Optimizing Energy Harvesting from Power Lines for Autonomous UAVs

[donstenx]

Hey, I have little knowledge in electrical so i thought it is the best to ask the experts. I am planning to design a mini uav for powerline surveillance. Since it should be autonomous, I thought it would be good if I can harvest some energy from the power line through electromagnetic induction. The uav should have some mechanism to perch on the lines & recharge the power supply.

The question:
1) I believe it is possible, but how the mechanism works?
2) How much energy can be harvested?
3) If I use (for example: 20V of batteries), how long it takes to charge up?

If possible, can you provide some references/diagram/calculations.

Thanks! (Sorry if my english is bad)

 

[Carl Pugh]

If there is current through the power line this should be possible.
Place a current transformer with appropiate turns ratio on the power line (This will probably have to be a custom designed transformer). Rectify the output from the current transformer. Use this rectified voltage to charge your battery.
With power line operating at maximum power, can probably harvest 1% of the power with no problem. With normal power lines this will be MUCH more power than required.
When the battery is charged, short the output from the current transformer. Do not open the current transformer output or remove the load.

 

[sophiecentaur]

I would say this is definitely not a DIY design job. There would be serious insurance issues, to say the least.
Rather than a magnetic coupling, with all the problems of open circuited current transformers, I could suggest a capacitative take-off. The advantage would be that you would have your supply at all times, compared with the current transformer, which would only work whilst power is being consumed. Even so, you would need some pretty highly specified capacitors. Electrical power is big business so you should expect an appropriate level of expense.

Are you proposing that your gear would be hung on the lines and would use a radio link for the data (the current transformer approach)? If that's the case, then could you guarantee having access for installation and maintenance - which would involve interrupting the power? Or would the equipment be mounted on a pole, 'near' the lines?
The solution would depend upon a lot of important details.
It may be better to use solar technology for your power - it's pretty universal for remote equipment, these days.

 

[skeptic2]

Can you explain what "powerline surveillance" is? Have you contacted the power company to ask permission use their power. This being in the near field, the power you recover is not otherwise wasted. Even the minute amount of energy you are extracting would otherwise go to the customer.

Why not consider the use of solar cells instead?

 

[Borek]

Thread temporarily closed for moderation.

 

[berkeman]

Hey, I have little knowledge in electrical so i thought it is the best to ask the experts. I am planning to design a mini uav for powerline surveillance. Since it should be autonomous, I thought it would be good if I can harvest some energy from the power line through electromagnetic induction. The uav should have some mechanism to perch on the lines & recharge the power supply.

The question:
1) I believe it is possible, but how the mechanism works?
2) How much energy can be harvested?
3) If I use (for example: 20V of batteries), how long it takes to charge up?

If possible, can you provide some references/diagram/calculations.

Thanks! (Sorry if my english is bad)

Thread temporarily closed for moderation.

Thanks Borek!

@donstenx -- Please send me a PM describing the scope of your project, including your sponsors and your intentions (concept development only, prototype development, or product deveopment with the intention of commercial production).

 

[berkeman]

After exchanging PMs with the OP, this appears to be a valid project, done for the benefit of the power companies. I'll go ahead and re-open the thread.

From my perspective, BTW, using a clamp-on current transformer (CT) would seem to be the best way to tap recharging power out of one wire of the powerline in this application.

 

[dlgoff]

I am planning to design a mini uav for powerline surveillance. Since it should be autonomous, I thought it would be good if I can harvest some energy from the power line through electromagnetic induction. The uav should have some mechanism to perch on the lines & recharge the power supply.

