Gaussian prior with electricity consumption time-series data

I am working with the change-point detection problem in time series (electricity consumption) data. I have detected the change in the consumption patterns by using DTW and k-NN. Now, I want to identify that the change is because of what reason, for example, due to the installation of PV, the introduction of EV, or something else.

Currently, I am focusing on the identification of change which is due to the installation of solar PV. In this regard, I have gone through the literature and come up with different techniques such as Cross-correlation, Gaussian prior, and ML methods. It seems that the Gaussian prior is well suited for my problem. I understood the concept that we have to use prior knowledge and evidence(data) to solve my problem. Unfortunately, I am unable to channel the concept to solve my problem at this point. In my case, the prior knowledge would be:

  1. There is a fall in electricity consumption during 7am-8am as generation from the PV starts.
  2. There is a rise in electricity consumption during 5pm-6pm as generation from the PV ends.
    The evidence would be the electricity consumption data.
    Summarily, I want to utilize the above two points as the prior knowledge and use the given data (electricity time-series) as evidence to prove whether the solar PV panel is present or not.

I need your kind help to incorporate the above two points and data to estimate the Posterior.
I fully understand if you are working on other projects at this time. I would highly appreciate your assistance. Please let me know if you have any further questions.

  • Thank you in advance