Submitted by Renata Lopes (renatasil82@gmail.com) on 2017-03-02T12:24:39Z No. of bitstreams: 1 annaclaudiamancinidasilvacarneiro.pdf: 1333903 bytes, checksum: a7b3819bb5b0e1adb8efd07bca0f9aa2 (MD5) / Approved for entry into archive by Adriana Oliveira (adriana.oliveira@ufjf.edu.br) on 2017-03-06T19:35:55Z (GMT) No. of bitstreams: 1 annaclaudiamancinidasilvacarneiro.pdf: 1333903 bytes, checksum: a7b3819bb5b0e1adb8efd07bca0f9aa2 (MD5) / Made available in DSpace on 2017-03-06T19:35:55Z (GMT). No. of bitstreams: 1 annaclaudiamancinidasilvacarneiro.pdf: 1333903 bytes, checksum: a7b3819bb5b0e1adb8efd07bca0f9aa2 (MD5) Previous issue date: 2014-09-26 / CAPES - Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / A previsão de cargas elétricas é fundamental para o planejamento das empresas de energia. O foco deste estudo são as previsões a curto prazo; assim, aplicamos métodos univariados de previsão de séries temporais a uma série real de cargas elétricas de 104 semanas no Rio de Janeiro, nos anos de 1996 e 1997, e experimentamos várias combinações dos métodos de melhor desempenho. As combinações foram feitas pelo método outperformance, uma combinação linear simples, com pesos fixos. Os resultados das combinações foram comparados ao de simulações de redes neurais artificiais que solucionam o mesmo problema, e ao resultado de um método de amortecimento de dupla sazonalidade aditiva. No geral, este método de amortecimento obteve os melhores resultados, e talvez seja o mais adequado e confiável para aplicações práticas, embora necessite de melhorias para garantir a extração completa da informação contida nos dados. / Forecasting the demand for electric power is crucial for the production planning in energy utilities. The focus of this study are the short-term forecasts. We apply univariate time series methods to the forecasting of a series containing observations of the energy consumption of 104 weeks in Rio de Janeiro, in 1996 and 1997, and experiment with several combinations of the methods which have the best performance. These combinations are done by the outperformance method, a simple linear combination with fixed weights. The results were compared to those obtained by neural networks on the same problem, and with the results of a exponential smoothing method for dual additive seasonality. Overall, the exponential smoothing method achieved the best results, and was shown to be perhaps the most reliable and suitable for practical applications, even though it needs improvements to ensure complete extraction of the information contained in the data.

La problématique développée dans la thèse est d'estimer, dans une démarche prospective et dans un but d'anticipation, les impacts en puissance induits par les ruptures technologiques et comportementales qui ne font pas aujourd'hui l'objet de mesures dans les panels. Pour évaluer les modifications sur les appels de puissance du parc résidentiel engendrées par ces profondes transformations,un modèle paramétrique, bottom-up, techno-explicite et agrégatif est donc nécessaire. Celui-ci serait donc destiné à la reconstitution, de manière non tendancielle, de la courbe de charge électrique résidentielle. Il permettrait ainsi de conduire la simulation de différents scénarios d'évolution contrastés. L'élaboration d'un tel modèle constitue le sujet de ce doctorat.Pour répondre à cette problématique, nous proposons une méthode conceptuelle originale de reconstitution de courbe de charge. Sa mise en application centrée sur la génération de foisonnement d'origine comportementale a conduit à la modélisation d'un certain nombre de concepts. Ce travail a abouti à l'élaboration d'un algorithme stochastique destiné à représenter le déclenchement réalistedes appareils domestiques. Différents cas d'application ont pu être testés et les résultats en puissance ont été étudiés. Plus particulièrement pour analyser le foisonnement visible à un niveau agrégé, nous avons mis en place une méthodologie nouvelle basée sur une distance adaptée aux courbes de charge. Finalement, nous avons cherché à identifier des comportements réels d'usage des appareils. Pour cela, nous avons conduit différents travaux de classification de courbes de charge. / In this dissertation, we focus on the estimation of the impacts in terms of power demand caused by the technological and behavioural breaks that will affect the domestic sector in the future. These deep changes are not measured in the existing panels and the estimation is required for prospective (long-term) studies. To evaluate the very likely modifications of the domestic power demand that will follow previous influences, a bottom-up, technically-explicit and aggregative model is needed. This one aims at reconstituting the electric residential load curve according to a non-trending manner. Thanks to it, various evolution scenarios can be simulated. The purpose of this PhD is the elaboration of such a model.A functional analysis was carried out to build up a new method to reconstitute the domestic electric load curve. Since the clarifying of the diversity represents one of the key points of our research, we decided to begin the modelling task with focus on it. More precisely, we elaborated a stochastic algorithm whose purpose is the realistic starting of domestic appliances. Some application cases have been tested. We studied the diversity affecting aggregated power demand and we propose a new methodology able to visualise and to analyse it. This method is based on a distance adapted to the load curve. Finally we tried to identify human behaviour concerning the use of appliances thanks to load curve classifications.

