Nous abordons dans ces travaux l’identification non intrusive des charges des bâtiments résidentiels ainsi que la prédiction de leur état futur. L'originalité de ces travaux réside dans la méthode utilisée pour obtenir les résultats voulus, à savoir l'analyse statistique des données(algorithmes de classification). Celle-ci se base sur des hypothèses réalistes et restrictives sans pour autant avoir de limitation sur les modèles comportementaux des charges (variations de charges ou modèles) ni besoin de la connaissance des changements d'état des charges. Ainsi, nous sommes en mesure d’identifier et/ou de prédire l'état des charges consommatrices d'énergie (et potentiellement contrôlables) en se basant uniquement sur une phase d'entrainement réduite et des mesures de puissance active agrégée sur un pas de mesure de dix minutes, préservant donc la vie privée des habitants.Dans cette communication, après avoir décrit la méthodologie développée pour classifier les charges et leurs états, ainsi que les connaissances métier fournies aux algorithmes, nous comparons les résultats d’identification pour cinq algorithmes tirés de l'état de l'art et les utilisons comme support d'application à la prédiction. Les algorithmes utilisés se différencient par leur capacité à traiter des problèmes plus ou moins complexe (notamment la prise en compte de relations entre les charges) et se ne révèlent pas tous appropriés à tout type de charge dans le bâtiment résidentiel / Smart metering is one of the fundamental units of a smart grid, as many further applicationsdepend on the availability of fine-grained information of energy consumption and production.Demand response techniques can be substantially improved by processing smart meter data to extractrelevant knowledge of appliances within a residence. The thesis aims at finding generic solutions for thenon-intrusive load monitoring and future usage prediction of residential loads at a low sampling rate.Load monitoring refers to the dis-aggregation of individual loads from the total consumption at thesmart meter. Future usage prediction of appliances are important from the energy management point ofview. In this work, state of the art multi-label temporal classification techniques are implemented usingnovel set of features. Moreover, multi-label classifiers are able to take inter-appliance correlation intoaccount. The methods are validated using a dataset of residential loads in 100 houses monitored over aduration of 1-year.

The transformation of the legacy electric grid to Smart Grid (SG) poses numerous challenges in the design and development of an efficient SG communications network. While there has been an increasing interest in identifying the SG communications network and possible SG applications, specific research challenges at the network protocol have not been elaborated yet. This dissertation revisited each layer of a TCP/IP protocol stack which basically was designed for a wired network and optimized their performance in IEEE 802.11s-based Advanced Metering Infrastructure (AMI) communications network against the following challenges: security and privacy, AMI data explosion, periodic simultaneous data reporting scheduling, poor Transport Control Protocol (TCP) performance, Address Resolution Protocol (ARP) broadcast, and network interoperability. To address these challenges, layered and/or cross-layered protocol improvements were proposed for each layer of TCP/IP protocol stack. At the application layer, a tree-based periodic time schedule and a time division multiple access-based scheduling were proposed to reduce high contention when smart meters simultaneously send their reading. Homomorphic encryption performance was investigated to handle AMI data explosion while providing security and privacy. At the transport layer, a tree-based fixed Retransmission Timeout (RTO) setting and a path-error aware RTO that exploits rich information of IEEE 802.11s data-link layer path selection were proposed to address higher delay due to TCP mechanisms. At the network layer, ARP requests create broadcast storm problems in IEEE 802.11s due to the use of MAC addresses for routing. A secure piggybacking-based ARP was proposed to eliminate this issue. The tunneling mechanisms in the LTE network cause a downlink traffic problem to IEEE 802.11s. For the network interoperability, at the network layer of EPC network, a novel UE access list was proposed to address this issue. At the data-link layer, to handle QoS mismatch between IEEE 802.11s and LTE network, Dual Queues approach was proposed for the Enhanced Distributed Channel Access. The effectiveness of all proposed approaches was validated through extensive simulation experiments using a network simulator. The simulation results showed that the proposed approaches outperformed the traditional TCP/IP protocols in terms of end to end delay, packet delivery ratio, throughput, and collection time.

