I am receiving a stream of time-series data/events from a manufacturing process machine. I need to be able to calculate the start of a cycle and pinpoint this event based on the trough of the data. What would be a recommended way to do this?
Thanks
Edit: I've attached an image of the data graph. Basically I have an ongoing stream of data that I need to be able to effectively identify the troughs in order to establish the start of the next cycle/end of the previous cycle. The only thing I'm guaranteed not to know is the period of the cycle - it will vary.
The only thing I'm guaranteed to know is the initial datapoint value - which could be anywhere in the cycle.
At the moment - I'm kind of thinking I have to do something with a sliding span of time, keeping tracking of the minimum temp seen so far.