Assistente AI
Trascrizione
00:05:240Jacopo Vivian: Okay, I think we can start the
00:08:580Jacopo Vivian: as you, as you see today, as you can see today, we talked about the pulling their systems.
00:19:55Jacopo Vivian: Yesterday you did the lecture about triggers. You know now what the trigger is.
00:27:260Jacopo Vivian: And today we see some concepts regarding the design and also the control of systems, including shimmers.
00:40:920Jacopo Vivian: 14 Hvac systems that are used to
00:45:800Jacopo Vivian: provide the space cooling today in the working model.
00:52:400Jacopo Vivian: So 1st of all.
01:02:260Jacopo Vivian: 1st of all, we just
01:04:879Jacopo Vivian: take some concepts back from the previous lectures. Remember the difference between the regulation and balancing. Today we speak more about the regulation rather than balancing. So we speak about how to control the systems during partial load operation. Okay?
01:30:210Jacopo Vivian: If you remember, we talked about the
01:34:540Jacopo Vivian: 2 withouts or 3 withouts, they can be used in different ways.
01:41:450Jacopo Vivian: So if you use a diverting valve on the supply or a mixing valve on the return.
01:50:650Jacopo Vivian: you are controlling the flow of the system?
01:54:580Jacopo Vivian: If so, you have a constant supply temperature
01:59:80Jacopo Vivian: vice versa. If you have a mixing value on the supply or a developer valve on the return.
02:08:570Jacopo Vivian: You are changing the flow to the
02:13:850Jacopo Vivian: you are. You are changing the supply temperature to the longer.
02:18:530Jacopo Vivian: Okay, and you have a constant flow to the to the load.
02:23:950Jacopo Vivian: So these are just 2.
02:25:910Jacopo Vivian: Stop the Jesus on that terminal unit side. Okay,
02:37:700Jacopo Vivian: of course, I was talking about 3 way valves
02:40:900Jacopo Vivian: in case we have 2 revolver, we only have the option of variable.
02:47:800Jacopo Vivian: Okay? So the supply temperature is the same.
02:50:970Jacopo Vivian: and we cannot only reduce the flow
02:53:980Jacopo Vivian: by introducing an A, an additional Delta P on the, on the line connecting the certain terminal unit.
03:06:310Jacopo Vivian: Change the flow, and then, yes, depending on where, how you position the the 3 way value.
03:16:30Jacopo Vivian: If you remember, I I draw that on the on the blackboard.
03:22:84Jacopo Vivian: So if you have a mix involved on the supply, or a diverter bar
03:31:200Jacopo Vivian: for a diverter valve. On the return
03:34:220Jacopo Vivian: you are changing the supply temperature to the terminal level
03:40:460Jacopo Vivian: vice versa. You are changing the flow.
03:46:470Jacopo Vivian: If you have a mixing valve on the return.
03:49:830Jacopo Vivian: or a divertor valve on the supply. In that case you are changing the no.
03:58:350Jacopo Vivian: the flow rate to the terminal unit. And you have the same supply temperature to the terminal unit.
04:04:450Jacopo Vivian: Okay?
04:07:320Jacopo Vivian: Okay.
04:09:210Jacopo Vivian: We said also that when you have such regulation on the load side. We can do something that
04:21:29Jacopo Vivian: to reduce the energy consumption of auxiliary systems like pumps.
04:28:140Jacopo Vivian: And 1st of all, we see what happens. If you. If you don't
04:33:660Jacopo Vivian: change the speed of our pumper, basically at the partial load.
04:41:20Jacopo Vivian: you have an increase in the Delta P, and the reduction in the flow rate that shift
04:49:110Jacopo Vivian: the new operating point of the pump from the point A to point B, okay.
04:58:470Jacopo Vivian: this is the result of closing, for example, to without
05:05:630Jacopo Vivian: that will introduce further the PIN,
05:09:440Jacopo Vivian: and will reduce as a consequence, they will reduce
05:13:450Jacopo Vivian: the low, the sorry, the flow rate, the circulating flow. I need.
05:21:550Jacopo Vivian: We said that in order to reduce the energy consumption of banks.
05:29:50Jacopo Vivian: we can, if we have a viable speed bumps, we can
05:34:734Jacopo Vivian: regulate the speed so that instead of shifting the
05:41:110Jacopo Vivian: the new operating file from A to B, we shifted in this case to B first.st
05:47:630Jacopo Vivian: The 1st step, as you can see, is the intersection
05:51:500Jacopo Vivian: between the same characteristic curve of the Cf. With with the valves closed.
05:57:760Jacopo Vivian: the green la, the green dashed line if you see here.
06:03:470Jacopo Vivian: but we make the intersection with the new
06:07:250Jacopo Vivian: characteristic curve of the pump which is related to a lower speed.
06:13:470Jacopo Vivian: Okay?
06:14:890Jacopo Vivian: And of course, since the
06:17:510Jacopo Vivian: power consumption of the pump is given by the product of the delta P, and the flow rate
06:24:820Jacopo Vivian: in the new point B, first, st the electrical consumption of the pump
06:30:690Jacopo Vivian: is lower than the electrical consumption of the pump at the nominal speed.
06:38:380Jacopo Vivian: Okay.
06:42:950Jacopo Vivian: all right.
06:44:960Jacopo Vivian: In this case. Okay, it depends on how you set the curve
06:50:430Jacopo Vivian: that connects A and b 1
06:53:950Jacopo Vivian: be first.st So this line here.
07:00:960Jacopo Vivian: I take the pointer. So this
07:04:590Jacopo Vivian: line here connecting these 2 points
07:08:270Jacopo Vivian: is actually a line that can be set in the controller.
07:12:910Jacopo Vivian: Okay, so, depending on how you set this line in the Controller.
07:18:470Jacopo Vivian: you can have a constant flow or a reduction in the flow.
07:24:980Jacopo Vivian: Okay, it always depends on.
07:28:410Jacopo Vivian: In this case this is at the constant delta P. This is a constant delta p regulation, and therefore
07:38:210Jacopo Vivian: in this case you have only a very slight reduction of the flow, but you could have.
07:45:210Jacopo Vivian: if you take the slides, different lines that give you also a further reduction of the flow.
07:53:190Jacopo Vivian: Okay?
07:54:630Jacopo Vivian: And of course it would be better to reduce the flow as much as possible.
08:00:570Jacopo Vivian: I guess this was the purpose of your question in order to reduce the
08:06:750Jacopo Vivian: they have electrical consumption at partial load, but below a certain flow you
08:14:150Jacopo Vivian: you have the unit, not providing enough cooling or heating.
08:21:130Jacopo Vivian: and therefore there is always a trade off.
08:24:430Jacopo Vivian: Okay?
08:29:680Jacopo Vivian: So now we speak more explicitly about the pooling.
08:36:590Jacopo Vivian: because in pooling systems we have some constraints that are typical of pooling systems like
08:46:860Jacopo Vivian: basically, the the main limit we have is
08:51:660Jacopo Vivian: a limit on the mass flow
08:54:870Jacopo Vivian: rate that can circulate in the evaporator of the chiseler.
09:00:340Jacopo Vivian: Okay? So in the cold side of the team level.
09:04:960Jacopo Vivian: in particular, we have a range depending on a maximo
09:12:222Jacopo Vivian: or sorry minimum mass flow rate.
09:16:600Jacopo Vivian: because the masque rate circulating in the evaporator should be fine after
09:22:740Jacopo Vivian: to keep a high convective heat transfer in the evaporator.
09:28:580Jacopo Vivian: Otherwise you have possible ice formation in the evacuator.
09:38:660Jacopo Vivian: Okay?
09:43:250Jacopo Vivian: And the other, on the other hand, you need to
09:48:680Jacopo Vivian: you, you should consider having a maximum mass flow rate, because otherwise you have excessive pressure drops and excess excessive expenditure of the pubs.
10:03:280Jacopo Vivian: Basically, right?
