In case it is helpful, here are all my Control Theory videos in a single playlist ua-cam.com/play/PLxdnSsBqCrrF9KOQRB9ByfB0EUMwnLO9o.html. You can support this channel via Patreon at www.patreon.com/christopherwlum or by clicking the 'THANKS' button underneath the video. Please let me know what you think in the comments. Thanks for watching!
Thank you for the enlightening discussion. I was looking for a little guidance on tuning my PID for temperature control. Can't wait to get to the lab in the morning... Thanks again!
I'm glad it was helpful. There are several related videos on the channel. Please feel free to check them out and I would love to hear what you think in the comments. Thanks for watching!
Hi Chadd, Thanks for the kind words, I'm glad you enjoyed the video. If the find the these videos to be helpful, I hope you'll consider supporting the channel via Patreon at www.patreon.com/christopherwlum. 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. Thanks for watching! -Chris
Thank you, Dr Lum, for the informative, easy-to-understand video. One quick comment, at timestamp 29:52, It looks like the value of period(TU) of the oscillation is not around 0.96s from the zoomed-in oscillation degree plot. For me, it looks closer to 0.4s.
This could be a wonderful lecture and I sincerely appreciate the time and effort Mr. Lum has put into making this video. However, it would be so much better if he transcribed his notes onto the board BEFORE he started talking. Not having to frantically scribble while frantically talking and obstructing a view of the material would be a big improvement. Even better, would be for him to narrate over Power Point versions of his notes. Aside from that, he explained the material quite well.
Had to pause in the middle of the video to write. I seldom writes comments on UA-cam, but this explanation was so good it deserves a lot of praise! Thank you, you are much better than my teacher in system theory! Now i am a subscriber to your channel :)
Hi Daniel, Thanks for the kind words, I'm glad you enjoyed the video. If the find the these videos to be helpful, I hope you'll consider supporting the channel via Patreon at www.patreon.com/christopherwlum. 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. Thanks for watching! -Chris
This is a fucking great lecture. Although the pace is fast, it's at the suitable speed for us to capture what you're saying. I need a lecturer like you in my college life.
Hi Shawn, Thanks for the kind words, I'm glad you enjoyed the video. If the find the these videos to be helpful, I hope you'll consider supporting the channel via Patreon at www.patreon.com/christopherwlum. 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. Thanks for watching! -Chris
This is by far the best video I've seen about this topic! I have only one question how do you choose your Ki and Kd parameters if the neutral stability is a superposition of several frequencies? Thanx!
Hi Osama, 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 Inigo, 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. 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. Thanks for watching! -Chris
Hello, Thanks for this nice video, Can you please clear one of my doubt, You mentioned that At ku there will be Neutral stability, but for a system like y_doubledot + ydot=cos(t) the response will be unbounded, So I think instead of saying at ku means neutral stability it will be more good to say that ultimate gain is the one for which the system poles coincide with the Jomega axis and depending upon the input the system may attend neutral stability. Correct me if I'm wrong anywhere. Hoping for a respsone.
I'm glad it was helpful. There are several related videos on the channel. Please feel free to check them out and I would love to hear what you think in the comments. Thanks for watching!
I'm glad it was helpful. There are several related videos on the channel. Please feel free to check them out and I would love to hear what you think in the comments. Thanks for watching!
Hi, Thanks for the kind words, I'm glad you enjoyed the video. If the find the these videos to be helpful, I hope you'll consider supporting the channel via Patreon at www.patreon.com/christopherwlum. 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. Thanks for watching! -Chris
I'm glad it was helpful. There are several related videos on the channel. Please feel free to check them out and I would love to hear what you think in the comments. Thanks for watching!
Hi AJ, 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. 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. Thanks for watching! -Chris
I'm glad it was helpful. There are several related videos on the channel. Please feel free to check them out and I would love to hear what you think in the comments. Thanks for watching!
@@ChristopherLum I have watched almost all your videos in this channel about control theory. They are such great videos. Now I'm waiting for the TBDs :)
Great explanation! What I was wondering about is the use of the ideal algorithm with the results for Kc, Ti and Td. Should this not be the series algorithm, since it was more common in the days of Ziegler and Nichols? I have not been able to find any confirmation on this. Maybe you can shed your light on it?
