An increase in carbon emission which mostly caused by the transportation sector and electric power generation has been a hot topic nowadays in most countries in the world. To tackle this problem, the share of renewable energy use has been increased by up to 14% in the Netherlands. Moreover, the number of electric vehicles (EVs) on the road also reaches a total of 120,000 EVs in 2018. However, the high penetration of renewable energy sources (RESs) such as solar & wind power and the EVs charging in the distribution network could result in a severe problem. One of the solutions to avoid this problem is that switching the uncontrolled charging of EVs into a controlled charging or called as smart charging. Further, an integration between the EVs, RESs and the distribution grid could potentially lead to technical and economic benefits. The focus of this thesis is to develop an optimal power management system (PMS) between the EVs, PV system, and the distribution network. The goal of the power management system is to obtain the minimum operational cost while also considering the technical grid constraints, which subsequently could avoid the grid violation. The proposed power management system will be modeled in a mixed integer non-linear programming (MINLP) optimization problem and executed in General Algebraic Modelling System (GAMS) software. To evaluate the performance of the proposed power management system, a comparison between the with-grid and the no-grid constraints case will be performed through several case studies. This study shows that by implementing the proposed power management system of EVs charging from PV system considering the grid constraints, it could decrease the total operational cost remarkably by 18.16% - 214.08% when compared to the uncontrolled charging scheme. Besides, the grid problem caused by the uncontrolled charging process such as exceeding the allowable voltage deviation and the transformer rated power could be prevented. However, in comparison to the smart charging without considering the grid constraints, the operational cost is increased by 1.43% - 113.20%.

Electric Vehicles (EVs) provide a promising sustainable clean mode of transportation with much fewer emissions compared to the traditional Internal Combustion Engine (ICE) vehicles. However, this can only be considered true if the energy sources used for charging EVs are also from clean energy form. Various Renewable Energy Sources (RES) are available to provide the EVs charging demand. Nevertheless, due to intermittency power generation characteristic of RES, there is a high risk of supply and demand power mismatch. A grid-connected topology system is commonly used as a backup for lack or over power supply from RES generation. Even though, this configuration may lead to new problems of under/over voltage issues and maximum transformer/lines loading capacity due to RES fluctuation power production. The storage system is considered necessary to be installed in the system as a buffer of the power mismatch and also reducing the grid stress. Excellent approximation of storage sizing and placement is mandatory to provide satisfactory influence to the overall system technically and economically.

This study proposes optimization strategy for optimal battery sizing and placement with optimal power management system in the EVs charging station using photovoltaics (PV) power in a low voltage distribution network using Mixed Integer Linear Programming (MINLP) algorithm. The lithium-ion battery is considered in this project because of its relatively high power and energy density characteristic along with low self-discharging. The radial low voltage CIGRE benchmark is adopted as a model topology to evaluate the performance of the proposed optimization strategy. Dynamic daily electricity price, residential load, and PV power profile of the Netherlands are taken into account in this project. Furthermore, the research project implements several case studies using DICOPT (Discrete and Continuous OPTimizer) solver in GAMS (General Algebraic Modelling System) software with MATLAB interfacing. As the main result, proper EVs charging strategy with optimal battery sizing and placement with the appropriate power management system is successfully maintaining the system away from the grid violation while at the same time reducing the total cost of the system.

Keywords: EV-PV charging, optimal battery sizing and placement, MINLP

Airborne Wind Energy is a promising technology that is becoming the point of interest for many applications. One market in which the AWE systems can open a path as a renewable distributed generator is in offgrid communities or camps. Nowadays, this market is dominated by fossil fuel energy like diesel generators. Even though the diesel is cheap, in some cases the complexity to transport the diesel can increase its price considerably. A military camp is an example of this cases. AWE systems designed by the start-up Kitepower can offer a solution to this type of application. Nevertheless, there are some points in the AWE mechanism that should be improved. The most important of these aspects is the cyclic power production that generates a power gap.

