my use of the word 'incident' to mean an individual occurrence or event out side of the normal operating parameters
I'm not even sure they will be looking for that.
That would be an open-ended project that could take months of 'incident-free' flying before seeing something out of the ordinary. As Hazel says, this may be no more than an exercise in real-time data collection. Remember Boeing should have 100-150 batteries worth of data from the 'swap-outs' to go by, so this operation must be looking for something over and above what they already have.
I'm not sure what extra gear would need to be hooked up to the system over and above what is already there to 'feed' the algorithm. And remember, any further monitoring of things like sub-cell temperatures, voltages and currents would be fairly invasive of the actual battery itself to the point where you would be effectively looking at a different system.
