amazing This is exactly why I'm losing fate in the school teching system...just books formulas and exams that are needed to cut the number of students... I learned more about control system in these 4 days that I've been watching you that in one semester of Control system... You sir are just inspiring
its 2024 and i've learned and understood more about control system and finally deciphered my lecturers' cryptic slide in just 3 days watching Brian's video
Hi Daruko, thanks for the comment. Determining zeta through some white box method would be tough but it can be done. You'd have to find or develop an equation that uses the spring parameters like material type, number of loops, radius of the loop, etc in order to calculate it. Realistically though you'd just test for it (similar to how we found the spring constant). More often than not developing a model takes both methods - writing equations and testing for the parameter values.
Hi Glen, well you understood what I said correctly ... unfortunately what I said was wrong! I did mean to say that the vertical axis was the imaginary axis not the real axis. I've added an annotation to correct it plus I wrote it in the Errata section of the video description. Thanks for pointing that out :)
you know, your videos have helped me alot during my study in universty, and now i'm graduated and I have an interview tommorow in MBD which I got a very good understanding forming the big picture and working out the details with your illustrations, and if I am accepted and started my career, it will be thanks to you. So I just wanted to say how grateful I'm and wish you all the success and happiness.
Hi Brian, your videos are amazing.... I'm a PhD student from Brazil and I've been working with Multiobjective System Identification techniques I watched your videos about PID control, root locus and bode plots... all of them are great! Thx for these videos! Congrats!
Thank you System Identification Spring Linear and non linear region identification Spring constant K = N/m Simulation problem find y(t) Control system problem find u(t) System Identification find S Methods of System ID Black box find S based on sine sweep or step input or bump test White box find S based on mathematics (Soring constant, force = ma) missing here is damping factor, how to find damping factor?) Finally testing on real spring and mass. Such a wonderful explanation Thank you
Hi Brian, I'm not really sure how I got here but I'm damned glad I did. I'm an old school industrial electrician/ converted to I&E tech (not voluntarily) in the 90's. So much of this are things I've had to learn on my own on the down hill slide to retirement. I currently teach motor control at a tech school, something I'm very comfortable with. But these control methods are something I had to wrap my head around through books and other co-workers. I think you are going to be a great help for me in my current configuration.... I'll definitely contribute to your efforts, keep up the great work. John T
Suddenly, after entire studies of aerospace engineering, thanks to Your excellent teaching, Bode plots made sense to me. I'm so impressed in how all those things about roots and plots come together and help us build accurate models! WOW
You are pretty much single handedly going to get me through my control systems exam - Something I thought was boring as hell before I started watching your videos.. now I eagerly check youtube for your next vid.. Please keep going, One day who knows we could all end up building a mars rover together..
Wow, sir, your demonstration of the real life application of the Bode plot made so many concepts in my head combine into one big picture I could almost hear the clicking sound.
It's so amazing that the bode and pole-zero plot are nothing more than an other point of view of the complex laplace domain. Beautiful to see the connection.
Hi Eduardo, I apologize but I don't think I follow your question. I don't think what I did is a competitor to LSM. To me LSM is used for estimating oversampled signals (for example trying to curve fit a noisy measurement). I used LSM to get a bet fit linear line of the data. Also, when you're doing the black box method you can use LSM to determine the best coefficients for the transfer function that you're estimating. If I'm way off your question could you please state it differently? Thanks!
Ha! That's the best question yet :) It's from an online comic strip called XKCD, but I bought the shirt from thinkgeek. Man, TH-cam won't let me write dot com, they're on to me!
In Italy I experienced the same situation that I recognized in other comments. If only my Control Systems professor could explain the concepts in this visual way... Thank you Brian!
Final proof that my professor actually sucks at teaching is that he sliced this video lecture and put it on his lecture slide and launched video from this channel. P.S: The bes channel about control theory!
Amazing experiment ! We all have seen such actions of spring in our daily lives. But analyzing its behavior in terms of Bode was something great and innovative that even professors fail to do. I am keen to watch similar videos in near future.
