Hacker News Reader

Comment by tinco | Hacker News Reader

Off-topic question: is there some software part in nuclear energy systems that is restricting plants or research? I'd like to contribute to the industry as a non-physicist, but it's hard for me to imagine what kind of software might be missing or is being sold by too expensive specialist companies.

I imagine most software is tied to the specific devices they run on, but perhaps there's coupling or analytical software that could be better geared towards the problem domain. Is there any fundamental issue that is waiting for a good software solution?

View on HN

It just so happens that a nuclear fission reactor analysis framework recently emerged on github partially with hopes that people like you would be interested in contributing. It hasn't been widely publicized and is pretty esoteric but is there.

https://github.com/terrapower/armi

Some more open source things in the domain can be found here: https://github.com/paulromano/awesome-nuclear/blob/master/RE...

I'm not an expert at all, but I follow news about fusion energy with keen interest. Getting to conditions where fusion reactions can take place requires modeling the physics with supercomputers. I think the physicists have a handle on the software required to do that modeling; they also have access to the computers.

I think if you have a software background and want to contribute, you should consider applying for a job with one of the projects themselves. General Atomics lists 135 software jobs (https://www.ga-careers.com/search-jobs/software/499/1) for instance.

I think it must be mentioned that in addition to fission and fusion (ITER) work, General Atomics is the creator of the MQ-9 Reaper, which is commonly used for drone strikes. A software job at this company may involve work on these aircraft.

And that's bad because drones are evil? I'm sure I don't agree with a blanket statement to that effect. They're a tool of warfare and I don't see how they're particularly worse than other ways of killing your enemies. I can think of a few ways they're better.

While you may not agree, you can surely recognise that others may appreciate the heads-up. I can easily imagine some people may be pleased to work on fusion, but not on drones.

Your parent merely notified us that more than one type of product is in development at that company and one that is not fusion is specifically designed to kill people.

another factor is clearance. people may be unwilling or unable to obtain security clearances that General Atomics requires for all their engineers, if they're a major defense contractor.

I've dreamed about getting into fusion research myself, and concluded the best pivot would be to help with the plasma simulations at a university while obtaining a PhD.

Yes, that's fair. I think the subtext of the OPs post is that drones are evil, but he didn't write that.

lovemenot is correct. I do not think drones are evil.

I've been trying to find affordable software for modeling my proposed fusion reactor, but some of the software packages I've looked at cost upward of $50k, with no guarantee that they will be able to give the answers I need. It seems to be a problem of low demand for highly specialized software, and scale issues in the different regimes within the device.

In my device, sometimes, some ions will be traveling in a vacuum with a high mean free path, and at other times, the plasma will be as dense as lead. It's VERY difficult to simulate, to the point that it's almost better to just estimate and start building to see if I can start to get close to a solution through careful experimentation.

All kinds of simulations. For example: https://journals.sagepub.com/doi/full/10.1177/10943420177120...

Reply on news.ycombinator.com