Dear Buddhi Wimarshana ,

 Look the links, may be useful.

http://ieeexplore.ieee.org/document/1465484/?reload=true

https://www.uni-obuda.hu/journal/Grmanova_Laurinec_Rozinajova_Ezzeddine_Lucka_Lacko_Vrablecova_Navrat_66.pdf

http://cbl.eng.cam.ac.uk/pub/Public/Wolpert/Publications/AhmWol09.pdf

Regards, Shafagat

Thank you very much for your kind reply Shafagat Mahmudova. I will have a look at them.

Hi Buddhi,

You problem sounds very similar to my research problem. I suggest you to take a look at Linear Dynamical Systems (if continuous data) or Hidden Markov Models (if discrete or categorical data), or Bayesian models that allow you to perform prior/posterior update. Hybrid models (Bayesian Networks scenarios) can be very useful to you too (combining all above approaches).

I hope to be helpful,

 Hi Jhonathan,

Thank you very much for the informative reply. I am new to these so will go through your suggestions to see whether they can be any help. My concern is the predictions will be quite stochastic depending on the factors which lead to the load variations. Therefore, the system should be able to tackle them.

Thanks a lot.