Yep, one does indeed learn more by watching practical demonstrations than simply reading them in the books. Thank you for taking your time out and creating these videos ! Been helpful to me for understanding and appreciating the subject ❤.
Man, this is a great video. I've been watching and following your channel for some time now and I've found it excellent but this one just compelled me to write a thank you comment. It's just that great :) Thanks and I wish you all the best. Have a nice day. Ahmed
Awesome example to use Brian! The system gain plot matches transmissibility curves I learned in mechanical vibration classes. I love it when different theories link up!
there are actually three: 1) White box 2) Grey Box 3) Black Box i am also in your fb page. Syed Hamza Ali I loved the way you explained bode plots, and with hand spring.
oh screw Dos equis!!! you sir are most interesting man in the world! I cant wait till the next vid. U have inspired me into control systems and so far i have gotten all A's in my control class... i'm thinking of getting a masters degree in control systems ... actually i already applied for that!
Your explanation of the material you actually cover (which I would call modelling with some testing to do parameter identification) is superb. However, none of what you discuss in this video is system Id as it's typically thought of. Would love to see you do a dedicated lecture series on the detailed math behind response measurements, RLS, Kalman filtering for online parameter estimation, etc.
Awesome explanations! Thank you so much. I'd like to throw in my five cents on obtaining the harmonic oscillator equations. I think it's not the most transparent way to do it. Technically, in the F.B.D. there are only two forces acting upon the mass both in stationary case and in motion: Gravity and Spring Tension Fs(x2,x1). Reducing this to a single force k(x2-x1) (which equals to zero at steady-state with x1=0) is done my choosing the coordinate system so its origin coincides with the resting center of mass.
I have similar idea in my head. But i think if we analyze the stationary case, the tension in the spring us equal to the weight downwards. And the other upward force is caused by the input when it starts oscillating.
Brian, loving the videos, very thorough yet still accessible.As a fellow control engineer I'm interested in the sorts of projects/applications to which you've been able to apply the theory that you've covered in you're series so far?
Thank you so much, I've cleared out a lot of daubts from watching your videos. I yet have more to learn on system identification, particularly graphic Identification, refference model, varosimilitud maximum and instrumental variable. If you have any more videos on this topic it would be really helpful, thank you again.
Around 7:15, you are writing down equation of a spring as F= 94N/m* x, however I would expect an offset there(from the graph about 10N). Would that affect the analysis further down?
Visual inspection would give as you said. In context at to the verbal discussion, x in the equation refers to the linear region. It would be inferred that the "zero point" is the minimum displacement for linearity, say 10cm.
Great job. Now can you show how to take this data and build a simulation in Scilab? What is the maximum rate we can lift the paint bucket with 3N of load with minimal overshoot? What is the maximum rate we can fluid to the paint bucket without overshoot?
Hello brian, Your videos have really made a big difference by helping me get through the control subjects in chem engg! Please could you put up a video for cascade control, GM & PM settings! :) Really appreciate the effort you put into the videos. Cheers,
Hi Mr. Douglas, Do you know any online portal which provides the control system courses? Or is there any distant learning courses as I'm sitting in India and would like to learn this control systems. I have been searching for such courses but not found any. If no such thing exists, why don't you initiate the control system course something in Coursera or Udemy?
Hey Brian, great video! as a control enthusiast I love your channel and I always find myself going through your videos every time I get stuck with my course work. Anyways, next semester I will be starting my final year project and I have decide to propose a title on black box modelling of a rotary inverted pendulum, I decided on this because though it may seem easy it involves alot of topics which I personally think would give me a proper foundation in system modelling and control. What are your views on this? Is there something else apart from an inverted pendulum that would be a better option? Please enlighten me. Thanks!
"I don't always watch TH-cam videos, but when I do, I prefer Control System Lectures and funny cat videos" - The Most Interesting Man in the World
amazing
This is exactly why I'm losing fate in the school teching system...just books formulas and exams that are needed to cut the number of students...
I learned more about control system in these 4 days that I've been watching you that in one semester of Control system...