Hi, Thanks for the kind words, I'm glad you enjoyed the video. If the find the these videos to be helpful, I hope you'll consider supporting the channel via Patreon at www.patreon.com/christopherwlum. 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. Thanks for watching! -Chris
Hi Chris, 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. Thanks for watching! -Chris
I'm glad it was helpful. There are several related videos on the channel. Please feel free to check them out and I would love to hear what you think in the comments. Thanks for watching!
I'm glad it was helpful. I have another video showing how to design a PID controller using another technique (root locus). Please feel free to check it out if you are interested. Thanks for watching!
AE511: It looks like this technique could be used when you don't have a model of the system which is awesome! I suppose that means you could deploy this technique with any black box where technical documentation of the plant may not be available.
Hi, Thanks for the kind words, I'm glad you enjoyed the video. If the find the these videos to be helpful, I hope you'll consider supporting the channel via Patreon at www.patreon.com/christopherwlum. 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. Thanks for watching! -Chris
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
AE511: The example with the wheel in the lab was a cool practical application. What were you using to read that data into MATLAB and plot in real time? Thanks!
Hi Anitha, Thanks for reaching out, I'm glad you enjoyed the video. Unfortunately I'm unable to respond to questions on UA-cam due to the sheer volume of inquiries that I receive. That being said, I hope you'll consider supporting the channel via Patreon at www.patreon.com/christopherwlum as I'll be able to answer questions there. Given your interest in the topic, I'd love to have you as a Patron as I'm able to talk/interact personally with Patrons. Thanks for watching! -Chris
Thanks for the great explanation Sir And, I want to report some error at the tabel 11:32, about the KD value on "some overshoot" and "no overshoot" cmiiw
I'm glad it was helpful. There are other similar videos on the channel, please feel free to check them out and let me know what you think. Thanks for watching!
At 20:16 a text-comment was highlighted, showing the expansion "s^2 + 6*s^2 + 11*s + 6". I know it was commented out anyway, but it was meant to be cubic with the first term, right?
AE511: In industry or academia if there is a situation that would be well suited to Ziegler-Nichols where this is no mathematical model for a system but the response can be measured, is the typical approach to create an approximate mathematical model to enable one of the higher fidelity PID design techniques you mentioned?
Correct, the typical workflow is to get the model first which allows the linear analysis tools to be brought to bear. Ziegler Nichols is if you are too lazy to get the model
Hello. Is possible to obtain the period, TU (the 1.9), at marginally stable, using equations or formula? and not measuring the period from a simualtion? thanks
Thank you for the lecture and the examples with an existing model and no model/experimental data. For cases where we don't have a model, is the process of obtaining K just a simple trial and error? or is there any way of narrowing down to some range of possible K values?
Unfortunately without a model, trial an error is about as best as you can hope for unless you have some physical intuition/insight about the system. If you don't have a model then perhaps a simulation/approximation would suffice?
AE511. Is there a good rule of thumb when picking a "low" Kp at step 1? Or do you need to have some knowledge of your system dynamics? Also when you showed the Matlab plot of all of the different responses plotted together, how come the "no overshoot" configuration still seemed to have some overshoot?
Brad, that is a great question. "Low" is a completely subjective adjective and depends on your system. The "No Overshoot" control type also is not guaranteed to provide no overshoot for all systems. This is one of the major failings of Ziegler-Nichols. It is a sort of back-of-the-envelope approximation. If you need to guarantee no overshoot then you need to do something more mathematically defensible like root locus.
I'm glad it was helpful. There are other similar videos on the channel please feel free to check them out and let me know what you think in the comments. Thanks for watching!
Hi Christopher Lum! How do you know what family of PID controller in terms of P, PI or PID controller to choose depending on a system. What do you base your choice on?
Thank you so much for all your efforts! it is really helpful. But Matlab says that "ZieglerNichols" is unrecognized function . How can I solve this problem?
Great explanation man. Please can you give us the reference that you have used to obtain the required information, especially the pid table Thanks in advance
Hi Hanson, 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
Amazing explanation its simply awesome......... Introduction to PID control and Practical implementation issues with a PID controller, these two am i the only person unable to find? Can anyone help me out?
I'm glad it was helpful. I'm currently working on getting those videos completed and uploaded. Hopefully they will be up in a few weeks. Thanks for watching!
Thanks for the video. Should the Ziegler-Nichols method be repeated for a new set-point or is it designed for the max bound for the process variable that is being controlled?
