The objective of this research is to introduce a neural network based solution for the problem of measuring the actual amount of harmonic current injected into a power network by an individual nonlinear load. Harmonic currents from nonlinear loads propagate through the system and cause harmonic pollution. As a result, voltage at the point of common coupling (PCC) is rarely sinusoidal. The IEEE 519 harmonic standard provides customer and utility harmonic limits and many utilities are now requiring their customers to comply with IEEE 519. Measurements of the customer’s current at the PCC are expected to determine the customer’s compliance with IEEE 519. However, results in this research show that the current measurements at the PCC are not always reliable in that determination. In such a case, it may be necessary to determine what the customer’s true current harmonic distortions would be if the PCC voltage could be a pure sinusoidal voltage. However, establishing a pure sinusoidal voltage at the PCC may not be feasible since that would mean performing utility switching to reduce the system impedance. An alternative approach is to use a neural network that is able to learn the customer’s load admittance. Then, it is possible to predict the customer’s true current harmonic distortions based on mathematically applying a pure sinusoidal voltage to the learned load admittance. The proposed method is called load modeling. Load modeling predicts the true harmonic current that can be attributed to a customer regardless of whether a resonant condition exists on the utility power system. If a corrective action is taken by the customer, another important parameter of interest is the change in the voltage distortion level at the PCC due to the corrective action of the customer. This issue is also addressed by using the dual of the load modeling method. Topologies of the neural networks used in this research include multilayer perceptron neural networks and recurrent neural networks. The theory and implementation of a new neural network topology known as an Echo State Networks is also introduced. The proposed methods are verified on a number of different power electronic test circuits as well as field data. The main advantages of the proposed methods are that only waveforms of voltages and currents are required for their operation and they are applicable to both single and three phase systems. The proposed methods can be integrated into any existing power quality instrument or can be fabricated into a commercial standalone instrument that could be installed in substations of large customer loads, or used as a hand-held clip on instrument.

Στα σύγχρονα συστήματα ηλεκτρικής ενέργειας, η άεργος ισχύς που παράγουν τα αιολικά πάρκα είναι ένα πολύ σοβαρό ζήτημα που πρέπει να μελετηθεί τόσο στη μόνιμη κατάσταση όσο και στη δυναμική και μεταβατική μετά από σφάλματα και μεγάλες διαταραχές. Στην παρούσα διπλωματική εργασία εκπονείται μελέτη που αφορά την κατανεμημένη παραγωγή και ιδιαίτερο βάρος δίνεται στην ανεμογεννήτρια μεταβλητών στροφών με επαγωγική μηχανή διπλής τροφοδοσίας (DFIG) και τα όρια που παρουσιάζει η λειτουργία της από περιοριστικούς παράγοντες όπως: 1) ρεύμα του στάτη, 2) τάση του δρομέα, 3) ρεύμα του δρομέα. Αρχικά μελετάται η γενική δομή ενός συστήματος ηλεκτρικής ενέργειας καθώς και τα βασικά χαρακτηριστικά του δικτύου. Περιγράφεται επίσης σύντομα η κανονική λειτουργία και η ευστάθεια των συστημάτων ηλεκτρικής ενέργειας καθώς και πως μπορεί να γίνει ρύθμιση της τάσης και της αέργου ισχύος δύο μεγεθών με ισχυρή συσχέτιση. Στην συνέχεια προχωράμε στην περιγραφή της διεσπαρμένης παραγωγής και την διείσδυσή της στο δίκτυο. Παρουσιάζονται τα πλεονεκτήματα καθώς και τα μειονεκτήματά της όπου αυτά υπάρχουν και περιγράφονται κάποια πολύ βασικά είδη της δίνοντας έμφαση στα είδη των ανεμογεννητριών. Στο κύριο μέρος της εργασίας περιγράφεται αρχικά το μοντέλο της ανεμογεννήτριας μεταβλητών στροφών με επαγωγική μηχανή διπλής τροφοδοσίας (DFIG) και