Five frequently asked questions power utilities ask about LiDAR data
To round out the month here are some more commonly asked questions regarding LiDAR and the use of this technology within electricity utilities. Not sure what Aerial LiDAR is – read our blog post.
1) How often should you complete a LiDAR survey?
The answer to this question depends on the intention of the dataset. If you want to monitor change then frequent surveys are one method of doing this. The exact frequency will depend on the regulatory environment and internal policies. However as an example, a 4- year rolling thermal ratings program (25% a year) is common and seems to work, an annual survey using existing models in high-risk fire areas is used in places, a 2-3 year distribution LiDAR survey makes sense in lower-risk locations but allows a rolling program of using LiDAR to drive vegetation management OPEX savings. The key is working closely with a supplier to build an engineering survey schedule that best meets your networks unique needs. There is never a one-size-fits-all approach.
2) What point density would be ideal for detecting conductors?
This is a great question and depends on the kind of survey being conducted, the terrain, any other requirements and actually the surface of the conductors themselves (blackened copper conductor is always more challenging than a shiny new ACCC).
That being said, as a rule of thumb. If you are looking for powerlines for a vegetation management application, and you need to know where the conductors are relative to the trees we would recommend 10ppm2 or more. If you are doing a PLS-CADD or engineering led survey then something in the region of 15ppm2 to 25ppm2 would be ideal. You can go higher and in some cases, this may be required, but generally going higher is just going to give you bigger data overheads without any real value. The main advantage of going bigger would be for modelling structures in more detail. Usually though, the key consideration for point density is the ability to measure low reflectivity wires such as crossings, underbuild or low voltage distribution wires.
3) What do you see as the advantages and disadvantages of using Unmanned Aerial Vehicles (UAVs)?
We have worked with UAV systems and find they work well in the right circumstances. This is also true of helicopters and fixed-wing platforms. Essentially picking the right platform for the application is the key and that is something we spend a lot of time considering each project.
Regarding UAVs they are relatively inexpensive, can safely traverse hazardous areas and carry useful equipment (such as survey technology). Right now, the line of sight restrictions makes them most economical on smaller projects but less suitable for working at scale. Also, the bigger UAVs needed to complete larger projects are more expensive and often don’t work out cheaper than manned aircraft. That being said as legislation relaxes and sensor technology reduces in size we expect the use case to widen significantly.
4) Can data analyses be improved using multiple sensors?
The short answer is yes! There are lots of sensor options available that provide potential value, LiDAR, camera, video, UV, Infrared, NIR etc. Using the right remote sensing technology is a trade-off between what you want to capture today (and maybe for the future) versus the cost and carrying limits of the platform. You can only carry so many sensors so picking the right mix is essential. It isn’t simply a case of packing in more sensors as the operating conditions (height, speed, position next to OHL) will be different and aren't necessarily compatible. Working with a qualified supplier can help you meet this goal and ensure that you get a solution that provides a real return on investment.
5) Can you automatically extract features from LiDAR or images?
This is something that is talked a lot about and an area we are investing considerable R&D in as well. There are now semi-automated options which are fairly robust. However, given the importance of accuracy in engineering models and the lack of a complete market solution we still believe that using a highly trained technician or engineer is needed.
We believe that an iterative approach is needed. One that uses AI to remove the more manual, time-consuming tasks to minimise costs and turnaround time. For fine features in busy environments at the moment we need skilled people.