10:07:10Jacopo Vivian: So this mean means that, if possible, we should
10:13:690Jacopo Vivian: keep the masculpator very close to its nominal value in the in the operator.
10:22:920Jacopo Vivian: So that is a consequence.
10:26:400Jacopo Vivian: Which is it that the we we are constrained by this requirement of achievers?
10:35:410Jacopo Vivian: Because on the primary circuit of the cooling stations.
10:39:630Jacopo Vivian: we need to have an almost con- constant flow.
10:43:780Jacopo Vivian: Okay?
10:45:890Jacopo Vivian: So now we apply the concepts that we have seen on hydronic systems. To this specific situation.
10:56:840Jacopo Vivian: The consequence is that we have the 2 options.
11:01:190Jacopo Vivian: The 1st option is called capital distribution.
11:04:910Jacopo Vivian: It means that the flow is constant both on the primary and on the secondary circuit.
11:13:190Jacopo Vivian: Now we will check out.
11:15:860Jacopo Vivian: The second option is that you have a decoupled distribution, meaning that the flow on the secondary side
11:24:60Jacopo Vivian: depends on the load, and the flow in the primary side depends on this.
11:31:540Jacopo Vivian: Yeah need of the children. So the flow remains constant in the primary circuit.
11:41:680Jacopo Vivian: But in the secondary circuit we have an independent flow.
11:46:230Jacopo Vivian: That's why it's called decoupled, because the 2 2 circuits aren't decoupled.
11:53:940Jacopo Vivian: So oh, this is an example of coupled distribution problem.
12:05:640Jacopo Vivian: And then, basically, you see that in order to have.
12:14:530Jacopo Vivian: in order to have the same flow.
12:17:810Jacopo Vivian: even at partial load. We need to install 3 way valves
12:24:220Jacopo Vivian: on the return of the terminal unit. Imagine this to be a point right cool, so that the
12:34:450Jacopo Vivian: at partial loader.
12:36:940Jacopo Vivian: He's 3 way value on the return will open and back to the flow, and, like some
12:49:80Jacopo Vivian: some flow bypass the terminal unit.
12:53:800Jacopo Vivian: Okay,
13:00:580Jacopo Vivian: Here you have the tuners.
13:03:680Jacopo Vivian: Right? You have 3 refrigeration units.
13:08:710Jacopo Vivian: Each of them has enough circulation power.
13:12:940Jacopo Vivian: Okay, so that, for example, if the if this bank switches on
13:20:970Jacopo Vivian: this chiller will have exactly the fluorator given by this pump, and the sum of the of the fluorator will go always to the
13:32:920Jacopo Vivian: to the terminal units.
13:35:120Jacopo Vivian: So in this case the main disadvantage is that we have a lot of
13:44:730Jacopo Vivian: we have 2 disadvantages. The 1st is that we have a lot of
13:50:236Jacopo Vivian: flow circulating in the system also at partial load.
13:55:920Jacopo Vivian: and the circle is on the
13:58:490Jacopo Vivian: sizing side, because we need to size all the pipes
14:04:490Jacopo Vivian: according to the nominal flow rate.
14:09:600Jacopo Vivian: Okay, even if those even if those units do not require the
14:23:820Jacopo Vivian: the cooling simultaneously, you always need to size a project to the maximum to the nominal cooling power.
14:38:570Jacopo Vivian: Okay, what happens at partial flow at partial loader?
14:47:160Jacopo Vivian: We need to synchronize the their refrigeration units.
14:54:210Jacopo Vivian: Okay, because otherwise you would have an imbalance in the flow and paying each unit.
15:08:990Jacopo Vivian: I think I would show better with another slide what what I mean.
15:14:340Jacopo Vivian: But for now I just want to show you that
15:22:690Jacopo Vivian: you can have 2 options here. This is always couple distribution.
15:27:810Jacopo Vivian: So you always have the nominal flow going around all the equipment.
15:34:710Jacopo Vivian: What changes is the position of the parent?
15:38:630Jacopo Vivian: So here the main circulation pub that provides the head to the whole system are individual
15:49:240Jacopo Vivian: pumps, one for each chiller.
15:52:930Jacopo Vivian: And in this case you have. You have them grouped.
15:58:240Jacopo Vivian: Okay, here you might have. Maybe
16:03:70Jacopo Vivian: they. They don't need to be 3. Okay, there might they? There might be, for example, 2 of these working, and the 3rd to be a backup, for example. So it depends.
16:16:210Jacopo Vivian: And then I'm gonna flow rate
16:22:444Jacopo Vivian: in both cases the secondary statement must be sized according to the nominal.
16:30:460Jacopo Vivian: to the nominal cooling, to the nominal cooling power.
16:34:770Jacopo Vivian: Right?
16:43:630Jacopo Vivian: Yes, right now we're doing so. I'm gonna
16:50:731Jacopo Vivian: you need to oversize the pipes. Yes, revolve.
17:05:819Jacopo Vivian: Yes, which means having higher elect electrical expenditure for the South Pacific. Yes.
17:16:970Jacopo Vivian: Now, I think you can understand better what I mean by synchronizing.
17:23:79Jacopo Vivian: This is an example where you never
17:26:880Jacopo Vivian: the we are always talking about couple distribution. Now, okay, we will talk about the decoupled later.
17:35:170Jacopo Vivian: So this corresponds to the situation on the right here. Okay, so we have a common pump
17:45:290Jacopo Vivian: or 2 thrillers in this case.
17:50:830Jacopo Vivian: So now, I just want to show that if we switch off one shield, okay, what happens
18:03:390Jacopo Vivian: that the supply temperature on the oh, switch off to your side
18:10:380Jacopo Vivian: is the same of the return temperature in this case, in this example is 12.2,
18:17:820Jacopo Vivian: and therefore all the load. All the cooling load is supplied in this situation by chiller number one.
18:28:510Jacopo Vivian: So since there is always a there is always low circulation got it?
18:37:850Jacopo Vivian: It means that the true Masquelator will meet here.
18:43:790Jacopo Vivian: We'll meet, sir, and the corresponding supply temperature to the loads will be 9.4.
18:52:680Jacopo Vivian: Okay, okay, this is just an example. It depends on the master rates. But that you can.
18:58:420Jacopo Vivian: You can understand that we don't have the situation. The typical situation of a supply temperature of 7 Celsius degrees.
19:09:280Jacopo Vivian: and the return temperature of 12 Celsius degrees, which is what we want not always to.
19:18:710Jacopo Vivian: Why do we have these, these, typically this temperatures in cooling systems.
19:27:370Jacopo Vivian: Do you remember?
19:31:400Jacopo Vivian: Why do? Why do we typically have 7, 12 in some in cooling systems?
19:39:630Jacopo Vivian: Because we also want to dehumidify
19:43:570Jacopo Vivian: in order to dehumidify, we need to have low temperatures
19:49:360Jacopo Vivian: so that we reach the new point of the air, and they
19:54:570Jacopo Vivian: moisture can condensate in the terminal unit.
20:01:630Jacopo Vivian: So you can. Also, if we have an environment at the
20:08:460Jacopo Vivian: 26 degrees, and you want to keep it cool.
20:13:80Jacopo Vivian: Okay, you can also provide pulling at the 18 degrees.
20:19:740Jacopo Vivian: Right? We are always cooling this the room because it is 10 degrees lower than the air.
20:27:640Jacopo Vivian: So you are actually pulling that indoor space.
20:32:400Jacopo Vivian: but you are pulling only on the sensible side.
20:37:290Jacopo Vivian: You are not removing moisture from the air
20:41:130Jacopo Vivian: so that this could pose a problem of high relative humidity in the in the space, because you are not removing the excess moisture in the air.
20:53:670Jacopo Vivian: Yes, you've been trying.
21:01:50Jacopo Vivian: Yes, instead of selling, because the viewpoint temperature or well, for 26
21:13:500Jacopo Vivian: census degrees and 50% relative humidity, which is the normal set point in the cooling season.
