Renewable hydrogen electrolysis

hypervalve

Hello

I'm currently in the middle of a dissertation project, a feasibility study of a tidal/hydrogen generation system. At the moment I'm getting pretty bogged down trying to model the electrolyser. I'd like it to be more detailed than simply power in*conversion efficiencies. In particular I'd like to be able to comment on the ramifications of the transient nature of the tidal stream energy. From what I've read I would expect to see efficiency decrease as the input power increases.

I've been sifting through many research papers on the subject which attempt to mathematically model such systems with varying degrees of detail. I'm not sure if I'm more or less confused for it. If anyone has a grasp of the workings of electrolysers and is willing to answer a few of my (stupid) questions I'd be much obliged.

The technology I'm modelling is PEM. Which at the moment seems to be pretty small scale as far as production rates go (compared to alkaline). I'll basically be choosing an existing unit and scaling up to meet the power output of the tidal turbine. Which we'll say has a twice daily peak and minimum outputs of 500kW and 0kW respectively.

So I guess the first thing I'd like to know before I think of modelling the system, is the power conditioning arrangement. The electrolyser will require DC current, and it is my understanding that a stable voltage is also required? So I'm thinking some kind of double bridge rectifier could be used to supply stable voltage with varying current. The hydrogen output could then be calculated from he current flow if I'm not mistaken?

My apologies for the lack of understanding here, my background is in mechanical engineering so much of this is new to me!

Thanks in advance!

steifanc

hi hypervalve ,

Ive been playing around with hho cells for a while , great fun , quite dangerous and not quite efficient , although i have never tried a stack system , i have played around with injection dc with pwm and other tomfoolery , its part two of a project i was getting into ,
i can help with power and control , if i can .

bnt

The papers you've been sifting through - is one of them Frano Barbir's "PEM electrolysis for production of hydrogen from renewable energy sources", from the Solar Energy journal, Vol 78 issue 5 (661-669)? I'm not an expert on this topic, but this paper covers PEM electrolysis from intermittent supplies, though from photovoltaics, not tidal. It discusses the use of DC-DC converters for power conditioning, which impose some losses themselves. There also appear to be issues relating to operating temperatures of PEM electrolysers that also reduce system efficiency. But the author says it still looks good for further investigation.

hypervalve

Yep thats one of the papers I looked through. PEM is well suited because it can deal with variations in power really well. There seems to be many factors which have an effect on production rates- current, voltage, operating temperature, operating pressure seem to be the main players.

For my model I'd like to keep it as simple as possible. Pressure and temperature will remain constant, and I will choose a pre existing U-I curve so I don't have to calculate overvoltages myself (not that I have the resources to do so!)

So according to a typical I-V curve, as current density increases, so does voltage (due to overvoltages?) and thus efficiency decreases. So I guess I size the stack by active area to make the most of my 500kW resource- but I must find a compromise between capital cost and conversion efficiency. That is to say, I could go for a large active area to keep the current density (and overpotential) low, which would have a good efficiency. But the trade off is obviously the very capital cost of the electrolyser.

Once this is done, I have to model how it behaves at power inputs from 0kW to 500kW. For instance, what H2 output rate could I expect form 250kW on a system designed for 500kW.