Re: help needed PID control calculations
From: Edward Lee epl (me_at_linnix.com)
Date: 05/06/04
- Next message: Bryan Hackney: "Re: Serial port monitor"
- Previous message: Scott Brandt: "RTSS 2004 CFP (last call)"
- In reply to: Rene Tschaggelar: "Re: help needed PID control calculations"
- Next in thread: Jack Klein: "Re: help needed PID control calculations(BRUSHLESS DC MOTOR CONTROL)"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Date: 6 May 2004 07:29:24 -0700
Rene Tschaggelar <none@none.net> wrote in message news:<409a0c55$0$698$5402220f@news.sunrise.ch>...
> Sulaiman mohd wrote:
>
> > Hi ALL,
> >
> > I am mohd sulaiman. I am working on Brushless dc Motor
> > position control.In my application we are controlling the motor using
> > the dsp processor from TEXAS i.e TMS320LF2407A. my design have been
> > completed now i have startedd to write the code for the application.
> >
> > I have doubt regarding the implementatation of the PID
> > and PI control in software and how do we determine the Respective
> > constants of the PID or PI controller. could any one explain on what
> > parameters of the motor they are dependent and how do we determine
> > them.
>
> You're familiar with analog control loop theory ?
> Open loop response and closed loop response of the system ?
> If your sampling frequency is far above the bandwidth you can
> proceed if it was analog, else you have to pull it over
> to the digital domain by the means of the Z-tranform.
Z-transform is the discrete version of the lapplace transform. If
your sampling frequency is too low, you are cutting off the high
freqency responses. That's true for both digital and analog.
Let me try some simpler terms. Basically, any analog wave form can be
modeled in a sequence of fundamental frequencies. If you pick n of
the important ones, you can simulate the characteristic of the system
with N differential equations, or a Nth order matrix. Then your
system is in the form of
Y(t) = M(t)X(t) + A(t), where M(t) is NxN and A(t) is Nx1
You just need a do some matrix calcuations (computer can do that very
well).
If you need additional details, please email me (epl). I don't think
other people want to bother with these details.
>
> Rene
- Next message: Bryan Hackney: "Re: Serial port monitor"
- Previous message: Scott Brandt: "RTSS 2004 CFP (last call)"
- In reply to: Rene Tschaggelar: "Re: help needed PID control calculations"
- Next in thread: Jack Klein: "Re: help needed PID control calculations(BRUSHLESS DC MOTOR CONTROL)"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Relevant Pages
|