21:20:920Jacopo Vivian: is about 14 degrees.
21:24:440Jacopo Vivian: So you want to have both supply and return below 14 degrees. That's why we keep the 7 12.
21:33:120Jacopo Vivian: So if the if the supply is 9.4,
21:41:810Jacopo Vivian: we might have some terminal units that are unable to.
21:46:760Jacopo Vivian: They humidify the air.
21:49:610Jacopo Vivian: Okay, so
21:56:460Jacopo Vivian: or even on the sensitive, on the sensible pulling loader, we might have some areas that are not.
22:06:680Jacopo Vivian: No, that well, the the peak load is not fulfilled because we have a too high supply temperature. That
22:19:70Jacopo Vivian: so the the only option in this case would be to compensate for this hi Eric
22:31:10Jacopo Vivian: temperature by reducing by far the reduction of this temperature.
22:37:10Jacopo Vivian: Okay, so you only have children number one in operation.
22:42:390Jacopo Vivian: and in order to approach, to go to approach
22:46:520Jacopo Vivian: the temperature set point of 7 on the supply side.
22:51:370Jacopo Vivian: you reduce the supply temperature on of chiller number one.
22:57:620Jacopo Vivian: but this, of course, is not efficient, because the lower the supply temperature, then lower the
23:07:190Jacopo Vivian: cop or er of the chiller, because the difference between the source and the sink is higher.
23:19:750Jacopo Vivian: The sink is the outside, there 30 degrees
23:25:480Jacopo Vivian: the sync. Sorry. The source is our water of the cooling system.
23:34:340Jacopo Vivian: Of course, sir, have been going 7 degrees here is better on the than having 3.
23:47:780Jacopo Vivian: Okay, it will give us the higher efficiency of theater number one.
23:55:650Jacopo Vivian: So
23:56:980Jacopo Vivian: we can compensate this situation by reducing the supply temperature. But the er the energy efficiency ratio of the killer number one. We
24:07:810Jacopo Vivian: decrease. And also we have a lower limit
24:12:480Jacopo Vivian: after 3 Celsius degrees, because otherwise we are producing ice.
24:18:840Jacopo Vivian: Okay, if we go too close to 0 Celsius degrees we have the problem of ice formation.
24:27:380Jacopo Vivian: and this is something that needs to be avoided.
24:31:720Jacopo Vivian: Yes.
24:37:700Jacopo Vivian: we can reduce.
24:40:170Jacopo Vivian: How can we work with?
24:45:940Jacopo Vivian: It's a the internal regulation of the children.
24:51:420Jacopo Vivian: That depends, sir, on the internal parameters of the machine, like the refrigerant flow in the cycle.
25:05:335Jacopo Vivian: Yeah, how can I win the evaluation implementation of that pretty much.
25:18:580Jacopo Vivian: or I can change. If there is some ways like
25:29:880Jacopo Vivian: you can change the the speed of the compressor
25:35:689Jacopo Vivian: and the operating points of the cycle will change to a certain extent.
25:44:100Jacopo Vivian: I am not an expert on this, but you have some freedom in regulating the supply temperature
25:51:180Jacopo Vivian: at partial load.
25:53:280Jacopo Vivian: Of course you can't provide the
25:58:90Jacopo Vivian: completely different temperatures, but you have some freedom to change.
26:04:240Jacopo Vivian: It's the same of a heat pump. How can heat pump provide them 45 degrees instead of 50,
26:15:620Jacopo Vivian: because you change the operating points of the
26:21:340Jacopo Vivian: cycle on the cycle side, so that in case of the heat pump. You have
26:28:380Jacopo Vivian: a superheated point at the end of the compression, which is higher.
26:34:470Jacopo Vivian: so that in the condenser side it has it
26:38:410Jacopo Vivian: a higher heat flow rate that will heat up the water coming back from the heating system.
26:46:100Jacopo Vivian: Similarly, you can do something on the with cheer. Okay, it's just the the other way around instead of
26:55:250Jacopo Vivian: oh, instead of increasing the temperature on the condensation side.
27:02:640Jacopo Vivian: you reduce the temperature on the evaporator side.
27:08:390Jacopo Vivian: But the principle is the same other questions.
27:15:640Jacopo Vivian: Yes.
27:27:560Jacopo Vivian: Oh, if you have both of them on, you have 12 here, 7 here, and 7 here.
27:39:280Jacopo Vivian: and 7 at the exit, of course. And
27:42:570Jacopo Vivian: because of my, I was thinking about the selling point. So we can't if we shut off.
27:54:210Jacopo Vivian: So we keep it.
27:56:910Jacopo Vivian: The initial adversary lower number the global call.
28:06:210Jacopo Vivian: Because we can go to the hotel.
28:08:950Jacopo Vivian: we can go below 7. We can't go below 3 Celsius degrees. Yeah.
28:16:340Jacopo Vivian: there is not an option of the open portal recording.
28:21:70Jacopo Vivian: Yes, indeed, this is just an example I took from a chiller manufacturer.
28:35:20Jacopo Vivian: because this depends on what the chiller specific chiller can provide. I don't know, maybe can provide you 5.
28:43:920Jacopo Vivian: It's just an example, that of what what will happen. Okay, this tuner
28:51:200Jacopo Vivian: will be regulated can be regulated in order to reduce, compared to the previous situation, so that the this
29:00:890Jacopo Vivian: will be lower than 9.4.
29:05:650Jacopo Vivian: Don't look at the specific numbers for okay, okay.
29:14:553Jacopo Vivian: always, the couple distribution. In this case, we have what do we have?
29:34:990Jacopo Vivian: Okay.
29:36:250Jacopo Vivian: okay, this. This is different from the previous example. Because here we have individual pumps, individual circulation pumps. Okay?
29:47:450Jacopo Vivian: And in this case we don't only closer
29:54:210Jacopo Vivian: with the the chiller, the compressor. Okay, but we also stop the flow
30:01:980Jacopo Vivian: in China in this China, because it is off.
30:08:30Jacopo Vivian: we can do it. We could not do it before, because before we only have one pump for 2 children
30:15:670Jacopo Vivian: that was in common on the return side.
30:18:950Jacopo Vivian: Here you can see we have 2 parts.
30:22:550Jacopo Vivian: one for distributor and one for distributor.
30:26:290Jacopo Vivian: So in this case that we can decide to
30:31:410Jacopo Vivian: stop also the bank and to keep on the and to keep on providing them.
30:39:550Jacopo Vivian: The mass flow rater, the numer masquerade to the tumor that is still on.
30:47:600Jacopo Vivian: Okay, what happens in this case?
30:54:590Jacopo Vivian: It happens that when the cooling load is lower than 50%, there is a
31:03:230Jacopo Vivian: a 60 to 70% drop in the flow rate
31:07:980Jacopo Vivian: as a consequence of the new operating point.
31:12:390Jacopo Vivian: Okay, between now to banter and the characteristic curve of the seafood.
31:18:510Jacopo Vivian: Just the range.
31:20:100Jacopo Vivian: Okay?
31:21:890Jacopo Vivian: So the terminal units that are farthest away from the cooling station can
31:34:390Jacopo Vivian: can be can be a disadvantaged in terms of and no. Okay.
31:46:110Jacopo Vivian: In this case, we don't have a problem of supply temperature.
31:51:770Jacopo Vivian: Okay? Because now we are working at.
31:56:880Jacopo Vivian: imagine this working at the nominal loader.
32:00:150Jacopo Vivian: Okay, yeah, there is offer. So the supply temperature is okay.
32:05:160Jacopo Vivian: But the overall flow going to the terminal units is very low.
32:11:270Jacopo Vivian: The fights are oversized because it is still the couple distribution, then coupled distribution.
32:19:950Jacopo Vivian: So we have a lot of pressure drops, and then
32:26:980Jacopo Vivian: the very far away terminal. You need the wind shocker
32:30:930Jacopo Vivian: on the hydraulic side, because they will have a very low Delta P between supply and return.