This thesis compares solutions as 2 KP systems, other distributed generators (DG) and storage systems (SS) to find the best option to mitigate the power gap. For the integration of the components with the KP system, microgrids designs and layouts were studied. This led to one of the most important things in a microgrid that is the SS sizing. Here, also a method to size the SS to fill the power gap is proposed based on the distribution of power generated by the KP system each cycle. To test the sizing and the operating of the system, a model was built in Simulink. The results show that the method finds a power balance between the power produced and the power generated.

Finally, an efficient integrated system is proposed to mitigate the power gap of the KP system. Additionally, a method is developed to start the design and size additional DERs for future offgrid microgrid projects.

The world at large is passing through a phase of energy transformation – from a traditional centralized unidirectional to a decentralized and distributed bidirectional power system, with consumers becoming prosumers. Facilitated by the technological advancements in power electronics and growing concerns over climatic changes, this transformation lead to increasing deployment of renewable energy sources. All these converge to the evolution of a concept – Microgrid, which empowers Smart Grid of the Future.

In the thesis, the concept of Cell, a microgrid comprising of Distributed renewable energy sources like solar, wind etc., Battery energy storage systems and Controllable loads is conceived. Devoid of use of any communication equipment, the Cell relies on measurements of local variables for evaluating the State of Energy and Power. The primary focus is on the formulation of control strategy based on droop control that enables a seamless (dis)connection of Cell with External Grid, with still maintaining the operation of converter in voltage source mode. This control scheme is further extended for connection and disconnection of multiple Cells interconnected through a Backbone bus, referred to as Interconnected mode. Adding to this, a decision-making algorithm is developed for an autonomous operation of Cells, that determines the switching instants between Stand-alone and Interconnected modes.

The developed control strategy and automated decision-making algorithm was simulated in DIgSILENT PowerFactory software for varied scenarios to evaluate the performance and the overall stability of the system. The simulations show a seamless transition between different operating modes – Stand-alone, Interconnected and Grid-connected and the automated decision-making algorithm succeeded in achieving overall stability, increasing the reliability of power to end consumers. Experiments were performed on a standard converter to prove the practical implementation capability of the developed control scheme as an intermediate interface. The modular and distributed nature of the developed controls and algorithm, makes it's advantageous to apply such Cells to electrify remote areas, which have limited or no access to power. The system can be easily scaled and expanded by adding more Cells, to build a strong and robust microgrid network.

Continuous growth in electricity demand, increasing awareness for global warming and other en- vironmental issues have triggered attention to a new type of electricity generation-the so-called distributed generation consisting of various renewable energy sources. Despite the technological, economic and environmental benefits that the new generation system brings about, intermittent and uncontrollable nature of the renewable energy sources creates new type of challenges. The emergence of microgrids, prompted by such background, can offer a platform for effective integra- tion of distributed generation systems.
In this thesis, two microgrids network interconnected through a backbone bus is developed. The main feature of this system is the absence of ICT infrastructure. Instead, the so-called droop con- trol is used to regulate power outputs among the interconnected microgrids; frequency information obtained from local measurements, is the only signal used for communications.
Within this context, the thesis focuses on improvements of issues related to short-term dynamics of the droop control. To be more specific, it concerns the seamless transitions between islanded and backbone connected mode. The novel advanced controllers were developed in order to en- able these seamless transitions. They also aim to overcome the challenges found in the previous works within the same project. Three main functions exist in the controllers implemented for this thesis: synchronisation, seamless intentional(planned) and unintentional(unplanned) islanding. These developed controllers were verified and validated through both simulations and experimen- tal analysis. A significantly improved synchronisation performance was found, with the use of the newly introduced Synchronous Reference Frame-Phase Locked Loop (SRF-PLL) method. A fine tun- ing of the control parameters for voltage source inveters helped to shorten the preparation time for planned islandings. Last but not least, the introduction of the fuzzy droop controller reduced tran- sients in the microgrids during unexpected disconnections.
The software platform used for the development was LabVIEW, and GPIC FPGA was utilised as the hardware platform. The functionality of the controllers were tested under several base and extreme case studies and scenarios. The results were compared to that of the previous controllers and the agreement between simulation and experimental results were evaluated.