Moja snimka 14

0:00

S… Speaker 1 (Moja snimka 14)

in production.

0:01

S… Speaker 1 (Moja snimka 14)

So,

0:06

S… Speaker 1 (Moja snimka 14)

we have been making a study

0:11

S… Speaker 1 (Moja snimka 14)

on the topic of Potential Strip for Energy

0:15

S… Speaker 1 (Moja snimka 14)

Efficiency Report.

0:16

S… Speaker 1 (Moja snimka 14)

We could have regulations where it drives and

0:21

S… Speaker 1 (Moja snimka 14)

this study is made in

0:25

S… Speaker 1 (Moja snimka 14)

collaboration with SEMER for their request

0:29

S… Speaker 1 (Moja snimka 14)

and it's divided

0:34

S… Speaker 1 (Moja snimka 14)

into two parts.

0:38

S… Speaker 1 (Moja snimka 14)

analysis of the energy savings potential,

0:41

S… Speaker 1 (Moja snimka 14)

setting more stringent requirements

0:45

S… Speaker 1 (Moja snimka 14)

for the BST losses.

0:48

S… Speaker 1 (Moja snimka 14)

Second is investigating the

0:52

S… Speaker 1 (Moja snimka 14)

measurement uncertainty challenges related to the overall loss requirements.

0:55

S… Speaker 1 (Moja snimka 14)

And third is estimation of expected

1:00

S… Speaker 1 (Moja snimka 14)

increase in product cost as the function

1:04

S… Speaker 1 (Moja snimka 14)

of a lot of losses.

1:06

S… Speaker 1 (Moja snimka 14)

So about the motivation,

1:10

S… Speaker 1 (Moja snimka 14)

the current

1:12

S… Speaker 1 (Moja snimka 14)

high classifications for BSDs are quite loose

1:16

S… Speaker 1 (Moja snimka 14)

and therefore there might be decided to introduce

1:20

S… Speaker 1 (Moja snimka 14)

more new,

1:21

S… Speaker 1 (Moja snimka 14)

more strict FTC classes and regulations.

1:27

S… Speaker 1 (Moja snimka 14)

And the concern

1:31

S… Speaker 1 (Moja snimka 14)

among the BSD manufacturers has been that new regulations

1:35

S… Speaker 1 (Moja snimka 14)

could have adverse effects on the market,

1:38

S… Speaker 1 (Moja snimka 14)

so without the adoption of BSDs in the

1:42

S… Speaker 1 (Moja snimka 14)

worst case.

1:42

S… Speaker 1 (Moja snimka 14)

And in this study,

1:44

S… Speaker 1 (Moja snimka 14)

the potential

1:47

S… Speaker 1 (Moja snimka 14)

effects,

1:48

S… Speaker 1 (Moja snimka 14)

contact, potential,

1:49

S… Speaker 1 (Moja snimka 14)

which are repulations are analyzed.

1:52

S… Speaker 1 (Moja snimka 14)

So in the first part it

1:57

S… Speaker 1 (Moja snimka 14)

is

2:03

S… Speaker 1 (Moja snimka 14)

divided into two parts.

2:04

S… Speaker 1 (Moja snimka 14)

In the first part we look at how it would be seen

2:08

S… Speaker 1 (Moja snimka 14)

if only looking at the IE class limits can be current.

2:13

S… Speaker 1 (Moja snimka 14)

IE2

2:20

S… Speaker 1 (Moja snimka 14)

is the requirement currently.

