Brown out peak prediction and control system for predicting a time when energy consumption on a grid will exceed normal production capacity

Abstract

A method for predicting when a grid will exceed production capacity including generating time shifted versions of consumption data for each of the buildings, each version comprising consumption values along with time and temperature values, where the consumption values are shifted by lag values relative to the time and temperature values, and where each lag value is different; performing a regression analysis on each version to yield corresponding regression model parameters and a corresponding residual; determining a least valued residual indicating a corresponding energy lag for each of the buildings; using outside temperatures, regression model parameters, and lags for all of the buildings to estimate a cumulative consumption for the buildings, and to predict the time when the grid will exceed production capacity; and receiving the time when the grid will exceed production capacity, and preparing and commencing exceptional measures required to manage the consumption.