Thanks for reaching out. If you have questions or would like to request a video, please consider supporting the channel via Patreon at www.patreon.com/christopherwlum. I interact personally with Patrons at all levels. Thanks for watching!
4:55 You are using a Parallel form of pid controller in the diagram where P,I,D controllers are connected in parallel, but the mathematical equation you are describing is that of an Ideal form of PID controller. According to the parallel form Ki=1/Ti and Kd=Td. Please explain.
Thanks for reaching out. If you have questions or would like to request a video, please consider supporting the channel via Patreon at www.patreon.com/christopherwlum. I interact personally with Patrons at all levels. Thanks for watching!
Thank you sir. I have a really important question about an application of Gain(Kp,Ki,Kd) from autotuning. Assume that autotuned gain are Kp=3,Ki=5,Kd=1, then Control Output(CO) = 3*error + 5*∫error dt + 1*d(error)/dt. Also assume that I control a position of DC motor with Arduino. Command of Arduino will be " analogWrite( pin, CO) " which is a PWM command. In Arduino, the range of PWM is 0~255. **** If the range of PWM is 0~65535, Is CO still 3*error + 5*∫error dt + 1*d(error)/dt ??? I think it isn't. I think Scaling is needed to Gain or error=(Target degree - Current degree). But I really cannot find how to do it. Please please help me~~
Hi, Thanks for reaching out, I'm glad you enjoyed the video. Unfortunately I'm unable to respond to questions on UA-cam due to the sheer volume of inquiries that I receive. That being said, I hope you'll consider supporting the channel via Patreon at www.patreon.com/christopherwlum or via the 'Thanks' button underneath the video as I'll be able to answer questions there. Given your interest in the topic, I'd love to have you as a Patron as I'm able to talk/interact personally with Patrons. Thanks for watching! -Chris
Hi Christopher Lum when I ran the program through my matlab there is an error here. "Error using ZieglerNicholsMethod (line 59)" Invalid number of inputs Please help ASAP.
Great job 👍🏻, I have question regarding this method. I am designing PID controller using SIMULINK do i add the transfer functions of disturbances to the diagram and introduce a step function to them? Or they are neglected in this method? Thank you 🙏🏻
if you want to give disturbance tf or step of some magitude in disturbance, you can give. if u designed a good controller it will surely have characteristics of set-point tracking and disturbance rejection.
Hi Yingkai, Thanks for reaching out, I'm glad you enjoyed the video. Unfortunately I'm unable to respond to questions on UA-cam due to the sheer volume of inquiries that I receive. That being said, I hope you'll consider supporting the channel via Patreon at www.patreon.com/christopherwlum as I'll be able to answer questions there. Given your interest in the topic, I'd love to have you as a Patron as I'm able to talk/interact personally with Patrons. Thanks for watching! -Chris
AE511: will you be able to get an idea of the performance metrics using the PID controller obtained from this method? Also, how accurate does your "Tu" value need to be measured in order to have a good controller?
Unfortunately I don't believe you can get exact performance from this method. You can get "rough" ideas of performance but since this is an online method, it is difficult to predict performance. However if you have a model of your system you can take your Ziegler-Nichols designed gains and perform analytical predictions to get exact performance.
Alan, good point, I think my wording was misleading. I thought the previous question was related to trying to make exact/analytical predictions of a system controlled by a Ziegler-Nichols designed PID controller. Since you typically use Ziegler-Nichols when you don't have a mathematical model, you can't make exact predictions. As you point out, you can make exact measurements of the system to get the experimental performance. The only way you can make exact predictions is if you have a model of your system. That is the whole "can opener" discussion. If you have a mathematical model, you should be using another technique besides Ziegler-Nichols.
Thanks for reaching out. If you have questions or would like to request a video, please consider supporting the channel via Patreon at www.patreon.com/christopherwlum. I interact personally with Patrons at all levels. Thanks for watching!
In case it is helpful, here are all my Control Theory videos in a single playlist ua-cam.com/play/PLxdnSsBqCrrF9KOQRB9ByfB0EUMwnLO9o.html. You can support this channel via Patreon at www.patreon.com/christopherwlum or by clicking the 'THANKS' button underneath the video. Please let me know what you think in the comments. Thanks for watching!
I have a question
i wanna know how to get the Ku value. i found it only by trying a lot.
AE511 - The DC motor demonstration was superb and the use of pop ups to show the full table gives a great detailed view.
