A smart power system stabilizer for dynamic reduction of a power system model

Masrob, Mohamed Ali (2018) A smart power system stabilizer for dynamic reduction of a power system model. Doctoral (PhD) thesis, Memorial University of Newfoundland.

[English] PDF - Accepted Version Available under License - The author retains copyright ownership and moral rights in this thesis. Neither the thesis nor substantial extracts from it may be printed or otherwise reproduced without the author's permission. Download (9MB)

Abstract

This thesis focuses on studying the dynamic stability of power systems and improving them by the addition of smart power system stabilizers (PSSs). A conventional design technique of a power system stabilizer that uses a single machine connected to an infinite bus through a transmission line (SMIB) has been widely used for study of elecromechanical perturbations. This approach requires estimating the external equivalent impedance and the voltage at an external bus for each machine in a multi-machine system. This study will use the conventional mathematical method, which represents a power system with some modifications. The dynamic model is linearized by taking the high voltage side on the generation unit as a reference instead of the infinite bus voltage. By using this modification, several improvements are accomplished, the main ones of which are: the estimation of states is eliminated, the time consumed in estimating calculations is reduced, the parameters of the model are independent of the external system, and the PSS design for each machine is independent in a multi-machine environment system. This strategy enables a PSS to be designed for a single machine and then implemented in a multi-machine system. Power systems have advanced to the point that they now cover vast geographical areas. Consequently, they are not only quite complicated, but the system orders are also high. As the complexity of these systems increases, so does the difficulty of examining their dynamic stability and adjusting their controllers. In this research, to address these issues, the reduced model technique has been employed to mathematically define smaller system models from existing models, such that the properties of both systems are comparable properties. The parameters of the PSS are determined based on a modified Heffron- Phillips model of the power system at certain operating conditions where it can provide reliable performance. Since the power systems are highly nonlinear with configurations and parameters that change with time, a typical PSS design, which is based on a linearized model of the power system, cannot guarantee its performance in practical operating environments. The present study attempts to overcome this limitation by implementing smart power system stabilizers. In the context of this thesis the word smart means novel technique. An artificial neural network power system stabilizer (ANN-PSS), a novel multi input fuzzy logic power system stabilizer (FLPSS), and a modified multi-resolution proportional-integral-derivative power system stabilizer (MMR-PID-PSS), based on the dynamic reduction of a power system model. These PSSs have been developed to refine the power system dynamic performance by adjusting the regulator’s parameters in real-time simulation under various operating conditions. In the first part of this research, the digital simulations results using the proposed ANN-PSS and FLPSS are carried out on a single machine connected to a network and are then compared with conventional Lead-Lag PSS. The results show that the power system with FLPSS has a better dynamic response over a wide range of operating conditions and parameter changes. Next, the digital simulations results using the proposed MMR-PID-PSS is carried out on a single machine connected to the network, a 4-machine 10-bus power system, and a 10-machine 39-bus power system and then compared with FLPSS. The results validate the effectiveness of the proposed MMR-PID-PSS regarding reduced overshoot, undershoot, and settling time under a different type of disturbances.

Actions (login required)

View Item