![]() |
bonjour je te demande de m'envoyer les bloc de simumlation en matlab simulink de la regulation de la vitesse d'un moteur asynchrone en utilisnt logique flou adaptatif au sein de la commande vectorielle merci
Ne gardez plus qu'une seule adresse mail ! Copiez vos mails vers Yahoo! Mail |
![]() |
Dear Madame
I have made a model of a squirel-cage induction motor, a SV-PWM inverter and the algorithm of direct torque control (DTC) in SCICOS. Now, I am investigating a better control algorithm based on a synchronous reference of a stator current and voltage.
I hope someone can translate it in french.
Regards
Raymundo Cordero G.
CIEEP-Perú
2008/1/28, Anissa Raheme <[hidden email]>:
|
![]() |
Dear Raymundo,
my name is Simone Mannori. I'm Italian, so this means that I'm just a man and not a woman. Now that we have fixed this little detail, we can pass to our real core business :) With Scilab/Scicos you can model and simulate the full electro mechanical system: mechanical load, motor, power drive, PWM generation and control circuits. Several years ago I made a full simulation using "the other" non open source simulator ad I manually coding the control algorithm for a TI DSP (TMS 320F240) using C and assembly language. If you ask if we have a ready made Scicos model for this kind of application the answer is, unfortunately, no. If you have developed a model and you are looking for help or suggestions to optimize your simulation, please send us your files. After that we will be very pleased to discuss your specific requirements. Simone Mannori - Scilab/Scicos Embedded Applications Eng. On Mon, 2008-01-28 at 21:54 -0500, Cordero R. wrote: > Dear Madame > I have made a model of a squirel-cage induction motor, a SV- > PWM inverter and the algorithm of direct torque control (DTC) in > SCICOS. Now, I am investigating a better control algorithm based on a > synchronous reference of a stator current and voltage. > I hope someone can translate it in french. > Regards > Raymundo Cordero G. > CIEEP-Perú > > 2008/1/28, Anissa Raheme <[hidden email]>: > bonjour > je te demande de m'envoyer les bloc de simumlation en matlab > simulink de la regulation de la vitesse d'un moteur asynchrone > en utilisnt logique flou adaptatif au sein de la commande > vectorielle > merci > > > > > > > > ______________________________________________________________ > Ne gardez plus qu'une seule adresse mail ! Copiez vos mails > vers Yahoo! Mail > > |
![]() |
Dear Simone,
Firstly, I want to apologize for my error. I want to answer a email from Anissa Raheme (that is why I use madame). I made a model of a squirel cage induction motor for my thesis and because I work in a research center in my country, Perú: CIEEP,
I compared my model with the model in the SimPower system of MATLAB, with the same results.
In my model, I enter the values of resistors, inductances, pair of poles, friction, moment of inertia, etc. in the Diagram->context of SCICOS. There is a variable, Lx, which is automatically calculated and its part of the model. In the future, I would like to make a library like the SimPower System of MATLAB. Now, I am making libraries of field oriented control (FOC) and direct torque control (DTC) and SV-PWM inverter, and make lots of publications in the IEEE using SCILAB, because nobody in my country has made something like that.
I send you my model.
Raymundo Cordero
CIEEP-Perú
PD: On 30th January, I will give a videoconference in two universities of Perú about SCILAB as an open-source tool in electronic engineering. The objective is to promote the use of SCILAB in universities. 2008/1/29, Simone Mannori <[hidden email]>:
Dear Raymundo, |
![]() |
Dear Raymundo,
There is a recurring problem using Scilab/Scicos for engineering applications: the identification of systems parameters. One of the difficulties of the models used in SimPower is the mathematical/physical modeling used in the block. The Italian electrical engineers (and the industries that make motors and generators) use slightly different model from the Simulink one. Unfortunately there is not 1:1 correspondence between them. Usually, we prefer build the model from scratch using elementary blocks. If you are interested to develop a power system toolbox for Scicos you are very welcome :) Scicos-PST ? We are working in close contact with Microchip for Scicos-FLEX. Microchip have several product lines for power system applications and they have expressed their interest to extend FLEX to motor control. Another interesting point is the modeling of the power section (MOSFET or IGBT bridge) using Modelica. This kind of simulation could be extremely useful because you can study the behavior of the power section under stress (e.g. when you have a short circuits between phases or very long connection cables). I suppose that you are aware of the interaction between the control strategy and the EMC issues. In Europe the EMC is a pass/fail strict law. If you are interested in this field - as I am - we can start a productive collaboration. Thanks again for using Scilab/Scicos :) Simone Mannori - Scilab/Scicos Embedded Applications Eng. P.S. May I join - of just see - your video conference ? On Tue, 2008-01-29 at 16:34 -0500, Cordero R. wrote: > Dear Simone, > > Firstly, I want to apologize for my error. I want to answer a email > from Anissa Raheme (that is why I use madame). I made a model of a > squirel cage induction motor for my thesis and because I work in a > research center in my country, Perú: CIEEP, > > I compared