PID vibration controller simulation

[Md_Salem]

Hello, everyone,

I am using ANSYS 25 R1 to simulate PID vibration controller of a cantilever plate. The sensors measuring STRIAN and the actuators are piezoelectric patches attached to the surface of the plate. I used coupled field transient module to do the simulation. I input 20 N impact force at t=0.0005 s, and the simulation continued until t=0.1 s. I then used the next APDL macro at the solution section to simulate the controller :

"/post1

set,LAST

ex1=1675   ! element at which X-strain sensor 1

ex2=1680  ! element at which X-strain sensor 2

ey1=1153   !  element at which Y-strain sensor 1

ey2=1165   !  element at which Y-strain sensor 2

! Weights update at each time step based on strain magnitudes

*SET, fhigher, 185 ! Highest frequency (Hz)

DT = 1/(20*fhigher) ! Time step

*SET, TS , 0.1 ! Simulation time

*SET, integ , 0 ! Integral term

*SET, de , 0 ! Previous error

*SET, uL , -150

*SET, uH , 150 ! Voltage clamps

*SET, eps , 1e-12 ! Small value to avoid division by zero

*DO, t, DT, TS, DT

! --- Measure strains ---

*GET,StrainX1,ELEM,1675,EPTO,X ! X-strain sensor 1

*GET,StrainX2,ELEM,1680,EPTO,X ! X-strain sensor 2

*GET,StrainY1,ELEM,1153,EPTO,Y ! Y-strain sensor 1

*GET,StrainY2,ELEM,1165,EPTO,Y ! Y-strain sensor 2

! --- Compute dynamic weights based on absolute strains ---

*SET, absX1 = ABS(StrainX1)

*SET, absX2 = ABS(StrainX2)

*SET, absY1 = ABS(StrainY1)

*SET, absY2 = ABS(StrainY2)

*SET, total = absX1 + absX2 + absY1 + absY2 + eps

*SET, wX1 = absX1 / total ! Weight for X-sensor 1

*SET, wX2 = absX2 / total ! Weight for X-sensor 2

*SET, wY1 = absY1 / total ! Weight for Y-sensor 1

*SET, wY2 = absY2 / total ! Weight for Y-sensor 2

! --- Combine errors with dynamic weights ---

x = wX1*StrainX1 + wX2*StrainX2 + wY1*StrainY1 + wY2*StrainY2

! --- PID control ---

*SET,kp =1e3

*SET,ki =1e2

*SET,kd =1e1

*SET,kd =kd(0.00005/0.000263)

*SET,integ = 0

*SET, prev_err = 0

integ = integ + x*DT ! Integral term

diff = (x - de)/DT! Derivative term

U = kp*x + ki*integ + kd*diff ! Control voltage

! Clamp voltage

*IF, U, LT, uL, THEN, U = uL

*ELSEIF, U, GT, uH, THEN, U = uH

*ENDIF ! Apply voltage to PZT patch

ESEL,S,ENAME,,226 NSEL,R,LOC,Z,1e-3

D,ALL,VOLT,U

ALLSEL

de = x ! Update previous error

TIME, t

SOLVE

*ENDDO "

The first part of the simulation (the impact force effect) was done properly, and I could see the deformed shape in the solution section. But the previous code didn't run. When I look in the "solution information" section, I find an error. 

*** ERROR *** CP = 1.812 TIME= 17:40:29 Unknown label in field 5 ( EPTO ) of GET command. Line= GET,StrainX1,ELEM,1675,EPTO,X The *GET command is ignored. 

And then, the rest of the code didn't run. I need help to fix that error, please.

Regards

 

[Stella Peloni]

Hello,

Indeed, this issue/Error comes from the fact that *GET command for elements (ELEM) does not support EPTO (total strain). In the Documentation, in this link, in Table 161,you can check all the valid Items for Elements, when using *GET commnand. EPTO item is valif as averaged nodal result on selected elements as follows:

*GET,name,NODE, Node_Number,EPTO,X

To be more precise, you need to select prior issuing *get command, the element of interest and then select the node of this element to *get the EPTO value.

Thank you and have a great week ahead!

Kind Regards,

Stella

[ravef63622]

If you are looking for a reliable casino https://www.casino-sistersite.co.uk/jackpotmobile/ with good returns, this is the place to go. I've been playing here for a few months now, and everything works stably. I often win small amounts, but the main thing - honestly and without cheating. Bonuses come regularly, especially nice to get personalized ones. Suitable for beginners and pros alike.