Scientific Background, Problem Identification, and Objective

Scientific Background

The scientific backdrop of the Oproda project centers on the pivotal role of green hydrogen in facilitating the transition towards sustainable energy and decarbonizing the energy system, particularly in heavy industrial sectors such as steel and cement production. These sectors are significant contributors to climate change, and green hydrogen offers a pathway to drastically reduce their carbon footprint by serving both as a carbon-neutral energy carrier and as a fundamental raw material in various production processes. Solid oxide cell (SOC) technology, recognized for its high efficiency in both electricity generation and hydrogen production, stands at the forefront of this transformation. Its unique capability to operate in dual modes—fuel cell mode for power generation and electrolyser mode for hydrogen production—combined with the use of abundant materials, positions SOC as a highly promising solution for sustainable, large-scale hydrogen production. The project aims to address the challenges and optimize the performance of SOC systems, paving the way for broader commercialization and integration into the global effort to mitigate climate change.

Problem Identification

The operation of Solid Oxide Cell (SOC) systems is fraught with complexities that necessitate meticulous planning for their long-term optimal utilization. In the domain of performance optimization and degradation modeling, a variety of approaches have been employed, ranging from numerical simulations based on first principles to experimental methods like accelerated stress testing. However, these methods have their limitations. First principle models, while detailed, demand extensive background knowledge and sophisticated equipment for parameter estimation, making them costly and time-consuming. Conversely, purely data-driven models, despite their potential to bypass some of these limitations by leveraging available data, often result in models that are computationally intensive and whose parameters lack a clear physical interpretation.

The challenge is further compounded by the stochastic nature of SOC operations at the micro-level, driven by variations in environmental conditions such as humidity and temperature, fluctuations in fuel quality, and the imperfections of balance-of-plant components. These variables introduce a level of unpredictability that standard modeling approaches struggle to accommodate. For instance, electrochemical impedance spectroscopy (EIS) measurements, a common technique for SOC system analysis, can yield different results even when conducted consecutively under seemingly identical conditions due to these stochastic influences.

Moreover, the existing models and testing methods rarely cover the full spectrum of potential fault scenarios or the complete operational range of SOC systems. This limitation hampers the ability to fully understand and predict system performance and degradation, especially under untested conditions. While exhaustive testing could theoretically map out a comprehensive fault landscape, such an approach is practically unfeasible due to the immense time and financial costs involved. This issue is exacerbated in real-world applications, where the full characterisation of systems might be impossible, leaving significant gaps in our understanding and modeling capabilities of SOC technologies.

Addressing these challenges is critical for advancing SOC technology and making it a viable and reliable option for green hydrogen production and energy generation. The identification of a modeling and optimization strategy that can navigate the complexities of SOC systems, incorporate the stochastic nature of their operation, and predict performance under a wide range of conditions is crucial. This strategy must be robust enough to handle limited and noisy data while being flexible enough to adapt to new information and conditions, thus ensuring the long-term efficiency and sustainability of SOC applications in the energy sector.

Objective of the Proposed Research

The primary goal of the proposed research within the Oproda project is to develop and validate a comprehensive set of methodologies that can significantly advance the field of solid oxide cell (SOC) systems. The research aims to achieve two main objectives:

Develop Probabilistic Data-Driven Models for SOC Systems: The project seeks to establish probabilistic, data-driven models for the identification of SOC systems. These models are designed to function effectively even with limited datasets. A key aspect of this objective is to ensure that the models provide a clear linkage between the parameters they use and the physical processes within SOC systems. This involves moving beyond traditional black-box approaches to create models that are both explainable and capable of incorporating expert knowledge into their structure. This will allow for the models to be more readily interpreted and trusted by engineers and scientists working in the field. Validate the Models Under Various Operational Scenarios: Validation will be conducted on typical SOC-based systems under a range of operation scenarios that include different fault modes and operating conditions. This is crucial for demonstrating the models’ accuracy and reliability across the spectrum of potential SOC applications. The validation process will test the models’ ability to predict performance and identify faults, thereby proving their utility in real-world settings. Achieving these objectives is expected to break new ground in the understanding and optimization of SOC technologies, enabling more reliable and efficient hydrogen production and energy conversion. This advancement is seen as a critical step toward the broader adoption of green energy technologies, contributing to global efforts to combat climate change by reducing reliance on carbon-intensive energy sources.

Energy transition diagram illustrating the role of green hydrogen.