2:22

S… Speaker 1 (Moja snimka 14)

This

2:28

S… Speaker 1 (Moja snimka 14)

is the introduce

2:32

S… Speaker 1 (Moja snimka 14)

or predict the next

2:36

S… Speaker 1 (Moja snimka 14)

IE2 class would probably reduce

2:40

S… Speaker 1 (Moja snimka 14)

the BST losses by another 25 % so the

2:45

S… Speaker 1 (Moja snimka 14)

class limits are based on the reference losses

2:49

S… Speaker 1 (Moja snimka 14)

of 100 % and IE2 is defined as 25

2:54

S… Speaker 1 (Moja snimka 14)

% less losses than the reference losses Our

2:58

S… Speaker 1 (Moja snimka 14)

guess is

3:02

S… Speaker 1 (Moja snimka 14)

that the next one would be IE3 that has 25

3:07

S… Speaker 1 (Moja snimka 14)

% less losses

3:09

S… Speaker 1 (Moja snimka 14)

for the sake of like not

3:13

S… Speaker 1 (Moja snimka 14)

setting any tolls we used to consider IH2 -25.

3:17

S… Speaker 1 (Moja snimka 14)

And we make analysis

3:21

S… Speaker 1 (Moja snimka 14)

in real world conditions on the level of DSP

3:26

S… Speaker 1 (Moja snimka 14)

and the rule drive system.

3:28

S… Speaker 1 (Moja snimka 14)

And the similar analysis is done from

3:32

S… Speaker 1 (Moja snimka 14)

the actual measured

3:36

S… Speaker 1 (Moja snimka 14)

DSP was this and what it

3:41

S… Speaker 1 (Moja snimka 14)

would like if you would like it.

3:50

S… Speaker 1 (Moja snimka 14)

analysis is based on three

3:54

S… Speaker 1 (Moja snimka 14)

actual lab systems,

3:58

S… Speaker 1 (Moja snimka 14)

and here using this visual

4:02

S… Speaker 1 (Moja snimka 14)

one example.

4:04

S… Speaker 1 (Moja snimka 14)

How

4:13

S… Speaker 1 (Moja snimka 14)

we make the comparison data,

4:15

S… Speaker 1 (Moja snimka 14)

we use the actual measured cross

4:19

S… Speaker 1 (Moja snimka 14)

-map. First we convert

4:24

S… Speaker 1 (Moja snimka 14)

it into a starting map by dividing

4:29

S… Speaker 1 (Moja snimka 14)

the whole map,

4:31

S… Speaker 1 (Moja snimka 14)

creating the data point losses,

4:34

S… Speaker 1 (Moja snimka 14)

so 100%.

4:35

S… Speaker 1 (Moja snimka 14)

Next,

4:37

S… Speaker 1 (Moja snimka 14)

you can then

4:41

S… Speaker 1 (Moja snimka 14)

achieve the loss map that would represent

4:47

S… Speaker 1 (Moja snimka 14)

the losses in the full operating range

4:51

S… Speaker 1 (Moja snimka 14)

for IE2 limit ESP and

4:55

S… Speaker 1 (Moja snimka 14)

similarly to IE2 minus 25%.

5:00

S… Speaker 1 (Moja snimka 14)

we need VST.

5:01

S… Speaker 2 (Moja snimka 14)

And using

5:05

S… Speaker 1 (Moja snimka 14)

the loss match,

5:06

S… Speaker 2 (Moja snimka 14)

we can then analyze

5:11

S… Speaker 1 (Moja snimka 14)

losses in real load scenarios

5:15

S… Speaker 1 (Moja snimka 14)

by applying the known

5:19

S… Speaker 2 (Moja snimka 14)

load periods on the

5:24

S… Speaker 1 (Moja snimka 14)

data.

5:24

S… Speaker 1 (Moja snimka 14)

And of course this would have been done in more accurate

5:29

S… Speaker 1 (Moja snimka 14)

terms for all these four cases.

5:32

S… Speaker 1 (Moja snimka 14)

There was very clear

5:37

S… Speaker 1 (Moja snimka 14)

Like,

5:37

S… Speaker 1 (Moja snimka 14)

so is both direct operating points.