παρουσιάζεται η θεωρητική επίδραση των παραπάνω περιοριστικών παραγόντων στην λειτουργία της. Στόχος της μελέτης είναι η προσομοίωση των παράπάνω σε μια πραγματική μηχανή κάτι που γίνεται στο Simulink του Matlab. Α) παρουσίαζονται τα όρια από τους τρείς περιοριστικούς παράγοντες. Β) περιγράφεται η επίδραση του μετασχηματισμού από αστέρα σε τρίγωνο στα τυλίγματα του στάτη. Γ) Σχεδιάζονται όλα τα όρια μαζί και παρατηρούμε ότι το όριο που κυριαρχεί σχετικά με την παραγωγή αέργου ισχύος είναι το ρεύμα του δρομέα ενώ το όριο που κυριαρχεί σχετικά με την απορρόφηση αέργου ισχύος είναι το ρεύμα του στάτη. Δ) Τέλος παρατηρούμε ότι η τάση του δρομέα επιδρά σαν όριο μόνο σε μεγάλες τιμές της ολίσθησης (είτε αρνητικής είτε θετικής ), αλλά κοντά στα όριά της είναι πολύ ευαίσθητη στις αλλαγές της τιμής της ολίσθησης. / In the modern regulations of transport of energy, reactive power by wind farms is a major concern that should be studied during both steady-state and dynamic-fault condition. This diploma thesis forms the distribution systems and particular weight is given in the wind turbine of variable speed with doubly fed induction generator (DFIG) and the operation limits from these factors: 1) stator current, 2) rotor voltage, 3) rotor current . Initially is studied the general structure of system of electric energy and the basic characteristics of the grid. Also are shortly described the regular operation and the stability of systems of electric and the regulation of voltage and reactive power- two sizes with powerful cross-correlation . Then we are going on with the description of distribution systems and the impact that wind turbines have in the grid where were connected. The advantages are presented as well as the disadvantages where these exist and we describe some of the very basic types of distributed generation giving accent in the types of wind generators. In the main part of work is described initially the model of wind turbine of variable speed with doubly fed induction generator (DFIG) and is presented the effect of the restrictive factors in the operation. Objective of study is the simulation in a real machine something that becomes in the Simulink of Matlab. A) The limitations in reactive power production, caused by the rotor current, the rotor voltage and the stator current are derived. B) The influence of switching from D to Y coupling of the stator is investigated. C) It is concluded that the limiting factor regarding reactive power production will typically be the rotor current limit, and that the limit for reactive power absorption will be the stator current limit. D) Further, it is concluded that the rotor voltage will only have a limiting effect at high positive and negative slips, but near the limitation, the reactive power capability is very sensitive to small changes in the slip.

Η παρούσα διπλωματική εργασία καταπιάνεται με τα προβλήματα Ποιότητας Ισχύος των Συστημά- των Ηλεκτρικής Ενέργειας και ειδικότερα με τη διαταραχή της Αρμονικής Παραμόρφωσης. Συγκεκρι- μένα με τη διείσδυση των Ανανεώσιμων Πηγών Ενέργειας και την τάση για Κατανεμημένη Παραγωγή Ενέργειας αυξήθηκε το ενδιαφέρον για την Ποιότητα Ισχύος των Συστημάτων Ηλεκτρικής Ενέργειας. Οι βασικές εξελίξεις στο πεδίο της ενέργειας και ο εκσυγχρονισμός του εξοπλισμού κυρίαρχα με τη χρήση των ηλεκτρονικών ισχύος αποτέλεσαν αιτία για έγχυση ανώτερων αρμονικών στο ηλεκτρικό δίκτυο. Επιδίωξη της διπλωματικής εργασίας είναι να εισάγει τον αναγνώστη στις βασικές έννοιες των Αρμονικών και της Ποιότητας Ισχύος, καθώς και να παρουσιάσει τη μελέτη έγχυσης αρμονικών μιας κατανεμημένης παραγωγής, συγκεκριμένα ενός φωτοβολταϊκού συστήματος, όταν αυτή συνδέεται με το δίκτυο χαμηλής τάσης. Αναλυτικότερα, αναπτύσσεται η έννοια της Ποιότητας Ισχύος και κατηγοριο- ποιούνται τα προβλήματα Ποιότητας Ισχύος που εμφανίζονται σε ένα ΣΗΕ. Περιγράφεται λεπτομερώς το πρόβλημα της Αρμονικής Παραμόρφωσης της τάσης και του ρεύματος λόγω της μη γραμμικής συ- μπεριφοράς των φορτίων, παρουσιάζεται