Hi Arash,

Observe that the step input that triggers at 1 sec (jumps from 0 to 0.1 within a very short time). The rate of change on error causes a steep slope and results in an undesirable control action known as derivative kick.

To solve this problem, you may add a tracking differentiator or a simple 1st-order low-pass filter to generate a smooth signal.

Dear Yew-Chung,

according to your advise, I design a low-pass filter, but it did not make much difference.

pleas look at following pictures:

Hi Arash,

What are the values of the PID gains?

Dear Yew-Chung,

I follow the 2 links below:

• http://ctms.engin.umich.edu/CTMS/index.php?example=Suspension§ion=SimulinkModeling
• http://ctms.engin.umich.edu/CTMS/index.php?example=Suspension§ion=SimulinkControl

Thanks for the links. In the control link, it doesn't say anything about the PID controller that you designed. But it tells that K = [0 2.3e6 5e8 0 8e6] is assigned to the feedback gain matrix K. This pretty much explains why the controller output is so high.

Yew-Chung Chak

You are welcome ;-)

I know that, because I design a PID control on my own ( as photo below ), and the output of that controller is high too. the parameter of the designed PID is:

• Kp = 4.5e+06
• Ki = 7.25e+07
• Kd = 6.98e+04

I want to know that system with this parameters can be implemented?

and for a vehicle, this outputs are normal ?

I'm no road vehicle expert, but I think it is possible to get a 50 - 100 hp engine to generate that high control output.

Dear Yew-Chung,

Thanks for your help.

I suggest you to use optimization technique, mainly one of evolutionary algorithms to tune the PID parameters.

Best regards

Dear Nizar, Thanks for your help.

Look at your PID gains and tune them optimally.