Periodic Reporting for period 2 - VIRTUAL-FCS (VIRTUAL & physical platform for Fuel Cell System development)
Período documentado: 2021-07-01 hasta 2023-04-30
Fuel cells have promise in transport applications ranging from delivery drones via ocean going ships to large scale power production, all in which they are competing with well-established technologies. As a complicated and disruptive technology, fuel cells require specialized knowledge to integrate into devices and systems. The mixture of interconnected sub‐systems that must precisely control gases, liquids, heat and electricity to the keep the fuel cell system running optimally is challenging to design and operate. The design and prototyping process is often prohibitively complex, risky, long and expensive, hampering and slowing the adoption of fuel cell technologies in many fields. The VIRTUAL‐FCS project will address these challenges thereby accelerating the market introduction of the environmentally friendly fuel cell technology.
The overall objective of the VIRTUAL-FCS project is to make the design process of hybrid fuel cell and battery systems easier, cheaper and quicker through the development of a toolkit combining software and hardware-interfacing parts for designing and optimizing PEM fuel cells and battery hybrid systems. The platform will be entirely open-source, allowing everyone in both industry and research to benefit from and contribute to the future development of the framework. Machine learning will be integrated to process data from real PEM fuel cells hybrid powertrains, to improve parameterization of the models and to help identify critical aspects on lifetime and performance. The software tools will be developed in close calibration with end users and system integrators, securing a widespread use.
The overall objective of the VIRTUAL-FCS project is to make the design process of hybrid fuel cell and battery systems easier, cheaper and quicker through the development of a toolkit combining software and hardware-interfacing parts for designing and optimizing PEM fuel cells and battery hybrid systems. The platform will be entirely open-source, allowing everyone in both industry and research to benefit from and contribute to the future development of the framework. Machine learning will be integrated to process data from real PEM fuel cells hybrid powertrains, to improve parameterization of the models and to help identify critical aspects on lifetime and performance. The software tools will be developed in close calibration with end users and system integrators, securing a widespread use.
Developed and validated the VirtualFCS Modelica Library containing plug-and-play models for building hybrid fuel cell battery systems including energy management strategies (EMS). This allows for a more efficient and cost-effective way to design reliable and high-performance hybrid system.
Developed two algorithms for computing the degradation and to predict the remaining useful lifetime has been developed. The first one is using machine learning while the other are using parameter estimation.
Implemented rule-based and optimization-based control strategies in the VirtualFCS Library and tested it on a physical hybrid fuel-cell battery system.
Developed an XInTheLoop Library and used it to integrate components from a physical hybrid system into the virtual system simulated in the software. In Virtual-FCS this capability has been demonstrated in hardware system tests where the control system (EMS) has been emulated in OpenModelica.
Development of user cases for maritime vessels, trucks and buses. The end users in the project have each defined and contributed with a use case that aligns with their respective areas of expertise and utilized the VirtualFCS library to develop and present their relevant use cases. These models have also been used to verify/validate the library. These use cases were showcased in the form of explanatory videos.
Successful releases of the VirtualFCS library, documentation, and explanatory webinars. The VirtualFCS library is available for download on GitHub (https://github.com/Virtual-FCS/) and can be used together with the free modelling environment Open Modelica. Explanatory webinars and blog posts have been released and explaining both the science and engineering behind the models.
Dissemination, communication and exploitation actions organized by the project partners allowed the VirtualFCS library to be known by the fuel cell community while the training sessions, tutorials, and summer school allowed to educate users on the basics of fuel cell and batteries modelling and the developed VirtualFCS library utilization.
Developed two algorithms for computing the degradation and to predict the remaining useful lifetime has been developed. The first one is using machine learning while the other are using parameter estimation.
Implemented rule-based and optimization-based control strategies in the VirtualFCS Library and tested it on a physical hybrid fuel-cell battery system.
Developed an XInTheLoop Library and used it to integrate components from a physical hybrid system into the virtual system simulated in the software. In Virtual-FCS this capability has been demonstrated in hardware system tests where the control system (EMS) has been emulated in OpenModelica.
Development of user cases for maritime vessels, trucks and buses. The end users in the project have each defined and contributed with a use case that aligns with their respective areas of expertise and utilized the VirtualFCS library to develop and present their relevant use cases. These models have also been used to verify/validate the library. These use cases were showcased in the form of explanatory videos.
Successful releases of the VirtualFCS library, documentation, and explanatory webinars. The VirtualFCS library is available for download on GitHub (https://github.com/Virtual-FCS/) and can be used together with the free modelling environment Open Modelica. Explanatory webinars and blog posts have been released and explaining both the science and engineering behind the models.
Dissemination, communication and exploitation actions organized by the project partners allowed the VirtualFCS library to be known by the fuel cell community while the training sessions, tutorials, and summer school allowed to educate users on the basics of fuel cell and batteries modelling and the developed VirtualFCS library utilization.
Significant reduction of development times for new fuel cell battery hybrid systems. The advanced modelling, simulation and emulation tools developed in VIRTUAL‐FCS will enable end users with limited experience of fuel cell systems to design and implement new systems more quickly. Crucially for developing tools suitable for a wide range of end users the consortium contains three partners developing fuel cell systems in three different industries who has provided feedback during development to make the tools as useful as possible. The three industries involved are maritime, industrial vehicles and bus. The tools developed have different levels of complexity for different end users making the tools useful for both experienced and novice fuel cell users.
A better understanding of hybridization strategies on the performance, reliability and durability including a benchmark of current methods. Both the lifetime modelling and real time simulation of systems help to predict the impact that different hybridisation and energy management strategies will have on performance. State of the art reliability and durability predictions of individual components are key for results to be meaningful; these has been addressed thoroughly in the project. The end users implemented system models using the developed tools and training material to evaluate and demonstrate the approach and seek improvements in their hybridisation strategies. This has involved investigating the hybridisation strategies for four use cases, including systems with multiple stacks.
A development platform for hybrid fuel cell systems with integration capabilities and corresponding simulation models. The real‐time software platform combined with a full range of emulated components enables end users to seamlessly integrate of real, simulated, and emulated components together in a mixed software‐hardware system. Because the range of possible end user applications is so broad and component specifications will vary so widely in the future rather than just producing the system/hardware to be used in the project, VIRTUAL‐FCS will publish the necessary steps for people to generate their own emulated components and systems using the techniques used in this project.
A better understanding of hybridization strategies on the performance, reliability and durability including a benchmark of current methods. Both the lifetime modelling and real time simulation of systems help to predict the impact that different hybridisation and energy management strategies will have on performance. State of the art reliability and durability predictions of individual components are key for results to be meaningful; these has been addressed thoroughly in the project. The end users implemented system models using the developed tools and training material to evaluate and demonstrate the approach and seek improvements in their hybridisation strategies. This has involved investigating the hybridisation strategies for four use cases, including systems with multiple stacks.
A development platform for hybrid fuel cell systems with integration capabilities and corresponding simulation models. The real‐time software platform combined with a full range of emulated components enables end users to seamlessly integrate of real, simulated, and emulated components together in a mixed software‐hardware system. Because the range of possible end user applications is so broad and component specifications will vary so widely in the future rather than just producing the system/hardware to be used in the project, VIRTUAL‐FCS will publish the necessary steps for people to generate their own emulated components and systems using the techniques used in this project.