Robust Nonlinear Feedback Linearizing Control for Power Systems to Enhance Transient Stability

摘要

Stability is an important issue for the secure operation of power systems as an unsecured system faces a lot of unusual operating conditions and due to these unusual operating conditions, power systems can undergo blackouts which cause ahuge losses. In addition, the demand on power systems is increasing consistentlyfor which the recent trend is to operate the system near to the extreme conditions.The changing characteristics of power systems continue to provide new challengesto the system designers and operators. To mitigate the constraints related to powersystem stability, the design and implementation of high-performance controllers arecrucial which is the key focus of this dissertation.The first contribution of this work is to design a nonlinear excitation controller toenhance transient stability. To clearly identify the necessity of new controllers, analyseshave been performed to determine the limitations of existing linear controllers.In this thesis, the feedback linearization technique which transforms a nonlinear systeminto a fully (exactly) or partially linear system based on the selection of outputfunction, is proposed. The proposed feedback linearizing is mainly based on thepartial feedback linearization technique which transforms the power system into apartially linearized system and is novel in power system applications. The proposedcontrol scheme offers several benefits as compared to the feedback linearization thattransforms a power system into a fully linear one.To implement a control algorithm in an accurate and effective manner, it is essentialto express the control law in terms of all measured variables. The control lawobtained through feedback linearization is in terms of the states of the power systemand some assumptions must be satisfied to express the state variables in terms ofmeasured variables. To validate the accuracy of these assumptions, it is essentialto design an effective observer which is capable to estimate the states of power systems under different changing conditions. The second and other unique contributionof this dissertation is to design a novel nonlinear state observer for power systemsbased on the feedback linearization technique and the observed states are used asthe feedback variables for feedback linearizing controllers. In this contribution, boththe full and partial feedback linearization are considered to provide a deep insightregarding the proposed scheme. The comparisons between the performances of the controller with observed states and measured variables feedbacks are presented witha view to look at the suitability of the assumptions.The other significant contribution of this research is to design a robust nonlinearcontroller for power systems to improve the performance by considering the mismatchbetween the original system model and the considered mathematical model.This part of the dissertation considers the uncertainty within the power systemmodel where the uncertainties are bounded in such a way that the proposed controllercan guarantee the stability and enhance the performance for all possibleperturbations within the given upper bounds on the modeling error of nonlinearpower systems. The effectiveness of the proposed controller is tested on small andlarge power systems.The current trend is the inclusion of renewable energy sources into the distributionnetwork to get some technical and economical benefits from the existingsystems. Since the integration of photovoltaic systems is increasing, this dissertationproposes the design of a novel controller based on partial feedback linearizationto ensure the dynamic stability under changing conditions of the system.The final contribution of this dissertation is to develop the relationship betweenthe amount of generation that can be integrated into the distribution networks andthe voltage variation which is a prominent issue for distribution networks. To developthis relationship, the worst-case scenario of the distribution network is consideredand based on the formulation, some effective solutions have been recommended.Finally, a partial feedback linearizing controller is designed for a distributed staticcompensator (DSTATCOM) to maintain the voltage stability through reactive powercompensation.