η θεωρία ισχύος υπό την παρουσία ανώτερων αρμονικών, καθώς επίσης παρατίθεται η ανάλυση Fourier ως βοηθητικό και χρήσιμο εργαλείο για την ανάλυση του φασματικού περιεχομένου ενός σήματος τάσης/ρεύματος. Στη συνέχεια αναλύεται ο όρος της Κατανε- μημένης Παραγωγής και παρουσιάζονται οι βασικές ανανεώσιμες και μη-συμβατικές πηγές ενέργειας που χρησιμοποιούνται σε συστήματα Κατανεμημένης Παραγωγής Ενέργειας, υπό το πρίσμα στατιστι- κών στοιχείων για την παγκόσμια ενεργειακή κατάσταση. Για την εξυπηρέτηση των στόχων της διπλωματικής εργασίας επιλέχθηκε η προσομοίωση μιας Κα- τανεμημένης Παραγωγής, εκείνης ενός φωτοβολταϊκού συστήματος 5.5kW συνδεδεμένου στο δίκτυο χαμηλής τάσης, μέσω ενός μετατροπέα υποβιβασμού τάσης και ενός τριφασικού αντιστροφέα. Η μο- ντελοποίηση και προσομοίωση του συστήματος έγινε με τη χρήση του πλέον διαδεδομένου προγραμ- ματιστικού περιβάλλοντος MATLAB/Simulink. Τελικά, παρουσιάζεται μια σειρά αποτελεσμάτων που αφορούσαν στην απόκριση του συστήματος σε απότομες διαταραχές, στη μελέτη σωστής λειτουργίας του Maximum Power Point Tracking ελεγκτή και στη μελέτη έγχυσης αρμονικών του φωτοβολταϊκού συ- στήματος στο δίκτυο με χρήση φίλτρου και χωρίς χρήση φίλτρου. Το εργαλείο του φασματογραφήματος αποδείχτηκε χρήσιμο για την κατανόηση των βασικών χαρακτηριστικών των σημάτων της τάσης και του ρεύματος και του αρμονικού τους περιεχομένου στην έξοδο του αντιστροφέα. Στο Κεφάλαιο 1 δίνεται ο ορισμός της Ποιότητας Ισχύος, καταγράφονται τα Διεθνή Πρότυπα για την Ποιότητα Ισχύος, καθώς επίσης κατηγοριοποιούνται τα διάφορα προβλήματα Ποιότητας Ισχύος βάσει του χαρακτηριστικού της τάσης που επηρεάζουν. Στο Κεφάλαιο 2 ερμηνεύεται το φαινόμενο της παρουσίας αρμονικών τάσεων και ρευμάτων στο δίκτυο. Παρατίθενται παράλληλα βασικές έννοιες της ανάλυσης Fourier, η οποία αποτελεί εργαλείο ανάλυσης σημάτων και συστημάτων στο πεδίο της συχνότητας. Καταγράφονται επίσης οι πηγές που παράγουν αρμονικά ρεύματα και τάσεις, ενώ παράλληλα αναπτύσσεται η θεωρία ισχύος υπό μη ημιτο- νοειδείς συνθήκες. Το κεφάλαιο κλείνει με την παράθεση των βασικών δεικτών που περιγράφουν την Αρμονική Παραμόρφωση. Στο Κεφάλαιο 3 παρουσιάζεται στατιστική μελέτη που αφορά στην παγκόσμια ενεργειακή κατά- σταση, πιο ειδικά στην παγκόσμια παραγωγή και κατανάλωση ηλεκτρικής ενέργειας και το μερίδιο των Ανανεώσιμων Πηγών Ενέργειας επί αυτών. Εισάγεται η έννοια της Κατανεμημένης Παραγωγής και αναπτύσσονται οι επιπτώσεις της διείσδυσής της στο δίκτυο. Επιπλέον, αναφέρονται οι βασικές τεχνολογίες Κατανεμημένης Παραγωγής. Στο Κεφάλαιο 4 αναπτύσσεται η προσομοίωση ενός Φωτοβολταϊκού Συστήματος συνδεδεμένου στο δίκτυο χαμηλής τάσης με τη χρήση του λογισμικού MATLAB/Simulink. Αναλύονται διεξοδικά τα επιμέρους υποσυστήματα, δηλαδή η φωτοβολταϊκή γεννήτρια, ο DC-DC μετατροπέας, ο ανιχνευτής σημείου μέγιστη ισχύος-MPPT, αντιστροφέας, το φίλτρο αποκοπής ανώτερων αρμονικών, το δίκτυο χαμηλής τάσης και παρουσιάζονται τα μοντέλα τους και οι παράμετροί τους. Τέλος, στο Κεφάλαιο 5 μετά από μια παρουσίαση του ολικού μοντέλου προσομοίωσης, παρουσιάζο- νται οι διάφορες μελέτες προσομοίωσης, ήτοι μελέτη απόκρισης της φωτοβολταϊκής γεννήτριας, μελέτη προσομοίωσης του συστήματος και ανάλυση αρμονικών για σταθερή και μεταβαλλόμενη ένταση της ακτινοβολίας, μελέτη μεταβολής φορτίου και μελέτη διακοπής. / This Diploma Thesis addresses the problem of Power Quality in Power Systems and particularly the disturbing effect of harmonic distortion. The field of power quality analysis gained further attention due to the extended spread of renewable power sources and distributed generation. The technological breakthroughs in the field of power electronics enabled the efficient utilization of distributed power systems but at the same time increased the undesired infusion of higher harmonics to the power network. The main goal of this thesis is to provide a concrete introduction to the concepts and methodological tools of harmonics and power quality analysis as well as to present a simulation-based study of