You sir are just inspiring
its 2024 and i've learned and understood more about control system and finally deciphered my lecturers' cryptic slide in just 3 days watching Brian's video
Sir Brian, only one request, don't let this channel die, keep it alive and of course you've our support ✌️👏👏
Hi Daruko, thanks for the comment. Determining zeta through some white box method would be tough but it can be done. You'd have to find or develop an equation that uses the spring parameters like material type, number of loops, radius of the loop, etc in order to calculate it. Realistically though you'd just test for it (similar to how we found the spring constant). More often than not developing a model takes both methods - writing equations and testing for the parameter values.
I don't know how to thank you man, God bless you, if only all our teachers were like you....
My God where were you when I was in college? Could have saved me tons of headache. Oh, that's right...TH-cam didn't exist yet.
Hi Glen, well you understood what I said correctly ... unfortunately what I said was wrong! I did mean to say that the vertical axis was the imaginary axis not the real axis. I've added an annotation to correct it plus I wrote it in the Errata section of the video description. Thanks for pointing that out :)
you know, your videos have helped me alot during my study in universty, and now i'm graduated and I have an interview tommorow in MBD which I got a very good understanding forming the big picture and working out the details with your illustrations, and if I am accepted and started my career, it will be thanks to you. So I just wanted to say how grateful I'm and wish you all the success and happiness.
10:41 - 12:02 This was extremely satisfying, thanks!
That demonstration was mind blowing. The universe truly is written in math, the universal language.
Why, oh why I did not meet such teacher when I was in school...
So simply and graphically - in a word, great job!
Wow, one of the most useful videos I've seen on TH-cam. Thanks!
Hi Brian,
your videos are amazing.... I'm a PhD student from Brazil and I've been working with Multiobjective System Identification techniques
I watched your videos about PID control, root locus and bode plots... all of them are great!
Thx for these videos!
Congrats!
Thank you
System Identification
Spring
Linear and non linear region identification
Spring constant K = N/m
Simulation problem find y(t)
Control system problem find u(t)
System Identification find S
Methods of System ID
Black box find S based on sine sweep or step input or bump test
White box find S based on mathematics
(Soring constant, force = ma)
missing here is damping factor, how to find damping factor?)
Finally testing on real spring and mass.
Such a wonderful explanation
Thank you
Hi Brian, I'm not really sure how I got here but I'm damned glad I did.
I'm an old school industrial electrician/ converted to I&E tech (not voluntarily) in the 90's. So much of this are things I've had to learn on my own on the down hill slide to retirement. I currently teach motor control at a tech school, something I'm very comfortable with. But these control methods are something I had to wrap my head around through books and other co-workers. I think you are going to be a great help for me in my current configuration....
I'll definitely contribute to your efforts, keep up the great work.
John T
Suddenly, after entire studies of aerospace engineering, thanks to Your excellent teaching, Bode plots made sense to me. I'm so impressed in how all those things about roots and plots come together and help us build accurate models! WOW
You are pretty much single handedly going to get me through my control systems exam - Something I thought was boring as hell before I started watching your videos.. now I eagerly check youtube for your next vid..
Please keep going, One day who knows we could all end up building a mars rover together..
Wow, sir, your demonstration of the real life application of the Bode plot made so many concepts in my head combine into one big picture I could almost hear the clicking sound.
12:09 The crux of the experiment, perfect explanation for in phase and out of phase response
dear Brian, I'm astonished by your engineering skills ... Great
It's so amazing that the bode and pole-zero plot are nothing more than an other point of view of the complex laplace domain. Beautiful to see the connection.
Hi Eduardo, I apologize but I don't think I follow your question. I don't think what I did is a competitor to LSM. To me LSM is used for estimating oversampled signals (for example trying to curve fit a noisy measurement). I used LSM to get a bet fit linear line of the data. Also, when you're doing the black box method you can use LSM to determine the best coefficients for the transfer function that you're estimating. If I'm way off your question could you please state it differently? Thanks!
No words to describe this channel, simply awesome!
