Using a mathematical model to figure out the hysteresis of the temperature output to the temperature measured at the sensor and delay of heat transfer from one location to another is one way to approach this issue.
In this case I can see where LR or KNN will do the job.
LR= Linear Regression
KNN = Nearest Known Neighbor.
The way I see it is that some research needs to be done on the heat propagation from source to target. Now whatever solution you decide to use, in the measurements there should be enough data gathered anyways that a math model could be used, if so desired.
I've not been able to see any of your attachments (photos, diagrams). This makes understanding the difficulty of the thermal calculations.
For simple shapes it is not too difficult to calculate the propagation of temperature from one area to another.
Ignoring convection cooling for them moment and discounting radiation cooling as too small to matter (at the temperatures you cited).
Heating a homogeneous material is very much like an electrical R and C.
Thermal capacity is the same as a capacitor
Thermal conductivity is the same as a resistor.
You can use a spice type program to calculate the times etc. You can do it by hand if the form factor is simple enough. However the most accurate solution will likely entail some finite element analysis. Sounds advanced but again depending on your form factor, you can break the conduction path into smaller pieces estimating the resistance by the cross section and capacitance by the volume of material of the element. I'm not sure I am communicating it clearly enough, its crystal clear in my mind but difficult to put into words. Especially since I don't know your background.
Google thermal calculations for heat sinks and you should get a wealth of information. Perhaps even an on line calculator to help you.
You could perform some experiments to determine the time constants from point to point, however the addition of some calculations would likely add to your project acceptance. Perhaps do the calculations and verify with some testing.
I have a fundamental problem with understanding this…
…neutralize the temperature coming from region A to have 23C at maximum depending on how fast the heat is been transferred. from the region A. The signal will be givn to peltier element in region B…
That suggests you are heating both ends of the sample?
The only temperature that ‘moves’ is heat, so your system appears to be chasing itself.
Peltier element is a device that produce cold at one end and hot at the other end literally. so I'm using one as heater and the second one as a cooling device
Electrical background and I graduated a while ago, just started my MSc basically. I've gathered few things to progress on this work.
Thank you guys
I will keep you posted
Trying to get some data from both end and see how fast heat is been transferred to the point. Another I'm currently having is getting data with Macbook. PLC DAQ isn't working neither is data streamer on excel working for mac as well
what can you suggest please
Ok, sorry, so your heating A, and following / cooling B, unless you’re trying to slow the transfer? which would involve heating B…
Movement of energy implies migration of the states from ‘hot’ to ‘cold’, unless you intend to reverse the flow by heating B, and letting A follow by effect of conduction.
Hello guys. I'm using 7 DBS1820 sensors and 'm unable to print device address for my sensors. which i couldn't locate which sensors is which how my material. can you please work me around this
Yes. you are right. I saw a sample code from AZ-delivery to get the device address of each sensor but its not working. saying error compiling for arduino... below is the code please kindly help me check the errors
Hi Everyone. Thanks for the advice, I really appreciate. I'm able to progress but at the same time i am currently hanging . I want to take decision from the data I got but it's confusing. Heating is moving from region A to B at a constant temperature and my aim to stabilize the temperature from B before it reaches 24.5C in the midpoint such that it won't exceed 24.5. Attached are the graph and conceptual diagram of the temperature control
How can I justify when to switch on my controller so it will be able to regulate the temperature to 24.5C considering how fast the heat is transfering (time constant). Thanks
Hi..
I'm working on a project and it's confusing on how to do the prediction or when to start cooling. I'm using an aluminum material such that heat is been transferred from one end and I'm cooling from the other end of the material. I want to control the cooling such that the temperature won't reach the setpoint before switching on my controller... NB: the controller shouldn't ON the cooling instantly until it feels the temperature and I have to use pid controller...
What your setup is and what you are trying to do is not very clear. What's the "aluminium material"? A heat sink? What are you doing to control the cooling? You operate a fan?
I'm working on a stabilizing temperature on aluminium material which the dimension is shown in the picture. A peltier element is placed at both end, A is for heating and B is for cooling. A is heating the material and I'm not controlling this TEC at anytime and this can heat up to 25.5C in the midpoint but the cooling TEC which is B that I'm controlling to cool down the material and can cool down to 24.5C. the numbering on the material is position of the sensors to monitor how fast the heat is transferring in the system from A and cooling from B.
I'm struggling with transfer function of this system
And implement PID to control this system. Please I need advice on this. Thanks
I've now merged 3 topics on the same subject from you and found that you posted the same question on Stack Exchange. Please stop posting duplicate questions. If you continue to do so you will have a time out from the forum.