From my perspective, BTW, using a clamp-on current transformer (CT) would seem to be the best way to tap recharging power out of one wire of the powerline in this application.

berkemans suggestion of using a clamp-on Current Transformer (CT) sounds like the way to go if you are talking about relative low current power lines. There are dangers when disconnecting a CT while there's high current in the primary/power line. In the power industry, CTs are use to measure the amperage in the line and shorting blocks are used to insure you don't have an open secondary. Here are some considerations for this approach from Current Transformers:
A Tester Survival Guide.

https://docs.google.com/viewer?pid=bl&srcid=ADGEEShaMtbskWL0OEC5wMdchODmrLy3-p5qeh3hyZapSb_BBogJc39Y9hiBJ_QGNdNF0VaqV8pal9sKnErNGA9xZUMtdV0kXfWQ0-1tOAIedKfwhe4OVgkJOyenZnj1XlR1gtobeP1e&q=cache%3A85VtEWY4xMsJ%3Apscal.ece.gatech.edu%2FFDAPRC%2Ffiles%2FPRC%2520Website%2520material%2FCurrent%2520Transformers%2520-%2520A%2520Tester%2520Survival%2520Guide.ppt%20&docid=dd181f2567023f8fa782f013123c91ec&a=bi&pagenumber=12&w=888

 

[sophiecentaur]

It could be worth asking what voltage these power lines are operating on. That could have an enormous bearing on what's the best solution. I seem to be the only one to have introduced the fact that the lines could be at hundreds of kV.
Btw, is there any reason why a Current Transformer couldn't be protected against unintentional loss of load by built-in hefty crossed zener diodes?

 

[dlgoff]

It could be worth asking what voltage these power lines are operating on. That could have an enormous bearing on what's the best solution. I seem to be the only one to have introduced the fact that the lines could be at hundreds of kV.
Btw, is there any reason why a Current Transformer couldn't be protected against unintentional loss of load by built-in hefty crossed zener diodes?

Bold by me.

Well, by "high current" in my post, I was indicating that transmission line (kVs) would be a problem using CTs. Also, I've never seen a diode for those kinds of potentials. If the OP uses a CT, they will have to connect and disconnect the thing, so some sort of shorting needs to be done.

 

[berkeman]

Bold by me.

Well, by "high current" in my post, I was indicating that transmission line (kVs) would be a problem using CTs. Also, I've never seen a diode for those kinds of potentials. If the OP uses a CT, they will have to connect and disconnect the thing, so some sort of shorting needs to be done.

Good point about the disconnection transient... Perhaps they should use solid state devices to make and break the contact. That way the contact could be broken at the zero-crossing of the current in the line...

 

[dlgoff]

Good point about the disconnection transient... Perhaps they should use solid state devices to make and break the contact. That way the contact could be broken at the zero-crossing of the current in the line...

I'm sure that could be done but consider this. See @ 1 minute. These are the transmission line voltages I think would be a problem for the electronics.https://www.youtube.com/watch?v=9tzga6qAaBA

 

[donstenx]

It could be worth asking what voltage these power lines are operating on. That could have an enormous bearing on what's the best solution. I seem to be the only one to have introduced the fact that the lines could be at hundreds of kV.
Btw, is there any reason why a Current Transformer couldn't be protected against unintentional loss of load by built-in hefty crossed zener diodes?

Well, its about 110kV power lines if I'm not mistaken. The project aims for bushfires prevention due to hotspots.
Thank you for all your ideas & kind reply. I'll have to look at them carefully. :)

 

[donstenx]

If possible, can i ask about ir sensor under this topic? (if not, i'll delete this post).
The question is, I've look into Melexis IR thermopile, which has 90deg field of view(fov). But after some reading, some says that high fov doesn't always good. Is it a good idea to use this thermopile for hotspot detection? My current idea is to use one(or multiple) thermopile, plus some fresnel lens+amplifier+filter (maybe) for detection.

Or should i go with thermal imaging camera? But i think that camera was very expensive. I would like to have an affortable sensors ($300-$500 maximum). Any good sensors out there that i can use?

 

[jim hardy]

I think your clamp-around CT will work okay. Presumably your secondary will go into a bridge rectifier with large filter capacitor... Just don't open circuit the secondary ...

i'm more worried about the machine's wingspan being a substantial fraction of the distance between lines.
There's minimum approach distances for power lines.
Basically be sure the distance between wires after subtracting your wingspan is not less than about two of the line's insulator lengths.