Man you have no idea how much help you've done to me and my friends, #respect from #Algeria , thank you so much
Your best video, by a long shot, and that is saying something.
I enjoyed a lot how you use physical experiments to illustrate control theory. So amazing
Best System Identifications videos online
That's simply brilliant. I wish I had had these videos during my lectures, life would have been lot easier :)
It's the best practical way I've ever seen to describe the Bode magnitude and phase plot! Thanks, Brian.
Best physical interpretation ever! The natural frequency experiment is awesome. Thanks Mr. 💐
Ha! That's the best question yet :) It's from an online comic strip called XKCD, but I bought the shirt from thinkgeek. Man, TH-cam won't let me write dot com, they're on to me!
You are the Richard Feynman of Controls Engineering...well done, sir.
Thanks Brian. I love the way you present your lectures. I wish you were my control system course teacher at university.
In Italy I experienced the same situation that I recognized in other comments.
If only my Control Systems professor could explain the concepts in this visual way...
Thank you Brian!
Final proof that my professor actually sucks at teaching is that he sliced this video lecture and put it on his lecture slide and launched video from this channel. P.S: The bes channel about control theory!
Amazing experiment ! We all have seen such actions of spring in our daily lives. But analyzing its behavior in terms of Bode was something great and innovative that even professors fail to do. I am keen to watch similar videos in near future.
That's great teaching. Way better than some university profs. Congrats!
Yep, one does indeed learn more by watching practical demonstrations than simply reading them in the books. Thank you for taking your time out and creating these videos ! Been helpful to me for understanding and appreciating the subject ❤.
I really like this type of video where I can actually see and relate equations and concepts to what you talked about in the previous videos!
Your videos are incredible and your channel is awesome. Very good job.
Thanks Brian! you are a gifted teacher :) and on top of that you understand the subject matter so well.. these videos have been immensely helpful
Wow I'm watching this after 10 years and it's so intuitive. By the way, where is Brian Douglas these days ?
I'm around still! I post to the MATLAB channel now. You can find everything I make at engineeringmedia.com. Cheers!
Just wanted to tell you that this is still very useful and well made content, have a great year!
a clear introduction to system identification
Wow... the demo at the end really tied things together for me. Thank you Brian. Will you make anymore videos?
Man, this is a great video. I've been watching and following your channel for some time now and I've found it excellent but this one just compelled me to write a thank you comment. It's just that great :)
Thanks and I wish you all the best. Have a nice day.
Ahmed
Awesome example to use Brian! The system gain plot matches transmissibility curves I learned in mechanical vibration classes. I love it when different theories link up!
this is just the best! Really appreciate all your videos, may the universe in turn help you and your family forever!
there are actually three: 1) White box 2) Grey Box 3) Black Box
i am also in your fb page. Syed Hamza Ali
I loved the way you explained bode plots, and with hand spring.
I really liked the experiment you did and the explanations you gave.
THIS IS AWESOME
Ulysses ... I think I have beard envy.
oh screw Dos equis!!! you sir are most interesting man in the world!
I cant wait till the next vid. U have inspired me into control systems and so far i have gotten all A's in my control class... i'm thinking of getting a masters degree in control systems ... actually i already applied for that!
It's nice to see the face behind my control lectures after all these weeks!!
Your explanation of the material you actually cover (which I would call modelling with some testing to do parameter identification) is superb. However, none of what you discuss in this video is system Id as it's typically thought of. Would love to see you do a dedicated lecture series on the detailed math behind response measurements, RLS, Kalman filtering for online parameter estimation, etc.
I just was sitting with open mouth during this video...excellent explanation !
Finally I get to know why do we always use Step input to check all systems in all problems we study :)
Why...can u explain it to me
your video is really awesome, helping a lot for people confused by obscure textbook knowledge. wish know you sooner!
High quality video. Good job.
A question Mr. Douglas
In ur opinion, what is the best book for Control theory for beginners ?
Great work Mr. Brian Douglas!!! Thank you!
Great content Brian
Superb Brian.
Best explanation ever......