Alan, great, I'm glad it was helpful. It was fun to get the system working to film this.
I've seen Ziegler-Nichols previously but this is a much better / straight-forward explanation
I appreciated the DC motor example. It was a helpful visualization.
Thank you for the enlightening discussion. I was looking for a little guidance on tuning my PID for temperature control. Can't wait to get to the lab in the morning... Thanks again!
I'm glad it was helpful. There are several related videos on the channel. Please feel free to check them out and I would love to hear what you think in the comments. Thanks for watching!
AE 511. I liked the lab example of the DC motor. It gave a good real life reference.
1:46 "allow you to do this in iterative online fashion"
This guy really knows what I'm doing in this year,,,
first PID tutorial i can understand, ty.
Wonderful explanation! One of the best and simplest explanation I have seen on youtube. Kudos to you sir!!!
My hope for good lectures are restored
I'm glad it was helpful thanks for watching!
Your lessons are brilliant - thank you very much!
Hi Chadd,
Thanks for the kind words, I'm glad you enjoyed the video. If the find the these videos to be helpful, I hope you'll consider supporting the channel via Patreon at www.patreon.com/christopherwlum. 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. Thanks for watching!
-Chris
Thank you, Dr Lum, for the informative, easy-to-understand video. One quick comment, at timestamp 29:52, It looks like the value of period(TU) of the oscillation is not around 0.96s from the zoomed-in oscillation degree plot. For me, it looks closer to 0.4s.
This could be a wonderful lecture and I sincerely appreciate the time and effort Mr. Lum has put into making this video. However, it would be so much better if he transcribed his notes onto the board BEFORE he started talking. Not having to frantically scribble while frantically talking and obstructing a view of the material would be a big improvement. Even better, would be for him to narrate over Power Point versions of his notes.
Aside from that, he explained the material quite well.
Had to pause in the middle of the video to write. I seldom writes comments on UA-cam, but this explanation was so good it deserves a lot of praise! Thank you, you are much better than my teacher in system theory! Now i am a subscriber to your channel :)
Hi Daniel,
Thanks for the kind words, I'm glad you enjoyed the video. If the find the these videos to be helpful, I hope you'll consider supporting the channel via Patreon at www.patreon.com/christopherwlum. 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. Thanks for watching!
-Chris
This is a fucking great lecture. Although the pace is fast, it's at the suitable speed for us to capture what you're saying. I need a lecturer like you in my college life.
Hi Shawn,
Thanks for the kind words, I'm glad you enjoyed the video. If the find the these videos to be helpful, I hope you'll consider supporting the channel via Patreon at www.patreon.com/christopherwlum. 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. Thanks for watching!
-Chris
Thank you very much. It was very nice and clear explained.
good lecture. thank you a lot for your sharing, it is helpful
This is by far the best video I've seen about this topic!
I have only one question how do you choose your Ki and Kd parameters if the neutral stability is a superposition of several frequencies?
Thanx!
Great explanation. Thanks for sharing the video.
Hi Osama,
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
Really a great explanation. Thanks for sharing the video.
AE511: The practical examples are really helpful, looking at output traces in MATLAB/Simulink doesn't always give the best comprehension
This video was so useful. Thanks Sir Lum!
Hi Inigo,
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. 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. Thanks for watching!
-Chris
Hello,
Thanks for this nice video, Can you please clear one of my doubt, You mentioned that At ku there will be Neutral stability, but for a system like y_doubledot + ydot=cos(t) the response will be unbounded, So I think instead of saying at ku means neutral stability it will be more good to say that ultimate gain is the one for which the system poles coincide with the Jomega axis and depending upon the input the system may attend neutral stability.
Correct me if I'm wrong anywhere.
Hoping for a respsone.
Thanks very much for this excellent Video. THANKS!
I'm glad it was helpful. There are several related videos on the channel. Please feel free to check them out and I would love to hear what you think in the comments. Thanks for watching!
That was quite good explaination of the method!
I'm glad it was helpful thanks for watching!
Practical explanation! unique from other tutorials!
I'm glad it was helpful. There are several related videos on the channel. Please feel free to check them out and I would love to hear what you think in the comments. Thanks for watching!
Very nice. Have to watch your other videos.
Hi,
Thanks for the kind words, I'm glad you enjoyed the video. If the find the these videos to be helpful, I hope you'll consider supporting the channel via Patreon at www.patreon.com/christopherwlum. 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. Thanks for watching!
