Taxiing with the same configuration in the same airport produce a very different behavior of computer fans speed and noise. The Longitude taxi very quietly, but the Pilatus Pc-24 or the Citation X make the fans spin at top speed and produce great noise. This difference is also present in mid flight.
I cannot understand this huge difference in my fans speed. Someone can explain it? And most important, can it be solved in any way?
Which fans? GPU fans? Case fans? CPU AIO radiator fans?
Do you have fan curves set up in BIOS?
I’ve got 4 fans, three in my RTX 4070 Ti and one that I think is linked to the CPU. I think is this last one that goes crazy. I have no curves in BIOS, but I use Omen Game hub and I have not set curves there.
You don’t have any intake fans or separate exhaust fans?
I’d suggest installing HWiNFO64 and monitoring temps and fan speeds. You can even do some very granular graphing, so you can correlate those metrics.
As @BegottenPoet228 suggests, get some monitoring software and check on what your hardware is doing.
Typically, PC fans - whether installed in your case, on your CPU cooler/radiator, or your GPU - are designed to speed up as temperatures increase. The faster they spin, the more air they’re trying to move, and the more cooling they can provide. So if your fans are spinning up at any time, it is usually a sign your system is under some increased load, causing it to work harder and thus produce more heat, which those faster-spinning fans are trying to remove.
Remaining generally aware of how hot your system is running (CPU package temperature; GPU temperature; chipset, memory and storage temperatures; etc.) is always a good idea for any kind of modern PC stuff these days. If things get too hot, your system will begin to throttle or slow itself down to reduce heat and may even shut down entirely. Regularly overheating a system is also a good way to damage something expensive like your CPU or GPU.
There are plenty of good options for monitoring things though, ranging from simple overviews to extremely granular, very detailed. The simple options will give you something like CPU temp, GPU temp, and fan speeds. Those are fine for most people unless they are either concerned with something specific or simply data nerds (yes, those exist!). The more complex ones are good if you’ve just built a new system and want to be sure everything is working as expected, or if you make major changes to your case, cooler or fans are configured and arranged, or if you change our a major piece of hardware like your CPU, GPU or RAM.
Anyway, the tl;dr here is download a monitoring program and see why the fans are spinning up. It’s probably normal but at least you’ll know what’s going on.
I normally use OpenHardwareMonitor. It has a neat overlay gadget that can display chosen metrics. I monitored CPU temp, GPU temp, and GPU power.
Unfortunately, it doesn’t have the granular fan monitoring I want.
For everyday use after setting up my fan and pump curves, it’s great.
But like you said, for a deeper dive, HWiNFO64 is the gold standard.
Personally I have used MSI Afterburner for many years. I generally monitor a lot of stuff but only overlay the basics: overall CPU load expressed as a single percentage; CPU package temperature; total system memory use; GPU load percentage; GPU temperature and FPS. If I’m really trying to characterize something weird in practice, I’ll add items like per-core load and clock speeds, GPU and case fan speeds, and maybe 1% low measurements for FPS. But in something like flight sim, I find that too much info is never a road to happiness or enjoyment of the sim. Obsessing over minutia is fine for product development but not fine for just having fun. 
Afterburner is great. Is took me a while figure out how to get the overlay working properly with RTSS - but that’s a ‘Me’ thing…
One issue I’m having with it now is keeping the overlay confined to one monitor when I use POPM to push Garmin screens to my touchscreen monitors. The Afterburner overlay is forced onto all three monitors, and flickers wildly. But I’m getting a bit off topic.
Your point about using Afterburner to monitor different things is a good one.