PREDICTION OF POWER-SYSTEM FREQUENCY-RESPONSE AFTER GENERATOR OUTAGESUSING NEURAL NETS

A new methodology is presented for estimating the frequency behaviour of power systems necessary for an indication of underfrequency load sh edding in steady-state security assessment. It is well known that larg e structural disturbances such as generator tripping or load outages c an initiate cascading outages, system separation into islands, and eve n the complete breakup. The underfrequency load shedding takes place d uring the beginning phase of a dynamic change of the system frequency initiated by these distrubances. In this context the authors examine t he ability of neural nets to properly interpolate among training data sets and to accurately predict the system frequency variations. The ne utral-net approach provides a fairly accurate method of estimating the system average frequency response without making simplifications or n eglecting non-linearities and small time constants in the equations of generating units, voltage regulators and turbines. The understanding and selection of the input features comes from the developed, simple l ow-order system frequency response model. Additional features are defi ned in terms of the centre of inertia acceleration and admittance dist ances. The efficiency of the new procedure is demonstrated using the N ew England power system model for a series of characteristic perturbat ions. The validity of the proposed approach is verified by comparison with the simulation of short-term dynamics including effects of contro l and automatic devices.