Support

This project is supported by the Estonian Doctoral School in ICT under interdisciplinary project “FOMCON” and Estonian Science Foundation under grant nr. 8738.

12 Comments

  1. Posted February 4, 2014 at 16:46 | Permalink

    Respected Sir

    i want to know can we add our program for optimization in it or not for any fractional order transfer function
    please suggest us

    • Posted February 5, 2014 at 09:35 | Permalink

      Yes, this is possible. Please note, that optimization based tuning for models of the form
      \begin{eqnarray}
      G(s)=\frac{b_{m}s^{\beta_{m}}+b_{m-1}s^{\beta_{m-1}}+\cdots+b_{0}s^{\beta_{0}}}{a_{n}s^{\alpha_{n}}+a_{n-1}s^{\alpha_{n-1}}+\cdots+a_{0}s^{\alpha_{0}}}.
      \tag{1}\end{eqnarray}
      is already implemented in the optimization tool.

  2. Posted May 3, 2014 at 11:11 | Permalink

    Dear Aleksei,
    Is there any toolbox available for tuning of fraction order lead-lag controller in matlab environment?
    if there is…. please give the details.
    thanks in advance

  3. Posted December 22, 2014 at 06:44 | Permalink

    While designing fractional order controller for a conical tank system( linearized conical tank transfer function, for that I performed pade approximation), the final controller equation have negative coefficients in denominator. Is there any problem in stability because of that, since negative poles leads to instability.

  4. Posted April 11, 2015 at 14:03 | Permalink

    Sir, How to use fomcon for following transfer function:
    :
    G(s)=(-25)/(s^2-2180)

  5. Posted August 18, 2015 at 16:13 | Permalink

    Dear Alexsei,
    How can I plot step response of a “delayed” fractional order system? in Simulink, I have used frac Tf(s) block but I can’t use delay block (z^-1) with it to add the delay to plan. also my delay value is not an integer value.

    Thanks.

  6. Posted September 30, 2015 at 21:35 | Permalink

    Dear Aleksei
    I have problem in mobile robot control with FOPID controller optimization in PSO algorithm if the sampling time 0.5 the simulation is stopped but if the sampling time 0.001 the simulation is run with out problem
    How can I run the simulation with sampling time 0.5

  7. Posted February 19, 2016 at 20:58 | Permalink

    Dear sir:
    i have a simulink model and i tried to use the toolbox for tuning the FOPID controller parameters’. An error is always appeared
    ?? Attempted to access y_sim_all(:,1); index out of bounds because size(y_sim_all)=[0,0].

    Error in ==> fpid_optimize_sim at 175
    y = y_sim_all(:,1); % System response

    Error in ==> fpid_optimfun at 80
    [y, ~, t] = fpid_optimize_sim([Kp,Ki,lam,Kd,mu], …

    Error in ==> fpid_optimize>@(x)fpid_optimfun(x,G,opt) at 292
    x_opt = fmincon(@(x) fpid_optimfun(x,G,opt), …

    Error in ==> fmincon at 574
    initVals.f = feval(funfcn{3},X,varargin{:});

    Error in ==> fpid_optimize at 292
    x_opt = fmincon(@(x) fpid_optimfun(x,G,opt), …

    Error in ==> fpid_optim>btnOptimize_Callback at 270
    [Kp Ki Kd lam mu Z] = fpid_optimize(fsim, fopt, [], usesim);

    Error in ==> gui_mainfcn at 96
    feval(varargin{:});

    Error in ==> fpid_optim at 24
    gui_mainfcn(gui_State, varargin{:});

    Error in ==>
    @(hObject,eventdata)fpid_optim(‘btnOptimize_Callback’,hObject,eventdata,guidata(hObject))

    Caused by:
    Failure in initial user-supplied objective function evaluation. FMINCON cannot continue.

    ??? Error while evaluating uicontrol Callback

    • Posted February 21, 2016 at 14:01 | Permalink

      Your Simulink model does not correctly specify an output which is to be used for optimization.
      Please use the GUI for the fpid_optim() tool to create a new model and see how it is configured.

  8. Posted January 9, 2017 at 18:10 | Permalink

    Dear Aleksei,
    Is it possible to convert oustaloup refined filter object (zpk) to the state space format (frac_ss) or to fotf object format.
    I am looking forward to hearing from you soon.
    Victor

    • Posted January 14, 2017 at 15:28 | Permalink

      If you want to convert a simple FO operator approximation, then the procedure is straightforward.
      However, for more complicated objects there is no direct way for getting back the FO description. You can consider frequency domain identification since the complete response is known.

Post a Comment

You must be logged in to post a comment.