harmonic infusion of a distributed generation example. In particular, the case of a Photovoltaic array connected to the low voltage grid is considered. Initially, the concept of power quality is presented and the power quality problems that arise in power systems are categorized. The effect of harmonic distortion is described in detail and the analytical methods of power system analysis subject to the disturbing effects of higher harmonics are presented. The basic theory as well as the tools of frequency domain Fourier-based analysis of the voltage and current harmonics are described in detail, followed by an overview of the concept of distributed power generation and the dominal renewable and unconventional power sources found in such systems as concluded from statistical analysis of worldwide power generation. To enable numerical analysis, a simulation model that captured the behaviour and dynamics of a 5.5kW Photovoltaic array connected to the low voltage power grid through a buck converter cascaded with a 3-phase inverter, was developed. The MATLAB/Simulink(R) environment was chosen to perform the simulation studies due to its extended spread within the engineering community. Based on this model, several results were derived that present the functionality of the photovoltaic array itself, the role of the Maximum Power Point Tracking controller and the response of the overall system to input and external disturbances. Extended studies of the harmonic context of the voltage/current signals both in the case of using an LC filter a well as without were conducted employing frequency-domain signal processing tools. Apart from the well established method of Fast Fourier Transform calculations, the intuitive method of spectrograms was also developed and utilized for the analysis. This thesis is structured as follows: In Chapter 1, the definition of Power Quality is given, followed by a literature review of the international standards and regulations regarding power quality analysis. Moreover, categorization of the several power quality problems according to the voltage properties is provided. Chapter 2 deals with the presence of higher harmonics in the voltage and current signals of the power grid. An overview of the Fourier Analysis is provided and the specific tools used to aid in frequency-domain power systems analysis are further elaborated. The sources of voltage and current higher harmonics are listed and the theory of power systems in non perfect sinusoidal conditions is derived. This chapter ends with a brief overview of metrics describing the level of harmonic distortion. In Chapter 3, statistical analysis on the worldwide power systems status is presented with a special focus on the role and share of renewable power sources. The concept of distributed power generation is introduced followed by its undesired effects on the public power grid. The dominant technologies for distributed generation are also detailed. Within Chapter 4, the development of a MATLAB/Simulink simulator of a Photovoltaic Array connected to the low voltage power grid is presented. The simulator is based on modules that provide the functionality of the corresponding physical systems including the photovoltaic array itself, the DC-DC converter, the Maximum Power Point Tracking (MPPT) controller, the inverter, the filter responsible for cutting-off higher harmonics and the low voltage network. Each module, its simulator model and parameters as well as the interconnections that form the overall simulator are described in detail. Finally, in Chapter 5 the simulation based studies are presented. Particularly, the conducted studies include analysis of the photovoltaic array response, study of the efficiency of the MPPT controller and analysis of the 3-phase voltage and current signals with a special focus on examining their harmonic context. The conducted simulations contain cases under constant as well as varying irradiance, load variation and short-circuit faults.