It is known that Sweden is categorised by being one of the regions that experience low solar radiation because it is located in the northern hemisphere that has a low potential of solar radiation during the colder seasons. The government of Sweden aim to promote a more sustainable future by applying more renewable initiative in the energy sector. One of the initiatives is by applying more renewable energy where PV panels will play a greater role in our society and in the energy sector. However, the produced energy from the PV panels is unpredictable due to changes in radiation throughout the day. One great way to tackle this issue is by combining PV panels with different energy storage system. This thesis evaluates an off-grid rowhouse in Eskilstuna Sweden where the PV panels are combined with a heat pump, thermal storage tank, including batteries and hydrogen system. The yearly electrical demand is met by utilizing PV panels, battery system for short term usage and hydrogen system for long-term usage during the colder seasons. The yearly thermal demand is met by the thermal storage tank. The thermal storage tank is charged by heat losses from the hydrogen system and thermal energy from heat pump.The calculations were simulated in Excel and MATLAB where OPTI-CE is composed with different components in the energy system. Furthermore, the off-grid household was evaluated from an economic outlook with respect to today’s market including the potential price decrease in 2030.The results indicated that the selected household is technically practicable to produce enough energy. The PV panels produces 13 560 kWh annually where the total electrical demand reaches 6 125 kWh yearly (including required electricity for the heat pump). The annual energy demand in terms of electricity and thermal heat reaches 12 500 kWh which is covered by the simulated energy system. The overproduction is stored in the batteries and hydrogen storage for later use. The back-up diesel generator does not need to operate, indicating that energy system supplies enough energy for the off-grid household. The thermal storage tank stores enough thermal energy regarding to the thermal load and stores most of the heat during the summer when there are high heat losses due to the charge of the hydrogen system. The simulated energy system has a life cycle cost reaching approximately k$318 with a total lifetime of 25 years. A similar off-grid system has the potential to reduce the life cycle cost to k$195 if the energy system is built in 2030 with a similar lifespan. The reduction occurs due to the potential price reduction for different components utilized in the energy system.

