Some aspects of smart grid

Abstract

Modern power systems worldwide are faced up with a series of challenges transforming towards a more intelligent and resilient electricity network commonly known as smart grid. Myriads of techniques and measures have been introduced to the existing grid ranging from generation side to end-user side. Our work focuses on two aspects of smart grid namely photovoltaic (PV) energy and smart meter. Due to the increasing demand for sustainable resources, PV energy has been widely adopted in various landscapes. Partial shading can drastically alter the output power-voltage characteristics of PV systems, resulting in a modified profile with both global and local optimal power points. Many studies of global maximum power point tracking (GMPPT) are conventionally based on one or several fixed (steady-state) shading conditions. In reality, partial shading is changing dynamically and so is the location of the GMPP. In this thesis, we present a simple and yet effective method for modeling dynamic partial shading of PV systems. The shaded areas on the PV arrays can change shape and position with time. This dynamic partial shading modeling approach enables the dynamics of GMPPT controllers to be evaluated. The dynamics of this modeling method has been successfully confirmed in Matlab simulation and experimentally verified with a sliding-mode controller and a proportional-integral controller. The proposal can be used as a design tool for GMPPT controller as well as an educational tool to illustrate the dynamic changes of the output power-voltage profiles. Smart meter is increasingly important in providing essential communication between utilities and consumers. However, traditional smart meter requires professional installation which significantly encumbers the replacement process. In this thesis, we describe a non-intrusive power measurements method that is suitable for a new type of low-cost and easy-to-install smart meters in (i) measuring current and power from parallel electric cables based on non-contact magnetic flux and (ii) differentiating the power consumption of each phase. By placing non-contact magnetic flux sensors close to the parallel electric cables that carry electric currents, standard techniques obtain the current information in the cables by measuring the magnetic flux values and use cancellation algorithms to reduce the mutual coupling effects among the magnetic fields generated by these cables. Based on the use of the mains voltage of the electronic control as a reference, a new and simple method for accurately differentiating the phase currents and power in the cables and estimate the power delivered through these cables is proposed. The proposed method has been verified with practical measurements with good accuracy.