Islanding Detection in Microgrids Using Harmonic Signatures

Islanding Detection in Microgrids Using Harmonic Signatures

Islanding detection in Microgrids using Harmonic Signature

In recent years, there has been a growing interest in incorporating microgrids in electrical power networks. This is due to various advantages they present, particularly the possibility of working in either autonomous mode or grid connected, which makes them highly versatile structures for incorporating intermittent generation and energy storage. However, they pose safety issues in being able to support a local island in case of utility disconnection. Thus, in the event of an unintentional island situation, they should be able to detect the loss of mains and disconnect for self-protection and safety reasons. Most of the anti-islanding schemes are implemented within control of single generation devices, such as dc-ac inverters used with solar electric systems being incompatible with the concept of microgrids due to the variety and multiplicity of sources within the microgrid. In this paper, a passive islanding detection method based on the change of the 5th harmonic voltage magnitude at the point of common coupling between grid-connected and islanded modes of operation is presented. Hardware test results from the application of this approach to a laboratory scale microgrid are shown. The experimental results demonstrate the validity of the proposed method, in meeting the requirements of IEEE 1547 standards.

Estudio y mejora de un microinterruptor e implementación de un método de detección de islanding pasivo (IDM) basado en la monitorización de armónicos

Las microrredes son agrupaciones de fuentes de generación distribuida, cargas y elementos almacenadores de energía, que pueden actuar individualmente o conectadas a una red mayor. Debido a que presentan una serie de ventajas se están implantando de forma progresiva y creciente en los sistemas eléctricos de potencia. Son sistemas totalmente autónomos y, por ello, altamente flexibles. Ante una situación anómala que pueda producirse en la red principal, ya sea un fallo eléctrico, una disrupción física en la topología o una subida desproporcionada de precio, la microrred debe ser capaz de detectar esa situación y desconectarse a fin de poder protegerse. La situación de islanding se produce cuando por cualquier circunstancia hay una pérdida de la red principal y la microrred queda operando de forma aislada. Los métodos anti-islanding, que ayudan a la detección del fenómeno en las microrredes y a la desconexión de las mismas presentan dificultades a la hora de realizar esa tarea en determinados puntos del plano P-Q. Con este trabajo fin de Máster se pretende profundizar en el conocimiento y aplicabilidad de los métodos de detección de islanding extendidos a microrredes. Se analiza la viabilidad del microinterrumptor inteligente diseñado por los investigadores de la Universidad de Wisconsin - Madison para desarrollar la función de detección y actuación ante una situación de islanding inintencionado. A continuación se realiza un estudio de armónicos en el punto de conexión entre la red principal y la microrred. Por último se utilizan los resultados de este análisis para implementar y validar un nuevo método pasivo de detección de islanding en el microinterruptor basado en la medida y monitorización del nivel de tensión de 5º armónico en el PCC.

An accurate model for simulating energetic behavior of PV grid connected inverters

This paper proposes a new model for characterizing the energetic behavior of grid connected PV inverters. The model has been obtained from a detailed study of main loss processes in small size PV inverters in the market. The main advantage of the used method is to obtain a model that comprises two antagonistic features, since both are simple, easy to compute and apply, and accurate. One of the main features of this model is how it handles the maximum power point tracking (MPPT) and the efficiency: in both parts the model uses the same approach and it is achieved by two resistive elements which simulate the losses inherent to each parameter. This makes this model easy to implement, compact and refined. The model presented here also includes other parameters, such as start threshold, standby consumption and islanding behavior. In order to validate the model, the values of all the parameters listed above have been obtained and adjusted using field measurements for several commercial inverters, and the behavior of the model applied to a particular inverter has been compared with real data under different working conditions, taken from a facility located in Madrid. The results show a good fit between the model values and the real data. As an example, the model has been implemented in PSPICE electronic simulator, and this approach has been used to teach grid-connected PV systems. The use of this model for the maintenance of working PV facilities is also shown.