-Chris
thanks. you hang it on my brain compeletly
Really good explanation! Thank you so much! Your videos help a lot!
I'm glad it was helpful. There are several related videos on the channel. Please feel free to check them out and I would love to hear what you think in the comments. Thanks for watching!
Awesome! Exactly what I needed to know.
บรรยายดี เข้าใจง่ายดีครับ
สุดยอดมากครับ
Great explanation!
Your motivation is fascinating! Good luck
Yes, I am very much happy.
Thank you
I am subscribing your channel.
Hi AJ,
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. 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. Thanks for watching!
-Chris
Thanks you, sir for very helpful information.
Thank you so much for the tremendous video and the DC Motor example.
I'm glad it was helpful. There are several related videos on the channel. Please feel free to check them out and I would love to hear what you think in the comments. Thanks for watching!
@@ChristopherLum I have watched almost all your videos in this channel about control theory. They are such great videos. Now I'm waiting for the TBDs :)
@@ChristopherLum May I ask if you have planed to publish some videos about adaptive control and fuzzy logic?
Great Explanation, Thanks
I'm glad it was helpful thanks for watching
Great explanation! What I was wondering about is the use of the ideal algorithm with the results for Kc, Ti and Td. Should this not be the series algorithm, since it was more common in the days of Ziegler and Nichols? I have not been able to find any confirmation on this. Maybe you can shed your light on it?
thanks a lot. Works great on my software PID controller
great work , thank u sir
Hi,
Thanks for the kind words, I'm glad you enjoyed the video. If the find the these videos to be helpful, I hope you'll consider supporting the channel via Patreon at www.patreon.com/christopherwlum. 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. Thanks for watching!
-Chris
Great Examples Chris especially the DC Motor but I was hoping to see the "feed the baby" DC Motor experiment.
Haha, I forgot that you probably remember that from a few years back. I seem to remember that your system worked great.
Chris, wonderful explanation.
I'm glad it was helpful thanks for watching!
Nicely done. Thumbs-up
Hi Chris,
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. Thanks for watching!
-Chris
Great explanation, thanks!!
Very informative, big thanks.
This was very helpful, thank you for taking the time . Great quality
I'm glad it was helpful. There are several related videos on the channel. Please feel free to check them out and I would love to hear what you think in the comments. Thanks for watching!
Yes that is something initiative sir ,you always come with practical aspects 😊😊😊
What"s your use book for reference PID ziegler nichlos this videos ?
Great video and simple explanation. Thank you for your afford.
I'm glad it was helpful thanks for watching
Crystal clear. thanks
I'm glad it was helpful. I have another video showing how to design a PID controller using another technique (root locus). Please feel free to check it out if you are interested. Thanks for watching!
AE511: It looks like this technique could be used when you don't have a model of the system which is awesome! I suppose that means you could deploy this technique with any black box where technical documentation of the plant may not be available.
Great explanation
Thank you. This helped a lot.
I'm glad it was helpful thanks for watching!
Great explanation of Ziegler Nichols...keep banging on the can of beans, you'll open it eventually haha!
Thank you very much sir !
I implemented this method with Visuino and also auto tuning , and managed to fit this code on a Arduino Nano. i have also wrote the code for ESP32.
Amazing video !!!
Hi,
Thanks for the kind words, I'm glad you enjoyed the video. If the find the these videos to be helpful, I hope you'll consider supporting the channel via Patreon at www.patreon.com/christopherwlum. 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. Thanks for watching!
-Chris
te amo, saludos desde méxico
thank man, really helpful !
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
@@ChristopherLum sure thing
AE511: The example with the wheel in the lab was a cool practical application. What were you using to read that data into MATLAB and plot in real time? Thanks!
your explanations are very clear but could you tel me how you got Simulink model in this case.thank you
Hi Anitha,
Thanks for reaching out, I'm glad you enjoyed the video. Unfortunately I'm unable to respond to questions on UA-cam due to the sheer volume of inquiries that I receive. That being said, I hope you'll consider supporting the channel via Patreon at www.patreon.com/christopherwlum as I'll be able to answer questions there. Given your interest in the topic, I'd love to have you as a Patron as I'm able to talk/interact personally with Patrons. Thanks for watching!
-Chris
Thanks for the great explanation Sir
And, I want to report some error at the tabel 11:32, about the KD value on "some overshoot" and "no overshoot"
cmiiw
amaaaaaaaaazing. Just thanks.
I'm glad it was helpful. There are other similar videos on the channel, please feel free to check them out and let me know what you think. Thanks for watching!
Thanks for your amazing LECTURE =) Gracias
At 20:16 a text-comment was highlighted, showing the expansion "s^2 + 6*s^2 + 11*s + 6". I know it was commented out anyway, but it was meant to be cubic with the first term, right?
thank you I spotted that and checked comments because surely others saw that too. You did, well done.
AE511: In industry or academia if there is a situation that would be well suited to Ziegler-Nichols where this is no mathematical model for a system but the response can be measured, is the typical approach to create an approximate mathematical model to enable one of the higher fidelity PID design techniques you mentioned?
Correct, the typical workflow is to get the model first which allows the linear analysis tools to be brought to bear. Ziegler Nichols is if you are too lazy to get the model
Hello. Is possible to obtain the period, TU (the 1.9), at marginally stable, using equations or formula? and not measuring the period from a simualtion? thanks
Great content, thanks!
Also looking forward to the video about PID using Root Locus Method! :)
Tomas, thanks for watching. I'm working on this video right now. I hope to have it posted in the next 24 hours so please stand by.
Thank you for the lecture and the examples with an existing model and no model/experimental data. For cases where we don't have a model, is the process of obtaining K just a simple trial and error? or is there any way of narrowing down to some range of possible K values?
Unfortunately without a model, trial an error is about as best as you can hope for unless you have some physical intuition/insight about the system. If you don't have a model then perhaps a simulation/approximation would suffice?
Nicely explained good 👍
AE511. Is there a good rule of thumb when picking a "low" Kp at step 1? Or do you need to have some knowledge of your system dynamics? Also when you showed the Matlab plot of all of the different responses plotted together, how come the "no overshoot" configuration still seemed to have some overshoot?
Brad, that is a great question. "Low" is a completely subjective adjective and depends on your system. The "No Overshoot" control type also is not guaranteed to provide no overshoot for all systems. This is one of the major failings of Ziegler-Nichols. It is a sort of back-of-the-envelope approximation. If you need to guarantee no overshoot then you need to do something more mathematically defensible like root locus.
@@ChristopherLum Got it, thank you!
Spot on👌🏾thank you
I'm glad it was helpful. There are other similar videos on the channel please feel free to check them out and let me know what you think in the comments. Thanks for watching!
Hi Christopher Lum! How do you know what family of PID controller in terms of P, PI or PID controller to choose depending on a system. What do you base your choice on?
AE511: @27:08 your "No overshoot" curve looks like it has some overshoot...?
ae511, good lecture on implementing the PID controller
thanku sir, your video may help me designing PI controller for bidirectional buck boost charger.
thanks for the video sir
Thank you so much for all your efforts! it is really helpful.
But Matlab says that "ZieglerNichols" is unrecognized function . How can I solve this problem?
Point of confusion for me @20:06 on line 10 of the algebraic expansion you have "s^2 + 6*s^2 + 11*s + 6"... should not that first term be s^3 ??
Great explanation man.
Please can you give us the reference that you have used to obtain the required information, especially the pid table
Thanks in advance
Hi, well explained. A minor note: @ time 20:03, the denominator in the comment should start with s^3, not s^2. Rest is ok :-).
thank you so much!
It is a very helpful video. But I am stuck by building the model. Is there a way I can directly get that model in Matlab?
Hi Hanson,
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
Amazing explanation its simply awesome......... Introduction to PID control and Practical implementation issues with a PID controller, these two am i the only person unable to find? Can anyone help me out?
I'm glad it was helpful. I'm currently working on getting those videos completed and uploaded. Hopefully they will be up in a few weeks. Thanks for watching!
Thanks for the video. Should the Ziegler-Nichols method be repeated for a new set-point or is it designed for the max bound for the process variable that is being controlled?
Thanks for reaching out. If you have questions or would like to request a video, please consider supporting the channel via Patreon at www.patreon.com/christopherwlum. I interact personally with Patrons at all levels. Thanks for watching!
4:55 You are using a Parallel form of pid controller in the diagram where P,I,D controllers are connected in parallel, but the mathematical equation you are describing is that of an Ideal form of PID controller. According to the parallel form Ki=1/Ti and Kd=Td.
Please explain.
