INVESTIGATIONS ON ENHANCEMENT OF POWER QUALITY OF GRID CONNECTED AND ISLANDED HYBRID DISTRIBUTED GENERATING SYSTEMS

Abstract

It is not possible to continue using nonrenewable energies like oil, gas, and coal because they release a lot of climate gases. Renewable energies, on the other hand, are better and last longer. The tools to collect and use these powers have also improved. In recent years, they've gotten better, which makes using them more practical and cost effective. One important way to fight climate change and lessen our reliance on fossil fuels is to use renewable energy sources. Systems that get their power from two or more green sources are called hybrid renewable energy systems. These systems are especially helpful in places that can't connect to the regular power source or where the link is weak or unreliable. The system that distributes electricity has problems with power quality (PQ), such as low power factor, unbalanced loads, harmonics, uneven voltage and current, and more. Today's electricity systems change so quickly that there are plenty of problems with the quality of the power. In terms of current, voltage, frequency, and other things, Poor power quality is usually caused by problems with the wiring, lightning, bad weather like storms, broken equipment, and other things. However, harmonics are one of the main problems with power quality. This can be caused by several loads that don't behave linearly, such as old loads like transformers, electrical machines, and furnaces,, and new loads like power converters in vapor lights. Also, in the distribution networks' power quality is rapidly declining as a result of the proliferation of power electronic converters in homes, businesses, and factories. There are a lot of issues that are being caused by this, including higher losses, inefficient use of distribution networks, sensitive equipment malfunctioning, and disruption to surrounding customers, protective devices, and communication systems. Injecting intermittent electricity

vi directly into the distribution grid from renewable sources compounds these issues. Because of the advantages of modern electric equipment, such as reduced maintenance requirements, ease of control, low wear and tear, size and cost reduction, and energy conservation, solid-state controllers have become more prevalent and have exacerbated power quality problems. Because they employ solid-state controllers, electronically controlled energy-efficient commercial and industrial electrical loads are particularly vulnerable to power quality issues and may cause them. In light of the foregoing, this thesis is to help engineers and scientists in the field develop better energy supply systems by identifying, classifying, analyzing, simulating, and quantifying the power quality problems that come along with them. There is a distinction between the methods used in freshly built and developed equipment and those used to enhance power quality in existing systems that are experiencing power quality issues. Electrical loads and supply systems have distinct types of power quality issues; hence these mitigation approaches are further classed accordingly. Over the last 25 years, there has been a tremendous increase in the amount of time and energy spent studying how to reduce power quality issues. For single phase two-wire, three-phase three-wire, and three-phase four-wire systems, there has been a lot of study on power filters of different kinds, including passive, active, and hybrid in shunt, series, or a mix of the two configurations, to reduce harmonics and other issues like reactive power, excessive neutral current, and balancing linear and nonlinear loads. Common ways to boost PQ in a distribution system include adding capacitors, changing the taps on transformers, reactors, capacitor banks, and more. But these methods of compensation are slow and don't offer active load compensation, so new special power devices have been made. The technology, called SAPF, is now fully developed and can be used to fix problems with harmonic current, reactive power, and

vii neutral current in AC distribution networks. In three-phase systems, shunt active power filters are also used to control the voltage at the terminals and stop voltage flicker. Each of these goals can be met on its own or together, based on the needs, the control strategy, and the setup that needs to be chosen correctly. Moreover, improved SAPF dynamic and steady-state performance is now within reach, thanks to advancements in SAPF that allow for the application of various control algorithms. These algorithms include PI (proportional-integral), variable structure, fuzzy logic, and neural network based control algorithms. These enhancements allow APFs to quickly respond to changing nonlinear loads and take remedial action. More than that, these APFs may counteract harmonics of a higher order, usually up to the 25th harmonic. As part of the proposed work, design creation and study of both new and old control techniques have been investigated. With both linear and nonlinear loads, simulation and experiment data have been analyzed and put into tables. The sample system has been tried with standard control methods, such as Synchronous Reference Frame Theory (SRFT) and Instantaneous Reactive Power Theory (IPRT). The SRFT control method with ANN controller is developed and verified with a three phase-four wire SAPF system in MATLAB/Simulink environment. Harmonic current and reactive power adjustment were made possible by the photovoltaic integrated SAPF. Comparisons between the SAPF modified with ANN controller and PI and FLC are made. The DC-link voltage of Shunt APF was maintained constant by the PI, FUZZY, and ANN controllers. The ANN controller-based SAPF delivers better performance than PI and fuzzy controllers. Simulation has been performed in MATLAB/SIMULINK environment.

viii The other controllers were made for a grid-connected solar/hybrid distribution system. They are referred to as the second-order generalized integrator (SOGI) and the third-order generalized integrator (TOGI). To modify the energy storage of dc voltage, an artificial neural network controller (ANN), Fuzzy logic is employed with shunt active power filter (SAPF) for PQ improvement. Also, the SAPF is compared and assessed based on PI control, FUZZY controller, and Artificial Neural Network (ANN). Moreover, Fuzzy and ANN techniques are employed with conventional SRFT, SOGI techniques as well as with TOGI control algorithms. As well as being generated in MATLAB/Simulink, the system has also been tested in the real world with a distorted grid and distorted linear and nonlinear loads. An adaptive fourth-order based frequency locked loop (AFOGI-FLL) controller is developed to provide switching pulses for a three-phase IGBT-based voltage source converter. Accurate frequency synchronization, lower and higher-order harmonic elimination, power quality refinement, and reactive power compensation are some of the features of the proposed controller. In addition, a comparative analysis has been performed between the proposed controller and the third-order generalized integrator (TOGI). The comparative analysis results show that fourth-order generalized integrator have higher DC offset elimination capability, and better dynamic performance as well and THD is less compared with the TOGI controllers. Maximum Solar PV system extraction is achieved using the updated perturb and observed method. The proposed controller robustness is validated by an experimental prototype setup and test results exhibit the performances under nonlinear loading. Additionally, the feasibility of the proposed control scheme is demonstrated by using MATLAB/SIMULINK software. The system is robust. The system is well executed under irradiance variations, load fluctuations, and frequency variations as well. The

ix laboratory prototype is used to demonstrate the proposed system's simulated behaviour and experimental test results. Next, the stochastic-gradient-based adaptive control algorithms have been discussed and employed for power quality enhancement in a PV-integrated distribution system. Least mean square (LMS), Least mean fourth (LMF), sign-error LMS and ϵ Normalized LMS (ϵ-NLMS) have been implemented as control algorithms for the estimation of fundamental load current. The performances of these adaptive algorithms have been compared under steady-state and dynamic conditions under the non-linear load conditions in a closed-loop three-phase system. The most interesting part of this thesis is how to solve PQ problems in a three-phase grid-connected solar/hybrid distribution system with different SAPF configurations and both new and conventional algorithms.