RE: [TRNSYS-users] fuel cells/electrolyser

[Todd Houstein <toddh@postoffice.utas.edu.au>]

Dear George,
I understand exactly where you are coming from as I was attempting to 
tackle the same problem when your email arrived in my inbox! I am only 
working with electrolysers at this stage, not fuel cells. Generally, the 
technical information available from manufacturers is severely lacking.

There is a paper written by the author of HYDROGEMS, Oystein Ulleberg, 
which explains the electrolyser model Type 160 in detail:

Oystein Ulleberg, Modeling of advanced alkaline electrolyzers: a system 
simulation approach, International Journal of Hydrogen Energy, Volume 28, 
Issue 1, January 2003, Pages 21-33.
(http://www.sciencedirect.com/science/article/B6V3F-45X2MNY-1/2/ffc2e3da  
c26a69277d8280a9d39ea659)

You should be able to obtain the 11 parameters in the TRNSYS deck (area of 
electrode, number of cells, etc) from information from the manufacturer. 
The 10 parameters in the external file are much more difficult to 
determine. (Ulleberg 2003) describes a systematic method for determining 6 
of them (R1, R2, S1, T1, T2 and T3), however it requires fairly extensive 
experimental or operational data (voltage, current and temperature). From 
the information in (Ulleberg 2003) you can also work out how to determine 
the remaining 4 parameters in the external file, however again fairly 
extensive experimental or operational data is required. I can think of only 
two ways of obtaining this data:

1) Find a particularly cooperative manufacturer who will give you the 
required data (and good luck with that!)

2) Purchase the electrolyser and conduct the experiments yourself (which is 
very expensive, and for me it defeats the purpose as I am using modeling in 
order to work out which electrolyser to use).

So, whilst Type 160 is an excellent model as far as accuracy goes, it is 
not practical for many applications due to the difficulty in determining 
the necessary parameters. What we need is a more simplified model. I agree 
that a model with constant efficiency is probably not accurate enough. My 
initial suggestion is to specify pairs of input power (kW) and hydrogen 
production rate (Nm3/h) (similar to the WECS model (Type 190), which 
specifies pairs of wind speed (m/s) and output power (kW)). It should be 
much easier to obtain this information from manufacturers.

(Ulleberg 2003) also refers to six other papers that describe electrolyser 
models. So first of all I will read these papers to see if there is a more 
suitable model. I'll let you know how I go.

One question before I go: What inputs and outputs do you require from the 
model? At this stage I think my only input will be input power (kW), and 
the outputs will be hydrogen production rate (Nm3/h), oxygen production 
rate (Nm3/h) and water consumption rate (L/h).

Regards,
Todd Houstein
PhD Candidate
Hydrogen & Allied Renewable Technologies (HART) research group
School of Engineering
University of Tasmania

Private Bag 65 Hobart
Tasmania Australia 7001
Telephone: (03) 6226 2094
Facsimile: (03) 6226 7863


-----Original Message-----
From:	George Kyriakarakos
Sent:	Wednesday, 13 October 2004 5:06
To:	trnsys-users@engr.wisc.edu
Subject:	[TRNSYS-users] fuel cells/electrolyser

Dear all,
The detailed simulation of electrochemical components such as fuel cells 
and electrolysers needs very detailed parameters, some of which aren't even 
provided by the companies themselves. I have been trying to model a small 
system (fuel cell about 1 kW and electrolyser tops 5 kW) and I haven't been 
able to find the parameters of any commercial products. Luckily I found a 
publication concerning the evaluation of a fuel cell by Ballard (model 
Nexa) and I managed to get all the parameters from there, but so far I have 
been unable to find parameters for a relatively small electrolyser. Has 
anyone got any data and wants to share them? In order to actually do some 
simulations I have built a new type which simply takes as input the 
provided power, uses efficiency of the electrolyser as a parameter and 
calculates the hydrogen output, but that's not enough.
Thank you in advance,
George Kyriakarakos
PS. The publication I mentioned above is "Uniformity analysis at MEA and 
stack Levels for a Nexa PEM fuel cell system" by Wenhua H. Zhu, Robert U. 
Payne, Donald R. Cahela, Bruce J. Tatarchuk, Journal of Power Sources 128 
(2004) 231-238
------------------------------------------------------------------------  
----------------------------------------------------------------------
George Kyriakarakos, Agricultural Engineer
Agricultural University of Athens,
Dept. of Natural Resources and Agricultural Engineering
Iera odos street, 75, Athens 11855, Greece
Tel. +30.210.5294046 (direct), Fax: +30.210.5294023
Mobile +30.6942.046895
e-mail: <gk@aua.gr <mailto:gk@aua.gr>>