System and method for distribution load forecasting in a power grid

Patent metadata

Field	Value
Publication number	US-10700523-B2
Application number	US-201715688370-A
Country	US
Kind code	B2
Filing date	Aug 28, 2017
Priority date	Aug 28, 2017
Publication date	Jun 30, 2020
Grant date	Jun 30, 2020

Abstract

A method for short term load forecasting in a power grid includes obtaining historical data comprising power data, load data and weather data corresponding to time index data recorded from a location in a power distribution network of the power grid. The method further includes receiving power grid data comprising a plurality of power values, and a plurality of weather parameter values corresponding to a plurality of recent time instant values. The method also includes generating modified historical data using statistical techniques to rectify error conditions. The method further includes estimating one or more power values at a future time instant based on the modified historical data and the power grid data. The method also includes balancing load of the power distribution network based on the estimated one or more power values.

First claim

The invention claimed is:

1. A method for short term load forecasting in a power grid, comprising:
obtaining historical data comprising power data, load data and weather data corresponding to time index data recorded from a location in a power distribution network of the power grid;
receiving power grid data comprising a plurality of power values, and a plurality of weather parameter values corresponding to a plurality of recent time instant values;
generating modified historical data using statistical techniques to rectify error conditions, wherein generating modified historical data comprises:
selecting a first plurality of power values recorded in a first time period and a second plurality of power values recorded in a second time period, wherein the first time period precedes the second time period;
determining a second plurality of power estimates based on the first plurality of power values; and
rectifying the second plurality of power values by replacing at least one of the second plurality of power values by corresponding second plurality of power estimates when a difference between the two is larger than a predetermined threshold value to generate modified second plurality of power values;

estimating one or more power values at a future time instant, based on the modified historical data using a linear model and the power grid data at a more recent time instant, wherein the estimation comprises determining a third plurality of power values at a plurality of time instant values in a third time period, and wherein the first time period precedes the third time period and the second time period precedes the third time period; and
balancing load of the power distribution network based on the estimated one or more power values.

2. The method of claim 1 , wherein obtaining the historical data comprises recording a plurality of historical power values and a plurality of historical temperature values at the location corresponding to time index data, wherein the time index data comprises hourly time stamp values.

3. The method of claim 2 , wherein receiving the historical data comprises recording a correlation between the plurality of historical temperature values and the plurality of historical power values.

4. The method of claim 1 , wherein receiving the power grid data comprises measuring a plurality of power values at the location at a plurality of time stamp values in a specified time period.

5. The method of claim 1 , wherein estimating the one or more power values at the future time instant comprises
receiving a pair of recorded power values corresponding to a pair of time instants in the second time period, wherein the pair of time instants comprises a first time instant and a second time instant subsequent to the first time instant;
determining a pair of forecast error values based on the pair of recorded power values and a corresponding pair of estimated power values;
determining a regression model based on the pair of forecast error values and the corresponding pair of estimated power values; and
modifying the third plurality of power values based on the modified second plurality of power values using the regression model.

6. The method of claim 1 , wherein estimating the one or more power values at the future time instant comprises determining a plurality of linear models corresponding to the plurality of time instant values using a least mean squared estimation technique.

7. The method of claim 1 , wherein receiving the power grid data comprises receiving data from supervisory control and data acquisition (SCADA) architecture, and weather forecast data.

8. The method of claim 1 , wherein the more recent time instant is a real-time instant.

9. A system for short term load forecasting in a power grid, comprising: a data acquisition module configured to:
obtain historical data comprising power data, load data and weather data corresponding to time index data recorded from a location in a power distribution network of the power grid;
receive, from power grid sensors, power grid data comprising a plurality of power values, a plurality of load values and a plurality of weather parameter values corresponding to a plurality of recent time instant values;
a pre-processing module communicatively coupled to the data acquisition module and configured to:
determine an error condition in the historical data based on statistical techniques;
generate modified historical data using statistical techniques to rectify error conditions, wherein generating modified historical data comprises:
selecting a first plurality of power values recorded in a first time period and a second plurality of power values recorded in a second time period, wherein the first time period precedes the second time period;
determining a second plurality of power estimates based on the first plurality of power values; and
rectifying the second plurality of power values by replacing at least one of the second plurality of power values by corresponding second plurality of power estimates when a difference between the two is larger than a predetermined threshold value to generate modified second plurality of power values;

a forecasting module communicatively coupled to the pre-processing module and configured to estimate one or more power values at a future time instant based on the modified historical data using a linear model and the power grid data at a more recent time instant wherein the estimation comprises determining a third plurality of power values at a plurality of time instant values in a third time period, wherein the first time period precedes the third time period and the second time period precedes the third time period; and
a control module communicatively coupled to the forecasting module and configured to balance load of the power distribution network based on the estimated one or more power values.

10. The system of claim 9 , wherein the data acquisition module is configured to record a plurality of historical power values and a plurality of historical temperature values at the location corresponding to time index data, wherein the time index data comprises hourly time stamp values.

11. The system of claim 10 , wherein the data acquisition module is configured to record a correlation between the plurality of historical temperature values and the plurality of historical power values.

12. The system of claim 9 , wherein the data acquisition module is configured to measure a plurality of power values at the location at hourly time stamp values in a specified time period.

13. The system of claim 9 , wherein the forecasting module is configured to:
receive a pair of recorded power values corresponding to a pair of time instants in the second time period, wherein the pair of time instants comprises a first time instant and a second time instant subsequent to the first time instant;
determine a pair of forecast error values based on the pair of recorded power values and a corresponding pair of estimated power values;
determine a regression model based on the pair of forecast error values and the corresponding pair of estimated power values; and
modify the third plurality of power values based on the modified second plurality of power values using the regression model.

14. The system of claim 9 , wherein the forecasting module is configured to determine a plurality of linear models corresponding to the plurality of time instant values using a least mean squared estimation technique.

15. The system of claim 9 , wherein the data acquisition module is configured to receive data from supervisory control and data acqui

Assignees

• Gen Electric

Inventors

• Pan Yan
• Premerlani William James

Classifications

• Y04S10/50

  Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications · CPC title

• G06Q50/06

  Energy or water supply · CPC title

• H02J3/28 Primary

  Arrangements for balancing of the load in networks by storage of energy · CPC title

• H02J3/003 Primary

  Load forecast, e.g. methods or systems for forecasting future load demand · CPC title

• G06Q10/04 Primary

  Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem" (market predictions or forecasting for commercial activities G06Q30/0202) · CPC title