Time-Series intelligence?

Hey there,

I was just asking myself if there are any approaches for intelligent analysis of device data?
Such as energy consumption: I would like to detect “sessions” (or anomalies) in the data of eg my dishwasher: How much does he consume per session?

Also, it would be neat to have some sort of aggregation functions for the stored events (eg: only last month, - I know that this would be possible through an hour threshold) and display for example the top 10 consuming devices descending by their consumption.

I think the last topic is easier than the first. For the first - is there any approach? Or would one do this via influx?