State estimation of synchronous generator mode parameters based on Kalman filter

Relevance. of the research is to develop a methodology for the Kalman filter tuning for the problem of synchronous generator control in the operation mode. This allows us to synthesize optimal and adaptive controllers. The complication consists in the absence of the possibility of direct measurement of all the necessary quantities. The Purpose. To determine the approach of Kalman filter tuning for the control problems in the operating mode of a synchronous generator. To determine the filtering quality of the output signals and estimation of the operating parameters with disturbance in the input channels and noise in the measuring ones. Methods. To achieve the aim, the methods of mathematical modeling implemented by MatLab/Simulink tools were used. The methods of mathematical statistics and optimization were used. Results. The paper describes the algorithm for Kalman filter tuning that allows to determine the vector of state variables without linearizing the synchronous generator model. The research was conducted on the influence of noise covariance matrix Q and R for quality filtration and parameters estimation of rotor angle and speed. For the research, a digital model with a synchronous generator from the MatLab/Simulink library was designed and another one was designed with a generator in the state space. Conclusion. It has been established that matrices Q and R influence on the speed of movement of the estimated parameters by the true values. If the matrices components are equal the power spectral densities of the disturbances and noise in these measuring channels, the measurement error will tend for zero. The same result is obtained if the components of the Q and R matrices do not have the necessary power spectral densities of disturbances and noise, but at the same time, between the components of the Q and R matrices, it is necessary to maintain the same proportion as when they are created by the above-mentioned method.