In the last two decades, power systems have faced a thundering transformation due to the widespread penetration of Distributed Generation (DG) which revolutionized the traditional paradigms of network operation and planning. Driven by a multitude of factors (including the diffusion of renewable sources in distribution grids, the progress in storage technologies and their price reduction, the environmentally friendly policies that promoted green sources to reduce fossil fuels consumption, the improvements in Information and Communication Technology (ICT) field allowing remote monitoring and control) the traditional passive distribution networks are turning into active grids, therefore requiring far more complex control schemes, but at the same time offering the opportunity to implement strategies that improve the efficiency, reliability and quality of power supply, if remote devices are shrewdly exploited for grid services. This thesis focuses on the modelling approaches for optimal operation and planning of “smart” electrical transmission and distribution networks, able to integrate Distributed Energy Resources (DERs) for ancillary services provision. Among possible DERs and their applications, renewable sources and storage devices are targeted, either in stand-alone operation or as part of aggregations. The reported analysis starts indeed from the difficulty of planning in the modern scenario, dealing with uncertainty management due to stochastic resources and the guarantee of reliability requirements, that can be possibly overcome through storage devices, which however need high initial investments moving from Operating Expense (OpEx) to Capital Expenditure (CapEx) structure. Then, the challenge of network modelling is addressed: since complete power flow equations are computationally heavy, several methodologies for grid behaviour representation are proposed, sorted by increasing accuracy, comparing the pros and cons that each approach implies. As will be evident from these considerations, there is no claim to find a perfect technique for all purposes, but the best trade-off between robustness and approximation depends on the actual assumptions and goals of each problem. The proposed formulations incorporate also the main characteristics of all the network elements involved in power flows, e.g. branch maximum current bounds and both minimum and maximum bus voltage limits, power factor, energy balance, state of charge, etc. depending on the case. Some elements need a particular focus, like the above-mentioned charge and discharge plans of storage devices, which are discussed in details. The last theoretical aspect proposed is the design of Volt/Var Optimization (VVO) procedures, which are strongly related to the chosen power flow formulation, thus particularly hard to implement from a mathematical point of view: Volt/Var control aims to require optimized contributions of reactive power from DERs in order to regulate voltage at all nodes and relieve line congestions together with power losses. Finally, some applications of these techniques are described, showing examples of multi-target optimizations, of optimal sizing and allocation algorithms or real-time operation strategies, of coordinated approaches to control several dispatchable devices. The first chapter of this thesis describes power system evolution, from the original structure to modern challenges to be coped with, analysing emerging problems as well as leading solutions to take advantage of technological progress and presenting common planning techniques. In the end, the objectives and contributions of this thesis are illustrated. In the second chapter, the core point of electrical networks studies is deepened: power flow modelling, i.e. an overview of possible approaches to the strongly non-linear equations which rule the system behaviour. Chapter three focusses on some aspects of active networks management, like the challenges and possible solutions for storage systems inclusion in grid modelling, the approaches to distributed volt/var control formulation, and the integration of multi-target optimization in integrated systems. The fourth chapter describes the applications of proposed techniques on different test cases and with different goals, presenting performed simulations and discussing their results. Finally, in the last chapter, the conclusions of this work are drawn, analysing the strong points emerging from the simulations described but also open points yet to be resolved.

Optimal operation and planning of transmission and distribution networks, towards renewable sources and storage integration

PONGIGLIONE, PAOLA
2020-05-29

Abstract

In the last two decades, power systems have faced a thundering transformation due to the widespread penetration of Distributed Generation (DG) which revolutionized the traditional paradigms of network operation and planning. Driven by a multitude of factors (including the diffusion of renewable sources in distribution grids, the progress in storage technologies and their price reduction, the environmentally friendly policies that promoted green sources to reduce fossil fuels consumption, the improvements in Information and Communication Technology (ICT) field allowing remote monitoring and control) the traditional passive distribution networks are turning into active grids, therefore requiring far more complex control schemes, but at the same time offering the opportunity to implement strategies that improve the efficiency, reliability and quality of power supply, if remote devices are shrewdly exploited for grid services. This thesis focuses on the modelling approaches for optimal operation and planning of “smart” electrical transmission and distribution networks, able to integrate Distributed Energy Resources (DERs) for ancillary services provision. Among possible DERs and their applications, renewable sources and storage devices are targeted, either in stand-alone operation or as part of aggregations. The reported analysis starts indeed from the difficulty of planning in the modern scenario, dealing with uncertainty management due to stochastic resources and the guarantee of reliability requirements, that can be possibly overcome through storage devices, which however need high initial investments moving from Operating Expense (OpEx) to Capital Expenditure (CapEx) structure. Then, the challenge of network modelling is addressed: since complete power flow equations are computationally heavy, several methodologies for grid behaviour representation are proposed, sorted by increasing accuracy, comparing the pros and cons that each approach implies. As will be evident from these considerations, there is no claim to find a perfect technique for all purposes, but the best trade-off between robustness and approximation depends on the actual assumptions and goals of each problem. The proposed formulations incorporate also the main characteristics of all the network elements involved in power flows, e.g. branch maximum current bounds and both minimum and maximum bus voltage limits, power factor, energy balance, state of charge, etc. depending on the case. Some elements need a particular focus, like the above-mentioned charge and discharge plans of storage devices, which are discussed in details. The last theoretical aspect proposed is the design of Volt/Var Optimization (VVO) procedures, which are strongly related to the chosen power flow formulation, thus particularly hard to implement from a mathematical point of view: Volt/Var control aims to require optimized contributions of reactive power from DERs in order to regulate voltage at all nodes and relieve line congestions together with power losses. Finally, some applications of these techniques are described, showing examples of multi-target optimizations, of optimal sizing and allocation algorithms or real-time operation strategies, of coordinated approaches to control several dispatchable devices. The first chapter of this thesis describes power system evolution, from the original structure to modern challenges to be coped with, analysing emerging problems as well as leading solutions to take advantage of technological progress and presenting common planning techniques. In the end, the objectives and contributions of this thesis are illustrated. In the second chapter, the core point of electrical networks studies is deepened: power flow modelling, i.e. an overview of possible approaches to the strongly non-linear equations which rule the system behaviour. Chapter three focusses on some aspects of active networks management, like the challenges and possible solutions for storage systems inclusion in grid modelling, the approaches to distributed volt/var control formulation, and the integration of multi-target optimization in integrated systems. The fourth chapter describes the applications of proposed techniques on different test cases and with different goals, presenting performed simulations and discussing their results. Finally, in the last chapter, the conclusions of this work are drawn, analysing the strong points emerging from the simulations described but also open points yet to be resolved.
29-mag-2020
Network optimization; network operation; network planning; tso dso; distribution networks; microgrids; distributed storage; der; linear load flow; volt var; monte carlo
File in questo prodotto:
File Dimensione Formato  
phdunige_3657535.pdf

accesso aperto

Descrizione: PhD thesis, full version
Tipologia: Tesi di dottorato
Dimensione 8.06 MB
Formato Adobe PDF
8.06 MB Adobe PDF Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11567/1006216
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact