In case it is helpful, here are all my Control Theory videos in a single playlist ruclips.net/p/PLxdnSsBqCrrF9KOQRB9ByfB0EUMwnLO9o. Please let me know what you think in the comments. You can support this channel via Patreon at www.patreon.com/christopherwlum or by clicking on the 'Thanks' button underneath the video. Thanks for watching!
Hi James, thanks for the kind words. I have a more similar videos at ruclips.net/p/PLxdnSsBqCrrF9KOQRB9ByfB0EUMwnLO9o . Please feel free to check them out, I'd love to hear your thoughts as an academic. Thanks for watching!
This is how everything should be taught. Your lesson is surely more valuable than the whole control theory course at my university 🤣. Universities should be RUclipsrized in my opinion.
Hi Dr. Lum. As an first-year student who is learning about PID control, I would say that your videos about PID are really helpful for me. Thank you so much.
Thank you very much for this explanation. I'm not a math person so I could finally got this. But, as a beekeeper, I lose all after the honey waste... I'll watch again. Thanks from Brazil!
AE511: 47:07 It was great to see the comprehensive table of what manipulating Kp/KI/Kd gains will do for the overall system performance gains. When we briefly went over this in undergrad, my limited exposure was trial and error but this was useful in actually making targeted estimations (by hand, obviously there are tools like PID tuner that help out with this)
Thank you for breaking it down into understandable chunks. sometimes it's hard to understand abstract ideas without a visual representation! I'm a visual learner so your visual example really helped me understand and gave me that ah ha moment! great stuff. This will really help me further my career as a Multicraft technician.
Thank you Wery much! Tomorrow I have the final exam related to PID adjustments and I have been studying the subject for a really long time now. I understand the P and I and D controls separately, but it was the combination of these that seemed very challenging to figure out in practice. This video put the rest of the pieces in place and it's completely self-evident to me now! Thank you!
Hi, Thanks for the kind words, I'm glad you enjoyed the video. If you find these videos helpful, I hope you'll consider supporting the channel via Patreon at www.patreon.com/christopherwlum or via the 'Thanks' button underneath the video. Given your interest in this topic, I'd love to have you a as a Patron as I'm able to talk/interact personally with all Patrons. I can also answer any questions, provide code, notes, downloads, etc. on Patreon. Thanks for watching! -Chris
Quite strange but all videos about this topic where anyone is trying to give analogies with something are just confusing me more, but when you are on the white board everything became much clearer and for the first time a understood it.:)
39:10 Regarding the D component, it seems to me that something is off about the model. With the P and the I components, the force exerted was in the direction of reduced error (i.e., in the direction of "moving towards the set point"), yet with the D, the force that gets applied OPPOSES the motion towards the setpoint. Is this distinction relevant? Should we not be imagining D as a force that is proportional to the rate of change of position of the object AND moving the object towards the setpoint (rather than resisting said motion)?
I have a question about 27:30, wouldn’t the u(t) of the D response be negative after the initial delta response to the discontinuity? As the rest of the control drives the error down, wouldn’t it be acting as negative feedback to lower the control response to minimize overshoot?
Excellent demonstration - I knew functionally nothing about PID controllers but need to have some understanding not only for work but also for home (wanting to use PID for a home built coffee roaster). Honestly...I would've used oil - e.g. mineral oil instead of honey...much less messy but point taken LOL
Hi, Thanks for the kind words, I'm glad you enjoyed the video. If you find these videos helpful, I hope you'll consider supporting the channel via Patreon at www.patreon.com/christopherwlum or via the 'Thanks' button underneath the video. Given your interest in this topic, I'd love to have you a as a Patron as I'm able to talk/interact personally with all Patrons. I can also answer any questions, provide code, notes, downloads, etc. on Patreon. Thanks for watching! -Chris
Hi Dr. Lum! I love your videos, have you ever considered lecturing a course in Estimation and Tracking with applications in Aerospace Engineering? Topics could include Kalman Filtering, GNC, state estimation, etc. Would love to watch lectures from you on that topic!
Any hints on how to tune a PID controller for a recirculating chiller with a hot and cold PID controller? The default settings don't work (large steady state error). Randomly tweaking the PID parameters hasn't work so far.
In your analysis, the error signal starts out as a step signal and then decays. In the derivative case, when the error goes from zero to some value, ud(t) is large and positive, but when the error decays, ud(t) should be negative because the slope of the error signal is negative therefore the derivative is negative ...
One part I don't understand is where does the function equation {e.g. e(t)} come from which will be integrated or differentiated by the PID controller. The reason I'm asking this question is because the signal coming from the transmitter (feedback) to the controller is just an integer (4mA- 20mA). My guess is the equation are build into the controller based on models of the system behaver.
Dear Professor i think the spike of the derivative part has the sign inverted respect to the rest of the deirivative part because the sign of the derived step error is a positive delta instead the derived after the step goes to zero and then goes to negative values ....so changing the sign respect to the sign of the delta....it means that after a reset....a step error brings suddenly the derivative pid to slow down the control and after the step the control tends slowly to increaase ita own pushing capability.....anyway great video thanlk you
Hi, Thanks for the kind words, I'm glad you enjoyed the video. If you find these videos helpful, I hope you'll consider supporting the channel via Patreon at www.patreon.com/christopherwlum or via the 'Thanks' button underneath the video. Given your interest in this topic, I'd love to have you a as a Patron as I'm able to talk/interact personally with all Patrons. I can also answer any questions, provide code, notes, downloads, etc. on Patreon. Thanks for watching! -Chris
can you please tell the name of the signal after using PID controller? Sometimes, we use error between altitude and set point, and after using PID, it becomes speed . I wonder how? Am I the only one who cannot understand it properly.. please explain
It depends on what system specifically you are controlling. For instance, if you are controlling a motor to have a specific speed, it would be speed. But if you are controlling a motor to maintain a given position, like a servomotor, it would be position. The device after the PID controller in general, is called the plant, as in factory. It's a representation of the dynamics of the process you are controlling. It represents how the output of your controller is turned into a signal to drive the actuator (motor, piston, etc) that controls the system being controlled, and turns it into the physical measurement that is of interest to control. As an example, if you are controlling a DC servomotor to maintain a certain position, the plant dynamics would consist of a model for the coil inductance (L), the coil resistance (R), the rotational inertia of the motor and all that is attached to it (J), the motor field constant (K) that relates current to torque (and likewise speed to back EMF), and the drag it experiences (D). The controller would produce a voltage that would then indirectly determine the current that the motor coil receives, and the torque that propels it forward. Here's a diagram that show a DC motor's dynamics, and the interaction among all these terms: forums.mikeholt.com/attachments/1691248709486-png.2566732/ For a free-spinning motor (no external mechanical load), you can use this diagram to show that the dynamics of the motor can be represented by this transfer function, that relates input voltage U(s) to speed Omega(s): Omega(s)/U(s) = 1/((D + J s)*(L s + R) + K^2) This would be the plant transfer function, if you were interested in controlling the speed. If you were interested in controlling the position like a servomotor, then we'd integrate to find the plant transfer function would be: Theta(s)/U(s) = 1/(s*((D + J s)*(L s + R) + K^2)) If your goal is for Theta(s) to track a desired input angle X(s), then you'd put this plant in series with its controller, and with a feedback loop. This controller would translate the error of X(s) - Theta(s), and use it to generate the voltage necessary to control this motor.
Thanks, that is a huge compliment. I see some of your videos are also in the Brian Douglas style so it is good to see other like-minded people here on RUclips. Thanks for watching!
In case it is helpful, here are all my Control Theory videos in a single playlist ruclips.net/p/PLxdnSsBqCrrF9KOQRB9ByfB0EUMwnLO9o. Please let me know what you think in the comments. You can support this channel via Patreon at www.patreon.com/christopherwlum or by clicking on the 'Thanks' button underneath the video. Thanks for watching!
Professors like you, that share to the world their knowledge. Are really a rare Gem.
As a retired ME professor, this video took me back to when I taught the PID course at the university. Thanks so much for the memory
Hi James, thanks for the kind words. I have a more similar videos at ruclips.net/p/PLxdnSsBqCrrF9KOQRB9ByfB0EUMwnLO9o . Please feel free to check them out, I'd love to hear your thoughts as an academic. Thanks for watching!
This is how everything should be taught. Your lesson is surely more valuable than the whole control theory course at my university 🤣. Universities should be RUclipsrized in my opinion.
Hi Dr. Lum. As an first-year student who is learning about PID control, I would say that your videos about PID are really helpful for me. Thank you so much.
after years in college you are the only person that make everything make sense for me
[AE 511] 24:26
The breakdown of the PID controller is great. Especially the paradox explanation of the integration component.
Thank you very much for this explanation. I'm not a math person so I could finally got this.
But, as a beekeeper, I lose all after the honey waste... I'll watch again.
Thanks from Brazil!
AE511: I find it interesting how complementary the three components are. The "past, present, and future" analogy was very helpful.
AE511: 47:07
It was great to see the comprehensive table of what manipulating Kp/KI/Kd gains will do for the overall system performance gains. When we briefly went over this in undergrad, my limited exposure was trial and error but this was useful in actually making targeted estimations (by hand, obviously there are tools like PID tuner that help out with this)
That being said, never discount the power of guess and check 🙂
Thank you for breaking it down into understandable chunks. sometimes it's hard to understand abstract ideas without a visual representation! I'm a visual learner so your visual example really helped me understand and gave me that ah ha moment! great stuff. This will really help me further my career as a Multicraft technician.
we owe this guy for the enlightment
Beautiful video, finally i can understand the base of the pid control, keep making videos about this!!!
Thank you Wery much!
Tomorrow I have the final exam related to PID adjustments and I have been studying the subject for a really long time now.
I understand the P and I and D controls separately, but it was the combination of these that seemed very challenging to figure out in practice.
This video put the rest of the pieces in place and it's completely self-evident to me now!
Thank you!
Hi,
Thanks for the kind words, I'm glad you enjoyed the video. If you find these videos helpful, I hope you'll consider supporting the channel via Patreon at www.patreon.com/christopherwlum or via the 'Thanks' button underneath the video. Given your interest in this topic, I'd love to have you a as a Patron as I'm able to talk/interact personally with all Patrons. I can also answer any questions, provide code, notes, downloads, etc. on Patreon. Thanks for watching!
-Chris
AE 511: Loved the demo! I've always had to google the impacts of each component of a PID controller, but now I'll forever remember! :)
Helpful explanation. Respect to you for your contribution to academaic environment. Prof aziz
Thank you very much for the refresher. I had to tune a temp control system tonight and I had a memory lapse.
Quite strange but all videos about this topic where anyone is trying to give analogies with something are just confusing me more, but when you are on the white board everything became much clearer and for the first time a understood it.:)
Well done sir!
I love the way u present the idea very straightforward and short informative.
I hope u the best!
Hey bud - watching your vids because you helped me with my fridge lol. Good luck with your videos!
Thank you Sir. Your video is Gold. Keep blessing the humanity with what you got🙏
What an amazing video. Well explained, and managed at answer all my questions i had previously about pids.
great physical demo of PID
Very informative and clear, it will be hard to forget what PID is
Thank you soo much sir for the physical demonstration. The demo has all my heart ❤️❤️
Very good. I was wondering how someone could explain it so well for me, then I noticed we are both left handed ;D
39:10 Regarding the D component, it seems to me that something is off about the model. With the P and the I components, the force exerted was in the direction of reduced error (i.e., in the direction of "moving towards the set point"), yet with the D, the force that gets applied OPPOSES the motion towards the setpoint. Is this distinction relevant? Should we not be imagining D as a force that is proportional to the rate of change of position of the object AND moving the object towards the setpoint (rather than resisting said motion)?
I have a question about 27:30, wouldn’t the u(t) of the D response be negative after the initial delta response to the discontinuity? As the rest of the control drives the error down, wouldn’t it be acting as negative feedback to lower the control response to minimize overshoot?
Not all heroes wear capes. I've already graduated. But still love to watch all those videos
Great efforts, Mr Lum. Keep Going.
Excellent demonstration - I knew functionally nothing about PID controllers but need to have some understanding not only for work but also for home (wanting to use PID for a home built coffee roaster). Honestly...I would've used oil - e.g. mineral oil instead of honey...much less messy but point taken LOL
This lecture is so good.
Hi,
Thanks for the kind words, I'm glad you enjoyed the video. If you find these videos helpful, I hope you'll consider supporting the channel via Patreon at www.patreon.com/christopherwlum or via the 'Thanks' button underneath the video. Given your interest in this topic, I'd love to have you a as a Patron as I'm able to talk/interact personally with all Patrons. I can also answer any questions, provide code, notes, downloads, etc. on Patreon. Thanks for watching!
-Chris
Perfectly explained
Amazing explanation!!
It was very useful. Thank you!
Too good explanation sir
Excellent lecture thanks!
Yay! Control theory!
AE 511 - Thought this was a very good intro to PID controllers
This video is amazing thank you so much
Great Explaination ever
Отличный урок! Спасибо огромное
The arrows in the stability columns should be in upward directions as the stability increases (more stable) in the table at the end of lecture?
Hi Dr. Lum! I love your videos, have you ever considered lecturing a course in Estimation and Tracking with applications in Aerospace Engineering? Topics could include Kalman Filtering, GNC, state estimation, etc. Would love to watch lectures from you on that topic!
