Refine
Document Type
- Doctoral Thesis (6)
Language
- English (6)
Has Fulltext
- yes (6)
Is part of the Bibliography
- no (6)
Keywords
- energy system design (2)
- power transmission (2)
- CO2 emission reduction targets (1)
- Chaostheorie (1)
- Complex networks (1)
- Datenanalyse (1)
- Energy system design (1)
- Evolutionäre Spieltheorie (1)
- Gefangenendilemma (1)
- Globale Optimierung (1)
Institute
The present work deals with the integration of variable renewable energy sources, wind and solar energy into the European and US power grid. In contrast to other networks, such as the gas supply mains, the electricity network is practically not able to store energy. Generation and consumption therefore always have tobe balanced. Currently, the load curve is viewed as a rigid boundary condition, which must be followed by the generation system. The basic idea of the approach followed here is that weather-dependent generation causes a shift of focus of the electricity supply. At high shares of wind and solar generation, the role of the rigid boundary condition falls to the residual load, that is, the remaining load after subtraction of renewable generation. The goal is to include the weather dependence as well as the load curve in the design of the future electricity supply.
After a brief introduction, the present work first turns to the underlying weather-, generation and load data, which form the starting point of the analysis. In addition, some basic concepts of energy economics are discussed, which are needed in the following.
In the main part of the thesis, several algorithms are developed to determine the load flow in a network with a high share of wind and solar energy and to determine the backup supply needed at the same time. Minimization of the energy needed from controllable power plants, the capacity variable power plants, and the capacity of storing serve as guiding principles. In addition, the optimization problem of grid extensions is considered. It is shown that it can be formulated as a convex optimization problem. It turns out that with an optimized, international transmission network which is about four times the currently available transmission capacity, much of the potential savings in backup energy (about 40%) in Europe can be reached. In contrast, a twelvefold increase the transmission capacity would be necessary for a complete implementation of all possible savings in dispatchable power plants.
The reduction of the dispatchable generation capacity and storage capacity, however, presents a greater challenge. Due to correlations in the generation of time series of individual countries, it may be reduced only with difficulty, and by only about 30%.
In the following, the influence of the relative share of wind and solar energy is illuminated and examined the interplay with the line capacitance. A stronger transmission network tends to lead to a higher proportion of wind energy being better integrated. With increasing line capacity, the optimal mix in Europe therefore shifts from about 70% to 80% wind. Similar analyses are carried out for the US with comparable results.
In addition, the cost of the overall system can be reduced. It is interesting at this point that the advantages for the network integration may outweigh higher production costs of individual technologies, so that it is more favourable from the viewpoint of the entire system to use the more expensive technologies.
Finally, attention is given to the flexibility of the dispatchable power plants. Starting from a Fourier-like decomposition of the load curve as it was a few years ago, when hardly renewable generation capacity was present, capacities of different flexibility classes of dispatchable power plant are calculated. For this purpose, it is assumed that the power plant park is able to follow the load curve without significant surplusses or deficits. From this examination, it is derived what capacity must at least be available without having to resort to a detailed database of existing power plants.
Assuming a strong European cooperation, with a stronger international transmission network, the dispatchable power capacity can be significantly reduced while maintaining security of supply and generating relatively small surplusses in dispatchable power plants.
In this work the flexibility requirements of a highly renewable European electricity network that has to cover fluctuations of wind and solar power generation on different temporal and spatial scales are studied. Cost optimal ways to do so are analysed that include optimal distribution of the infrastructure, large scale transmission, storage, and dispatchable generators. In order to examine these issues, a model of increasing sophistication is built, first considering different flexibility classes of conventional generation, then adding storage, before finally considering transmission to see the effects of each.
To conclude, in this work it was shown that slowly flexible base load generators can only be used in energy systems with renewable shares of less than 50%, independent of the expansion of an interconnecting transmission network within Europe. Furthermore, for a system with a dominant fraction of renewable generation, highly flexible generators are essentially the only necessary class of backup generators. The total backup capacity can only be decreased significantly if interconnecting transmission is allowed, clearly favouring a European-wide energy network. These results are independent of the complexity level of the cost assumptions used for the models. The use of storage technologies allows to reduce the required conventional backup capacity further. This highlights the importance of including additional technologies into the energy system that provide flexibility to balance fluctuations caused by the renewable energy sources. These technologies could for example be advanced energy storage systems, interconnecting transmission in the electricity network, and hydro power plants.
It was demonstrated that a cost optimal European electricity system with almost 100% renewable generation can have total system costs comparable to today's system cost. However, this requires a very large transmission grid expansion to nine times the line volume of the present-day system. Limiting transmission increases the system cost by up to a third, however, a compromise grid with four times today's line volume already locks in most of the cost benefits. Therefore, it is very clear that by increasing the pan-European network connectivity, a cost efficient inclusion of renewable energies can be achieved, which is strongly needed to reach current climate change prevention goals.
