Dear Peter Mbamaluikem

Can you clarify your question, in general, this depends on several variables, for example: What is the maximum load and less load, can make a random program of higher and lower load capacity and the probability ratio, the probability of high load increases in peak hours and less in other hours can be trained neural network parameters on that.

Good luck

Dear Peter Mbamaluikem

In January 2015 Gerro J. Prinsloo answered a similar question, hope his answer will be useful for you:

Gerro J. Prinsloo answer in January 30, 2015

A number of mathworks load forecasting models are available, many of these with data in .xls, .csv formats. You can download the matlab models from this case study link:

http://www.mathworks.com/matlabcentral/fileexchange/28684-electricity-load-and-price-forecasting-webinar-case-study/content//Electricity%20Load%20&%20Price%20Forecasting/Load/html/LoadScriptNN.html

There are some webinars on this link also

http://www.mathworks.com/discovery/load-forecasting.html?refresh=true

Thank you Sameer Alguri

Dear Peter

Also, a Research Article: A Simple Method of Residential Electricity Load Forecasting by Improved Bayesian Neural Networks published in Mathematical Problems in Engineering Volume 2018, Article ID 4276176, 16 pages https://doi.org/10.1155/2018/4276176 with available data will be useful.

It was concluded that Traditional time-series BNN load forecast model has some problems when applied in the residential load forecast area, such as long-running time and relatively strong dependence on time and weather factors. To solve these problems, based on basic BNN training method and simple FFNN structure, an improved BNN forecast model is built by augmenting historical load data as inputs through correlation analysis of electricity consumption at different delayed time scales.

This table can be obtained from Taza control network .