O presente trabalho apresenta uma comparação de duas técnicas para a análise tempo - frequência em análise de qualidade de energia elétrica para sinais de tensão que contenham distúrbios individuais ou simultâneos. Dessa forma, o objetivo, desta dissertação, é encontrar uma ferramenta que forneça as características e parâmetros para a localização, identificação e classificação de tais distúrbios. O estudo consiste na análise do desempenho da Transformada Wavelet Discreta e da Transformada-S, principalmente, quando os sinais são analisados na presença de múltiplos distúrbios. Ambas as transformadas fornecem informação importante nos domínios do tempo e da frequência. No entanto, essas ferramentas não tem sido amplamente exploradas para análise de múltiplos distúrbios. Neste contexto, ambas as transformadas são testadas para conhecer seus desempenhos e suas capacidades de identificação e localização de eventos de qualidade de energia elétrica. Para finalizar, é projetado um sistema classificador baseado em arvore de decisão capaz de reconhecer quinze tipos de distúrbios diferentes. / This work presents a comparison of two methods for time-frequency analysis applied in Power Quality signals containing single or multiple disturbances. In this way, the aim of this work is to apply tools that supply the parameters and characteristics to identify, locate and classify Power Quality disturbances. For that, the proposed method analyzes the performance of the Wavelet and S transforms, mainly when the signals are with more than one disturbance type. Both mathematical tools supply important information on the time and frequency domain. However, these tools have not been thoroughly used to analyze multiple events locate Power Quality events. In this contest, both transforms are tested in order to assess their performance to identify and locate electrical power quality events. According to a decision tree classifier, fifteen types of single and combined power disturbances are well recognized.

Este trabalho trata de conceitos relacionados à qualidade da Energia Elétrica (EE) e, neste contexto, apresenta a proposta de técnicas para a compressão da representação de transitórios rápidos e da análise tempo-frequência de distúrbios elétricos em geral. A qualidade da Energia Elétrica é medida pelo coeficiente de desvios que os sinais de tensão e corrente apresentam em relação ao sinal senoidal ideal. Tais desvios são denominados de distúrbios, podendo ser classificados como quase estacionários (e.g. distorção de harmônicas) e eventos (e.g. transitórios rápidos). No contexto de EE, os transitórios rápidos possuem pequena duração (i.e. na ordem dos microssegundos), são detectados por altas taxas de amostragem (i.e. na ordem dos MHz) e possuem difícil parametrização. Portanto, as representações das formas de onda geralmente são armazenadas para auxiliar a avaliação subjetiva dos transitórios e dos parâmetros de interesse. Consequentemente, a compressão destas formas de onda torna-se de extrema importância para armazenar dados adquiridos por longos períodos de tempo, e estes modos de compressão são tratados nesta dissertação. Em virtude das altas taxas de amostragem utilizadas, uma técnica baseada em Análise de Componentes Principais (PCA – Principal Component Analysis) é proposta para esta representação mais compacta de transitórios. Resultados mostram que o desempenho em compressão versus qualidade de reconstrução é semelhante ao de trabalhos relacionados com a vantagem de atender aos requisitos de altas taxas de amostragem. A análise tempo-frequência é um mecanismo que auxilia na classificação e caracterização dos distúrbios elétricos. Neste trabalho, a Transformada de Hilbert-Huang é estudada e uma proposta de melhoria na Decomposição Empírica de Modos (EMD – Empirical Mode Decomposition) é apresentada. Nossos resultados mostram que a técnica proposta economiza o custo computacional se comparada com o estado da arte. Em virtude disso, a técnica proposta apresenta uma taxa de redução no tempo médio de execução de 99,76 % em relação à técnica do estado da arte. Além disso, uma verificação acerca do desempenho em eficiência de compressão versus qualidade de reconstrução de trabalhos anteriores é também desenvolvida nesta dissertação. Foi utilizada uma sistemática de avaliação experimental com base em amostras de sinais AC, de forma a avaliar as taxas de compressão atingidas pelas técnicas estudadas, como a Transformada Wavelet Discreta. Resultados mostram que a Transformada Wavelet falha para compressão de todo e qualquer tipo de distúrbio elétrico quando analisado o compromisso entre acuidade de reconstrução versus eficiência de compressão. / This work deals with concepts related to the AC Power Quality theoretical framework and, in this scope, proposes techniques for the representation of fast transient data compression and for the power line disturbances time-frequency analysis. The AC power quality is measured by the differences between actual and ideal sinusoidal voltage/current signals. These differences are known as electrical disturbances, which can be classified as quasi-stationary (e.g. harmonic distortion) or events (e.g. surge or fast transients) disturbances. In the AC Power Quality scope, the fast transients have short duration (i.e. typically on the order of microseconds), are detected by high sampling rates (i.e. typically on the order of MHz), and are hard to characterize and parameterize. Hence, the resultant representation of the waveforms is in general stored to help in the subjective evaluation of these fast transients and their parameters of interest. As a consequence the compression turns out to be of main concern, in order to store this information acquired over long periods of time (like weeks or months). In this work, a compression technique is proposed taking into account the high sampling rates. The proposed technique makes use of the Principal Component Analysis (PCA) for such compact representation of fast transients. The Compression efficiency versus reconstruction accuracy results show a similar performance for the proposed technique when compared to the related works. On the other hand, the proposed technique can handle the large amount of data provided by the high sampling rates. The time-frequency analysis helps in the classification and characterization of AC power quality disturbances. In this work, the Hilbert-Huang Transform is studied and a modification is proposed in order to improve the Empirical Mode Decomposition (EMD) performance. Our results show that the proposed modification can save computational cost when compared to the state-of-the-art. Therefore, the average execution time is reduced to 99.76 % in comparison with the state-of-the-art technique. Besides that, this work also revisits previous techniques based on the Discrete Wavelet Transform (DWT) in order to verify the trade-off between reconstruction accuracy versus compression efficiency under a more systematic experimental evaluation setup, considering samples of real AC signals. Results show that DWT fails as a general-purpose technique in AC Power Quality scope.

