what are the alternative PID controllers?

crutschow

Joined Mar 14, 2008
34,285
You can use Fuzzy Logic which is a digital technique, or PID which is an analog technique (but can be emulated in a digitial processor).
Fuzzy PID is a mishmash of the two.

To compensate the control loop you need to know the response parameters of your particular converter, and design the compensation circuit (either with Fuzzy Logic or PID) accordingly.
It's not a one-size-fits-all.
 

Papabravo

Joined Feb 24, 2006
21,159
Just for the record, stable is NOT the same thing as constant. A system can be STABLE and have an output that changes with time, especially if the change is periodic. In the context of controllers UNSTABLE means specifically that the magnitude of the output increases WITHOUT BOUND.. This increase is usually exponential. Let's get down with using the correct terminology.
 

ci139

Joined Jul 11, 2016
1,898
makes the mean value (The output mean value) almost constant
basically -- all times you assume you have the constant output /!\ you are wrong /!\ <- there are more than one way such can manifest (by surprise) ,
+ all times you try to reach one - you'll be wasting more resources than it is appropriate for the reached output precision
————————————————
it is somewhat a bad definition/idea -- there is always a possible transient event the adjustment loop can't cope with = you can define it "constant" for a specified range of conditions only
 
Last edited:

Deleted member 115935

Joined Dec 31, 1969
0
I have a DC output voltage that I generate from full bridge converter with a rectifier then sensed and sent to a processor to calculate its mean value.
The output mean value that I calculate, I would like to use a controller that makes the mean value almost constants or change in very slow mode.
I have tried to activate the I only, then I tried to deactivate both P And I to watch the reading of my value and It seems even when I have a constant output controller I have little variation(around 100mV) in the reading of my mean value.
I am reading about fuzzy controller and fuzzy PID controller if there are possibilities to make more stable
to go back to the original post,
any chance of a schematic please,

It sounds to me ,

can I confirm , your taking the AC, and rectifying it ?


ac rectify .JPG

If so , is this "mains" AC of from another signal source ?

How are you measuring and calculating the mean ?

Once you have this mean,
you say you want t control its level, how are you affecting it ?
 

Deleted member 115935

Joined Dec 31, 1969
0
You are right , I have a mains AC that I can control ,then pass it through a transformer, then through a rectifier and then a capacitor that provides me a DC output with small oscillations, then I have a voltage divider that I pass the signal again through a DSP chip that calculate the mean value of my signal. it is this mean vale that I want to be constant at 99.95%.
Now, I opted for adaptive fuzzy PID controller but I am having difficulty in writing the membership function (triangular form) in C code,snce I am implementing it in Microcontroller

out of interest, how are you controlling the AC mains
Are you adjusting the amplitude, or what ?
 

Deleted member 115935

Joined Dec 31, 1969
0
So What is your means for adjusting the AC mains in that you mention your trying to control ?
normal I'd expect not to need to do the rectification etc your mentioning, but just sample the mains with the processor system direct, using a suitable transformer to drop the AC down to that needed by your circuit.
 

soyez

Joined Aug 17, 2020
51
I have a DC output voltage that I generate from full bridge converter with a rectifier then sensed and sent to a processor to calculate its mean value.
The output mean value that I calculate, I would like to use a controller that makes the mean value almost constants or change in very slow mode.
I have tried to activate the I only, then I tried to deactivate both P And I to watch the reading of my value and It seems even when I have a constant output controller I have little variation(around 100mV) in the reading of my mean value.
I am reading about fuzzy controller and fuzzy PID controller if there are possibilities to make more stable
use some fuzzy digital technique it would be helpful
 
Top