Minimum Frequency Predictive Model For Remote Isolated Power System Of Hydro Generation

Isolated grids in remote areas of southwest China are powered by hydro energy and faced with large load fluctuation. The low inertia and dominant water hammer effect deteriorate the frequency stability. Thus, a minimum frequency predictive model for power system containing hydro generation unit is proposed. First, based on the frequency response model of multi-machine system, two first-order transfer functions of numerator and denominator sharing the same order are adopted to simulate governor and turbine. The effect of zeros in transfer function, especially the water hammer effect, on the frequency response of power system is reflected. Then, the closed-loop of frequency response model is broken by approximating the feedback to governor using the exponential function. The amount of computation is thus reduced. The coefficients of the exponential function are obtained by implementing the balanced reduction method on the simplified closed-loop response of power system. Finally, the system of nonlinear equations is established through the open-loop transfer function. The frequency nadir of disturbed power system is obtained by solving the system of nonlinear equations using Newton method. The case study of an isolated grid in the southwest shows that the proposed method provides a quite accurate frequency nadir of disturbed power system containing hydro generation unit. The adoption of exponential fitting to approximate the feedback to governor improves the computation accuracy more than the linear fitting.