Thanks for reaching out. If you have questions or would like to request a video, please consider supporting the channel via Patreon at www.patreon.com/christopherwlum. I interact personally with Patrons at all levels. Thanks for watching!
Wait. You mean you're not supposed to bash the can with the handle? How do can openers work?
/s
👍
best video thanks
Thank you sir. I have a really important question about an application of Gain(Kp,Ki,Kd) from autotuning. Assume that autotuned gain are Kp=3,Ki=5,Kd=1, then Control Output(CO) = 3*error + 5*∫error dt + 1*d(error)/dt. Also assume that I control a position of DC motor with Arduino. Command of Arduino will be " analogWrite( pin, CO) " which is a PWM command. In Arduino, the range of PWM is 0~255. **** If the range of PWM is 0~65535, Is CO still 3*error + 5*∫error dt + 1*d(error)/dt ??? I think it isn't. I think Scaling is needed to Gain or error=(Target degree - Current degree). But I really cannot find how to do it. Please please help me~~
Hi,
Thanks for reaching out, I'm glad you enjoyed the video. Unfortunately I'm unable to respond to questions on UA-cam due to the sheer volume of inquiries that I receive. That being said, I hope you'll consider supporting the channel via Patreon at www.patreon.com/christopherwlum or via the 'Thanks' button underneath the video as I'll be able to answer questions there. Given your interest in the topic, I'd love to have you as a Patron as I'm able to talk/interact personally with Patrons. Thanks for watching!
-Chris
Thank you Sir
Would you please do one on I-PD ?
Are Ku and Pu values same for all the controllers like P, Pi and Pid or different for different controllers
Hi
Christopher Lum
when I ran the program through my matlab there is an error here.
"Error using ZieglerNicholsMethod (line 59)"
Invalid number of inputs
Please help ASAP.
Dear Professor, May I share your video link to my students for learning purposes? I seek your permission. Thanks and regards-A.Krishnamoorthy
Absolutely, please feel free to share the UA-cam links, I hope it is helpful to your students.
Great job 👍🏻, I have question regarding this method. I am designing PID controller using SIMULINK do i add the transfer functions of disturbances to the diagram and introduce a step function to them? Or they are neglected in this method? Thank you 🙏🏻
if you want to give disturbance tf or step of some magitude in disturbance, you can give. if u designed a good controller it will surely have characteristics of set-point tracking and disturbance rejection.
please upload frequency response analysis
Why nuetral stability can achieve? I mean if you only consider proportional control, then you will always get a steady state error, right?
Hi Yingkai,
Thanks for reaching out, I'm glad you enjoyed the video. Unfortunately I'm unable to respond to questions on UA-cam due to the sheer volume of inquiries that I receive. That being said, I hope you'll consider supporting the channel via Patreon at www.patreon.com/christopherwlum as I'll be able to answer questions there. Given your interest in the topic, I'd love to have you as a Patron as I'm able to talk/interact personally with Patrons. Thanks for watching!
-Chris
Thank you!
AE511: will you be able to get an idea of the performance metrics using the PID controller obtained from this method? Also, how accurate does your "Tu" value need to be measured in order to have a good controller?
Unfortunately I don't believe you can get exact performance from this method. You can get "rough" ideas of performance but since this is an online method, it is difficult to predict performance. However if you have a model of your system you can take your Ziegler-Nichols designed gains and perform analytical predictions to get exact performance.
I'm slightly confused on the response. Couldn't you get the exact performance experimentally by measuring the output with the controller in the loop?
Alan, good point, I think my wording was misleading. I thought the previous question was related to trying to make exact/analytical predictions of a system controlled by a Ziegler-Nichols designed PID controller. Since you typically use Ziegler-Nichols when you don't have a mathematical model, you can't make exact predictions. As you point out, you can make exact measurements of the system to get the experimental performance. The only way you can make exact predictions is if you have a model of your system. That is the whole "can opener" discussion. If you have a mathematical model, you should be using another technique besides Ziegler-Nichols.
@@ChristopherLum If I have mathematical model then which algorithm, I should use to get values of Kp, Ki and Kd.
What is the reference of procedure mentioned by you sir??
I mean in which book or research article this procedure is given??
Thanks for reaching out. If you have questions or would like to request a video, please consider supporting the channel via Patreon at www.patreon.com/christopherwlum. I interact personally with Patrons at all levels. Thanks for watching!