5:41

S… Speaker 1 (Moja snimka 14)

Average

5:45

S… Speaker 1 (Moja snimka 14)

relative losses were 48%,

5:49

S… Speaker 1 (Moja snimka 14)

creating point losses

5:53

S… Speaker 1 (Moja snimka 14)

for whatever it was.

5:55

S… Speaker 1 (Moja snimka 14)

55 % hit 1 % for linear loss,

5:59

S… Speaker 1 (Moja snimka 14)

39 % for constant power,

6:01

S… Speaker 1 (Moja snimka 14)

84 % for constant torque growth.

6:04

S… Speaker 2 (Moja snimka 14)

And from

6:09

S… Speaker 2 (Moja snimka 14)

that data.

6:15

S… Speaker 1 (Moja snimka 14)

Of course,

6:16

S… Speaker 1 (Moja snimka 14)

in this video we have determined also the

6:20

S… Speaker 1 (Moja snimka 14)

losses for IE3 motor

6:25

S… Speaker 1 (Moja snimka 14)

in a similar manner and we are using the same

6:29

S… Speaker 1 (Moja snimka 14)

losses for both IE2 and IE2

6:34

S… Speaker 1 (Moja snimka 14)

-25 % cases so this includes the assumption that both

6:39

S… Speaker 1 (Moja snimka 14)

these VSTs will produce exactly the same type

6:44

S… Speaker 1 (Moja snimka 14)

of voltage for the motor

6:48

S… Speaker 1 (Moja snimka 14)

And here we can see that the reduction of 25 % in

6:53

S… Speaker 1 (Moja snimka 14)

the BSD losses translates to around

6:58

S… Speaker 1 (Moja snimka 14)

12 % of reduction in the EDS

7:03

S… Speaker 1 (Moja snimka 14)

losses,

7:04

S… Speaker 1 (Moja snimka 14)

almost across the whole.

7:07

S… Speaker 2 (Moja snimka 14)

So there is a

7:11

S… Speaker 1 (Moja snimka 14)

whole actually surprisingly difficult

7:15

S… Speaker 1 (Moja snimka 14)

effect on the load scenario in these terms.

7:27

S… Speaker 2 (Moja snimka 14)

Here are data from both assignments

7:32

S… Speaker 1 (Moja snimka 14)

from the beginning.

7:33

S… Speaker 1 (Moja snimka 14)

This was the IH2 -25 % case and

7:37

S… Speaker 1 (Moja snimka 14)

then here the actual DST

7:42

S… Speaker 1 (Moja snimka 14)

was system measured and the 25

7:46

S… Speaker 1 (Moja snimka 14)

% reduction from those is included.

7:50

S… Speaker 1 (Moja snimka 14)

So the real DST

7:55

S… Speaker 1 (Moja snimka 14)

already is allowed.

7:57

S… Speaker 1 (Moja snimka 14)

It's far below.

8:01

S… Speaker 1 (Moja snimka 14)

IE2 limit that is here,

8:02

S… Speaker 1 (Moja snimka 14)

then this would be 25 %

8:06

S… Speaker 1 (Moja snimka 14)

where we currently are.

8:11

S… Speaker 1 (Moja snimka 14)

And in the study,

8:12

S… Speaker 1 (Moja snimka 14)

we have small

8:17

S… Speaker 1 (Moja snimka 14)

market research using about values

8:22

S… Speaker 1 (Moja snimka 14)

for ESPs and actually many are even better than this.

8:26

S… Speaker 1 (Moja snimka 14)

So around here,

8:29

S… Speaker 1 (Moja snimka 14)

this is not an exception.

8:32

S… Speaker 2 (Moja snimka 14)

Here

8:37

S… Speaker 1 (Moja snimka 14)

is a table for all the three

8:41

S… Speaker 2 (Moja snimka 14)

cases.

8:49

S… Speaker 1 (Moja snimka 14)

illustrated,

8:50

S… Speaker 1 (Moja snimka 14)

the Bosch -Gleece is similar

8:54

S… Speaker 1 (Moja snimka 14)

in the old cases and this is for the actual minus 25 % case.