The big data framework applied to smart meters offers an exception platform for data-driven forecasting and decision making to achieve sustainable energy efficiency. Buying-in consumer confidence through respecting occupants' energy consumption behavior and preferences towards improved participation in various energy programs is imperative but difficult to obtain. The key elements for understanding and predicting household energy consumption are activities occupants perform, appliances and the times that appliances are used, and inter-appliance dependencies. This information can be extracted from the context rich big data from smart meters, although this is challenging because: (1) it is not trivial to mine complex interdependencies between appliances from multiple concurrent data streams; (2) it is difficult to derive accurate relationships between interval based events, where multiple appliance usage persist; (3) continuous generation of the energy consumption data can trigger changes in appliance associations with time and appliances. To overcome these challenges, we propose an unsupervised progressive incremental data mining technique using frequent pattern mining (appliance-appliance associations) and cluster analysis (appliance-time associations) coupled with a Bayesian network based prediction model. The proposed technique addresses the need to analyze temporal energy consumption patterns at the appliance level, which directly reflect consumers' behaviors and provide a basis for generalizing household energy models. Extensive experiments were performed on the model with real-world datasets and strong associations were discovered. The accuracy of the proposed model for predicting multiple appliances usage outperformed support vector machine during every stage while attaining accuracy of 81.65\%, 85.90\%, 89.58\% for 25\%, 50\% and 75\% of the training dataset size respectively. Moreover, accuracy results of 81.89\%, 75.88\%, 79.23\%, 74.74\%, and 72.81\% were obtained for short-term (hours), and long-term (day, week, month, and season) energy consumption forecasts, respectively.

The aim of this thesis is to perform an analysis of possibilities and benefits of installing the Smart Water Metering technology at campus of Faculty of Civil Engineering, Brno University of Technology. The main goal was to provide specific solutions of Smart Metering systems from different companies. One part of this thesis is dedicated to a detailed analysis of a water consumption measurement campaign that took place at the beginning of this academic year using dataloggers. Real-time flow rate and consumed volume was recorded every five minutes. Therefore it was possible to determine water consumption patterns during the day and also tell the minimum, maximum and average flow rate in different parts of the campus. Thanks to these measurements, irregular water consumption during the night was observed. This would not have been easily detected without recording real-time data. This system of recording real-time flow rates with dataloggers is considered to be suitable for the needs of faculty. Thanks to a GSM module, recorded values are sent to an FTP server once a day. From there the data can be downloaded to any kind of analysis software. Installation of Smart Water Metering technology at campus of Faculty of Civil Engineering would be beneficial for both operation of the water supply network and also for purposes of an academic research.

It is commonly agreed among energy experts that smart meters (SMs) are the key component that will facilitate the transition towards the smart grid. Fast-peace innovations in the smart metering infrastructure (AMI) are exposing countless benefits that network operators can obtain when they integrate SMs applications into their daily operations.  Following the amendment in 2017, where the Swedish government dictated that all SMs should now include new features such as remote control, higher time resolution for the energy readings and a friendly interface for customers to access their own data; network operators in Sweden are currently replacing their SMs for a new model, also called the second generation of SMs. While the replacement of meters is in progress, many utilities like Hemab are trying to reveal which technical and financial benefits the new generation of SMs will bring to their operations.    As a first step, this thesis presents the results of a series of interviews carried out with different network operators in Sweden. It is studied which functionalities have the potential to succeed in the near future, as well as those functionalities that are already being tested or fully implemeneted by some utilities in Sweden. Furthermore, this thesis analyses those obstacles and barriers that utilities encounter when trying to implement new applications using the new SMs. In a second stage, an alarm system for power interruptions and voltage-quality events (e.g., overvoltage and undervoltage) using VisionAir software and OMNIPOWER 3-phase meters is evaluated. The results from the evaluation are divided into three sections: a description of the settings and functionalities of the alarm, the outcomes from the test, and a final discussion of potential applications. This study has revealed that alarm functions, data analytics (including several methods such as load forecasting, customer segmentation and non-technical losses analysis), power quality monitoring, dynamic pricing, and load shedding have the biggest potential to succeed in Sweden in the coming years. Furthermore, it can be stated that the lack of time, prioritization of other projects in the grid and the integration of those new applications into the current system seem to be the main barrier for Swedish utilities nowadays. Regarding the alarm system, it was found that the real benefits for network operators arrive when the information coming from an alarm system is combined with a topology interface of the network and a customer notifications server. Both applications could improve customer satisfaction by significantly reducing outage time and providing customers with real-time and precise information about the problems in the grid.

This master’s thesis is dealing with remote measuring systems and their utilization in power engineering. In the first part there are described requirements on an autonomous measuring system and description of AMM and AMR systems. Communicating and data flows working on accurately defined communicating layers are also depicted in this part. The attention is devoted to the possibility of data processing and functions of devices offered by these attributes. In the following part there are described reasons for utilization of remote measuring systems in power engineering and the main advantages of connection of more devices into one unit. After finding of required parametres and functions of systems, a suggestion of terminal device is created in the same way. The suggestion is described in the measuring part with A/D converter, processing in microprocessor, measuring and evaluating alogorithms and attributes of communication of bus used. The last part is devoted to utilization of remote analysis in small power stations in dispersed production. Formation of deformation of voltage, harmonic analysis of signal and its application for data processing are described there. Described analysis was tested on data acquired from a cogeneration unit.