Wow, You Brought it all together Awesome Job.
awesome ...thanks for giving a practical example
Best explanation ever thanks so much. Looking forward to your book on control theory !!
Man... you're good. Very good.
Nice explanation. Question, what hardware and software do you use to create the presentation?
Thank you so much for this courses. I'm an engineer and need to identify a system and make a control for it. I hoping for more on this topic! Thanks.
Your videos are fantastic! Thank you!
Great Lectures!!! Nice Experiment!!! Cool shirt!!!
Excellent job Brian, I just found your videos, great job, congratulations!
Great job. I'm just wondering though, how would you go about determining zeta using the white box method to get a more realistic bode plot?
Excellent video !
Awesome explanations! Thank you so much. I'd like to throw in my five cents on obtaining the harmonic oscillator equations. I think it's not the most transparent way to do it. Technically, in the F.B.D. there are only two forces acting upon the mass both in stationary case and in motion: Gravity and Spring Tension Fs(x2,x1). Reducing this to a single force k(x2-x1) (which equals to zero at steady-state with x1=0) is done my choosing the coordinate system so its origin coincides with the resting center of mass.
I have similar idea in my head. But i think if we analyze the stationary case, the tension in the spring us equal to the weight downwards. And the other upward force is caused by the input when it starts oscillating.
Brian, loving the videos, very thorough yet still accessible.As a fellow control engineer I'm interested in the sorts of projects/applications to which you've been able to apply the theory that you've covered in you're series so far?
Please make more!!! You are amazing :)
Very good video! Thanks!
As always, brilliant.
Thank you for the good work
Thank you so much, I've cleared out a lot of daubts from watching your videos. I yet have more to learn on system identification, particularly graphic Identification, refference model, varosimilitud maximum and instrumental variable. If you have any more videos on this topic it would be really helpful, thank you again.
Around 7:15, you are writing down equation of a spring as F= 94N/m* x, however I would expect an offset there(from the graph about 10N). Would that affect the analysis further down?
Visual inspection would give as you said. In context at to the verbal discussion, x in the equation refers to the linear region. It would be inferred that the "zero point" is the minimum displacement for linearity, say 10cm.
Amazing explanation about this such important topic. I was wondering if you have any videos about Matlab Ident Toolbox?
Really nice way of presenting the topic. Even interesting for engineers. Please keep up this great work!
What a good explain! Thanks!
Another input with that can be sent to the system which covers all frequencies is white noise.
Great job. Now can you show how to take this data and build a simulation in Scilab? What is the maximum rate we can lift the paint bucket with 3N of load with minimal overshoot? What is the maximum rate we can fluid to the paint bucket without overshoot?
Hello brian,
Your videos have really made a big difference by helping me get through the control subjects in chem engg! Please could you put up a video for cascade control, GM & PM settings! :)
Really appreciate the effort you put into the videos.
Cheers,
Hi Brian, fantastic job. I am interested in learning more about system identification, what literatures do you recommend to start with?
Thanks
Edgard
I pressed Like because of the great explanatory video (and definitely not because of the xkcd t-shirt, no, it barely affected my judgement at all).
Wow imazing
Hi Mr. Douglas,
Do you know any online portal which provides the control system courses? Or is there any distant learning courses as I'm sitting in India and would like to learn this control systems. I have been searching for such courses but not found any. If no such thing exists, why don't you initiate the control system course something in Coursera or Udemy?
dude you are the best!
very explanatory 👍
is there a video for identifying the system with system response data for step input?
excellent video.
You're the best! Thanks for your videos.
Thank you so much! I really love your channel. It's gonna help me a lot in getting my thesis work done :)
Hey Brian, great video! as a control enthusiast I love your channel and I always find myself going through your videos every time I get stuck with my course work.
Anyways, next semester I will be starting my final year project and I have decide to propose a title on black box modelling of a rotary inverted pendulum, I decided on this because though it may seem easy it involves alot of topics which I personally think would give me a proper foundation in system modelling and control. What are your views on this? Is there something else apart from an inverted pendulum that would be a better option? Please enlighten me.
Thanks!