my model with the model in the SimPower system of MATLAB, > with the same results. > In my model, I enter the values of resistors, inductances, pair of > poles, friction, moment of inertia, etc. in the Diagram->context of > SCICOS. There is a variable, Lx, which is automatically calculated and > its part of the model. > > In the future, I would like to make a library like the SimPower System > of MATLAB. Now, I am making libraries of field oriented control (FOC) > and direct torque control (DTC) and SV-PWM inverter, and make lots of > publications in the IEEE using SCILAB, because nobody in my country > has made something like that. > > I send you my model. > > Raymundo Cordero > CIEEP-Perú > > PD: On 30th January, I will give a videoconference in two > universities of Perú about SCILAB as an open-source tool in electronic > engineering. The objective is to promote the use of SCILAB in > universities. > > > 2008/1/29, Simone Mannori <[hidden email]>: > Dear Raymundo, > > my name is Simone Mannori. I'm Italian, so this means that I'm > just a > man and not a woman. Now that we have fixed this little > detail, > we can pass to our real core business :) > > With Scilab/Scicos you can model and simulate the full electro > mechanical system: mechanical load, motor, power drive, PWM > generation > and control circuits. > > Several years ago I made a full simulation using "the other" > non open > source simulator ad I manually coding the control algorithm > for a TI DSP > (TMS 320F240) using C and assembly language. > > If you ask if we have a ready made Scicos model for this kind > of > application the answer is, unfortunately, no. > > If you have developed a model and you are looking for help or > suggestions to optimize your simulation, please send us your > files. > > After that we will be very pleased to discuss your specific > requirements. > > Simone Mannori - Scilab/Scicos Embedded Applications Eng. > > > On Mon, 2008-01-28 at 21:54 -0500, Cordero R. wrote: > > Dear Madame > > I have made a model of a squirel-cage induction motor, a > SV- > > PWM inverter and the algorithm of direct torque control > (DTC) in > > SCICOS. Now, I am investigating a better control algorithm > based on a > > synchronous reference of a stator current and voltage. > > I hope someone can translate it in french. > > Regards > > Raymundo Cordero G. > > CIEEP-Perú > > > > 2008/1/28, Anissa Raheme <[hidden email]>: > > bonjour > > je te demande de m'envoyer les bloc de simumlation > en matlab > > simulink de la regulation de la vitesse d'un moteur > asynchrone > > en utilisnt logique flou adaptatif au sein de la > commande > > vectorielle > > merci > > > > > > > > > > > > > > > > > ______________________________________________________________ > > Ne gardez plus qu'une seule adresse mail ! Copiez > vos mails > > vers Yahoo! Mail > > > > > |
![]() |
Dear Simone
In 1 month I will start my postgrade in Brazil and I am going to use SCILAB for modeling power system. As a result, I will make the "SCICOS-PST". Meanwhile, I almost finished a model of a SVPWM, SPWM.Park, Clarke transform and THD. The video conference finished today. But I can organize another one in february, where you can participate, even as an expositor. The main problem is that if you speak spanish. Regards Raymundo 2008/1/30, Simone Mannori <[hidden email]>: Dear Raymundo, |
![]() |
Bonjour Raymundo,
> In 1 month I will start my postgrade in Brazil and I am going to use SCILAB for modeling power system. > As a result, I will make the "SCICOS-PST". Very good :) > > Meanwhile, I almost finished a model of a SVPWM, SPWM.Park, Clarke > transform and THD. OK. > The video conference finished today. But I can organize another one in > February, where you can participate, even as an expositor. The main > problem is that if you speak Spanish. Not yet :(. (English, French, Italian). Back to our core business Scicos-PST. For the moment you can develop using the latest Scilab 4.1.2 but, before start to develop a serious toolbox we need to to sure that it will be compatible with the next Scilab 5.0 in the form and in the substance. SUBSTANCE A good starting point is to discuss the application field. I'm understanding that you are working on electrical motor drive. In this field ( ops :) ) motor's modeling is the most critical point. Define a pretty theoretical model could be a good idea for a paper, but the real challenge is to define a model sufficiently close to reality for the applications with parameters easy to extract from real world, clearly defined, measurements. "Simple, but not simpler". >From this point of view I'm not sure that the Simulink approach is the best one. Anyway, if you have positive experience on that, we can use a dual/mode Scicos model or two different blocks with the same layout but with different computational functions. FORM: It is up to you decide the modality of development, distribution and licensing of you toolbox. If you will join the community of Scilab/Scicos developers you will have more options and opportunities respect to the isolate individual. Feel be free to directly contact Sylvestre Ledru (he is in CC) for any question about joining the community. Best Regards Simone Mannori - Scilab/Scicos Embedded Applications Eng. |
![]() |
Dear Simone
At the moment, I also am working in the proyect about symbolic computation (difficult but interesting),
However, motor control and modeling is part of my thesis and a previous research work, so this is something I will do.
