Sea water underground pipe leak detection

[npower0073]

On our power plant we take sea water through a 6 km underground line. This line has faced a lot of leaks from vents and flanges. We are thinking of a way to detect leaks using some intrumentation. One way could be to use flow sensors on the intkae line, one on the pwoer plant and one near the sea. The differential flow could give us a signal if there is a leak somewhere.
Can you suggest on this issue? The pipe is GRP DN 800.
What type of flow sensor could be used in order to avoid cutting? E.g. ultrasonic ..

 

[rconner]

This is probably a stupid question, but if you suspect the leaks are at the locations you identify, are the "vents" etc. not in some kind of structure like a vault that might be accessed to either look to see if they are leaking or put some kind of sensors at those accessible locations?

 

[gerhardl]

What amount of leakage do you want to detect? When is the leakage becoming critical?

A flow measurement at regular or at least, at one midway station, could help detect a larger leakage within the accuracy parameters of the measuring equipment and help you to locate the stretch with leakage.

With water running you could perhaps, if visual is impossible, try to detect different levels of moisture in the ground, or run a low degree radioctive isotop through the line and detect eccessive levels ov reminding radioactivity.

But then again: a quality pipeline in GRP with properly selected vents and flanges, properly laid and constructed and with not to high pressure peaks caused by pumps or valves will not give any leaks.

Perhaps you should lok into the actual cause of the leakage and think a major overhaul or a complete new pipeline?

What is the cost/risk analysis over the next ten years? A power plant with a 800 cooling line will be of some size, and it will cost money if you repeatedly and unexpected have to shut down, dig and repair.

 

[BigInch]

if you have inlet flow and pressure readings at inlet and outlet, you an deduce the location of a leak via hydraulic modeling. If you have more than one leak, you'll need more measured data points to isolate affected segments.

Otherwise, you might resort to tracers introduced near the regions of suspected leaks, then mesuring the concentrations of the tracer at the plant to determine where the active intake points are located.

http://virtualpipeline.spaces.msn.com

"What gets us into trouble is not what we don't know, its what we know for sure" - Mark Twain

 

[MortenA]

Question: Does it matter? I mean - u take water from 6 km and closer - and so what? Doesit mean you get more debris, bio fouling etc?

Best regrds

Morten