Forums before death by AOL, social media and spammers... "We can't have nice things"
|    comp.ai.fuzzy    |    Fuzzy logic... all warm and fuzzy-like    |    1,275 messages    |
[   << oldest   |   < older   |   list   |   newer >   |   newest >>   ]
|    Message 557 of 1,275    |
|    Bruce Postlethwaite to All    |
|    Re: Two question about fuzzy logic    |
|    02 Feb 06 15:44:42    |
      From: NOSPAM_b.postlethwaite@strath.ac.uk              My interest is mainly in model-based control, but I've played with       rule-based controllers too.              1. Fuzzy logic is only really worth thinking about if you have significant       non-linearity in the control problem. If the system you are controlling is       pretty much linear, then you will get to an acceptable controlled response       quicker and easier using standard linear controllers (like PID). Note that       even some very non-linear systems, like pH control, can be successfully       handled using standard PID controllers if the range of variation in the       system isn't too large (the gains stay pretty much the same in the region of       interest). Where fuzzy logic is really useful, and this is the area I'm       interested in, is when you have a multivariable non-linear system to       control.              2. In a non-linear application fuzzy control can give good performance       across the whole range of operation. A PID controller will give good       performance in the region it is tuned for, but will break down as the system       moves away from this point. This can, of course, be dealt with in many cases       by gain scheduling, and this is often easier to do than setting up a fuzzy       controller.              Bruce               |
[   << oldest   |   < older   |   list   |   newer >   |   newest >>   ]
(c) 1994, bbs@darkrealms.ca