Search

The decarbonization of energy systems worldwide requires a transformation of their design and operation across all sectors, that is, the residential and commercial, industrial, and transportation sectors. Energy system models are frequently employed for assessing these changes, providing scenarios on potential future system design and on how new technologies and a modified infrastructure will meet future energy demands. Thus, they support investment decisions and policy-making. The Python-based Framework for Integrated Energy System Assessment (ETHOS.FINE) is a software package that provides a toolbox for modeling, analyting and evaluating such energy systems using mathematical optimization. ETHOS.FINE is part of the Energy Transformation paTHway Optimization Suite (ETHOS) , a collection of modeling tools developed by the Institute of Energy and Climate Research - Techno-Economic System Analysis (IEK-3) at Forschungszentrum J\"ulich. ETHOS offers a holistic view on energy systems at arbitrary scales providing tools for geospatial renewable potential analyses, time series simulation tools for residential and industrial sector, discrete choice models for the transportation sector, modeling of global energy supply routes, and local infrastructure assessments, among others. The ETHOS model suite is, e.g., used for analyzing the energy transition of Germany (Stolten et al., 2022).

Recent global events emphasize the importance of a reliable energy supply. One way to increase energy supply security is through decentralized off-grid renewable energy systems, for which a growing number of case studies are researched. This review gives a global overview of the levelized cost of electricity (LCOE) for these autonomous energy systems, which range from 0.03 \$_{2021}/kWh to over 1.00 \$_{2021}/kWh worldwide. The average LCOEs for 100% renewable energy systems have decreased by 9% annually between 2016 and 2021 from 0.54 \$_{2021}/kWh to 0.29 \$_{2021}/kWh, presumably due to cost reductions in renewable energy and storage technologies. Furthermore, we identify and discuss seven key reasons why LCOEs are frequently overestimated or underestimated in literature, and how this can be prevented in the future. Our overview can be employed to verify findings on off-grid systems, to assess where these systems might be deployed and how costs evolve.

The growing share of intermittent renewable energy sources, storage technologies, and the increasing degree of so-called sector coupling necessitates optimization-based energy system models with high temporal and spatial resolutions, which significantly increases their runtimes and limits their maximum sizes. In order to maintain the computational viability of these models for large-scale application cases, temporal aggregation has emerged as a technique for reducing the number of considered time steps by reducing the original time horizon down to fewer, more representative ones. This study presents advanced but generally applicable clustering techniques that allow for ad-hoc improvements of state-of-the-art approaches without requiring profound knowledge of the individual energy system model. These improvements comprise the optimal tradeoff between the number of typical days and inner-daily temporal resolutions, as well as constituting a representation method that can reproduce the value distribution of the original time series. We prove the superiority of these approaches by applying them to two fundamentally different model types, namely a single-node building energy system and a European carbon-neutral energy scenario, and benchmark these against state-of-the-art approaches. This is performed for a variety of temporal resolutions, which leads to many hundreds of model runs. The results show that the proposed improvements on current methods strictly dominate the status quo with respect to Pareto-optimality in terms of runtime and accuracy. Although a speeding up factor of one magnitude could be achieved using traditional aggregation methods within a cost deviation range of two percent, the algorithms proposed herein achieve this accuracy with a runtime speedup by a factor of two orders of magnitude.

Currently, Deep Learning (DL) components within a Case-Based Reasoning (CBR) application often lack the comprehensive integration of available domain knowledge. The trend within machine learning towards so-called Informed machine learning can help to overcome this limitation. In this paper, we therefore investigate the potential of integrating domain knowledge into Graph Neural Networks (GNNs) that are used for similarity assessment between semantic graphs within process-oriented CBR applications. We integrate knowledge in two ways: First, a special data representation and processing method is used that encodes structural knowledge about the semantic annotations of each graph node and edge. Second, the message-passing component of the GNNs is constrained by knowledge on legal node mappings. The evaluation examines the quality and training time of the extended GNNs, compared to the stock models. The results show that both extensions are capable of providing better quality, shorter training times, or in some configurations both advantages at once., Comment: Accepted at the IJCAI-21 workshop on Deep Learning, Case-Based Reasoning, and AutoML: Present and Future Synergies

Energy system models are challenged by the need for high temporal and spatial resolutions in or-der to appropriately depict the increasing share of intermittent renewable energy sources, storage technologies, and the growing interconnectivity across energy sectors. This study evaluates methods for maintaining the computational viability of these models by ana-lyzing different temporal aggregation techniques that reduce the number of time steps in their in-put time series. Two commonly-employed approaches are the representation of time series by a subset of single (typical) time steps, or by groups of consecutive time steps (typical periods). We test these techniques for two different energy system models that are implemented using the Frame-work for Integrated Energy System Assessment (FINE) by benchmarking the optimization results based on aggregation to those of the fully resolved models and investigating whether the optimal aggregation method can, a priori, be determined based on the clustering indicators. The results reveal that typical time steps consistenly outperform typical days with respect to cluster-ing indicators, but do not lead to more accurate optimization results when applied to a model that takes numerous storage technologies into account. Although both aggregation techniques are ca-pable of coupling the aggregated time steps, typical days offer more options to depict storage oper-ations, whereas typical time steps are more effective for models that neglect time-linking con-straints. In summary, the adequate choice of aggregation methods strongly depends on the mathematical structure of the considered energy system optimization model, and a priori decisions of a sufficient temporal aggregation are only possible with good knowledge of this mathematical structure.

The determination of environmentally- and economically-optimal energy system designs and operations is complex. In particular, the integration of weather-dependent renewable energy technologies into energy system optimization models presents new challenges to computational tractability that cannot only be solved by advancements in computational resources. In consequence, energy system modelers must tackle the complexity of their models daily and introduce various methods to manipulate the underlying data and model structure, with the ultimate goal of finding optimal solutions. As which complexity reduction method is suitable for which research question is often unclear, herein we review some approaches to handling complexity. Thus, we first analyze the determinants of complexity and note that many drivers of complexity could be avoided a priori with a tailored model design. Second, we conduct a review of systematic complexity reduction methods for energy system optimization models, which can range from simple linearization performed by modelers to sophisticated multi-level approaches combining aggregation and decomposition methods. Based on this overview, we develop a guide for modelers who encounter computational limitations.

The controlled modulation of an optical wavefront is required for aberration correction, digital phase conjugation or patterned photostimulation. For most of these applications it is desirable to control the wavefront modulation at the highest rates possible. The digital micromirror device (DMD) presents a cost-effective solution to achieve high-speed modulation and often exceeds the speed of the more conventional liquid crystal spatial light modulator, but is inherently an amplitude modulator. Furthermore, spatial dispersion caused by DMD diffraction complicates its use with pulsed laser sources, such as those used in nonlinear microscopy. Here we introduce a DMD-based optical design that overcomes these limitations and achieves dispersion-free high-speed binary phase modulation. We show that this phase modulation can be used to switch through binary phase patterns at the rate of 20 kHz in two-photon excitation fluorescence applications.