Este trabalho apresenta uma metodologia para análise e monitoração da qualidade da energia elétrica, através da identificação e quantificação dos distúrbios eletromagnéticos. A metodologia utiliza técnicas de processamento digital de sinais, possibilitando a construção de filtros digitais, a detecção de eventos e a estimativa da freqüência dos sinais elétricos analisados. Os principais distúrbios da qualidade definidos pelas normas da ANEEL e do ONS são quantificados através do algoritmo proposto. O programa desenvolvido foi testado usando formas de ondas com distúrbios previamente conhecidos para sua validação. Analisando formas de onda obtidas de medições em campo verificou-se a robustez do algoritmo frente a ruídos e outros fenômenos vinculados à qualidade da energia presentes em medições reais. Esta pesquisa apresenta um programa eficaz e prático que pode ser utilizado no desenvolvimento de um novo equipamento de medição dos distúrbios da QEE. Os resultados obtidos através da análise de dados (sintetizados e medições de campo) validaram o programa proposto. / This dissertation presents a methodology for detection and quantification of power quality disturbances. Digital signal processing (DSP) is applied to simulated and measured disturbances. The use of DSP enables the construction of digital filters for the detection of events, and the estimation of the frequency of voltage and current signals. The disturbances set by ANEEL and ONS standards are quantified by the proposed algorithm. The program developed was tested using simulated disturbance for its validation. Field measurements were used to assess the robustness of the algorithm against signal noise and other disturbances. This research presents effective and practical software that can be used to develop a new device for measuring the disturbances of Power Quality. The results obtained through the analysis of data (simulated and field measurements) validated the proposed algorithm.

Este trabalho apresenta um modelo para alocação e dimensionamento ótimo da geração distribuída em sistemas elétricos de potência. Com o objetivo de minimizar o custo da concessionária devido as perdas ativas e desvios de tensão em regime permanente, ambos foram transformados em valores monetários utilizando a normativa existente no Brasil. O problema de otimização considera uma curva de carga de 24 níveis com o intuito de simular uma curva de carga diária em intervalos de uma hora. Foram considerados limites de tensão de cada barra e corrente máxima em cada trecho do alimentador. O fluxo de potência foi estimado através do algoritmo clássico de Newton Raphson. A alocação da geração distribuída, a qual é considerada em mais de um local do alimentador, é tratada como uma variável binária no modelo desenvolvido. O modelo de otimização não-linear inteira mista é escrita mediante um algoritmo em Matlab na linguagem GAMS e enviado para o servidor de otimização NEOS e solucionado pelo solver KNITRO. O resultado é a obtenção do valor dos custos evitados para a concessionária facilitando o diagnóstico para a tomada de decisões. Por fim para verificação do método, realiza-se um estudo de caso em um sistema de distribuição de 33 barras da IEEE, sendo os resultados analisados e discutidos. / This work presents a model for optimal distributed generation sizing and allocation in power systems. The main objectives are to reduce costs for the power distribution company by power losses and compensation for voltage levels violation, both are converted in monetary values based on the Brazilian normative. The optimization problem considers a load curve with 24 levels to simulate one day in intervals of one hour. Also the model considers voltage limits for each bus and maximum currents for every line in the feeder. The power flow was formulated by the classical Newton Raphson theory. The distribution generation allocation is modeled as binary variables and can be allocated in more than one bus in the feeder. The mixed integer nonlinear model is written by a Matlab algorithm in GAMS language and solved by KNITRO through NEOS solver for optimization. The model was tested using the IEEE 33 buses, and the results were evaluated and discussed. The model lets power distribution companies reduce operational cost and penalties with optimal placement and sizing of distributed generation.

