基于RLS的永磁同步电机在线参数辨识仿真模型

function[sys,x0,str,ts]=RLS_ident2(t,x,u,flag) theta0=[0.05,0.01]; %被辨识参数的初值，取充分小的实向量 Pn0=0.0008*eye(2); % 协方差矩阵，取充分大的实数为0.5 switch flag, case 0, [sys,x0,str,ts]=mdlInitializeSizes(theta0,Pn0); case 2, sys=mdlUpdate(t,x,u); case 3, sys=mdlOutputs(t,x,u); case {1,4,9} sys=[]; otherwise error(['Unhanded flag=',num2str(flag)]); end %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% function[sys,x0,str,ts]=mdlInitializeSizes(theta0,Pn0) sizes=simsizes; sizes.NumContStates=0; sizes.NumDiscStates=6; sizes.NumOutputs=2; sizes.NumInputs=3; sizes.DirFeedthrough=0; sizes.NumSampleTimes=1; sys=simsizes(sizes); x0=[theta0';Pn0(:)];% 需要更新的状态向量的初值 str=[];ts=[-1,0]; % 继承输入信号的采样时间 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% function sys=mdlUpdate(t,x,u) h=[u(1),u(2)]'; %采集数据 Pn0=reshape(x(3:end),2,2); %从状态变量中分离出p0 K=Pn0*h/(1+h'*Pn0*h); %计算增益矩阵 Pn1=(Pn0-K*h'*Pn0); %计算下一个协方差阵 theta0=x(1:2); %从状态变量中分离出theta0 theta1=theta0+K*(u(3)-h'*0.999*theta0); %计算下一个theta sys=[theta1;Pn1(:)]; %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% function sys=mdlOutputs(t,x,u) theta1=x(1:2); sys=theta1;