open_system('scdspeedctrl') How can I get the state space equations for this system?

Question

Aikaterini Tsahalou 2016-8-13

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/299513-open_system-scdspeedctrl-how-can-i-get-the-state-space-equations-for-this-system

回答： Arkadiy Turevskiy 2016-8-19

Hello,

I am a student and I have a question about one of your control system designs:

Link: http://www.mathworks.com/help/slcontrol/examples/single-loop-feedback-prefilter-compensator-design.html

How can you get the state space equations? I tried linearizing the system to get the state space equations but I came up with some very big numbers in the matrix.

These are the steps that I am following:

Linearize the speed control system by removing the: 1. Reference Filter 2. Summing junction to the right of the reference filter 3. PID controller 4. External disturbance 5. Summing junction under the external disturbance

I make it "Open loop input" and "Open loop output" and I select: Analysis --> Control Design --> Linear Analysis...

(See picture "PIC1" for details of the system)

(See picture "PIC2" for details of the state space equations from Simulink after linearization)

Please, let me know if I am doing something wrong....

Best Regards,

Katerina Tsahalou

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Arkadiy Turevskiy 2016-8-19

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/299513-open_system-scdspeedctrl-how-can-i-get-the-state-space-equations-for-this-system#answer_232270

Big numbers don't mean there is a problem. There is nothing really wrong with what you are doing. To linearize the plant model you do not need to remove the elements you did remove. Simply place open loop linearization input at plant input, open loop linearization output at plant output, like you did. Then open Linear Analysis tool, and click "Edit Models IOs". Leave the open loop input and open look output you specified as active ones, and uncheck the rest. Now you can linearize the section of the model you want. Again,nothing wrong with numbers you are getting. If you want to convince yourself that the linear system you got captures model dynamics accurately, you can validate linearization results as discussed here .