From: walter@bytecraft.com   
      
   FL normalizes parameters of interest into a Degree of Membership which often   
   makes FL solutions less computationally intensive. FL rules tend to produce   
   smooth results that are a composite of  many considerations. It is common for   
   example to include rules   
   of marginal interest that will improve the overall solutions. Adding rules to   
   a home environment control system that takes time and outside conditions as it   
   makes decisions on furnace and air condition operations is quite easy in FL.   
      
      
   FL is very well suited to non linear control applications. PID controllers are   
   difficult to set-up for cases  where some of the parameters are changing. I   
   posted a fuzzy control system to replace some PID controllers here a few days   
   ago. It has about a   
   half dozen rules and is very flexible.   
      
   w..   
      
      
      
      
   sally.zahav@gmail.com wrote:   
      
   > Hi All.   
   > I've recently developed two fuzzy logic controllers for my workplace.   
   > They work well, but some people are still not convinced it's for us.   
   > Since I'm new to FL (and controls in general) I don't have answers for   
   > the skeptical.  The company I work for has applications used in   
   > irrigation, that control electroconductivity and pH levels, and we are   
   > considering using FL to control temperature/cooling.  Here are the   
   > questions:   
   > 1. Which kind of applications are especially suited to fuzzy logic, and   
   > which are not?   
   > 2. How does FL compare with PID, as far as obtaining the target value   
   > quickly with a minimum of overshoot and minimal steady state error?   
   > Thanks,   
   > Sally   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)