Case: Upscaling of lab-size reactor to industrial application
Case: Upscaling of lab-size reactor to industrial application
Related Initiative
What is this about?
Project context: A reactor for a conversion step in the refinement of biofuels (an element of a circular bioeconomic process) demonstrates efficient performance in the lab on a small scale. It shall now be scaled up to industrial size with high throughput of material. The team of engineers, highly experienced in risk assessment and safety measures, applies for a novel method of effective upscaling that ensures safety for both research workers and those working with the industrial reactor later: Building a digital twin of the device and running simulations employing machine learning and AI routines that are capable of handling the many variables of the application. They apply all common risk assessment and safety protocols. They argue for the budget allocation into the digital twin work package with safety and saving material and costs by ruling out design pathways that are doomed to fail before an actual prototype is built.
Ethics self-assessment, part 7: Environment
Safety protocols are maintained and applied throughout the research. The digital twin model reduces the exposure risk of researchers in the prototype design phase. Moreover, it will enable us to manage risks of the operation of the final design in industrial applications. Material converted and produced in the research phase will be handled in compliance with all disposal and recycling regulations [empirical and methodological details in supplementary information].
Our reactor enables conversion processes that contribute to bioeconomic innovation in industrial production and manufacturing. In the long run, the effective production of biofuels as part of circular bioeconomic processes contributes to tackling ecological challenges by reducing the demand of fossil fuels and an economic dependence on them. On the one hand, it improves environmental health, on the other hand it doesn’t reduce or restrict economic agency for the sake of environmental integrity (maintaining freedom of choice). These benefits need to be weighed against technological and ecological risks such as critical failures, disposal of end-of-lifecycle material, and competing demands of input material (for example, maize and turnip) with other industries sharing the same ecological space. In our research, we will identify multi-stakeholder perspectives and integrate them into our reactor design discussions [description in respective work package description].
A critical element of the project is the handling of data and the machine learning involved in the digital twin modelling. Quality of base data for the model will be controlled. Safety-sensitive performance data of the digital twin will be stored on offline devices in order to avoid breeches that could compromise the security of operating the physical reactor facility or pose bioterrorism risks. Yet, principles of transparency demand that we build the system towards explainability and reproducibility.