Approximating Trajectory Constraints with Machine Learning -- Microgrid Islanding with Frequency Constraints
This addresses the problem of ensuring stable frequency response during microgrid islanding for power system operators, representing an incremental improvement by applying a deep learning method to encode constraints in scheduling.
The paper tackles the microgrid scheduling problem with frequency constraints by approximating the nonlinear relationship between operating conditions and frequency nadir using a neural network, which is integrated into a mixed-integer formulation to ensure secure islanding; validation on a modified 33-node system shows successful islanding with secure frequency response, especially when wind turbine inertia emulation is included.
In this paper, we introduce a deep learning aided constraint encoding method to tackle the frequency-constraint microgrid scheduling problem. The nonlinear function between system operating condition and frequency nadir is approximated by using a neural network, which admits an exact mixed-integer formulation (MIP). This formulation is then integrated with the scheduling problem to encode the frequency constraint. With the stronger representation power of the neural network, the resulting commands can ensure adequate frequency response in a realistic setting in addition to islanding success. The proposed method is validated on a modified 33-node system. Successful islanding with a secure response is simulated under the scheduled commands using a detailed three-phase model in Simulink. The advantages of our model are particularly remarkable when the inertia emulation functions from wind turbine generators are considered.