32:38:410Jacopo Vivian: So the phone calls that are far away, for example, can receive a low flow.
32:46:00Jacopo Vivian: and they, instead of providing I don't know 2 kilowatts of cooling, they provide warm.
32:55:810Jacopo Vivian: If the cooling load is higher than one, the room temperature will increase as simple as that.
33:05:780Jacopo Vivian: Okay.
33:14:290Jacopo Vivian: so now we talk about the other option that we have that is called decouple distribution.
33:23:290Jacopo Vivian: the decoupled in the decoupled distribution. We also have some different options here.
33:30:980Jacopo Vivian: Here. You can see that the the 1st option I'm showing still has only the pumps on the primary with
33:41:630Jacopo Vivian: I'm not the. There is no pamper on the secondary circuit here.
33:47:380Jacopo Vivian: See?
33:49:680Jacopo Vivian: So what are the changes that you can see in this there are 2 main changes.
33:58:610Jacopo Vivian: Oh, yes, 1st of all, we have different valves. We don't have the bypass on the terminal units
34:08:760Jacopo Vivian: right here we have that
34:12:110Jacopo Vivian: this rectangle above the valve means that this is the one we use for regulation.
34:18:920Jacopo Vivian: Okay, and we only have the 2 way valve in series with the terminal unit
34:25:679Jacopo Vivian: that will reduce the flow coming to this specific terminal unit at partial load, and the second difference.
34:41:230Jacopo Vivian: the bypass piper, that the will decouple this flow from the flow circulating on the secondary side.
34:53:870Jacopo Vivian: All the couple. How does this decoupling work?
34:59:460Jacopo Vivian: Simply by adjusting the Delta P between the supply and the return.
35:08:20Jacopo Vivian: So that if we need this.
35:13:320Jacopo Vivian: 70% of the flow at partial load, 30% will go through the bypass.
35:22:710Jacopo Vivian: Okay, instead of going to the terminal units.
35:28:330Jacopo Vivian: So the opening of the two-way valve on the bypass
35:34:900Jacopo Vivian: is responsible for the regulation of the flow to the terminal units.
35:50:90Jacopo Vivian: In this case, we still have individual pumps, one for each chiller okay and
36:06:820Jacopo Vivian: the consequence of the decoupled distribution is that now, we can size
36:14:90Jacopo Vivian: the pipes according to the maximum flow that they they will receive simultaneously.
36:24:130Jacopo Vivian: Good.
36:26:190Jacopo Vivian: So
36:44:580Jacopo Vivian: it's not exactly like this, because in this case
36:49:440Jacopo Vivian: we can size according to the maximum simultaneous flow.
36:55:890Jacopo Vivian: And not today, maximum flow of which part of the building.
37:02:730Jacopo Vivian: Okay.
37:08:710Jacopo Vivian: of course, this has an impact on big buildings with bigger schooling systems
37:16:660Jacopo Vivian: where you have a pulling loads that are not necessarily simultaneous.
37:22:190Jacopo Vivian: and in the couple distribution you need to size of
37:28:910Jacopo Vivian: the unit, the the piping for the flow rate that is equal to the one of the chiller.
37:37:320Jacopo Vivian: And in the second case it can be lower, depending on the load. Okay?
37:44:350Jacopo Vivian: Oh, okay.
37:52:210Jacopo Vivian: Don't want depending on start.
37:56:750Jacopo Vivian: Yes.
38:10:980Jacopo Vivian: so
38:20:50Jacopo Vivian: So in this situation, at the full load all the chillers receive.
38:28:950Jacopo Vivian: If they are on 33% of the design flow rate.
38:35:580Jacopo Vivian: Okay?
38:37:840Jacopo Vivian: So each chiller has its nominal flow rated, and the, if they are
38:46:790Jacopo Vivian: of the same size, occur, imagine we have 3 chillers of 200 kilowatt each.
38:54:730Jacopo Vivian: Okay, they will have.
38:57:410Jacopo Vivian: When they are all of them. When they are on they will receive 33%.
39:04:490Jacopo Vivian: So 1 3rd of the overall flow rate.
39:08:50Jacopo Vivian: Now we look well.
39:12:450Jacopo Vivian: Yes.
39:13:980Jacopo Vivian: Now we look what happens at the and partial load.
39:20:480Jacopo Vivian: Imagine the the part. The heat load is 66%
39:28:380Jacopo Vivian: of the nominal of the design loader.
39:31:810Jacopo Vivian: In this case the 3 chillers still receive 33% of the flow.
39:41:470Jacopo Vivian: So we, we fulfill the requirement of constant flow to the evaporator of the chillers. Also in the decoupled distribution.
39:52:440Jacopo Vivian: But not all of this flow will circulate through the system.
39:57:550Jacopo Vivian: because 1 3rd of the flow will circulate through the bypass.
40:03:580Jacopo Vivian: and the rest will go to the risers or to the terminal units.
40:19:760Jacopo Vivian: If you switch off one chiller.
40:25:820Jacopo Vivian: You need to regulate the bypass
40:29:240Jacopo Vivian: so that 0% of the flow will go through the bypass
40:36:170Jacopo Vivian: because you always want the 33% of the design flu rate. They're going in the chillers.
40:43:920Jacopo Vivian: So there is always a need for synchronization.
40:47:690Jacopo Vivian: When we have a different dealers in parallel.
40:51:770Jacopo Vivian: which is a common situation in pulling systems. Okay, you bet.
41:05:360Jacopo Vivian: Alright.
41:09:400Jacopo Vivian: In general it would be better to switch off the chiller so that we
41:18:250Jacopo Vivian: make the chillers work at full load.
41:22:530Jacopo Vivian: and we avoid having excessive flow rate circulating.
41:29:50Jacopo Vivian: because in this case, you see, 66% of the flow rate in the primary circuit is also the same
41:38:470Jacopo Vivian: going to the, to, the, to the risers, even the increase.
41:50:320Jacopo Vivian: Yes, but but on the primary side you have much more flow rate.
41:59:460Jacopo Vivian: Okay.
42:01:10Jacopo Vivian: it depends on the chiller. So if the chiller manufacturer tells you, it is better to work always at full load.
42:12:00Jacopo Vivian: because you are not able to gain
42:15:320Jacopo Vivian: efficiency and pack her load. It would be better to switch off as much as possible as long as the
42:23:450Jacopo Vivian: pulling load the decreases.
42:27:270Jacopo Vivian: If the chillers are able to work at partial load, then you can also.
42:36:110Jacopo Vivian: So of course it.
42:39:50Jacopo Vivian: It is strictly dependent on what the manufacturer of the chiller.
42:44:700Jacopo Vivian: Oh, is is telling you shoot it the full load of the
42:58:770Jacopo Vivian: okay. Here I'm talking about the the load. So in in both cases we are working at partial load.
43:08:340Jacopo Vivian: Okay, in one case 66. So imagine the load is 100 kilowatt.
43:18:100Jacopo Vivian: Now, I'm talking about the situation where you have 66 kilowatt of load.
43:24:550Jacopo Vivian: You can provide the 66 kilowatt, either with 2 are with 3 chillers.
43:30:910Jacopo Vivian: If you provide them with 3 children, you need to have a
43:36:550Jacopo Vivian: the some flow some some of the flow rate going through the bypasser.
43:43:520Jacopo Vivian: If you switch 1 1 of the 3 chillers off, you provide the same load with 2 chillers, and you have no flow in the bikes. My guess is that this option is more efficient than the other.
44:00:20Jacopo Vivian: because the 2 chillers provide exactly
44:04:20Jacopo Vivian: the load that they were meant to provide a design load.
44:08:700Jacopo Vivian: So you can never happen. And then, in this case
44:17:207Jacopo Vivian: no, because it could not. If the design load is
44:21:960Jacopo Vivian: If if the design no, if the design load is 100 overall, it means that the design load for each chiller is 33,
44:34:550Jacopo Vivian: just need to scale the numbers accordingly.
44:39:820Jacopo Vivian: Oh.
