Hello Reddit!<\/p>\n
In the following weeks I’ll have to start writing and conducting research for my Master’s thesis titled “Pattern recognition in industrial systems for fault detection using artificial intelligence algorithms.” My tutor has given some example datasets like Tennessee Eastman Process, CSTR, DAMADICS… But honestly I have no interest whatsoever in the field they’re in (maybe DAMADICS).<\/p>\n
I have been searching the web for other datasets and NASA’s C-MAPSS (Commercial Modular Aero-Propulsion System Simulation) and NASA’s ADAPT (Advanced Diagnostics and Prognostics Testbed) appear more interesting to us: windturbine lifespan, failures in spacecraft, etc. <\/p>\n
My question is, which dataset would you recommend us focusing on? This thesis will be done in group and one of my colleagues knows a lot about machine learning since she has been working in the field quite some time, while the other colleague and I have worked with some things but not in depth. We want something that is interesting and challenging, but not excessively hard or complicated to work around.<\/p>\n
Any insights would be appreciated! Thank you!!<\/p>\n<\/div>\n
submitted by \/u\/mustakit <\/a>
[link]<\/a><\/span> [comments]<\/a><\/span><\/p>
