No description available.

This research work is devoted to the analysis, design and implementation of a shunt connected Hybrid Active Power Filter by the use of a lower rated voltage source PWM converter, and a series connected LC passive filter. In recent years, voltage and current harmonics have become a serious problem both in transmission and distribution systems, due to the widespread usage of non-linear loads such as diode/thyristor rectifiers, electric arc furnaces and motor drives. In order to obtain a better performance than those of the conventional passive filter solutions, active power filters (APF) have been worked on and developed. Among various configurations listed in the literature, conventional shunt connected voltage source active power filter is widely used in industrial applications. Unfortunately, for large power applications, the losses and the rating of the APF increase considerably. As a result, various hybrid filter topologies have been developed which combine the advantages of both passive and active filters. In this thesis, a shunt connected hybrid active power filter is developed by combining a 4.7 kVA voltage source converter and a 30kVAR 7th harmonic passive filter. The developed system has been implemented to eliminate the most dominant 5th, 7th and 11th current harmonic components existing at 400V low voltage bus of TUBITAK SPACE Technologies Institute. The theoretical and experimental results have shown that the DC link voltage of the converter and the rating of APF are minimized while keeping the filtering performance satisfactory.

Ce travail de thèse propose des approches d’identification et de reconnaissance des harmoniques de courant qui sont basées sur des stratégies d’apprentissage automatique. Les approches proposées s’appliquent directement dans les dispositifs d’amélioration de la qualité de l’énergie électrique.Des structures neuronales complètes, dotées de capacités d’apprentissage automatique, ont été développées pour identifier les composantes harmoniques d’un signal sinusoïdal au sens large et plus spécifiquement d’un courant alternatif perturbé par des charges non linéaires. L’identification des harmoniques a été réalisée avec des réseaux de neurones de type Multi–Layer Perceptron (MLP). Plusieurs schémas d’identification ont été développés, ils sont basés sur un réseau MLP composé de neurones linéaire ou sur plusieurs réseaux MLP avec des apprentissages spécifiques. Les harmoniques d’un signal perturbé sont identifiées avec leur amplitude et leur phase, elles peuvent servir à générer des courants de compensation pour améliorer la forme du courant électrique. D’autres approches neuronales a été développées pour reconnaître les charges. Elles consistent en des réseaux MLP ou SVM (Support Vector Machine) et fonctionnent en tant que classificateurs. Leur apprentissage permet à partir des harmoniques de courant de reconnaître le type de charge non linéaire qui génère des perturbations dans le réseau électrique. Toutes les approches d’identification et de reconnaissance des harmoniques ont été validées par des tests de simulation à l’aide des données expérimentales. Des comparaisons avec d’autres méthodes ont démontré des performances supérieures et une meilleure robustesse. / This thesis proposes identifying approaches and recognition of current harmonics that are based on machine learning strategies. The approaches are applied directly in the quality improvement devices of electric energy and in energy management solutions. Complete neural structures, equipped with automatic learning capabilities have been developed to identify the harmonic components of a sinusoidal signal at large and more specifically an AC disturbed by non–linear loads. The harmonic identification is performed with multilayer perceptron neural networks (MLP). Several identification schemes have been developed. They are based on a MLP neural network composed of linear or multiple MLP networks with specific learning. Harmonics of a disturbed signal are identified with their amplitude and phases. They can be used to generate compensation currents fed back into the network to improve the waveform of the electric current. Neural approaches were developed to distinguish and to recognize the types of harmonics and is nonlinear load types that are at the origin. They consist of MLP or SVM (Support Vector Machine) acting as classifier that learns the harmonic profile of several types of predetermined signals and representative of non–linear loads. They entry are the parameters of current harmonics of the current wave. Learning can recognize the type of nonlinear load that generates disturbances in the power network. All harmonics identification and recognition approaches have been validated by simulation tests or using experimental data. The comparisons with other methods have demonstrated superior characteristics in terms of performance and robustness.

