (RoCoF). By reformulating the ANN mathematical model, these indicators are then integrated as mixed-integer non-linear constraints into a classical AC security-constrained optimal power flow (AC SCOPF), resulting in the proposed AC-F-SCOPF problem. The results of the proposed AC-F-SCOPF on the IEEE 39-bus system show that the problem identifies accurately the synchronous condensers which must run to ensure the frequency stability.