In the last 10 years, no significant increase in the peak electricity consumption of the region of Stockholm has been observed, despite new customers being connected to the grid. But, as urbanization continues and with electrification being a decisive step of decarbonization pathways, more growth is expected in the future. However, the Swedish Transmission System Operator (TSO), Svenska Kraftnat, can only supply a limited power to Stockholm region. Distribution System Operators (DSOs) such as Vattenfall Eldistribution, which operates two thirds Stockholm region's distribution grid, need to find solutions to satisfy an increasing demand with a limited power supply. In these times, forecasting the worst-case scenarios, i.e., the highest possible loads, becomes a critical question. In Sweden, peak loads are usually triggered by the coldest temperatures, but the recent winters have been mild: this brings uncertainty about a possible underlying temperature adjusted growth that would be masked by relatively warm winters. Answering the question 'What would be the highest loads in 2022 with -20°C in Stockholm region ?' could help Vattenfall Eldistribution estimating the flexibility needed nowadays and designing the future grid with the necessary grid reinforcements. This master thesis uses a data-driven approach based on eleven years of hourly data on the period 2010-2021 to investigate the temperature sensitivity of aggregated electricity load in Stockholm region. First, an exploratory analysis aims at quantifying how large the growth has been in the past ten years and at understanding how and when peak loads occur. The insights obtained help design two innovative regression techniques that investigate the evolution of the loads across years and provide first estimates of peak loads. Then, a Seasonal Autoregressive Integrated Moving Average with eXogenous regressors (SARIMAX) process is used to model a full winter of load as a function of temperatures. This third method provides new and more reliable estimates of peak loads in 2022 at e.g. -20°C. Eventually, the SARIMAX estimates are kept and a synthesis of the global outlooks of the three methods and possible extensions of the SARIMAX method is presented in a final section. The results conclude on a significant increase in the load levels in southern Stockholm ('Stockholm Sodra') between 2010 and 2015 and stable evolution onwards, while the electric consumption in Northern Stockholm remained stable during the period 2010-2021. During a very cold winter, the electricity demand is expected to exceed the subscription levels during about 300h in Stockholm Sodra and 200h in Stockholm Norra. However, this will be a rare occurrence, which suggests that short-term solutions could be privileged rather than costly grid extension work. Many questions arise, and the capability of local heat & power production and electricity prices signals to regulate today's demand are yet to investigate. Additional work exploring future demand scenarios at a smaller scale could also be contemplated. / Under den senaste årtionden har Stockholms toppkonsumtion av el inte ökat markant trots nya elkunder som ansluter till elnätet. Med en snabb urbanisering, är ökad elektrifiering en huvudlösning för att uppnå ett fossilfritt samhälle och denna trend förväntas fortsätta under kommande årtionden. Samtidigt börjar den svenska transmissionsnätoperatören (TSO) Svenska kraftnät få problem med att leverera elkraft till Stockholmsregionen, på grund av en begränsad överföringskapacitet. Därför måste lokala eldistributörer (DSO), liksom Vattenfall Eldistribution, som är Sveriges största DSO med systemansvar för distributionssystem, undersöka nya lösningar för att uppfylla den ökande efterfrågan på el. Det blir dessutom mycket viktigt att identifiera de värsta tänkbara scenario, som att göra prognos av högsta möjliga elförbrukning. Stockholm konsumerar exempelvis mest el när det är som kallast – men de senaste vintrarna har varit milda jämfört med till exempel vintrarna 2010 – 2011 eller 2012 – 2013 då temperaturer i Stockholmsregion mättes till under -20°C grader för flera dagar i sträck. Detta resulterar i en relevant frågeställning: ” Vad skulle Stockholms elkonsumtion vid -20°C bli 2021 eller 2022?”. Att kvantitativt kunna besvara denna fråga skulle hjälpa Vattenfall med att designa framtidens elnät samt se till att det finns rätt mängd flexibilitet i reserv i nuvarande Stockholm Flex elmarknad. Detta examensarbete utgår från att kvantitativt analysera denna frågeställning. Utgångsläget är ett datadrivet tillvägagångssätt baserat på tio års tidseriedata för att undersöka temperaturkänsligheten för det aggregerade elbehovet i Stockholmsregionen, och dra slutsatser om dess utveckling genom åren. I första hand, utförs en explorativ analys för att förstå när och hur toppbelastning kan hända. Då hjälper dessa insikter till att utforma två innovativa regressionsmetoder för att undersöka utvecklingen av elförbrukning under det senaste decenniet och uppskatta värdet på toppbelastningen. Därefter används ett säsongmässigt autoregressivt integrerat rörligt genomsnitt med exogena faktorer (SARIMAX) för att modellera en vinter som en funktion av temperaturerna. Denna tredje metod behandlar nya och mer tillförlitliga beräkningar av toppbelastning värden i 2022 på -20°C. Huvudslutsatser från examensarbetet är att elförbrukningen skulle öka i området Stockholm Södra speciellt mellan 2010 och 2015, medan elförbrukningen skulle vara stabil under hela perioden i området Stockholm Norra. Det finns en risk för att under ett antal timmar vid riktigt kall vinter, ha ett elbehov högre än Vattenfall Eldistributions summa av abonnemang. Dock är det väldigt låg sannolikhet att detta händer, vilket innebär att det förmodligen finns andra sätt att hantera denna efterfråga på el än att öka överföringskapaciteten i elnätet. Examensarbetet resulterar i flera frågor. Exempelvis att utreda möjligheter i att utnyttja lokala el och värmekraftverk och använda elprissignaler. Ytterligare arbete kan också undersöka scenarier av den framtida elförbrukning i en mindre skala.