Eine angemessene Qualität der Elektroenergie ist Grundvoraussetzung für den störungsfreien Betrieb aller angeschlossenen Geräte und Anlagen und spielt in den Verteilungsnetzen moderner Industriegesellschaften wie Deutschland eine zentrale Rolle. Die Elektroenergiequalität (EEQ) wird in Strom- und Spannungsqualität unterteilt. Während die Stromqualität maßgeblich im Verantwortungsbereich der Hersteller von Geräten und Anlagen liegt, sind für die Sicherung einer angemessenen Spannungsqualität im Wesentlichen die Netzbetreiber verantwortlich. Durch die technische Weiterentwicklung bspw. neuer Gerätetechnologien und die zunehmende Integration dezentraler Erzeugungsanlagen wie Photovoltaikanlagen ist zu erwarten, dass die EEQ auch künftig weiter an Bedeutung gewinnt. Die EEQ im Niederspannungsverteilungsnetz ist abhängig von Ort und Zeit und wird durch verschiedene Qualitätskenngrößen beschrieben. Die örtliche und zeitliche Abhängigkeit resultieren aus einer Vielzahl verschiedener Einflussfaktoren, welche sich entweder der elektrischen oder der nicht-elektrischen Umgebung des betrachteten Verteilungsnetzes zuordnen lassen. Die elektrische Umgebung wird durch die Art und Anzahl angeschlossener Verbraucher bzw. Erzeuger (Abnehmer- bzw. Erzeugerstruktur) sowie Struktur und technische Parameter des Verteilungsnetzes (Netzstruktur) bestimmt. Die nicht-elektrische Umgebung umfasst u.a. Einflüsse der klimatischen Umgebung wie bspw. Temperatur oder Globalstrahlung. Ziel dieser Arbeit ist die systematische Identifikation korrelativer Zusammenhänge zwischen den genannten Umgebungseinflüssen und der EEQ sowie deren Quantifizierung auf Basis geeigneter Indizes und Kenngrößen. Die Ergebnisse der Arbeit helfen grundlegende Prinzipien der Ausprägung der Elektroenergiequalität im öffentlichen Verteilungsnetz besser zu verstehen sowie die Verteilungsnetze im Hinblick auf die Elektroenergiequalität zu charakterisieren und zu klassifizieren. Analog zu den Standard-Lastprofilen erfolgt die Definition von Standard-Qualitätsprofilen. / Power quality levels in public low voltage grids are influenced by many factors which can either be assigned to the electrical environment (connected consumers, connected genera-tion, network characteristics) or to the non-electrical environment (e.g. climatic conditions) at the measurement site. Type and amount of connected consumers (consumer topology) are expected to have a very high impact on power quality (PQ) levels. The generation topology is characterized by number and kind of equipment and generating installations like photovoltaic systems which are connected to the LV grid. The electrical parameters of the grid define the network topology. The parameters which are most suitable to describe each of the three topologies and the climatic environment will be identified. Voltage and current quality in public low voltage (LV) grids vary depending on location and time. They are quantified by a set of different parameters which either belong to events (e.g. dips) or to variations (e.g. harmonics). This thesis exclusively addresses continuous parameters describing variations. Continuous phenomena like harmonics are closely linked to an one-day-cycle which implies a more or less periodic behavior of the continuous power quality parameters. Consumer topologies such as office buildings or residential areas differ in their use of equipment. Time series analysis is used to distinguish between different consumer topologies and to identify characteristic weeks. The clustering of one-day time series is applied to identify characteristic days within the weeks of certain topologies. Based on the results, emission profiles for certain current quality parameters of different consumer topologies will be defined. Due to the characteristic harmonic current emission of certain consumer topologies which represents the typical user behaviour a classification system is developed. It is used to automatically classify the emission profiles of harmonic currents for unknown measurements and to estimate a likely consumer topology. A classification measure is introduced in order to identify unusual or false classified emission profiles. The usage behaviour of equipment by customers usually varies over the year. Subsequently, the levels of PQ parameters like harmonics may show seasonal variations which are identified by using newly defined parameters. The introduction of new device technologies on a large scale like the transition from incandescent to LED lamps might result in long-term changes to the levels of PQ parameters (e.g. harmonics). The analysis of the long-term behavior (trend) will be applied in order to quantify global trends (looking on the measurement duration as a whole) and local trends (looking on individual segments of the whole time series).