8:58

S… Speaker 1 (Moja snimka 14)

So in energy savings this is calculated as

9:02

S… Speaker 1 (Moja snimka 14)

the reduction of the total power consumption.

9:07

S… Speaker 1 (Moja snimka 14)

So on the total power consumption 25

9:12

S… Speaker 1 (Moja snimka 14)

% reduction in the current.

9:17

S… Speaker 1 (Moja snimka 14)

ESP losses result in 1 % reduction in

9:22

S… Speaker 1 (Moja snimka 14)

the operating energy savings.

9:25

S… Speaker 2 (Moja snimka 14)

Here

9:34

S… Speaker 1 (Moja snimka 14)

is the same thing we illustrated as the quite

9:41

S… Speaker 1 (Moja snimka 14)

small share for the overall picture.

9:45

S… Speaker 2 (Moja snimka 14)

And as a conclusion of

9:49

S… Speaker 1 (Moja snimka 14)

part 1, introducing a new ESTI

9:53

S… Speaker 1 (Moja snimka 14)

class with 25 % lower limit.

10:01

S… Speaker 1 (Moja snimka 14)

From the current eye to no effect

10:05

S… Speaker 1 (Moja snimka 14)

on real products,

10:06

S… Speaker 1 (Moja snimka 14)

the share of the VST losses of the

10:10

S… Speaker 1 (Moja snimka 14)

total energy power consumption is small.

10:13

S… Speaker 1 (Moja snimka 14)

Reduction in VST losses have

10:18

S… Speaker 1 (Moja snimka 14)

a beautiful effect in the thick picture,

10:20

S… Speaker 2 (Moja snimka 14)

and the share of the VST losses of

10:24

S… Speaker 1 (Moja snimka 14)

the total losses of the motor drive system depends on the application

10:28

S… Speaker 1 (Moja snimka 14)

setup.

10:29

S… Speaker 3 (Moja snimka 14)

Some degree,

10:30

S… Speaker 2 (Moja snimka 14)

but the general picture is the same.

10:34

S… Speaker 2 (Moja snimka 14)

in all applications to be analyzed.

10:37

S… Speaker 3 (Moja snimka 14)

And the second part was

10:41

S… Speaker 2 (Moja snimka 14)

the seventh uncertainty issues or

10:46

S… Speaker 1 (Moja snimka 14)

challenges.

10:46

S… Speaker 3 (Moja snimka 14)

So actually related

10:51

S… Speaker 1 (Moja snimka 14)

to the previous presentation here

10:55

S… Speaker 1 (Moja snimka 14)

is actually included an indirect or basically

11:00

S… Speaker 2 (Moja snimka 14)

if you have one percent accuracy with the colorimeter.

11:05

S… Speaker 1 (Moja snimka 14)

you already at 95 % efficiency

11:09

S… Speaker 1 (Moja snimka 14)

are in the same accuracy as Yokokawa

11:14

S… Speaker 1 (Moja snimka 14)

BP5000 that is considered the

11:18

S… Speaker 1 (Moja snimka 14)

current state of PR typing so this is what

11:24

S… Speaker 1 (Moja snimka 14)

is achievable with the best power analyzers

11:28

S… Speaker 3 (Moja snimka 14)

correctly so when

11:35

S… Speaker 1 (Moja snimka 14)

we get higher and higher efficiency

11:38

S… Speaker 3 (Moja snimka 14)

get into trouble with the uncertainty.

11:41

S… Speaker 3 (Moja snimka 14)

This is then

11:46

S… Speaker 2 (Moja snimka 14)

the uncertainty across different power levels of

11:51

S… Speaker 1 (Moja snimka 14)

DSPs and the curves represent

11:55

S… Speaker 3 (Moja snimka 14)

reference CPM,

11:58

S… Speaker 1 (Moja snimka 14)

IE2,

11:59

S… Speaker 1 (Moja snimka 14)

so basically IE2 -25.

