Important PID Concepts | Understanding PID Control, Part 7

[@BrianBDouglas]

Thanks for watching! If you have questions for me, leave them on this comment so I get notified and can do my best to get them answered. Cheers!

[@zainahmad7502]

how can i built hybrid fuzzy pid controller ? on simulink and connecting to port (arduino...)

[@BrianBDouglas]

For fuzzy PID check out: www.mathworks.com/help/fuzzy/examples/using-lookup-table-in-simulink-to-implement-fuzzy-pid-controller.html And for connecting Simulink to an Arduino check out: www.mathworks.com/hardware-support/arduino-simulink.html and a short overview video: www.mathworks.com/videos/introduction-to-simulink-support-for-target-hardware-68955.html

[@manuelaugustosisko9758]

Hi! Awesome series! It's been really helpful. I need guidance on a system with multiplexed blocks. If you could point me to a useful source I'd be grateful. I need to code a Controller for a Pump + 2 Valves closed pneumatic system (one valve for feeding air flow into the container and one for releasing air to the atmosphere). The resultant input flow would be A x B, where A is the pump motor control output flow and B is the valve open duty. I need to control both variables with the same input. Any kind of hints could help as I haven't found any info on such a topic.

[@AliHassan-xt1xb]

would you please do a series on Model predictive control.

[@BrianBDouglas]

MATLAB already has a series on Model predictive control: kzbin.info/www/bejne/boaTqZyFoKmWo9k

[@i8tweak]

I earned my EE degree 3 years ago and went to sea running engine rooms. This series was an AWESOME refresher for my interview tomorrow designing PID controllers. Wish me luck and thank you, Mr. Douglas!

[@BrianBDouglas]

Good luck! If you don't mind sharing here, I'd love to know how the interview goes. Cheers!

[@i8tweak]

Just had it today. It was great because using almost exclusively the content you covered in this series, I had an honest discussion about taking my PID/PLC knowledge to a professional setting without going off-base with any facts. I impressed them. Then right after the interview we drove to a water treatment plant and helped them fix a VFD controller. Luckily I proved myself, providing novel ideas leading to a quick solution just based on my experience handling ship plants. THANK YOU. I could not have pulled it off as well without your insight.

[@BrianBDouglas]

Well,​ that sounds like a pretty positive outcome. Congrats!

[@ericjw123]

As a retired EE now playing with drones and tuning PIDs this is an excellent and useful refresher. I especially like seeing the different methods you can use to model real devices.

[@monian9311]

3 months of struggles, learned in 7 videos! way too insightful Brian, Clean and crisp.

[@OrangeDurito]

I cannot thank you enough for these videos!!! In just 7 videos, you tied together all the concepts I have learned so far in my 2 semesters of graduate coursework and gave me an intuitive practical understanding of the mathematics we deal with. And it is going to be immensely helpful in deciding the direction of my thesis. You are an incredible teacher and I am eternally grateful to have come across your videos.

[@quoctrungtran8301]

I watched lots of videos from your own channel to this chanel and i have learned a lot, in a way that's simpler than any other. I just can't thank you enough for your knowledge and your explaination skills. Hope that i can apply more of your knowledge in the future and i am looking forward to see you in your new videos. Again, lots of appreciation and wish you the best Brian.

[@jmcarter9t]

This series is simple, clear, and quickly digestible -> great stuff! Thank you sir!

[@benjamincouillard8391]

For Power convectors (AC/DC, DC/AC and DC/DC) the cascaded loop architecture is also ubiquitous. Great video!

[@yasir9909]

Your lectures are quite informative, to-the-point and easy to grasp...

[@AliHassan-xt1xb]

Hats off to your explanation skills. would you please do a series on Model predictive control.

[@eng.athraa5707]

I wish that also

[@aryamallick5426]

enjoyed ur series brah ,really wow. ur quadcopter example 🤣🤣 stayed wid u and worked pretty awesome from start to the end. love u bro for telling us the story of pid controllers.

[@sashamuller9743]

"i'm brian and welcome to a matlab tech talk" is what i find myself saying on a daily basis now

[@shubhamkothari2709]

Thank you for this wonderful series!!You are awesome.

[@EssJay1095]

"If certain conditions are met, these imperfect representations are good enough!" - Engineering in a nutshell

[@100MillionThoughts]

⚡️ Tuning a PID Controller Using the Ziegler-Nichols Method | MATLAB code available kzbin.info/www/bejne/bn6wlIOCidyildU

[@khawar393]

Awesome Lectures ...simple and Informative

[@a.g.r.6578]

Thanks Brian excellent PID Series.I learnt a lot!

