ORCID

0000-0002-8516-9635

Date of Award

2024

Document Type

Doctoral Thesis

Degree Name

Doctor of Philosophy

Department

Process, Energy, and Transport Engineering

First Advisor

Dr Michael D. Murphy

Second Advisor

Prof. Dr.-Ing. Marco Braun

Abstract

To meet climate objectives, major energy consumers with local multi-energy systems (L- MESs) must transition to renewable energy sources soon. The variability of renewable energies requires increased operational flexibility for integration. In this context, demand response (DR), a strategy in which electricity consumers adjust their load profiles in response to incentives, has become crucial, offering cost-effective flexibility. It supports L-MESs in integrating renewable energy, reducing decarbonization costs, and enhancing resilience. However, quantifying the economic DR potentials for L-MESs under carbon emission constraints is complex, especially when considering investment options in distributed energy resources. This complexity hinders the rapid adoption of DR programs and the investment in distributed energy resources. Addressing this issue, this thesis introduces methods and tools for the economic and environmental evaluation of non-residential DR, applying them to real-world case studies.

To assess the environmental impact of DR, electricity carbon emission factors (CEFs) are required. There are two types of CEFs: average grid-mix CEFs and marginal CEFs. The first experimental study introduced methods for calculating both types of CEF for European electricity systems and applied them to data from 20 European countries for 2017–2019. Based on the CEFs, simulations of standardized price-based daily load shifts revealed that carbon emissions increased in 8 of the 20 countries, averaging a 2.1% rise. Load shifts based on marginal CEFs reduced average emissions by 35%, but with 56% lower monetary savings compared to price-based shifts. Further simulations at varying carbon price levels analyzed the impact of carbon pricing on emissions. The results confirmed that, with adequate carbon pricing, price-based DR could effectively enhance both economic and environmental outcomes. Elmada, a novel open-source Python package, was developed by the author to integrate these CEF calculation methods, enhancing its utility in environmental assessments.

The open-source tool Demand Response Analysis Framework (DRAF) was developed in Python, focusing on optimizing the design and operation of energy technologies. It considers aspects such as demand-side flexibility, electrification, and renewable energy sources. DRAF quantifies the potential for decarbonization and cost reduction using multi-objective mixed- integer linear programming. It provides decision-makers of L-MESs with optimal scenarios in terms of costs, carbon emissions, or Pareto efficiency. Supporting all steps of the energy system optimization process, DRAF includes functionalities ranging from time series analysis to interactive plotting and data export. It features several component templates for easy initiation, while also allowing expansive modeling possibilities through a generic model generator.

DRAF was applied to various non-residential case studies: In a detailed case study, the value of demand-side flexibility for decarbonizing a German beverage company’s L-MES was quantified. The optimal synthesis, design, and operation were identified, considering self-consumption optimization, peak shaving, and DR based on hourly prices and CEFs. Vehicle-to-X for electrical industrial forklifts, power-to-heat at multiple temperatures, wind turbines, photovoltaic systems, and energy storage systems were considered. Novel metrics to evaluate the intensity of price-based and CEF-based DR were proposed and applied. The results revealed that flexibility usage reduced decarbonization costs by 19–80% depending on electricity and carbon removal prices. Furthermore, it reduced total annual costs, operational carbon footprint, energy-weighted average prices and CEFs, as well as fossil energy dependency. The results also suggested that achieving net-zero operational carbon emissions for L-MESs required flexibility, which, in an economic case, is provided by a combination of different flexible technologies and storage systems that complement each other. In an industrial case study, the scheduling of two cement mills and storages were optimized under two DR programs resulting in a 13% annual electricity cost reduction in a real-time vs time-of-use pricing scenario. Another study highlighted the impact of fixed peak-demand grid fees on industrial DR, revealing a conflict between demand flattening and incentives for DR, energy storage, and renewable sources. Additionally, a case study on green hydrogen import hubs demonstrated the importance of considering landing interruptions within the design stage of hydrogen hubs, to increase its resilience through on-site flexibility.

A final study analyzed the use of time series aggregation (TSA) for reducing complexity in L-MES models. It explored hidden costs for investors resulting from TSA application in designing flexible and decarbonized MESs. These costs were quantified by backtesting the TSA-based design with unaggregated time series. Additionally, the study assessed the impact of carbon removal prices on L-MES design and compared the effects of using marginal CEFs instead of grid-mix CEFs on TSA’s efficacy. The findings indicated that with highly ambitious emission reduction targets, TSA can lead to backtesting errors over 10%. Thus, TSA should be applied cautiously in low carbon L-MES contexts, especially when outcomes guide investment decisions and the full model is solvable with a reasonable level of computational effort.

This research provides tools to quantify and enhance DR-induced cost and emission reductions, demonstrating the potential of demand-side flexibility in reducing emissions in electricity and local non-residential energy systems. Additionally, it offers novel findings that grant researchers in the field new insights. Ultimately, it promotes the adoption of non-residential DR by addressing the complexities of modern energy systems through integrated modeling.

Creative Commons License

Creative Commons Attribution 4.0 International License
This work is licensed under a Creative Commons Attribution 4.0 International License.

Access Level

info:eu-repo/semantics/openAccess

Download

Share

COinS