Este trabalho trata de conceitos relacionados à qualidade da Energia Elétrica (EE) e, neste contexto, apresenta a proposta de técnicas para a compressão da representação de transitórios rápidos e da análise tempo-frequência de distúrbios elétricos em geral. A qualidade da Energia Elétrica é medida pelo coeficiente de desvios que os sinais de tensão e corrente apresentam em relação ao sinal senoidal ideal. Tais desvios são denominados de distúrbios, podendo ser classificados como quase estacionários (e.g. distorção de harmônicas) e eventos (e.g. transitórios rápidos). No contexto de EE, os transitórios rápidos possuem pequena duração (i.e. na ordem dos microssegundos), são detectados por altas taxas de amostragem (i.e. na ordem dos MHz) e possuem difícil parametrização. Portanto, as representações das formas de onda geralmente são armazenadas para auxiliar a avaliação subjetiva dos transitórios e dos parâmetros de interesse. Consequentemente, a compressão destas formas de onda torna-se de extrema importância para armazenar dados adquiridos por longos períodos de tempo, e estes modos de compressão são tratados nesta dissertação. Em virtude das altas taxas de amostragem utilizadas, uma técnica baseada em Análise de Componentes Principais (PCA – Principal Component Analysis) é proposta para esta representação mais compacta de transitórios. Resultados mostram que o desempenho em compressão versus qualidade de reconstrução é semelhante ao de trabalhos relacionados com a vantagem de atender aos requisitos de altas taxas de amostragem. A análise tempo-frequência é um mecanismo que auxilia na classificação e caracterização dos distúrbios elétricos. Neste trabalho, a Transformada de Hilbert-Huang é estudada e uma proposta de melhoria na Decomposição Empírica de Modos (EMD – Empirical Mode Decomposition) é apresentada. Nossos resultados mostram que a técnica proposta economiza o custo computacional se comparada com o estado da arte. Em virtude disso, a técnica proposta apresenta uma taxa de redução no tempo médio de execução de 99,76 % em relação à técnica do estado da arte. Além disso, uma verificação acerca do desempenho em eficiência de compressão versus qualidade de reconstrução de trabalhos anteriores é também desenvolvida nesta dissertação. Foi utilizada uma sistemática de avaliação experimental com base em amostras de sinais AC, de forma a avaliar as taxas de compressão atingidas pelas técnicas estudadas, como a Transformada Wavelet Discreta. Resultados mostram que a Transformada Wavelet falha para compressão de todo e qualquer tipo de distúrbio elétrico quando analisado o compromisso entre acuidade de reconstrução versus eficiência de compressão. / This work deals with concepts related to the AC Power Quality theoretical framework and, in this scope, proposes techniques for the representation of fast transient data compression and for the power line disturbances time-frequency analysis. The AC power quality is measured by the differences between actual and ideal sinusoidal voltage/current signals. These differences are known as electrical disturbances, which can be classified as quasi-stationary (e.g. harmonic distortion) or events (e.g. surge or fast transients) disturbances. In the AC Power Quality scope, the fast transients have short duration (i.e. typically on the order of microseconds), are detected by high sampling rates (i.e. typically on the order of MHz), and are hard to characterize and parameterize. Hence, the resultant representation of the waveforms is in general stored to help in the subjective evaluation of these fast transients and their parameters of interest. As a consequence the compression turns out to be of main concern, in order to store this information acquired over long periods of time (like weeks or months). In this work, a compression technique is proposed taking into account the high sampling rates. The proposed technique makes use of the Principal Component Analysis (PCA) for such compact representation of fast transients. The Compression efficiency versus reconstruction accuracy results show a similar performance for the proposed technique when compared to the related works. On the other hand, the proposed technique can handle the large amount of data provided by the high sampling rates. The time-frequency analysis helps in the classification and characterization of AC power quality disturbances. In this work, the Hilbert-Huang Transform is studied and a modification is proposed in order to improve the Empirical Mode Decomposition (EMD) performance. Our results show that the proposed modification can save computational cost when compared to the state-of-the-art. Therefore, the average execution time is reduced to 99.76 % in comparison with the state-of-the-art technique. Besides that, this work also revisits previous techniques based on the Discrete Wavelet Transform (DWT) in order to verify the trade-off between reconstruction accuracy versus compression efficiency under a more systematic experimental evaluation setup, considering samples of real AC signals. Results show that DWT fails as a general-purpose technique in AC Power Quality scope.

