Systems and methods for improving the accuracy of day-ahead load forecasts on an electric utility grid
First Claim
1. A method of adjusting forecast load predictions on an electric grid using one or more operatively connected computers, the method comprising:
- obtaining, at the one or more computers from at least one of an energy generation computer system or an energy retailer computer system, electrical grid information comprising;
a day-ahead profile of a forecasted load for the electric grid,a day-ahead sparks ratio, wherein the day-ahead sparks ratio is the actual day-ahead electricity prices relative to the prices of fuel used to generate electricity,a day-ahead price profile, andan hourly forecasted load relative to maximum hourly forecasted load for the day, minimum hourly forecasted load for the day, and average forecasted load for the day;
accessing, from one or more databases operatively connected to the one or computers, electrical grid historical data;
calculating, by the one or more computers, coefficients for an electricity consumption equation by performing a regression analysis using the historical data, the electricity consumption equation being a function of day-ahead sparks ratio, day-ahead price profile, day-ahead profile of forecasted load, hourly forecasted load relative to the forecasted profile, and coefficients of variation and skewness in day-ahead prices and forecasted load;
calculating, by the one or more computers, forecast prediction errors by applying data from the obtained electrical grid information to the electricity consumption equation with the calculated coefficients;
providing, by the one or more computers to one or more computer systems associated with an electric grid system operator, one or more electronic reports containing forecast errors for the electric grid;
calculating, by the one or more computers, one or more residual terms as the differences between i) predicted errors from the historical data using the electricity consumption equation, and ii) actual errors from the historical data;
applying, by the one or more computers, the residual terms to an auto-regressive moving average analysis, so as to determine one or more disturbances at one or more times;
calculating, by the one or more computers, a new set of coefficients for the electricity consumption equation based on the auto-regressive moving average analysis; and
controlling operation of a plurality of electric power generators of the electric grid based on the electricity consumption equation with the new set of coefficients.
1 Assignment
0 Petitions
Accused Products
Abstract
Systems and methods improve the forecast of electricity consumption, and/or refining such predictions. Predictions may be refined by accounting for factors such as preliminary predictions, pricing and cost information associated with future supply of energy, the extent of anticipated changes in the predictions, the time of day and/or anticipated daylight for the period of time. Coefficient values are calculated for a forecast error model that takes into account factors related to electricity consumption using existing historical electrical grid data. Using the calculated values, the forecast error model may be applied to current electricity demand forecasts.
23 Citations
23 Claims
-
1. A method of adjusting forecast load predictions on an electric grid using one or more operatively connected computers, the method comprising:
-
obtaining, at the one or more computers from at least one of an energy generation computer system or an energy retailer computer system, electrical grid information comprising; a day-ahead profile of a forecasted load for the electric grid, a day-ahead sparks ratio, wherein the day-ahead sparks ratio is the actual day-ahead electricity prices relative to the prices of fuel used to generate electricity, a day-ahead price profile, and an hourly forecasted load relative to maximum hourly forecasted load for the day, minimum hourly forecasted load for the day, and average forecasted load for the day; accessing, from one or more databases operatively connected to the one or computers, electrical grid historical data; calculating, by the one or more computers, coefficients for an electricity consumption equation by performing a regression analysis using the historical data, the electricity consumption equation being a function of day-ahead sparks ratio, day-ahead price profile, day-ahead profile of forecasted load, hourly forecasted load relative to the forecasted profile, and coefficients of variation and skewness in day-ahead prices and forecasted load; calculating, by the one or more computers, forecast prediction errors by applying data from the obtained electrical grid information to the electricity consumption equation with the calculated coefficients; providing, by the one or more computers to one or more computer systems associated with an electric grid system operator, one or more electronic reports containing forecast errors for the electric grid; calculating, by the one or more computers, one or more residual terms as the differences between i) predicted errors from the historical data using the electricity consumption equation, and ii) actual errors from the historical data; applying, by the one or more computers, the residual terms to an auto-regressive moving average analysis, so as to determine one or more disturbances at one or more times; calculating, by the one or more computers, a new set of coefficients for the electricity consumption equation based on the auto-regressive moving average analysis; and controlling operation of a plurality of electric power generators of the electric grid based on the electricity consumption equation with the new set of coefficients. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23)
-
Specification