Any hints on how to tune a PID controller for a recirculating chiller with a hot and cold PID controller? The default settings don't work (large steady state error). Randomly tweaking the PID parameters hasn't work so far.
subscribed, good stuff here.
In your analysis, the error signal starts out as a step signal and then decays. In the derivative case, when the error goes from zero to some value, ud(t) is large and positive, but when the error decays, ud(t) should be negative because the slope of the error signal is negative therefore the derivative is negative ...
34:48 Uhhhhmmmm... Why not an even stiffer (i.e. shorter) elastic, like you did 15 seconds earlier?
Excellent video. What is considered a "small" Kd? Is there a numerical range that is considered "small"? @ChristopherLum
AE511: It's interesting how there is a physical example for both proportional and derivative, but not an obvious one for integral.
One part I don't understand is where does the function equation {e.g. e(t)} come from which will be integrated or differentiated by the PID controller. The reason I'm asking this question is because the signal coming from the transmitter (feedback) to the controller is just an integer (4mA- 20mA). My guess is the equation are build into the controller based on models of the system behaver.
Dear Professor i think the spike of the derivative part has the sign inverted respect to the rest of the deirivative part because the sign of the derived step error is a positive delta instead the derived after the step goes to zero and then goes to negative values ....so changing the sign respect to the sign of the delta....it means that after a reset....a step error brings suddenly the derivative pid to slow down the control and after the step the control tends slowly to increaase ita own pushing capability.....anyway great video thanlk you
Outstanding! Thanks for sharing.
Hi,
Thanks for the kind words, I'm glad you enjoyed the video. If you find these videos helpful, I hope you'll consider supporting the channel via Patreon at www.patreon.com/christopherwlum or via the 'Thanks' button underneath the video. Given your interest in this topic, I'd love to have you a as a Patron as I'm able to talk/interact personally with all Patrons. I can also answer any questions, provide code, notes, downloads, etc. on Patreon. Thanks for watching!
-Chris
Which book to follow along with your lecture on control theory .
Can an Op-Amp replace the PID controller?
Plz bring video on fractional order sliding mode control
Professor can you make a course on PID and make a couples of projects on PID on that course.
can you please tell the name of the signal after using PID controller? Sometimes, we use error between altitude and set point, and after using PID, it becomes speed . I wonder how? Am I the only one who cannot understand it properly..
please explain
It depends on what system specifically you are controlling. For instance, if you are controlling a motor to have a specific speed, it would be speed. But if you are controlling a motor to maintain a given position, like a servomotor, it would be position.
The device after the PID controller in general, is called the plant, as in factory. It's a representation of the dynamics of the process you are controlling. It represents how the output of your controller is turned into a signal to drive the actuator (motor, piston, etc) that controls the system being controlled, and turns it into the physical measurement that is of interest to control.
As an example, if you are controlling a DC servomotor to maintain a certain position, the plant dynamics would consist of a model for the coil inductance (L), the coil resistance (R), the rotational inertia of the motor and all that is attached to it (J), the motor field constant (K) that relates current to torque (and likewise speed to back EMF), and the drag it experiences (D). The controller would produce a voltage that would then indirectly determine the current that the motor coil receives, and the torque that propels it forward. Here's a diagram that show a DC motor's dynamics, and the interaction among all these terms:
forums.mikeholt.com/attachments/1691248709486-png.2566732/
For a free-spinning motor (no external mechanical load), you can use this diagram to show that the dynamics of the motor can be represented by this transfer function, that relates input voltage U(s) to speed Omega(s):
Omega(s)/U(s) = 1/((D + J s)*(L s + R) + K^2)
This would be the plant transfer function, if you were interested in controlling the speed. If you were interested in controlling the position like a servomotor, then we'd integrate to find the plant transfer function would be:
Theta(s)/U(s) = 1/(s*((D + J s)*(L s + R) + K^2))
If your goal is for Theta(s) to track a desired input angle X(s), then you'd put this plant in series with its controller, and with a feedback loop. This controller would translate the error of X(s) - Theta(s), and use it to generate the voltage necessary to control this motor.
thank you so much
There is a new Brian Douglas in town!
Thanks, that is a huge compliment. I see some of your videos are also in the Brian Douglas style so it is good to see other like-minded people here on RUclips. Thanks for watching!
28:30
Line follower gonna go krazy
Why de/dt is minus then u(t) not minus
In D control
chekov's gun of the honey on the table 😂
You are describing hysteresis.
Si es chino tiene que ser bueno.
As long you not describe what is happening in G(s), all this video is poor and useless