Hometown: Sunnyvale CA
September 21, 2009
I'm exploring using AI algorithm to detect and diagnose faults in solar power plants.
Electrical Engineer with systems and software background
"This is a tool meant for technical person to get a output value given a static set of solar site parameters, and two time and environment dependent parameters --Irradiance and module temperature. We want to help detect under-performance in sites. "
Sorry to get to take so long to answer. My email is out of control. A few questions come to mind.
What kind of data is present from the manufacturing and defects? What form does the current knowledge representation take? Along what dimensions would you like to refine it? What is the outcome, preferably somewhat quantifiable, that you want to achieve by this effort?
I don’t know if you have been tracking any of my threads or the comments I have made regarding the thoughts of others. My focus is that of creating self-motivated entities. I don’t know if this is what you have in mind. I think what applies here are the elements of predictability that the Heuristic approach offers, however. At this stage is it possible to ask what kind of data will be available from the power plants?
You asked about my interest in AI. I work in the fields of human capabilities augmentation, collaboration, collective intelligence, global sensemaking, solving complex, wicked problems. There are roles for varieties of AI technologies in there. ~Jack