44:46:800Jacopo Vivian: because each user, each user is one.
44:56:163Jacopo Vivian: Yes, in this case. Okay, in this case.
45:07:550Jacopo Vivian: you have 3 children working at their okay.
45:22:560Jacopo Vivian: Yes, yes.
45:26:650Jacopo Vivian: And in this case you have only 2 of them
45:36:650Jacopo Vivian: by in this case, instead of the device controller.
45:44:905Jacopo Vivian: Yes, exactly. This is the point.
45:59:380Jacopo Vivian: Oh, okay. So here we have a 6, 12.
46:06:570Jacopo Vivian: Okay?
46:08:370Jacopo Vivian: And in this situation
46:16:250Jacopo Vivian: we have 6, 10.
46:21:160Jacopo Vivian: So we are in this situation.
46:26:990Jacopo Vivian: We are not very close to the design point of the 3 chillers, as we are in this situation, because you don't only need to look at the flow. You also need to look at temperatures.
46:40:880Jacopo Vivian: Okay, that's why I'm telling that this situation here.
46:46:40Jacopo Vivian: in my opinion, is more efficient than the other one.
46:55:300Jacopo Vivian: You can think about it.
46:59:330Jacopo Vivian: Okay? So the other option is to have pumps also for the secondary side.
47:09:940Jacopo Vivian: In the 1st case we decouple the 2 splits by using a bypass.
47:16:610Jacopo Vivian: In this case we have a bypass.
47:21:90Jacopo Vivian: But there is a difference here, because
47:27:780Jacopo Vivian: what is the difference in your opinion between this bypass
47:33:880Jacopo Vivian: with the pumps on the secondary side, and the bypass that we have before.
47:47:280Jacopo Vivian: So in the bypass that we had before, I told you
47:53:780Jacopo Vivian: that, depending on the number of chillers that are in operation. You need to regulate these 2 wave valves so that the correct flow will go through the bikes
48:06:10Jacopo Vivian: right?
48:07:590Jacopo Vivian: And here.
48:10:740Jacopo Vivian: We don't need the regulation. Why.
48:13:900Jacopo Vivian: we saw it in the 1st lecture about hydraulics.
48:17:860Jacopo Vivian: because when you have 2 secrets that are connected.
48:23:530Jacopo Vivian: they are decapitated. If they bypass.
48:27:680Jacopo Vivian: If there is, that's the P
48:31:560Jacopo Vivian: almost 0 in the bypass pipe in the common 5.
48:36:950Jacopo Vivian: Okay, so here we can change it.
48:41:730Jacopo Vivian: The flow to the load by acting on the
48:48:180Jacopo Vivian: variable speed bumps that you see here right?
48:54:500Jacopo Vivian: And we don't care about the chillers.
48:58:660Jacopo Vivian: If we have a bypass that ensures that we have
49:03:160Jacopo Vivian: delta P equal to almost 0 between supply and return.
49:09:70Jacopo Vivian: Okay, because this flow is independent from this flow.
49:19:660Jacopo Vivian: and the same can be done
49:22:130Jacopo Vivian: which is more typical on with the bouncer on each
49:28:230Jacopo Vivian: section, or of the cooling system. For example, if they are risers, or if they are
49:35:780Jacopo Vivian: lines going to different parts of the building or to different buildings, even keeps the line of work
49:47:320Jacopo Vivian: is between this point and this point in bold.
49:53:50Jacopo Vivian: Yes, yes.
49:55:819Jacopo Vivian: they are the same. But here they are. Yes, it's not very clear, because they didn't separate this
50:05:840Jacopo Vivian: trade deck, anyway.
50:08:750Jacopo Vivian: Yes, these are the common pumps of the secondary circuit.
50:13:310Jacopo Vivian: And this is the more typical situation
50:16:220Jacopo Vivian: with the a pump for each line.
50:22:20Jacopo Vivian: So this is the another option where we have.
50:33:150Jacopo Vivian: Okay, we have. Okay, here, we want to make things complicated.
50:38:170Jacopo Vivian: We have a secondary pumps.
50:41:640Jacopo Vivian: Okay, where the flow goes to, this line comes back.
50:48:70Jacopo Vivian: But it also goes to the second line.
50:51:660Jacopo Vivian: The difference is that here we have also a pumper
50:57:430Jacopo Vivian: on the closer to the terminal units, and the flow between which makes this
51:06:740Jacopo Vivian: path here that I'm showing with the mouse.
51:10:890Jacopo Vivian: Okay, you can see the path.
51:13:630Jacopo Vivian: this path, this flow, is independent from the flow that will circulate here.
51:19:810Jacopo Vivian: Why? Because we have another bypass here.
51:25:950Jacopo Vivian: So the bypass, or, as we said, the hydraulic separator, because you either use a
51:33:820Jacopo Vivian: a bypass, or you can use a component which is
51:39:720Jacopo Vivian: exactly placed for this purpose, which is the hydraulic separator, which is basically like a pipe.
51:50:10Jacopo Vivian: Is exactly used to decouple
51:54:140Jacopo Vivian: circuits that otherwise would interfere with each other, meaning that the flow here.
52:02:880Jacopo Vivian: without the without the bypass, would cause a certain delta P. And this Delta P would affect the flow of this pump, and so on. So with this bypass, we solve the problem.
52:24:400Jacopo Vivian: Okay?
52:29:130Jacopo Vivian: So
52:33:200Jacopo Vivian: now we have seen from the hydraulic point of view how we can manage the
52:43:30Jacopo Vivian: the cooling supply to different in different situations
52:48:200Jacopo Vivian: and on the chiller side. So on the generation side.
52:52:380Jacopo Vivian: we also need to do something. As we said before.
52:56:570Jacopo Vivian: So we need to regulate the chiller itself.
53:00:680Jacopo Vivian: Okay, so to do it, we need to consider that the cooling load is not constant.
53:10:380Jacopo Vivian: typically is very fluctuating because it depends on the
53:17:540Jacopo Vivian: solar heat gains and on the internal heat gains. Okay?
53:22:730Jacopo Vivian: And of course, also with the contribution of the
53:27:60Jacopo Vivian: ventilation and transmission losses when the
53:31:700Jacopo Vivian: outdoor air temperature is higher than the indoor air temperature.
53:36:690Jacopo Vivian: Otherwise this is reducing the pulling load. Okay, this is called freak ring.
53:44:490Jacopo Vivian: When we have in summer.
53:47:460Jacopo Vivian: typically at night or early in the morning, we can have the outdoor air temperature, which is colder, lower than the indoor air temperature.
53:58:390Jacopo Vivian: In this case a good strategy would be to increase the ventilation during these hours to provide free pooling.
54:07:630Jacopo Vivian: Why, why is it called like that because we don't need to switch on a chiller.
54:17:360Jacopo Vivian: This is why it is called free pooling.
54:20:160Jacopo Vivian: We just work on the
54:23:700Jacopo Vivian: outdoor air on the ventilation rate to cool down the building when it is more energetically convenient to do so.
54:36:110Jacopo Vivian: So
54:42:180Jacopo Vivian: we need to control the units. We have seen
54:45:980Jacopo Vivian: that we might have some synchronization issues when we have different chillers in parallel, and that we also have.
54:55:670Jacopo Vivian: we need to consider to have an almost constant flow rate to the evaporators of the chillers. So these are the boundary conditions.
55:07:690Jacopo Vivian: Typically we to regulate the cooling systems, we use a temperature tensor placed on the return side.
55:20:190Jacopo Vivian: Why?
55:21:550Jacopo Vivian: Because it is a somehow considering the inertia of the system.
55:31:350Jacopo Vivian: Okay, so with the of course we have. We still have indoor control
55:44:170Jacopo Vivian: air control. Okay, depends. So if the temperature in a room increases to 30 degrees, what happens
55:56:390Jacopo Vivian: in this case?
55:58:250Jacopo Vivian: Imagine in building one, this is cooling down a certain room.