[@minghuayao5952]

brilliant! thank u so much! easy to understand in the video but also brought up much to lead us learn deeper. googleplex thumbs up!

[@ruslangorenko8917]

Nice video! Keep up the good lectures!

[@Antonioqwert3868]

Hi, Brian your videos about control theory is great, I hope you will talk about MPC controller

[@Drone.Robotics]

Thanks for sharing our knowledge

[@eng.athraa5707]

thanks for ur course to explain the PID controller ...After I washed all ur videos I understood what is PID controller. can u explain the model predictive controller???

[@MaciekDrozd]

Many thanks for your excelt video. How to discretize my PID design if I have no fixed sampling time? I mean, I use Hall for motor speed what gives speed update dependent on speed. Rotating at 60RPM I get uptade once per second and don't need to run PID more often. At 60kRPM I should run PID 1000x times more frequently. Discretization like Tustin gives very different results for such a diffesrent sampling times.

[@BrianBDouglas]

If you only need the PID controller to run at 1 Hz, then I wouldn't tie it to the asynchronous output of the hall sensor. It sounds like you only need speed information every second so if that meets your performance then I'd stick with that. To run the PID controller you need the current error, the rate of the change of the error and the integration of the error ... but even though you only need that 1 once per second, you could (if you want) have a sensor processing step that is keep track of those things asynchronously but returning the current value only once per second. So for example, you might average the speed for the last second and report that as the current error. If you can get away with it from a requirements perspective, you might be able to just always grab the latest measurement and assume it was constant for the last second.

[@MaciekDrozd]

@@BrianBDouglas At first I'd differentiate Current-PID and Speed-PID (they are exactly PI). They are cascaded. Current-PID takes part cyclic with calling frequence respecting electrical properties of motor. Due to low time constant of winding it is 40 kHz. Whatever the Speed-PID calling frequency is, this controller can not update the error until Hall signal changes or timeout happens (basic stop prediction). Hall edges come from single Hz to 1kHz in useful speed range. Otherside the Speed-PID execution with mentioned 40kHz is significant load for MCU without any benefit from this high rate. Therefore the idea to ececute Speed-PID with much reduced rate. I think 2kHz execution fixed should work for me. But I still stay courious, hot to model (z or s transform) non-fixed sampling or event based sampling of PID controller. Thank you so much for your answer and for really good learning stuff.

[@ruddymauricionishinojustin8738]

Nice videos, really helpful! I'm trying to build a model for the control of different levels of water in a tank, that is filled and emptied by two valves, respectively. It would be great if you can help me or advise me where I can find information about it.

[@rajurambha4611]

nice sir. Could you please upload PID control for non linear systems in matlab simulink. while auto tuning it is showing linearisation error for BLDC motor control.

[@phillipmaser132]

If I am running two pumps maybe cascaded them would be the answer? Only one pump needs to be on. If the distance of the setpoint is large big pump is on. otherwise small pump is on.

[@moseschuka7572]

I always hit the like button

[@hassanismail6120]

can this pid controlling apply on a robotic arm servo / stepper motor?

[@uchennaagujiobi4641]

Pls Brian, I need help on dc-dc boost converter for Incremental conductance in MPPT

[@Minji_Hanni_Dani_Haerin_Hyein]

7:00 & 7:09 state space or state based

[@luismesquita8202]

Video on PID control: kzbin.info/www/bejne/bWLSZZ98h6eNpM0

[@Amine-gz7gq]

Control system is hard 😭I'd better stick to software programming

[@wolfrobots118]

I've been watching all of these videos and to be honest with you you make it way too complicated.... And how many parts is there to this PID video anyways. You're tone makes it sound like it's super easy but to be honest with you it looks like a bunch of gibberish. I'm not a rocket scientist I'm just a robot creator that wants to know about PID the way you're explaining it I would need to be a rocket scientist to even remotely understand.

[@OrangeDurito]

What the heck are you saying? This series is the most intuitive explanation of PID control anywhere. Any control system engineer will tell you that. Maybe you don’t know the actually technicality of the topic and only know the basic implementation. This is not 5th standard stuff but he actually brought down the graduate level concepts to a high school level. There are thousands and thousands of pages on this topic, going deep into the mathematics, implementation, documentation, software - you can just keep on reading for years and not really understand what it all means in real life. He condensed those thousands of pages into an hour long series. This is teaching at its most awesome form.

[@nadiaakter9351]

Anyone please help? After using PID controller initially, how may we got propeller speed command. I mean, here deferences between altitude has been measured!! Please help

[@nadiaakter9351]

And in second pid, how did we get voltage after calculating the error of speed!! Please help

[@Amine-gz7gq]

control system is hard dude, rocket science is indeed control system