I want to design a neural network capable of detecting the degradation of the capacity of a battery. When the capacity is ok, the battery performs like this (x axis is the time in seconds and y axis is the voltage in volts):
However, if there is a problem in the battery, the capacity starts to decrease, which means the following for the voltage vs time:
So deterioration happens when the period of the signal decreases. I would like to know when this period has decreased a certain percentage so I can do some counter-actions.
I understand that I need an LSTM neural network to predict this situation, however I don't know where to start, and all examples I have read are not for this kind of problem. Any thoughts? Also It would be great If you can provide any example/tutorial of a similar problem.