56:06:160Jacopo Vivian: Imagine which is a big office space.
56:10:180Jacopo Vivian: Okay, if in this office we have 30 degrees
56:16:120Jacopo Vivian: this 3 way, valve will increase the opening so that more flow will circulate.
56:23:470Jacopo Vivian: And of course the same holds true in the couple, distribution
56:29:80Jacopo Vivian: with the the 2 way valve.
56:32:300Jacopo Vivian: Okay, and the variables be pumped the same.
56:36:120Jacopo Vivian: So we have a regulation on the room side.
56:42:600Jacopo Vivian: But this makes, as a consequence, in terms of her return temperature, because, depending on
56:51:760Jacopo Vivian: how much this valve is open. You have a mixer between 7 degrees coming from the supply.
57:01:160Jacopo Vivian: and 12 degrees, or 14, or what whatever it is in their return.
57:08:360Jacopo Vivian: So you have a return temperature that the
57:13:880Jacopo Vivian: depends on the regulation of all the terminal units in that specific moment.
57:20:680Jacopo Vivian: Okay, of course, if such, if we make such a controller, and there is no water.
57:33:620Jacopo Vivian: or there is a very small amount of water between this point and this pointer.
57:41:20Jacopo Vivian: we would have a continuous switching on and off offer that she learns.
57:48:520Jacopo Vivian: But we are talking about big systems with a lot of piping that
57:55:960Jacopo Vivian: accommodate a lot of a big amount of a bigger volume of water, which has a certain thermal inertia.
58:04:360Jacopo Vivian: and we leverage on this inertia to switch on and off, or to increase or decrease the cooling supply temperature of the chillers
58:15:680Jacopo Vivian: according to this return temperature that we measure just before coming back to the chillers.
58:24:170Jacopo Vivian: So what happens?
58:26:660Jacopo Vivian: It happens that if the return temperature
58:31:540Jacopo Vivian: increases, increases, increases at certain point, you switch on one more unit. Okay?
58:40:690Jacopo Vivian: And then this will make the return temperature reduce
58:46:830Jacopo Vivian: until you reach a point where the return temperature is much lower than 12.
58:56:210Jacopo Vivian: Imagine the you have 12 here
59:00:800Jacopo Vivian: and you have a plus minus one.
59:03:690Jacopo Vivian: Okay.
59:04:710Jacopo Vivian: In this situation it means that
59:07:480Jacopo Vivian: if we are providing a lot of cooling compared to the load, the temperature, the return temperature will decrease.
59:19:650Jacopo Vivian: And at a certain point, this situation, the the this controller, this thermostat, will say, Okay, we are providing too much cooling because the return temperature is too low.
59:33:300Jacopo Vivian: So we need to do something. In this case, we switch off, actually learn
59:42:210Jacopo Vivian: when we switch it off. It happens that after a certain amount of time, then
59:51:970Jacopo Vivian: you can either have the same situation.
59:54:990Jacopo Vivian: meaning that you have another step. Where the reduction temperature is.
00:01:390Jacopo Vivian: the the return temperature is decreasing, and then
00:05:70Jacopo Vivian: in that case you switch off another
00:08:150Jacopo Vivian: another one of the 3 chillers that we have here.
00:12:490Jacopo Vivian: Otherwise, if the return temperature is increasing again.
00:17:800Jacopo Vivian: it means that we are not providing enough pooling, and then we switch on again the chiller that we have switched off previously.
00:27:160Jacopo Vivian: If you look at this on the on a chart.
00:32:70Jacopo Vivian: You can. You can see here how the supply temperature
00:39:420Jacopo Vivian: changes according to the return temperature.
00:43:930Jacopo Vivian: Okay, so that the green difference is the the green delta t is there
00:54:490Jacopo Vivian: delta t of the hysteresis, the hysteresis differential.
00:59:340Jacopo Vivian: It is the delta t between on and off of a single modulation stack that means, for example.
01:11:100Jacopo Vivian: we, the hysterisk controller. The thermostat of this room is set to.
01:17:690Jacopo Vivian: I don't know 20 degrees, but we don't always have 20 degrees here 20 degrees plus minus one
01:27:850Jacopo Vivian: or plus minus something.
01:30:140Jacopo Vivian: This plus minus is the green delta. T, okay?
01:36:810Jacopo Vivian: And then we have different steps.
01:43:00Jacopo Vivian: which is the difference between on and off of consecutive modulation steps.
01:50:390Jacopo Vivian: So this is what the regulation can provide. If we have a chiller with different set points.
02:04:90Jacopo Vivian: or if we have different users.
02:07:570Jacopo Vivian: Okay in parallel, and each of them provide one cooling step.
02:14:490Jacopo Vivian: So basically, there is a chart. If the
02:20:220Jacopo Vivian: cooling load provided by the chiller is higher than the cooling load than the cooling load. We move to the left.
02:28:780Jacopo Vivian: so the return temperature is
02:32:143Jacopo Vivian: is reducing is decreasing. If the filler provides more lower cooling than the load. So we move to the right.
02:42:430Jacopo Vivian: So it's just that.
02:44:190Jacopo Vivian: Ha! Like having different thermostats connected to each other. But instead of having a
02:54:13Jacopo Vivian: to control all the terminal units in the room simultaneously.
02:59:270Jacopo Vivian: we switch on and off the terminal units
03:03:270Jacopo Vivian: with this order, okay, according to the return temperature.
03:08:510Jacopo Vivian: we do the same. But instead of having the terminal units, we have the chillers.
03:15:350Jacopo Vivian: and instead of having the indoor air temperature as a metric to decide that we have the return temperature
03:23:470Jacopo Vivian: of the cooling system.
03:26:730Jacopo Vivian: Okay, don't.
03:29:420Jacopo Vivian: It's it's not important to to remember this chart. Just remember the concept.
03:37:640Jacopo Vivian: And of course, the regulation changes depending on how we set these that these 2 Delta P, okay.
03:46:880Jacopo Vivian: so we have a faster or a slower reduction and increase in the
03:54:870Jacopo Vivian: supply temperature, and then, therefore, in the reduction in the return temperature.
04:01:630Jacopo Vivian: So you can think about our system in this way. There are the chidlers.
04:08:340Jacopo Vivian: Then there is the distribution system.
04:11:720Jacopo Vivian: Okay, remember, Delta T is between supplying a return is 5 degrees, and the cooling load. The cooling peak load of the buildings
04:21:10Jacopo Vivian: can reach values that are similar to the ones of the heating system.
04:25:940Jacopo Vivian: As in order open argently. Okay.
04:28:900Jacopo Vivian: therefore, we have a lot of
04:33:960Jacopo Vivian: water circulating in the system. We have a lot of volume of water in cooling systems.
04:41:410Jacopo Vivian: Good.
04:42:350Jacopo Vivian: So we can think about the cooling system in this distribution as a dynamic system that we can describe with this equation, which is simply
04:54:90Jacopo Vivian: the derivative of the internal energy of the system.
04:59:670Jacopo Vivian: Okay, this is density of the water in the cooling system
05:03:610Jacopo Vivian: times its volume. Times the specific heater
05:07:660Jacopo Vivian: multiplied by the derivative of the average temperature in the cooling system.
05:12:980Jacopo Vivian: Okay, cool.
05:14:70Jacopo Vivian: So this would be, the average temperature between supply and return is equal to then.
05:24:80Jacopo Vivian: Okay, here it is done like
05:27:110Jacopo Vivian: cooling, like heating. But it would be P in, okay. P. Chiller.
05:37:420Jacopo Vivian: it would. Okay. I just wrote it. For heating and not for cooling, but
05:46:220Jacopo Vivian: basically the thermal power of the loader minus the thermal power of the chiller minus the this is not a loss, but it is a gain.
05:58:200Jacopo Vivian: because in a pooling system the problem is when the heat comes into the system.
06:06:620Jacopo Vivian: we want to provide 7 Celsius degrees to the terminal unit. If we don't insulate the pipes
06:14:50Jacopo Vivian: between the cooling station and the terminal unit. You will have 8 instead of 7.