Now, I have a technical question. I could make scicos models using superblock, but I would prefer to make new blocks using C/C++. But I have problems because in my computer I don´t have a C compiler. Which one I could use?
Raymundo
2008/1/31, Simone Mannori <[hidden email]>:
Bonjour Raymundo, |
![]() |
Bonjour Raymundo,
On Thu, 2008-01-31 at 16:00 -0500, Cordero R. wrote: > Dear Simone > At the moment, I also am working in the proyect about symbolic > computation (difficult but interesting), > However, motor control and modeling is part of my thesis and a > previous research work, so this is something I will do. > Now, I have a technical question. At last ;) I like technical question :). > I could make Scicos models using super block, but I would prefer to > make new blocks using C/C++. ... and it is the best option ... > But I have problems because in my computer I don´t have a C compiler. If you want to do serious development with Scilab/Scicos would suggest you to switch to Linux. Linux has A LOT of advantages for the developers. The default "gcc" compiler is one of the most important. Scilab/Scicos is born in Unix, grow in Linux and ported to Windows and Mac OSX. I have done some development also using Windows but is just a bit more tricky for the missing default C compiler. My suggested solution for Scicos development under Windows it is to install Scilab 4.1.2 binary version and the free (free as "free beer") Visual Studio Express 2005. You need to install manually some packages and patch some files. Microsoft has prepared some usable help pages ... Start from here: http://wiki.scilab.org/Visual_Studio_C%2B%2B_2005_Express then go here http://msdn2.microsoft.com/en-us/express/aa700755.aspx for a detailed explanation of the "hack" After the installation you will able to compile Scicos block computational functions. BEWARE: with this set up you will not able to compile the full Scicos, but if you are interest mainly to Scicos, this is not a big limitation. The C compiler is useful (and necessary) for Modelica. How to write good and efficient Scicos computational functions will be the subject of the next email :). Simone Mannori - Scilab/Scicos Embedded Applications Eng. > Which one I could use? > Raymundo > > 2008/1/31, Simone Mannori <[hidden email]>: > Bonjour Raymundo, > > > In 1 month I will start my postgrade in Brazil and I am > going to use SCILAB for modeling power system. > > As a result, I will make the "SCICOS-PST". > > Very good :) > > > > > Meanwhile, I almost finished a model of a SVPWM, SPWM.Park, > Clarke > > transform and THD. > > OK. > > > The video conference finished today. But I can organize > another one in > > February, where you can participate, even as an expositor. > The main > > problem is that if you speak Spanish. > > Not yet :(. (English, French, Italian). > > Back to our core business Scicos-PST. > > For the moment you can develop using the latest Scilab 4.1.2 > but, before > start to develop a serious toolbox we need to to sure that it > will be > compatible with the next Scilab 5.0 in the form and in the > substance. > > SUBSTANCE > A good starting point is to discuss the application field. I'm > understanding that you are working on electrical motor drive. > In this > field ( ops :) ) motor's modeling is the most critical point. > Define a pretty theoretical model could be a good idea for a > paper, but > the real challenge is to define a model sufficiently close to > reality > for the applications with parameters easy to extract from real > world, > clearly defined, measurements. "Simple, but not simpler". > >From this point of view I'm not sure that the Simulink > approach is the > best one. > Anyway, if you have positive experience on that, we can use a > dual/mode > Scicos model or two different blocks with the same layout but > with > different computational functions. > > FORM: > It is up to you decide the modality of development, > distribution and > licensing of you toolbox. > If you will join the community of Scilab/Scicos developers you > will have > more options and opportunities respect to the isolate > individual. > Feel be free to directly contact Sylvestre Ledru (he is in CC) > for any > question about joining the community. > > Best Regards > > Simone Mannori - Scilab/Scicos Embedded Applications Eng. > > > > > > > |
Free forum by Nabble | Edit this page |