Weizhi Yao:

Lynton Mogridge:

It sounds like you are running a sub-critical two phase refrigeration cycle. This is similar to your domestic refrigerator. If you look up refrigeration cycle on Wikepedia it will give you the details and the normal performance measure COP (coefficient of performance), more detail in Cengel and Boles Chapter 10.

Lynton.

Thanks, I found that the COP in air conditioner is about 3.x,


If we use 1km copper line, radius is 900 mm2,

At 20 deg c

R = 0.0172*10^-6 * 1000 / (900*10^-6) = 0.0191(ohm)

At 90 deg c, temperature coefficient of resistivity is 0.00393 at 20 deg c

R = 0.0191*1.00393^70 = 0.02513(ohm)


The current is 1000A

P=1000*1000*0.0191=19100(W)

P90 = 1000*1000*0.02513 = 25130(W)

Psav = 25130-19100 = 6030(W)


Per year, the energy saving is

W = Psav* 24*365 = 52822.8(kw*h)   -- per kilometer


To keep the line 20 deg c, the input power of compressor should be

19100/3.5 = 5457.14(W)


It seems it is worth implementing the solution.

Am I right? Could anybody working in electricity domain point out the rough difference between the real value and the ideal value?

 

Ian Evans:

Firstly, your temperature adjustment for the resistance is incorrect: R(90°C) = R(20°C) × [1 + coeff × (90 - 20)]

Secondly, are you considering an a.c. or d.c. system; if a.c. then you need to take account of the skin and proximity effects which will further increase the conductor resistance.

Once you have the correct resistance you can calculate the losses from the conductor (joule losses).  All this energy needs to be removed from the conductor to maintain the temperature and you need to take account of all the heat transfer.  Your system would need to monitor the temperature because as soon as it gets above the required temperature the losses will increase and you will get thermal runaway.  As your system is 1 km long it is likely that the cooling gas will not be a constant temperature over the full length, it will be flowing around a system where the input will be colder than the output.  When considering cooling of a long conductor it is generally easier to think in terms of W/m (You are basically looking at a fluid dynamics problem to remove the W/m).

Other things to consider are costs resulting from probable reduced reliability and additional maintenance - it isn't just about the savings in terms of energy losses.

 

Ian Evans:

Firstly, your temperature adjustment for the resistance is incorrect: R(90°C) = R(20°C) × [1 + coeff × (90 - 20)]

Secondly, are you considering an a.c. or d.c. system; if a.c. then you need to take account of the skin and proximity effects which will further increase the conductor resistance.

Once you have the correct resistance you can calculate the losses from the conductor (joule losses).  All this energy needs to be removed from the conductor to maintain the temperature and you need to take account of all the heat transfer.  Your system would need to monitor the temperature because as soon as it gets above the required temperature the losses will increase and you will get thermal runaway.  As your system is 1 km long it is likely that the cooling gas will not be a constant temperature over the full length, it will be flowing around a system where the input will be colder than the output.  When considering cooling of a long conductor it is generally easier to think in terms of W/m (You are basically looking at a fluid dynamics problem to remove the W/m).

Other things to consider are costs resulting from probable reduced reliability and additional maintenance - it isn't just about the savings in terms of energy losses.

 

Ian Evans:

Firstly, your temperature adjustment for the resistance is incorrect: R(90°C) = R(20°C) × [1 + coeff × (90 - 20)]

Secondly, are you considering an a.c. or d.c. system; if a.c. then you need to take account of the skin and proximity effects which will further increase the conductor resistance.

Once you have the correct resistance you can calculate the losses from the conductor (joule losses).  All this energy needs to be removed from the conductor to maintain the temperature and you need to take account of all the heat transfer.  Your system would need to monitor the temperature because as soon as it gets above the required temperature the losses will increase and you will get thermal runaway.  As your system is 1 km long it is likely that the cooling gas will not be a constant temperature over the full length, it will be flowing around a system where the input will be colder than the output.  When considering cooling of a long conductor it is generally easier to think in terms of W/m (You are basically looking at a fluid dynamics problem to remove the W/m).

Other things to consider are costs resulting from probable reduced reliability and additional maintenance - it isn't just about the savings in terms of energy losses.