12:03

S… Speaker 3 (Moja snimka 14)

So these are basically the...

12:07

S… Speaker 3 (Moja snimka 14)

this is the like...

12:14

S… Speaker 1 (Moja snimka 14)

Yeah, this match is the real minus

12:21

S… Speaker 1 (Moja snimka 14)

25 % case of death.

12:23

S… Speaker 3 (Moja snimka 14)

So in

12:28

S… Speaker 3 (Moja snimka 14)

the IE2 class,

12:36

S… Speaker 1 (Moja snimka 14)

the uncertainty is near 10%,

12:40

S… Speaker 1 (Moja snimka 14)

which is still okay.

12:41

S… Speaker 1 (Moja snimka 14)

But the DST

12:46

S… Speaker 1 (Moja snimka 14)

is with half the losses of IE2.

12:49

S… Speaker 1 (Moja snimka 14)

But the uncertainty has to be even

12:54

S… Speaker 3 (Moja snimka 14)

high.

12:56

S… Speaker 3 (Moja snimka 14)

And then...

13:00

S… Speaker 1 (Moja snimka 14)

if we go outside the baby point.

13:03

S… Speaker 1 (Moja snimka 14)

These are the loss answers that is

13:07

S… Speaker 1 (Moja snimka 14)

calculated for those three

13:11

S… Speaker 1 (Moja snimka 14)

example VSTs that we have

13:15

S… Speaker 1 (Moja snimka 14)

measured for the wide operating rates

13:19

S… Speaker 1 (Moja snimka 14)

and basically the input output

13:24

S… Speaker 1 (Moja snimka 14)

loss answer that is 0 .2 % accuracy is

13:28

S… Speaker 1 (Moja snimka 14)

input level only in the

13:31

S… Speaker 1 (Moja snimka 14)

near the rapid operating point.

13:32

S… Speaker 3 (Moja snimka 14)

As

13:37

S… Speaker 2 (Moja snimka 14)

a conclusion of part two,

13:39

S… Speaker 2 (Moja snimka 14)

measuring the losses of high efficiency devices as the

13:43

S… Speaker 2 (Moja snimka 14)

difference of input output power is challenging from the uncertainty point of view.

13:47

S… Speaker 2 (Moja snimka 14)

In the previous rated point,

13:50

S… Speaker 2 (Moja snimka 14)

the loss alternative with typical testing

13:54

S… Speaker 2 (Moja snimka 14)

laboratory equipment is still at a decent level

13:58

S… Speaker 2 (Moja snimka 14)

and more accurate power analyzer exists.

14:03

S… Speaker 2 (Moja snimka 14)

Lower losses will increase the uncertainty and may cause some

14:07

S… Speaker 2 (Moja snimka 14)

challenges and require expensive testing equipment and

14:13

S… Speaker 2 (Moja snimka 14)

on a wider operating range,

14:15

S… Speaker 2 (Moja snimka 14)

especially in low load.

14:17

S… Speaker 2 (Moja snimka 14)

Low operating frequency region that measurement

14:21

S… Speaker 2 (Moja snimka 14)

accuracy can already be a challenge.

14:23

S… Speaker 3 (Moja snimka 14)

As

14:27

S… Speaker 2 (Moja snimka 14)

the last part of this study,

14:30

S… Speaker 3 (Moja snimka 14)

we made an analytical

14:36

S… Speaker 2 (Moja snimka 14)

model for frequency converter.

14:42

S… Speaker 2 (Moja snimka 14)

that was used to follow the losses and to

14:47

S… Speaker 2 (Moja snimka 14)

allow changing the IGBT module part for the device.

15:01

S… Speaker 1 (Moja snimka 14)

This is the original case of the integrator

15:05

S… Speaker 1 (Moja snimka 14)

loss distribution.

