help-octave
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Octave: System Identification for First order delay and dead time


From: Juan Pablo Carbajal
Subject: Re: Octave: System Identification for First order delay and dead time
Date: Fri, 30 Sep 2016 15:54:05 +0200

On Fri, Sep 30, 2016 at 3:05 PM, Noguchi_g-Mail <address@hidden> wrote:
> Dear Juan,
>
>
> Thank you for your information. It is a good news if I can identify the 
> process gain, first order delay time constant and dead time at onece as same 
> as Matlab system Identification tool box.
>
> Are there any example of the optimization code so that I should start from?
>
> I have no experence using Octave. I should start from now.
Does this help?
https://www.gnu.org/software/octave/doc/v4.0.0/Nonlinear-Programming.html

>
>
> Thanks.
>
>
> Noguchi
>
> ----- 元のメッセージ -----
> 差出人: "Juan Pablo Carbajal" <address@hidden>
> 宛先: "Yoshikazu Noguchi" <address@hidden>
> Cc: "Help GNU Octave" <address@hidden>
> 送信済み: 2016年9月30日(金曜日) 09:19:38
> 件名: Re: Octave: System Identification for First order delay and dead time
>
> On Thu, Sep 29, 2016 at 4:54 PM, Yoshikazu Noguchi
> <address@hidden> wrote:
>> Hello,
>>
>>
>>
>> I am looking for an Octave command which is similar to Matlab System ID tool
>> box.
>>
>>
>>
>> I would like to obtain values of parameters, Tau, K and Td for following
>> first order delay + Dead time model from time series data.
>>
>> ymodel = (K/(Tau*s+1))*exp(-Td*s)*u
>>
>> ymodel: process output, u: process input
>>
>> SISO continuoust time
>>
>>
>>
>> Object function: Min ( (y-ymodel)^2)
>>
>>
>>
>> Could you please tell me which Octave command I should use to solve this
>> issue?
>>
>>
>>
>> Best Regards,
>>
>>
>> _______________________________________________
>> Help-octave mailing list
>> address@hidden
>> https://lists.gnu.org/mailman/listinfo/help-octave
>>
>
> You can use the nonlinear optimization functions in the optim package,
> or sqp (and many others) in core.
> Also, I recommend to work on the log of your signals, y ymodel; if
> that is possible (i.e. only positive values of both).



reply via email to

[Prev in Thread] Current Thread [Next in Thread]