A Geração Distribuída no Brasil tem crescido ao longo dos anos, especialmente na fonte fotovoltaica e na categoria que a ANEEL denomina microgeração distribuída, que são centrais geradoras de energia elétrica com potência instalada menor ou igual a 75 kW e conectada na rede de distribuição por meio de instalações de unidades consumidoras. Apesar dos problemas tradicionais que podem ocorrer com a inserção em larga escala, como maiores perdas e desvios de tensão, as placas fotovoltaicas, conectadas à rede de distribuição por meio de um inversor, trazem a possibilidade de controlar a tensão no ponto de conexão, manipulando-se o fator de potência de saída do inversor, controlando a potência reativa. Neste contexto, considerando o uso dos inversores inteligentes com possibilidade de envio de sinal remotamente ao operador e a possibilidade de criação de Centros de Despacho de Geração Distribuída (CDGD), este trabalho tem por objetivo propor uma metodologia de regulação de tensão centralizada utilizando unidades de microgeração fotovoltaicas, em redes de distribuição com larga inserção. A metodologia proposta utiliza o conceito de despacho de potência reativa e fluxo de potência reativa ótimo, por meio de uma abordagem centralizada da rede, utilizando-se o software OpenDSS e MatLab, equipado com Algoritmo Genético (AG) para solucionar o problema de otimização. Para avaliar o desempenho do modelo, foram feitos estudos de caso no sistema IEEE 123 barras e em uma rede real em Alegrete. Os resultados obtidos validam a abordagem proposta, demonstra a aplicabilidade e limitações em casos reais. / The distributed generation in Brazil has increased in the past years, specially in photovoltaic power generation, at the category that Brazilian Electric Regulator ANEEL classifies as "distributed micro-generation", i.e., generation units in which the active power capacity is equal or bellow 75kW and conected on distribution network by the consumers. Despite the traditional problems that might occur due to high penetration of the units, e.g., higher losses and voltage sags and swells, they are conected to the network through an inverter that can be controlled in order to provide or consume reactive power, providing voltage regulation. In that context, considering the use of intelligent inverters that can be remotely operated by a centralized utility operator, this work aims to propose a centralized voltage regulation methodology using distributed micro-generation in distribution networks. The proposed methodology uses the concept of reactive power dispatch and optimal power flow, from a centralized view of the network, using the software MatLab and OpenDSS with Genetic Algorithm to solve the optimization problem. To evaluate the model, simulations were performed with the IEEE 123 bus system and with a real distribution network from the Alegrete city, in the Brazilian state of Rio Grande do Sul. The results show the proposed is valid, despite some limitations on real cases.

Este trabalho apresenta um estudo de Qualidade de Energia Elétrica (QEE) no sistema elétrico industrial de uma concessionária geradora de energia elétrica do Brasil. O estudo teve enfoque na distorção harmônica gerada nos serviços auxiliares da usina termoelétrica. Para tanto, foi necessária a simulação e análise da planta da termoelétrica, com atenção nos serviços auxiliares pertencentes a ela, devido aos altos níveis de distorções harmônicas registrados. A simulação foi feita com o programa ATP (Alternative Transients Program) no domínio do tempo, através de sua interface gráfica, o ATP Draw. Nele foram modelados os principais componentes que pertencem ao Sistema Elétrico de Potência (SEP) em questão, como geradores, transformadores, linhas e cargas não lineares. Com a modelagem do SEP da termoelétrica, é possível determinar os níveis da distorção harmônica em diversos locais da planta, e aprofundar o estudo para que seja possível fazer a mitigação desses distúrbios. A diminuição da distorção harmônica junto ao aumento do fator de potência permite um melhor desempenho da usina, aumentando a faturamento da empresa através do aumento da energia entregue ao sistema elétrico. / This dissertation presents a study of Power Quality in an electrical system of a thermoelectric power plant in Brazil. The study focused on the harmonic distortion generated in the auxiliary services of the plant. Therefore, it was necessary to simulate and analyze the thermal plant, with attention to auxiliary services belonging to it, due to high levels of harmonic distortion registered. The simulation was performed using the ATP (Alternative Transients Program) in the time domain through its interface, the ATP Draw. It were modeled the main components that belong to the Power System in question, such as generators, transformers, lines and non-linear loads. With the modeling of the thermoelectric plant, is possible to determine the levels of harmonic distortion at various locations within the plant. The reduction of harmonic distortion along with the increase of power factor allows a better plant performance, increasing company revenue by increasing the energy delivered to the electrical system.