15:08

S… Speaker 1 (Moja snimka 14)

Then,

15:09

S… Speaker 1 (Moja snimka 14)

if we start to

15:14

S… Speaker 1 (Moja snimka 14)

see all and use deeper components,

15:18

S… Speaker 1 (Moja snimka 14)

we would have higher losses and oversize,

15:24

S… Speaker 1 (Moja snimka 14)

we would have lower losses,

15:26

S… Speaker 1 (Moja snimka 14)

and then we also included an estimation for

15:31

S… Speaker 1 (Moja snimka 14)

the SIG module and put it

15:35

S… Speaker 1 (Moja snimka 14)

dramatically better off course.

15:38

S… Speaker 1 (Moja snimka 14)

also dramatically more expensive.

15:41

S… Speaker 1 (Moja snimka 14)

Then,

15:45

S… Speaker 1 (Moja snimka 14)

in order to evaluate

15:50

S… Speaker 1 (Moja snimka 14)

the cost of the cooling,

15:51

S… Speaker 1 (Moja snimka 14)

we make the thermal analysis and thermal

15:57

S… Speaker 1 (Moja snimka 14)

resistance will determine

16:01

S… Speaker 1 (Moja snimka 14)

the cost of the cooling and here the

16:05

S… Speaker 1 (Moja snimka 14)

system cost is calculated

16:09

S… Speaker 1 (Moja snimka 14)

with the assumption of

16:12

S… Speaker 1 (Moja snimka 14)

cooling system was being inversely

16:17

S… Speaker 1 (Moja snimka 14)

proportional to the cooling system thermal resistance and

16:21

S… Speaker 1 (Moja snimka 14)

actually the cheapest case

16:26

S… Speaker 1 (Moja snimka 14)

even though the losses are not very

16:31

S… Speaker 1 (Moja snimka 14)

bad it has very quite bad thermal

16:35

S… Speaker 1 (Moja snimka 14)

resistance and this probably would not

16:39

S… Speaker 1 (Moja snimka 14)

be even possible in this case to be cooled

16:46

S… Speaker 1 (Moja snimka 14)

and also this would become the other cheaper

16:51

S… Speaker 1 (Moja snimka 14)

option, a lot more expensive

16:57

S… Speaker 1 (Moja snimka 14)

for the cooling part and of course vice versa in the higher

17:02

S… Speaker 1 (Moja snimka 14)

efficiency cases.

17:04

S… Speaker 1 (Moja snimka 14)

In

17:10

S… Speaker 1 (Moja snimka 14)

order to estimate the cost for the

17:14

S… Speaker 1 (Moja snimka 14)

full BSD price we got a

17:18

S… Speaker 1 (Moja snimka 14)

couple of responses from BSD

17:22

S… Speaker 1 (Moja snimka 14)

manufacturers of

17:24

S… Speaker 1 (Moja snimka 14)

what is the share of the cost for their products.

17:28

S… Speaker 1 (Moja snimka 14)

So the ITVT module for both is

17:32

S… Speaker 1 (Moja snimka 14)

10 % of the water VSD cost and the cooling system is

17:37

S… Speaker 1 (Moja snimka 14)

also about 10 % of the cost.

17:39

S… Speaker 1 (Moja snimka 14)

And here is then the cost

17:45

S… Speaker 1 (Moja snimka 14)

of the VSD as the function of the

17:49

S… Speaker 1 (Moja snimka 14)

module of choice and as you can see

17:53

S… Speaker 1 (Moja snimka 14)

actually the original module

17:58

S… Speaker 1 (Moja snimka 14)

is the most effective and the price

18:02

S… Speaker 1 (Moja snimka 14)

increases in both ways.

18:04

S… Speaker 1 (Moja snimka 14)

If we try to use cheaper

18:08

S… Speaker 1 (Moja snimka 14)

components, a lot more

18:15

S… Speaker 1 (Moja snimka 14)

cost in the cooling part and

18:19

S… Speaker 1 (Moja snimka 14)

then the model rises,

18:22

S… Speaker 1 (Moja snimka 14)

the determining factor here.

