Abstract
Historically speaking, alternating current (ac) has been the standard for commercial electrical energy distribution. This is mainly because, in ac systems, electrical energy was easily transformed to diffierent voltages levels, increasing the efficiency of transmitting power over long distances. However, technological advances in, for example, power electronics, and societal concerns such as global warming indicate that a re-evaluation of the current distribution systems is timely. Direct current (dc) distribution systems are foreseen to have advantages over their ac counterparts in terms of efficiency, distribution lines, power conversion and control. Moreover, most renewable energy sources and modern loads produce or utilize dc, or have a dc link in their conversion steps. However, the stability, control, protection and standardization of these systems, and the market inertia of ac systems are major challenges for the broad adoption of dc distribution systems.
Steady-State, Dynamic and Transient Modeling
Adequate models of dc distribution grids are required for the analysis, design and optimization of these systems. In this thesis new and improved methods are proposed for steady-state and dynamic modeling. Two novel steady-state methods are presented, which are shown to be better than the methods in existing literature with respect to convergence, computational effort and accuracy. Furthermore, a dynamic state-space model is proposed that can be efficiently applied to any system topology, and can be used for the stability analysis of these systems. Moreover, an improved symmetrical component decomposition method is presented, which enables simplied (fault) analysis. Transient models for dc distribution systems are briefly discussed, but the development of transient models is outside of the scope of this thesis.
Algebraic and Plug-and-Play Stability
As a result of the decreasing conventional generation, the inertia of electrical grids is signicantly decreased. Furthermore, more and more tightly regulated load converters that have a destabilizing effect on the system's voltage (and frequency) are proliferated throughout the grid. Consequently, the stability of systems with substantial renewable generation is more challenging. In this thesis a method to algebraically derive the stability of any dc distribution system is presented. Moreover, utilizing a Brayton- Moser representation of these systems, two simple requirements are derived for plug-and-play stability (i.e., stability requirements that can be applied to any system, even systems that are subjected to uncertainty or change).
Decentralized Control Strategy and Algorithm
Decentralized control is essential to deal with the trend to decentralize generation and segment the distribution grid, and to manage the potential absence of a communication infrastructure. In this thesis a decentralized control scheme is proposed that ensures global stability and voltage propriety for dc distribution grids. The control scheme divides the acceptable voltage range into demand response, emission, absorption and supply response regions, and species the behavior of converters in these regions. Furthermore, it is shown that inadequate energy utilization can occur, when voltage dependent demand response is utilized. Therefore, the Grid Sense Multiple Access (GSMA) is proposed, which improves the system and energy utilization by employing an exponential backoff routine.
Decentralized Protection Framework and Scheme
Because of the absence of a natural zero crossing, low inertia, meshed topologies and bi-directional power ow, the protection of low voltage dc grids is more challenging than conventional ac grids. In this thesis a decentralized protection framework is presented, which partitions the grid into zones and tiers according to their short-circuit potential and provided level of protection respectively. Furthermore, a decentralized protection scheme is proposed, which consists of a modied solid-state circuit breaker topology and a specied time-current characteristic. It is experimentally shown that this protection scheme ensures security and selectivity for radial and meshed low voltage dc grids.
Steady-State, Dynamic and Transient Modeling
Adequate models of dc distribution grids are required for the analysis, design and optimization of these systems. In this thesis new and improved methods are proposed for steady-state and dynamic modeling. Two novel steady-state methods are presented, which are shown to be better than the methods in existing literature with respect to convergence, computational effort and accuracy. Furthermore, a dynamic state-space model is proposed that can be efficiently applied to any system topology, and can be used for the stability analysis of these systems. Moreover, an improved symmetrical component decomposition method is presented, which enables simplied (fault) analysis. Transient models for dc distribution systems are briefly discussed, but the development of transient models is outside of the scope of this thesis.
Algebraic and Plug-and-Play Stability
As a result of the decreasing conventional generation, the inertia of electrical grids is signicantly decreased. Furthermore, more and more tightly regulated load converters that have a destabilizing effect on the system's voltage (and frequency) are proliferated throughout the grid. Consequently, the stability of systems with substantial renewable generation is more challenging. In this thesis a method to algebraically derive the stability of any dc distribution system is presented. Moreover, utilizing a Brayton- Moser representation of these systems, two simple requirements are derived for plug-and-play stability (i.e., stability requirements that can be applied to any system, even systems that are subjected to uncertainty or change).
Decentralized Control Strategy and Algorithm
Decentralized control is essential to deal with the trend to decentralize generation and segment the distribution grid, and to manage the potential absence of a communication infrastructure. In this thesis a decentralized control scheme is proposed that ensures global stability and voltage propriety for dc distribution grids. The control scheme divides the acceptable voltage range into demand response, emission, absorption and supply response regions, and species the behavior of converters in these regions. Furthermore, it is shown that inadequate energy utilization can occur, when voltage dependent demand response is utilized. Therefore, the Grid Sense Multiple Access (GSMA) is proposed, which improves the system and energy utilization by employing an exponential backoff routine.
Decentralized Protection Framework and Scheme
Because of the absence of a natural zero crossing, low inertia, meshed topologies and bi-directional power ow, the protection of low voltage dc grids is more challenging than conventional ac grids. In this thesis a decentralized protection framework is presented, which partitions the grid into zones and tiers according to their short-circuit potential and provided level of protection respectively. Furthermore, a decentralized protection scheme is proposed, which consists of a modied solid-state circuit breaker topology and a specied time-current characteristic. It is experimentally shown that this protection scheme ensures security and selectivity for radial and meshed low voltage dc grids.
| Original language | English |
|---|---|
| Qualification | Doctor of Philosophy |
| Awarding Institution |
|
| Supervisors/Advisors |
|
| Award date | 3 Sept 2020 |
| Print ISBNs | 978-94-6384-152-8 |
| DOIs | |
| Publication status | Published - 2020 |
