pdmodel.asv

内模控制器（IMC）工具箱。包括参数整定、PID控制器参数转换等

function [pdnummt,pddenmt,pd_prime,pdxs]=pdmodel(Xpdnummt,Xpddenmt,vlb)
% function [pdnummt,pddenmt,pd_model,pds]=pdmodel(Xpdnummt,Xpddenmt,vlb)
% find the defult model of Pd for the qd calculation. It also claculate pds
% to be used in the partial sensitivity function for tuning of the
% disturbance rejection.
% Here, the model of Pd is a diagonal matrix containing the smallest lag on 
% each row of the original Pd process. Similarly, pds is a strig variable 
%that could be evaluated to have a  
% representation of diagnoal Pd with uncertainty.

[m,n]=size(Xpddenmt);
pddenmt=cell(m,n);
pd=pddenmt;
for i=1:m
   pddenmt{i,i}=mt2poly(Xpddenmt{i,i},vlb);
   for j=1:n
      temp=mt2poly(Xpddenmt{i,j},vlb);
      lentemp=length(temp);
      lenpdden=length(pddenmt{i,i});
      if lentemp >=2 & (temp(lentemp-1)/temp(lentemp) < ...
            pddenmt{i,i}(lenpdden-1)/pddenmt{i,i}(lenpdden))
         pddenmt{i,i}=temp;
      end
   end
   pd{i,i}=['1/' mtx2str(pddenmt{i,i},1)];
end