18:26

S… Speaker 1 (Moja snimka 14)

And here

18:32

S… Speaker 1 (Moja snimka 14)

is then one more picture

18:36

S… Speaker 1 (Moja snimka 14)

as the price system energy

18:40

S… Speaker 1 (Moja snimka 14)

consumption is in the

18:43

S… Speaker 1 (Moja snimka 14)

Here,

18:43

S… Speaker 1 (Moja snimka 14)

and the cost

18:48

S… Speaker 1 (Moja snimka 14)

of reducing

18:52

S… Speaker 1 (Moja snimka 14)

0 .8 % of energy consumption

18:58

S… Speaker 1 (Moja snimka 14)

would,

18:59

S… Speaker 1 (Moja snimka 14)

in

19:03

S… Speaker 1 (Moja snimka 14)

this analyzed case,

19:04

S… Speaker 1 (Moja snimka 14)

the BSD would be more than 20 % on

19:08

S… Speaker 1 (Moja snimka 14)

the more expensive end.

19:10

S… Speaker 1 (Moja snimka 14)

Here also,

19:11

S… Speaker 1 (Moja snimka 14)

and even these kinds of couple

19:16

S… Speaker 1 (Moja snimka 14)

of percent differences are hard

19:20

S… Speaker 1 (Moja snimka 14)

for manufacturers to understand.

19:24

S… Speaker 1 (Moja snimka 14)

This shows that

19:28

S… Speaker 1 (Moja snimka 14)

Or at least in this case there seems to be a clear,

19:31

S… Speaker 1 (Moja snimka 14)

clear,

19:33

S… Speaker 1 (Moja snimka 14)

was natural cost and was

19:38

S… Speaker 1 (Moja snimka 14)

minimal.

19:38

S… Speaker 1 (Moja snimka 14)

So as overall

19:42

S… Speaker 1 (Moja snimka 14)

conclusions,

19:43

S… Speaker 1 (Moja snimka 14)

introducing new future efficiencies required regulations

19:48

S… Speaker 1 (Moja snimka 14)

for VSDs would have only negligible benefits

19:52

S… Speaker 1 (Moja snimka 14)

and could even slow down adoption of VSDs due

19:56

S… Speaker 1 (Moja snimka 14)

to increased cost.

19:59

S… Speaker 1 (Moja snimka 14)

and had 25 % loss reduction from the current level.

20:03

S… Speaker 1 (Moja snimka 14)

Real BSDs could only be achieved

20:07

S… Speaker 1 (Moja snimka 14)

using silicon carbide technology and cause

20:12

S… Speaker 1 (Moja snimka 14)

significant loss increase.

20:13

S… Speaker 1 (Moja snimka 14)

Measurement uncertainty could become

20:18

S… Speaker 1 (Moja snimka 14)

a problem, making market surveillance very

20:22

S… Speaker 1 (Moja snimka 14)

difficult as an overall

20:26

S… Speaker 1 (Moja snimka 14)

comment.

20:28

S… Speaker 1 (Moja snimka 14)

there is no urgent need for tighter efficiency required

20:32

S… Speaker 1 (Moja snimka 14)

regulations for VSTs as there is

20:36

S… Speaker 1 (Moja snimka 14)

the, at least seems to be,

20:38

S… Speaker 1 (Moja snimka 14)

the natural cost minimum at a relatively high efficiency

20:43

S… Speaker 1 (Moja snimka 14)

as we already saw,

20:45

S… Speaker 1 (Moja snimka 14)

which leads to a sort of

20:49

S… Speaker 1 (Moja snimka 14)

self -regulation of the losses of the VSTs.

Moja snimka 14

خلاصه

از هوش مصنوعی در مورد این رونوشت بپرسید