The "What Are You Doing This Moment" Thread

Nah. What's happening is that when you get a disturbance the secondary process reads it and sends the signal to its controller to minimize the upset it would have caused. As the secondary process is initially disturbed, the primary process would also be affected and in turn its transmitter would send a signal back to the primary controller where it would try to compensate for the upset in the entire process.

This is a way to correct a big spike in temperature or an overflow of a particular input especially if product specification is important. It's also just one way of doing this. There are many other ways like inferential, feedforward and ratio instrumentation control. Cascade is basically having more than one feedback control loop.

If anyone even gets that.

I had to do some reading here to catch up on the terminology but at least it makes a little sense now.

It seems to me that the only time you would have multiple control loops (i.e. a cascade architecture) is if you have multiple disturbances in the system, right? Like for the water tank example they have, the first disturbance that's being corrected in the primary control loop is the drainage of water from the tank, and the second disturbance is the water pressure from the pipes.

Is there any way it would make sense to have a cascade architecture for a system like the A/C in a house? I can only think of one disturbance in that instance, the inflow of heat from outside.
 
zabu of nΩd;9798778 said:
I had to do some reading here to catch up on the terminology but at least it makes a little sense now.

It seems to me that the only time you would have multiple control loops (i.e. a cascade architecture) is if you have multiple disturbances in the system, right? Like for the water tank example they have, the first disturbance that's being corrected in the primary control loop is the drainage of water from the tank, and the second disturbance is the water pressure from the pipes.

Is there any way it would make sense to have a cascade architecture for a system like the A/C in a house? I can only think of one disturbance in that instance, the inflow of heat from outside.

Your main disturbance in an air conditioning system would be the hot air from the outside, I believe. You don't need to have multiple disturbances to be able to set up a cascade control system because as long as there is an external force that induces changes in the system, many other factors could be affected as well.

For example, the flow rate and pressure of freon gas within the coils, the operating conditions of the motor for the fans and compressor. All these and more probably factor in and to have a system that works at close to 100% efficiency it's probably best to set up a control system for it. It really is though up to capital and operating costs. All these things cost money and that's where design optimization comes in. That's another story.
 
Just got a Netflix subscription and set it up with my Xbox and I don't think I'll ever leave my room again now outside of for bathroom breaks and food
 
Your main disturbance in an air conditioning system would be the hot air from the outside, I believe. You don't need to have multiple disturbances to be able to set up a cascade control system because as long as there is an external force that induces changes in the system, many other factors could be affected as well.

For example, the flow rate and pressure of freon gas within the coils, the operating conditions of the motor for the fans and compressor. All these and more probably factor in and to have a system that works at close to 100% efficiency it's probably best to set up a control system for it. It really is though up to capital and operating costs. All these things cost money and that's where design optimization comes in. That's another story.

Ah right, I hadn't been thinking of it that way (not that I'm trained to of course). Very interesting stuff, thanks for sharing :)
 
Your main disturbance in an air conditioning system would be the hot air from the outside, I believe. You don't need to have multiple disturbances to be able to set up a cascade control system because as long as there is an external force that induces changes in the system, many other factors could be affected as well.

For example, the flow rate and pressure of freon gas within the coils, the operating conditions of the motor for the fans and compressor. All these and more probably factor in and to have a system that works at close to 100% efficiency it's probably best to set up a control system for it. It really is though up to capital and operating costs. All these things cost money and that's where design optimization comes in. That's another story.

So essentially an example of a cascade control system could be that of a cars on-board diagnostics system. It accounts for external and internal variables and adjusts accordingly. Also, one fucking thing goes wrong and you may be shit out of luck in regards to a few other things.
 
So essentially an example of a cascade control system could be that of a cars on-board diagnostics system. It accounts for external and internal variables and adjusts accordingly. Also, one fucking thing goes wrong and you may be shit out of luck in regards to a few other things.

Pretty much, except a car (which is a pretty complex piece of machinery) also has various other control systems running such as a ratio system that controls the pistons, fuel release that work in conjunction of the RPM meter and the odometer. Also, cars with cruise control work on a feedforward system that predicts disturbances such as friction against the tyres and slowing down during turns and compensates for this by maintaining velocity specified. I could go on about multi input multi ouput (MIMO) stuff as well but I should probably go back into doing my assignment and studying.
 
About to shower and then get some food. Supposed to be at class but didn't feel like it today...also have to umpire a high school baseball game tonight at 4:30. Bleh.
 
I spent the whole afternoon in bed sleeping / fapping. Now it's time for a calm night w/ death metal, Prokofiev, and maybe some poker.
 
Haven't been on UM for a while. Been committing heinous acts of drunkenness all weekend and then getting settled in at home. Missing my lady-friend. Fixed my computer, but the internet connection here is super slow.
 
Thinking about the hockey game between Sweden and USA tonight.

lbaqc.jpg
 
Doing some job searching. The job outlook doesn't seem good, and in my discipline it was comparatively weak pre-recession. That's something that I've always been aware of. I might have to look out of state for a job. The particular job market I'm dealing with is very specific and niche, and I don't know how it relates to the broader job outlook right now. I would like to do a semester or year-long contract and then come back to Arizona. If I plan on doing a Phd program, I won't want to be tied down by a long-term job. It looks like a lot of the offerings for philosophy instructor jobs are short-term contracts anyway. Would be just fine for me.

Well, two places I'm looking at applying to are the Pima Community College system in Tucson and Portland Community College (Portland, OR). Two problems with this idea: if I end up getting a job in Tucson, well, ummm, the girl I just broke up with lives there, and it ain't exactly hard to bump into people in Tucson. On the other hand, if I got a job in Portland, I would have to be bombarded by constant memories of her and all the stuff we did there..."oh look, here's where we had lunch that one day,""here's where we took that cheesy picture together," etc. etc. BUT I really like the Pacific Northwest and would like to spend more time there. This is all assuming that I get a job at any of these places, and I don't know how likely that is.