06:20:380Jacopo Vivian: So
06:21:410Jacopo Vivian: I just wrote it the other way around. But it is an energy balance. This is the this is the point.
06:28:110Jacopo Vivian: Yes.
06:36:730Jacopo Vivian: the
06:42:770Jacopo Vivian: please understand the password you mean. How do you decide between 2 and one here?
06:55:509Jacopo Vivian: Know that the are not so, said one of them.
07:10:880Jacopo Vivian: if possible.
07:13:840Jacopo Vivian: No, I think for you. The only thing that matters is that you know that there is this Delta T
07:21:80Jacopo Vivian: of Hysteres, and then, depending on how you set it, you have
07:28:996Jacopo Vivian: a faster or slower regulation. Mean that
07:33:580Jacopo Vivian: if you have a thermal discomfort problem, what would you do? Would you reduce or increase the Delta T of hysteresis?
07:44:590Jacopo Vivian: No, I would reduce it because you are closer to the set point.
07:49:440Jacopo Vivian: Imagine that. Imagine always imagine the the analogy with the room 20 plus minus 2,
07:59:330Jacopo Vivian: and you have someone calling you. It's cold.
08:03:430Jacopo Vivian: What would you do? Would you enlarge the Delta? P, so that 20 plus minus 5 or reduce it.
08:11:60Jacopo Vivian: I would reduce it.
08:13:820Jacopo Vivian: So this is basically what matters for you, because this is something that is typically set by the
08:22:401Jacopo Vivian: manufacturer that will commission the building. So
08:27:550Jacopo Vivian: you're not setting the controller for the chillers, because the manufacturers
08:32:870Jacopo Vivian: we'll do it. But you know that you have some space for improvement in case there are some thermal discomfort issues.
08:45:120Jacopo Vivian: Okay? So here we said that we are doing an energy balance by considering it day overall mass
08:57:120Jacopo Vivian: of water in the distribution circuit.
09:01:170Jacopo Vivian: Okay?
09:03:479Jacopo Vivian: And if we just consider a critical situation, what critical is
09:15:240Jacopo Vivian: when we imagine we have N regulation steps.
09:20:390Jacopo Vivian: Right?
09:22:640Jacopo Vivian: For example, 3 regulation steps.
09:27:439Jacopo Vivian: The most critical situation is when the load is half offer one regulation step.
09:37:270Jacopo Vivian: So if you have the design load is 300 kilowatt pulling.
09:48:330Jacopo Vivian: You have 3 regulation steps, meaning you can
09:52:310Jacopo Vivian: only modulate until 100 kilowatt.
09:57:750Jacopo Vivian: So the most critical situation is when you have 50 kilo.
10:03:670Jacopo Vivian: because in 50 kilowatt you can either switch off
10:07:660Jacopo Vivian: the last step, the last filler or switch it on.
10:13:120Jacopo Vivian: Okay.
10:14:600Jacopo Vivian: So I'm talking about this. So if you replace this
10:19:698Jacopo Vivian: into the into the equation of the energy balancer, you find something like this that the
10:29:40Jacopo Vivian: mass times the Cp times. The derivative of the temperature
10:35:450Jacopo Vivian: is equal to 0 point 5 times the design
10:41:320Jacopo Vivian: load of the chillers divided by the regulations tax.
10:47:310Jacopo Vivian: And you want this Delta T.
10:50:830Jacopo Vivian: Delta t not delta T temperature, but delta t time to be
10:58:850Jacopo Vivian: higher than a certain value that we set here to.
11:07:60Jacopo Vivian: I don't know why, it's 1. Okay, some meetings.
11:12:750Jacopo Vivian: Okay?
11:13:750Jacopo Vivian: Because this is a 0 point 8 h.
11:19:80Jacopo Vivian: So around 10 min,
11:25:360Jacopo Vivian: so we wanted this Delta P, this Delta T to be higher than the minimum value.
11:32:420Jacopo Vivian: In this case I don't know 10 min.
11:35:780Jacopo Vivian: So
11:37:280Jacopo Vivian: if you apply this rule, you obtain that the thermal, that the water volume mass in the cooling system
11:48:190Jacopo Vivian: must be higher than the ratio between the design cooling load.
11:57:960Jacopo Vivian: Okay, which is the same of the nominal
12:01:844Jacopo Vivian: power of the chillers. Of course, on the cooling side I'm always talking on the cooling side.
12:08:900Jacopo Vivian: divided by these, this product of 24
12:17:580Jacopo Vivian: cpn. Which is the number of regulation steps
12:22:110Jacopo Vivian: and the Delta T of hysteresis that you said.
12:27:170Jacopo Vivian: Okay, this is just a rule of thumb. Of course, if you change the minimum time between 2
12:37:30Jacopo Vivian: startups, you will have a different minimum mass.
12:43:590Jacopo Vivian: Okay, let let me just finish, and then I'll answer your question.
12:49:810Jacopo Vivian: So you need to check the water content
12:55:670Jacopo Vivian: especially for limited or circuits with limited hydraulic
13:03:110Jacopo Vivian: development, which means a low, low length of the pipes.
13:08:890Jacopo Vivian: Okay?
13:11:453Jacopo Vivian: When this minimum mass condition is not
13:18:120Jacopo Vivian: it is not fulfilled. You need to introduce an inertial storage tank where, on the return side.
13:29:470Jacopo Vivian: but before the temperature sensor of the thermostat.
13:37:310Jacopo Vivian: Okay?
13:39:980Jacopo Vivian: So this is where you would place the in actual storage tank to increase the volume. Number
13:54:980Jacopo Vivian: the water volume so that your hysteresis regulation, your thermostat regulation works properly.
14:04:930Jacopo Vivian: Otherwise, what is the problem that at the low
14:09:400Jacopo Vivian: load at partial loader you have frequent on offer of the last chiller unit.
14:17:650Jacopo Vivian: Okay, so you have a lot of cycling that is not good from a life cycle
14:28:790Jacopo Vivian: perspective. We don't want to have a lot of switch startups and switch off, first, st because
14:37:750Jacopo Vivian: the duration of the cheater itself will decrease. Second, because you have lower efficiency.
14:44:980Jacopo Vivian: Okay, because normally transient operation of the system always is always causing inefficiency.
14:56:230Jacopo Vivian: Okay, having the if you can choose between
15:03:433Jacopo Vivian: constant operation and intermittent operation.
15:08:270Jacopo Vivian: giving the same amount of energy. It would be better to have constant operation. Okay?
15:15:360Jacopo Vivian: Otherwise, it's like oversizing the chiller basically in instead of if
15:24:600Jacopo Vivian: if the cooling load is fixed
15:27:380Jacopo Vivian: and you have selected the chiller properly.
15:30:880Jacopo Vivian: But you had a lot of cycling.
15:33:690Jacopo Vivian: You can introduce an inertial storage tank to provide a better regulation.
15:39:870Jacopo Vivian: Okay, this this is typically done already in the design phase of the cooling system.
15:49:350Jacopo Vivian: Your question we do. Listening again.
15:57:910Jacopo Vivian: Oh, yes, because imagine that you have
16:05:20Jacopo Vivian: 4 units. Okay? And the load.
16:09:700Jacopo Vivian: Each of them gives you 25 kilowatts.
16:15:530Jacopo Vivian: and the nominal the design cooling load is
16:20:160Jacopo Vivian: 100 kilowatt. So at 100 kilowatt.
16:24:780Jacopo Vivian: the 4 of them are on
16:27:550Jacopo Vivian: at the 80 kilowatts. You probably switch off one of them right, and depending on
16:37:450Jacopo Vivian: the return temperature, you will keep it off, or you will switch it on again
16:43:60Jacopo Vivian: at 50 kilowatt. You switch off 2 of them.
16:48:250Jacopo Vivian: etc, etc. But what do you do at the 12 kilowatts
16:55:890Jacopo Vivian: you switch off. Either you switch off all of them.