It was also shown that a similarly cost efficient, highly renewable European electricity system can be achieved that considers a wide range of additional policy constraints and plausible changes of economic parameters.
Statistical physics of power flows on networks with a high share of fluctuating renewable generation
(2010)
Renewable energy sources will play an important role in future generation of electrical energy. This is due to the fact that fossil fuel reserves are limited and because of the waste caused by conventional electricity generation. The most important sources of renewable energy, wind and solar irradiation, exhibit strong temporal fluctuations. This poses new problems for the security of supply. Further, the power flows become a stochastic character so that new methods are required to predict flows within an electrical grid. The main focus of this work is the description of power flows in a electrical transmission network with a high share of renewable generation of electrical energy. To define an appropriate model, it is important to understand the general set-up of a stable system with fluctuating generation. Therefore, generation time series of solar and wind power are compared to load time series for whole Europe and the required balancing or storage capacities analyzed. With these insights, a simple model is proposed to study the power flows. An approximation to the full power flow equations is used and evaluated with Monte-Carlo simulations. Further, approximations to the distributions of power flows along the links are analytically derived. Finally, the results are compared to the power flows calculated from the generation and load data.
In nature, society and technology many disordered systems exist, that show emergent behaviour, where the interactions of numerous microscopic agents result in macroscopic, systemic properties, that may not be present on the microscopic scale. Examples include phase transitions in magnetism and percolation, for example in porous unordered media, biological, and social systems. Also technological systems that are explicitly designed to function without central control instances, like their prime example the Internet, or virtual networks, like the World Wide Web, which is defined by the hyperlinks from one web page to another, exhibit emergent properties. The study of the common network characteristics found in previously seemingly unrelated fields of science and the urge to explain their emergence, form a scientific field in its own right, the science of complex networks. In this field, methodologies from physics, leading to simplification and generalization by abstraction, help to shift the focus from the implementation's details on the microscopic level to the macroscopic, coarse grained system level. By describing the macroscopic properties that emerge from microscopic interactions, statistical physics, in particular stochastic and computational methods, has proven to be a valuable tool in the investigation of such systems. The mathematical framework for the description of networks is graph theory, in hindsight founded by Euler in 1736 and an active area of research since then. In recent years, applied graph theory flourished through the advent of large scale data sets, made accessible by the use of computers. A paradigm for microscopic interactions among entities that locally optimize their behaviour to increase their own benefit is game theory, the mathematical framework of decision finding. With first applications in economics e.g. Neumann (1944), game theory is an approved field of mathematics. However, game theoretic behaviour is also found in natural systems, e.g. populations of the bacterium Escherichia coli, as described by Kerr (2002). In the present work, a combination of graph theory and game theory is used to model the interactions of selfish agents that form networks. Following brief introductions to graph theory and game theory, the present work approaches the interplay of local self-organizing rules with network properties and topology from three perspectives. To investigate the dynamics of topology reshaping, coupling of the so called iterated prisoners' dilemma (IPD) to the network structure is proposed and studied in Chapter 4. In dependence of a free parameter in the payoff matrix, the reorganization dynamics result in various emergent network structures. The resulting topologies exhibit an increase in performance, measured by a variance of closeness, of a factor 1.2 to 1.9, depending in the chosen free parameter. Presented in Chapter 5, the second approach puts the focus on a static network structure and studies the cooperativity of the system, measured by the fixation probability. Heterogeneous strategies to distribute incentives for cooperation among the players are proposed. These strategies allow to enhance the cooperative behaviour, while requiring fewer total investments. Putting the emphasis on communication networks in Chapters 6 and 7, the third approach investigates the use of routing metrics to increase the performance of data packet transport networks. Algorithms for the iterative determination of such metrics are demonstrated and investigated. The most successful of these algorithms, the hybrid metric, is able to increase the throughput capacity of a network by a factor of 7. During the investigation of the iterative weight assignments a simple, static weight assignment, the so called logKiKj metric, is found. In contrast to the algorithmic metrics, it results in vanishing computational costs, yet it is able to increase the performance by a factor of 5.
