Final Report Summary - PEMTOOL (Development of novel, efficient and validated software-based tools for PEM fuel cell component and stack designers)
The project comprised of two main objectives: a scientific as well as a technical objective. The scientific objective was to obtain an understanding of the interaction of the wide range of physical phenomena that occur in a polymer electrolyte membrane (PEM) fuel cell, with a view to controlling that interaction in order to optimise the performance of the cell and arrangement of cells (stacks). The technical objective was to develop a validated modelling tool that is able, quickly and efficiently, to predict this interaction and to guide its user towards the optimisation of cells and stacks. The overall objective was that, at the end of the project, there should exist efficient and validated software tools that could be used by SMEs for optimising PEM fuel cell design. The specific measurable objectives are as follows:
1) to accelerate the development cycle, in terms of time, for PEM fuel cell product development by 50-60 %;
2) to cut the cost of PEM fuel cell product development by 50-60 %;
3) to improve PEM fuel cell performance by 30-50 %.
The project used a wide array of experimental techniques to determine the physical properties of PEFC components and their performance under actual operating conditions. In addition to standard techniques for measuring polarisation curves, other techniques were developed for determining the structural mechanical properties of the porous gas diffusion layer, and a method for measuring the flow of water through a polymer electrolyte membrane. The results of these were used as input for mathematical modelling activities; derived models were implemented into finite-element software. The work structure of the project was comprised by the following:
- three SMEs (Cellkraft, Environment Park and Hysytech) whose core business involved the design and development of polymer electrolyte fuel cells (PEFCs) and one (Comsol) which developed numerical software, a potential capability of which was to act as modelling tool for assisting in the design of PEFCs, have made a list of specifications regarding primary issues of interest in PEFCs;
- two RTD performers (KTH and INASMET) supplied theoretical knowledge;
- theory was then programmed into Comsol's software, Comsol Multiphysics;
- Cellkraft, Environment Park and Hysytech performed experiments on PEFCs, both to determine material input data necessary for modelling, and output necessary for model validation;
- a large end-user (Volvo) performed experiments on a PEFC stack, which the validated models help to optimise.
The achieved results were the following:
- extensive set of experimental data of the performance of different PEFCs using the same components under different operating conditions;
- derivation and analysis of theoretical models that encompass both the participating SMEs' specifications, as well as the latest literature;
- implementation of models in commercially available software (Comsol Multiphysics);
- numerical solutions for both one- and two-phase flows for one- and two-dimensional models;
- an up-to-date literature review of degradation mechanisms in PEFCs.
The unique combination of experimental work, theoretical modelling, mathematical analysis and numerical simulation ensured that the work performed in the project was state-of-the-art:
- the project took account of the most recently available experimental and theoretical literature on PEFCs in relevant conference proceedings and journals, in order derive the necessary models;
- where appropriate, the physics within these models was developed further for the particular needs of the project, i.e. with respect to current density range, physical phenomena;
- unlike in most previous models, mathematical analysis was conducted to enable systematic model reduction, which then leads to faster solution times;
- the models were programmed in a state-of-the art finite-element multiphysics software package.
The project had increased the knowledge base in polymer electrolyte fuel cells both experimentally and theoretically at the university, research institute, the SMEs and the end-user that participated. From the industrial perspective, the availability of a tool to help predict good cell design from the point of e.g. good structural mechanical principles was a significant step forward. From the research point of view, the gathering of accurate experimental results for the same components in different cells was perhaps unique, and should ensure that successful validation of the model would be forthcoming in the near future.
1) to accelerate the development cycle, in terms of time, for PEM fuel cell product development by 50-60 %;
2) to cut the cost of PEM fuel cell product development by 50-60 %;
3) to improve PEM fuel cell performance by 30-50 %.
The project used a wide array of experimental techniques to determine the physical properties of PEFC components and their performance under actual operating conditions. In addition to standard techniques for measuring polarisation curves, other techniques were developed for determining the structural mechanical properties of the porous gas diffusion layer, and a method for measuring the flow of water through a polymer electrolyte membrane. The results of these were used as input for mathematical modelling activities; derived models were implemented into finite-element software. The work structure of the project was comprised by the following:
- three SMEs (Cellkraft, Environment Park and Hysytech) whose core business involved the design and development of polymer electrolyte fuel cells (PEFCs) and one (Comsol) which developed numerical software, a potential capability of which was to act as modelling tool for assisting in the design of PEFCs, have made a list of specifications regarding primary issues of interest in PEFCs;
- two RTD performers (KTH and INASMET) supplied theoretical knowledge;
- theory was then programmed into Comsol's software, Comsol Multiphysics;
- Cellkraft, Environment Park and Hysytech performed experiments on PEFCs, both to determine material input data necessary for modelling, and output necessary for model validation;
- a large end-user (Volvo) performed experiments on a PEFC stack, which the validated models help to optimise.
The achieved results were the following:
- extensive set of experimental data of the performance of different PEFCs using the same components under different operating conditions;
- derivation and analysis of theoretical models that encompass both the participating SMEs' specifications, as well as the latest literature;
- implementation of models in commercially available software (Comsol Multiphysics);
- numerical solutions for both one- and two-phase flows for one- and two-dimensional models;
- an up-to-date literature review of degradation mechanisms in PEFCs.
The unique combination of experimental work, theoretical modelling, mathematical analysis and numerical simulation ensured that the work performed in the project was state-of-the-art:
- the project took account of the most recently available experimental and theoretical literature on PEFCs in relevant conference proceedings and journals, in order derive the necessary models;
- where appropriate, the physics within these models was developed further for the particular needs of the project, i.e. with respect to current density range, physical phenomena;
- unlike in most previous models, mathematical analysis was conducted to enable systematic model reduction, which then leads to faster solution times;
- the models were programmed in a state-of-the art finite-element multiphysics software package.
The project had increased the knowledge base in polymer electrolyte fuel cells both experimentally and theoretically at the university, research institute, the SMEs and the end-user that participated. From the industrial perspective, the availability of a tool to help predict good cell design from the point of e.g. good structural mechanical principles was a significant step forward. From the research point of view, the gathering of accurate experimental results for the same components in different cells was perhaps unique, and should ensure that successful validation of the model would be forthcoming in the near future.