17:00:60Jacopo Vivian: or you keep the last one in operation, and since you have
17:07:190Jacopo Vivian: many times the low load condition.
17:11:770Jacopo Vivian: This means switching on and off the last refrigeration unit many times.
17:18:970Jacopo Vivian: So you want to avoid this situation.
17:22:530Jacopo Vivian: Oh, I want to go
17:36:20Jacopo Vivian: in this case.
17:38:380Jacopo Vivian: If you have 4 units giving you 25.
17:42:490Jacopo Vivian: The the problem is when you have a 12.5 in this example, because you have
17:52:920Jacopo Vivian: 2 options, or you keep one on.
17:56:350Jacopo Vivian: or you just they are all off
18:00:700Jacopo Vivian: in case they are all offer. If you have 12.5 kilowatt of pulling load for sure. After a while you will need to switch it on again.
18:11:330Jacopo Vivian: because otherwise people feel discomfort.
18:16:70Jacopo Vivian: Okay.
18:20:380Jacopo Vivian: okay.
18:22:580Jacopo Vivian: Still,
18:26:560Jacopo Vivian: Another thing is that when the cooling capacity. When the cooling load exceeds
18:35:20Jacopo Vivian: 200 or 300 kilowatt, it is recommended to install more than one unit in parallel.
18:44:130Jacopo Vivian: Why? Because we want to guarantee a minimum reserve capacity right
18:49:780Jacopo Vivian: even in the event of failure.
18:54:940Jacopo Vivian: and also because if there is a domestic hot water demand.
19:00:400Jacopo Vivian: you might have one of the unit, also providing hot water by using
19:07:890Jacopo Vivian: the heat rejected, rejected on the condenser side.
19:13:570Jacopo Vivian: Okay, this case, you use the the the chiller.
19:23:20Jacopo Vivian: I don't know if if you saw this in yesterday's lecture, but chiller can also provide that hot water
19:32:890Jacopo Vivian: by using the deck super heater of the condenser.
19:40:990Jacopo Vivian: Okay, instead of rejecting all the heater to the outdoor air.
19:47:50Jacopo Vivian: You reject the heat to some technical water
19:51:190Jacopo Vivian: that will warm up the domestic hot water tank.
19:56:870Jacopo Vivian: Okay, so, having more than one unit is is recommended in general, and especially it's higher and loads.
20:16:960Jacopo Vivian: We still have 7 min. So I just want to
20:22:170Jacopo Vivian: remember to you that we don't only have thermostats to control Hvac sequence
20:30:330Jacopo Vivian: we can. We can have a different controllers.
20:36:890Jacopo Vivian: And in all cases we want to maintain a chosen set value.
20:44:830Jacopo Vivian: Okay, in this case, we want to maintain that the return temperature between
20:54:950Jacopo Vivian: we said 1212 plus minus something.
20:59:160Jacopo Vivian: Okay.
21:00:750Jacopo Vivian: So we either use it. If we use a hysterisk controller like a thermostat.
21:07:190Jacopo Vivian: we do it in a discrete way.
21:09:910Jacopo Vivian: meaning, when we reach the lower threshold, we switch on. We start up a new
21:16:720Jacopo Vivian: refrigeration unit in a chiller.
21:20:120Jacopo Vivian: When the return temperature increases too much, we switch it off and so on.
21:28:510Jacopo Vivian: But we have other ways to keep the temperature constant.
21:34:720Jacopo Vivian: And this is what I'm talking about. You just
21:38:610Jacopo Vivian: define the error between the measured variable, for example, the return temperature
21:45:960Jacopo Vivian: and the reference temperature. You want to reach this side. Pointer.
21:51:640Jacopo Vivian: Okay, this will give you an error.
21:56:90Jacopo Vivian: Okay, so, for example, if the cooling system is not providing enough cooling to the building you will have.
22:08:50Jacopo Vivian: hey? You will have a return temperature of 13 degrees.
22:13:60Jacopo Vivian: but your set point is 12,
22:15:380Jacopo Vivian: so you will have an error of one.
22:18:20Jacopo Vivian: 13 minus 12 is equal to one.
22:22:190Jacopo Vivian: So what can this controller do?
22:25:750Jacopo Vivian: It can control that by by an actuator it can control
22:38:220Jacopo Vivian: the chillers with either a proportional action.
22:43:260Jacopo Vivian: In this case the proportional action is
22:48:680Jacopo Vivian: proportional to the error. As you can see, we have a constant here. Kp, that multiplies the error.
22:57:650Jacopo Vivian: Where can we use, for example, this
23:08:460Jacopo Vivian: here you see here in this 2 way valve on the bypass of the couple, distribution we can
23:21:980Jacopo Vivian: reduce or increase the opening so that we maintain 12 degrees to the return side.
23:32:140Jacopo Vivian: So in this case the proportional controller will increase, will open the 2 way valve.
23:40:730Jacopo Vivian: so that we have more mixture between supply and return temperature.
23:47:190Jacopo Vivian: So that is at the next time step. We don't have a 13 degrees anymore. But we have a 12.5,
23:56:350Jacopo Vivian: and the next would be 12, so that we close
24:02:80Jacopo Vivian: the gap until we reach an error equal to 0. But then the load will change, and we will have another error.
24:12:220Jacopo Vivian: Okay? Because it's a dynamic system.
24:17:300Jacopo Vivian: Then we can have an integral collection
24:22:80Jacopo Vivian: where you have a Ki which is equal to one and divided by a time constant, which introduces
24:32:298Jacopo Vivian: a correction. For example, on the 2 way valve of the bypass that I'm I will, showing that depends on the integral of the error.
24:42:770Jacopo Vivian: Okay, so you have another constant, another parameter.
24:52:160Jacopo Vivian: and then you have a possible derivative action.
24:57:390Jacopo Vivian: What what does the integral action represent? It depends the past values of the error.
25:05:760Jacopo Vivian: Okay.
25:10:20Jacopo Vivian: the derivative action represents the future values of the action, meaning that if the error is decreasing.
25:18:390Jacopo Vivian: you have a negative derivative.
25:21:870Jacopo Vivian: and therefore you forecasted that in the next time step you will not need to close the valve anymore to open the valve anymore. Okay.
25:32:80Jacopo Vivian: so this is why you have
25:34:890Jacopo Vivian: these 3 components of controllers that are called.
25:40:180Jacopo Vivian: either you have 2, 3 types of controllers.
25:44:160Jacopo Vivian: If you don't have thermostats, you have the proportional controllers only using the proportional collection.
25:53:570Jacopo Vivian: You have pi controllers using the proportional and the integral correction.
26:02:760Jacopo Vivian: And you have a Pid controllers
26:05:530Jacopo Vivian: that use the 3 type of corrections to open and close the cloud.
26:14:840Jacopo Vivian: This is, I think, this is useful to remember, because in Hvac systems you don't always have statuses, controllers
26:23:910Jacopo Vivian: where you just to set the delta T of the hysteresis like in a thermostat, you might have this
26:31:280Jacopo Vivian: more advanced controllers, especially when you want to make a continuous regulation.
26:39:420Jacopo Vivian: If you have a 2 way valve like the one of the bypass, you don't want to hide there, and it's fully closed or fully open.
26:48:430Jacopo Vivian: In that case a control like this can be useful
26:53:270Jacopo Vivian: in another case where that we will see maybe tomorrow. Yes, sir.
26:59:90Jacopo Vivian: that where we talk about heat pumps
27:03:160Jacopo Vivian: we will have a situation where you want to have a fully open or fully closed condition.
27:09:970Jacopo Vivian: In that case it does not make sense to have a pad controller.
27:14:330Jacopo Vivian: because you not just need to have that
27:17:630Jacopo Vivian: a thermostat telling you if this valve should be fully open or fully closed.
27:24:620Jacopo Vivian: But we will see the example tomorrow.
27:27:990Jacopo Vivian: Okay, I think that's it for today.