Dynamics of chaotic strings
(2011)
The main topic of this thesis is the investigation of dynamical properties of coupled Tchebycheff map networks. At every node of the network the dynamics is given by the iteration of a Tchebycheff map, which shows strongest possible chaotic behaviour. By applying a coupling between the various individual dynamics along the links of the network, a rich structure of complex dynamical patterns emerges. Accordingly, coupled chaotic map networks provide prototypical models for studying the interplay between local dynamics, network structure, and the emergent global dynamics. An exciting application of coupled Tchebycheff map lattices in quantum field theory has been proposed Beck in Spatio-temporal chaos and vacuum fluctuations of quantized fields' (2002). In this so-called chaotic string model, the coupled map lattice dynamics generates the noise needed for the Parisi-Wu approach of stochastic quantization. The remarkable obversation is that the respective dynamics seems to reproduce distinguished numerical values of coupling constants that coincide with those observed in the standard model of particle physic. The results of this thesis give insights into the chaotic string model and its network generalization from a dynamical point of view. This leads to a deeper understanding of the dynamics, which is essential for a critical discussion of possible physical embeddings. Apart from this specific application to particle physics, the investigated concepts like synchronization or a most random behaviour of the dynamics are of general interest for dynamical system theory and the science of complex networks. As a first approach, discrete symmetry transformations of the model are studied. These transformations are formulated in a general way in order to be also applicable to similar dynamics on bipartite network structures. An observable of main interest in the chaotic string model is the interaction energy. In Spatio-temporal chaos and vacuum fluctuations of quantized fields' (2002) it has been observed that certain chaotic string couplings, corresponding to a vanishing interaction energy, coincide with coupling constants of the standard model of elementary particle physics. Since the interaction energy is basically a spatial correlation measure, an interpretation of the respective dynamical states in terms of a most random behaviour is tempting. In order to distinguish certain states as most random', or evoke another dynamical principle, a deeper understanding of the dynamics essential. In the present thesis the dynamics is studied numerically via Lyapunov measures, spatial correlations, and ergodic properties. It is shown that the zeros of the interaction energy are distinguished only with respect to this specific observable, but not by a more general dynamical principle. The original chaotic string model is defined on a one-dimensional lattice (ring-network) as the underlying network topology. This thesis studies a modification of the model based on the introduction of tunable disorder. The effects of inhomogeneous coupling weights as well as small-world perturbations of the ring-network structure on the interaction energy are discussed. Synchronization properties of the chaotic string model and its network generalization are studied in later chapters of this thesis. The analysis is based on the master stability formalism, which relates the stability of the synchronized state to the spectral properties of the network. Apart from complete synchronization, where the dynamics at all nodes of the network coincide, also two-cluster synchronization on bipartite networks is studied. For both types of synchronization it is shown that depending on the type of coupling the synchronized dynamics can display chaotic as well as periodic or quasi-periodic behaviour. The semi-analytical calculations reveal that the respective synchronized states are often stable for a wide range of coupling values even for the ring-network, although the respective basins of attraction may inhabit only a small fraction of the phase space. To provide analytical results in closed form, for complete synchronization the stability of all fixed points and period-2 orbits of all chaotic string networks are determined analytically. The master stability formalism allows to treat the ring-network of the chaotic string model as a special case, but the results are valid for coupled Tchebycheff maps on arbitrary networks. For two-cluster synchronization on bipartite networks, selected fixed points and period-2 orbits are analyzed.
Defossiliation of the energy system is crucial in the face of the impending risks of climate change. Electricity generation by burning fossil fuels is being displaced by renewable energy sources like hydro, wind and solar, driven by support schemes and falling costs from technological advances as well as manufacturing scale effects. The unavoidable shift from flexibly dispatchable generation to weather-dependent spatio-temporally varying generators transforms the generation and distribution of electricity into highly interdependent complex systems in multiple dimensions and disciplines:
In time, different scales, stretching from intra-day, diurnal, synoptic to seasonal oscillations of the weather interact with years and decades of planning and construction of capacity. In space, long-range correlations and local variations of weather systems as well as local bottlenecks in transmission networks affect solutions. The investment decisions about technological mix and spatial distribution of capacity follow economic principles, within restrictions which adapt in social feedback loops to public opinion and lobbyist influences.
In this work, a family of self-consistent models is developed which map physical steady-state operation, capacity investments and exogeneous restrictions of a European electricity system, in higher simultaneous spatial and temporal detail as well as scope than has previously been computationally tractable. Increasing the spatial detail of the renewable resources and co-optimizing the expansion of only a few transmission lines, reveals solutions to serve the European electricity demand at about today’s electricity cost with only 5% of its carbon-dioxide emissions; and importantly their electricity mix differs from the findings at low spatial resolution.
As important intermediate steps,
• new algorithms for the convex optimization of electricity system infrastructure are derived from graph-theoretic decompositions of network flows. Only these enable the investigation of model detail beyond previous computational limitations.
• a comprehensive European electricity network model down to individual substations at the transmission voltage levels is built by combining and completing data from freely available sources.
• a network reduction technique is developed to approximate the detailed model at a sequence of spatial resolutions to investigate the role of spatial scale, and identify a level of spatial resolution which captures all relevant detail, but is still computationally tractable.
• a method to trace the flow of power through the network, which is related to a vector diffusion process on a directed flow graph embedded in a network, is used to analyse the resulting technology mix and its interactions with the power network
The open-source nature of the model and restriction to freely available data encourages an accessible and transparent discussion about the future European electricity system, primarily based on renewable wind and solar resources.