Indiana University-Purdue University Indianapolis (IUPUI) / The compressed air system is widely used throughout the industry. Air compressors are one of the most costly systems to operate in industrial plants in terms of energy consumption. Therefore, it becomes one of the primary targets when it comes to electrical energy and load management practices. Load forecasting is the first step in developing energy management systems both on the supply and user side. A comprehensive literature review has been conducted, and there was a need to study if predicting compressed air system’s load is a possibility. System’s load profile will be valuable to the industry practitioners as well as related software providers in developing better practice and tools for load management and look-ahead scheduling programs. Feed forward neural networks (FFNN) and long short-term memory (LSTM) techniques have been used to perform 15 minutes ahead prediction. Three cases of different sizes and control methods have been studied. The results proved the possibility of the forecast. In this study two control methods have been developed by using the prediction. The first control method is designed for variable speed driven air compressors. The goal was to decrease the maximum electrical load for the air compressor by using the system's full operational capabilities and the air receiver tank. This goal has been achieved by optimizing the system operation and developing a practical control method. The results can be used to decrease the maximum electrical load consumed by the system as well as assuring the sufficient air for the users during the peak compressed air demand by users. This method can also prevent backup or secondary systems from running during the peak compressed air demand which can result in more energy and demand savings. Load management plays a pivotal role and developing maximum load reduction methods by users can result in more sustainability as well as the cost reduction for developing sustainable energy production sources. The last part of this research is concentrated on reducing the energy consumed by load/unload controlled air compressors. Two novel control methods have been introduced. One method uses the prediction as input, and the other one doesn't require prediction. Both of them resulted in energy consumption reduction by increasing the off period with the same compressed air output or in other words without sacrificing the required compressed air needed for production. / 2019-12-05

No description available.

<div>The compressed air system is widely used throughout the industry. Air compressors are one of the most costly systems to operate in industrial plants in therms of energy consumption. Therefore, it becomes one of the primary target when it comes to electrical energy and load management practices. Load forecasting is the first step in developing energy management systems both on the supply and user side. A comprehensive literature review has been conducted, and there was a need to study if predicting compressed air system’s load is a possibility. </div><div><br></div><div>System’s load profile will be valuable to the industry practitioners as well as related software providers in developing better practice and tools for load management and look-ahead scheduling programs. Feed forward neural networks (FFNN) and long short-term memory (LSTM) techniques have been used to perform 15 minutes ahead prediction. Three cases of different sizes and control methods have been studied. The results proved the possibility of the forecast. In this study two control methods have been developed by using the prediction. The first control method is designed for variable speed driven air compressors. The goal was to decrease the maximum electrical load for the air compressor by using the system's full operational capabilities and the air receiver tank. This goal has been achieved by optimizing the system operation and developing a practical control method. The results can be used to decrease the maximum electrical load consumed by the system as well as assuring the sufficient air for the users during the peak compressed air demand by users. This method can also prevent backup or secondary systems from running during the peak compressed air demand which can result in more energy and demand savings. Load management plays a pivotal role and developing maximum load reduction methods by users can result in more sustainability as well as the cost reduction for developing sustainable energy production sources. The last part of this research is concentrated on reducing the energy consumed by load/unload controlled air compressors. Two novel control methods have been introduced. One method uses the prediction as input, and the other one doesn't require prediction. Both of them resulted in energy consumption reduction by increasing the off period with the same compressed air output or in other words without sacrificing the required compressed air needed for production.</div><div><br></div>

The forecasting of Eskom transmission electrical network demands is a complex task. The lack of historical data on some of the network components complicates this task even further. In this dissertation a model is suggested which will address all the requirements of the transmission system expansion engineers in terms of future loads and market trends. Suggestions are made with respect to ways of overcoming the lack of historical data, especially on the point loads, which is a key factor in modelling the electrical networks. A brief overview of the transmission electrical network layout is included to provide a better understanding of what is required from such a forecast. Lastly, some theory on multiple regression, neural networks and qualitative forecasting techniques is included, which will be of value for further model developments. / Computing / M. Sc. (Operations Research)

The forecasting of Eskom transmission electrical network demands is a complex task. The lack of historical data on some of the network components complicates this task even further. In this dissertation a model is suggested which will address all the requirements of the transmission system expansion engineers in terms of future loads and market trends. Suggestions are made with respect to ways of overcoming the lack of historical data, especially on the point loads, which is a key factor in modelling the electrical networks. A brief overview of the transmission electrical network layout is included to provide a better understanding of what is required from such a forecast. Lastly, some theory on multiple regression, neural networks and qualitative forecasting techniques is included, which will be of value for further model developments. / Computing / M. Sc. (Operations Research)