If you are looking to commission a survey of solar PV, then a question you may ask contractors is ‘what height do you fly at’ to try and get a bearing on what data may be delivered, but to be able make an informed judgement you will need to know more.
What is the best height from which to survey solar PV?
The short answer to this would be ‘as close as practicable’. The aim should be to capture the best quality data but within the boundaries that make sense economically.
What we really want to know is what is the level of detail in our data. To estimate this, we not only need to know the height, but the size of the sensor as well as the focal length of the lens. With these 3 items of information we can then calculate the ground sampling distance (GSD). GSD is the distance between two consecutive pixel centers measured on the ground and the lower the GSD the higher the resolution of the data.
So the question to ask should be: What GSD will your data be collected at?
GSD = (sensor width mm x flight height m x 100) / (focal length mm x image width pixels)
For thermal surveys the IEC state a minimum of 5 pixels per cell. As a cell is usually 150mm x 150mm we therefore know that a GSD of 3cm or lower would be good enough to achieve this.
High resolution thermal imagers by Flir are very popular with aerial surveyors. An imager fitted with a 13mm lens need to be flown at a height lower than 24m to achieve 3cm GSD. A camera of different specification or different lens choice will be required to be flown at a different height to achieve the same GSD.
Here is a thermal image taken at around 6cm GSD. There should be 5 anomalies visible within the marked area but it is very difficult to see them all, they could easily be missed and some of those that can be seen are very difficult to analyse. It should be said that the most obvious fault that can be seen here is the one that is losing the most money!
Here is the same area captured at around 2cm GSD. you can quite clearly see all 5 anomalies and already a clearer picture can be seen with regards to the type of potential faults. For example, when looking at the 6cm GSD image, No.2 looks like a potential diode fault (as per No.1). Now at 2 cm GSD it looks more like a problem on the busbar connecting the cells, and heat is spreading along it. This could well be a warranty claim as opposed to a general maintenance concern. You can also clearly see the Junction boxes on the right hand side of each module which are not visible in the higher level image. Junction boxes are a prime candidate for potential fires.
RGB images (digital jpegs) need to be collected at the same time as the thermal to be able to classify faults accurately. Digital cameras are usually far higher spec when compared to IR, however standards should still be high so the data contains the fine details that is needed.
Here is an RGB image taken at 1.2 cm GSD. With the target module enlarged the picture is very fuzzy. You cannot see any module clips for example. You can of course see a rather excessive ‘bird deposit’ which explains anomaly No.5 but the area in which No.4 is looks discoloured and inconclusive. If the soiling was less prominent or front glass was shattered it would be very hard to tell.
Here is the same area at 0.65cm GSD. Now we can clearly see the module clips, we could tell if the glass was shattered, we can see what is causing the hotspot of No.4 and we can even tell that that is coconut flavoured yoghurt and not a bird deposit – amazing!
To find out more, and have your asset inspected to the correct standard please get in touch.
