I'm given a hourly heat consumption data for many apartments. I'm actually trying to figure out their peak consumption timings. I know its not that difficult but I'm furthermore interested in knowing if there is any specific pattern that exists in a group of apartments like for eg., may be there are certain apartments that have peak consumption during 05:00 - 08:00 in the morning & then their peak consumption is during the evening form 06:00 - 08:00, on the other hand there might be certain apartments which have moderate consumption throughout the day with no specific peaks.
Is there any specific statistic technique that I should apply on the data, It would be nice if someone could enlighten a bit on this.
**What I have tried so far: I made clusters of the apartments based on their peak consumption. I considered two peaks for each apartment & results are something that can be seen in the attachment. I'm not very much satisfied with the results therefore thought if someone here could give his/her word on this.apartment_clusters_first_peak.pdf (24.8 KB) apartment_clusters_second_peak.pdf (24.6 KB) **