Modeling a gas-turbine unit with prognostic regulators of voltage and speed

Abstract

Object. The article presents the results of computer research, the purpose of which was to develop dynamic models of low-power twin-shaft gas turbine units (GTU) equipped with automatic excitation and speed regulators with predictive links. Methods. Setting up the regulators of the commissioned gas turbine requires complex calculations. The use of algorithms that make it possible to improve the classical regulators with minimal costs allows us to effectively solve the problems of their settings. These include prognostic algorithms that allow you to configure the automatic control system using one parameter - forecast time. The models presented in the article are implemented using the Simulink package of the MATLAB system. Results. The influence of prognostic algorithms on the quality of controlling the voltage and rotor speed of a gas turbine generator in the modes of connecting and dumping loads is studied. Studies have shown that increasing the coefficient of the amplifier of the auto-predictive speed regulator can significantly reduce overshoot and the transition process, and also has a positive effect on voltage regulation. Conclusions. The results made it possible to formulate the following conclusions: the use of prognostic regulators when connecting an additional load to the gas turbine allows you to remove the oscillation, reduce voltage dips, reduce the transient time by 2.5 s; with a sharp load shedding due to the use of prognostic algorithms, it is possible to completely remove the oscillation, reduce the overshoot of the rotor speed and overvoltage at the generator terminals, and also significantly reduce the transient time for speed compared to classical regulators; the use of prognostic algorithms allows us to obtain acceptable quality indicators of transients without the use of complex regulator tuning procedures.