Method for predicting power demand and controlling ess charge/discharge based on the predicted demand, and apparatus using the same

Patent metadata

Field	Value
Publication number	US-2019011970-A1
Application number	US-201816018053-A
Country	US
Kind code	A1
Filing date	Jun 26, 2018
Priority date	Jul 6, 2017
Publication date	Jan 10, 2019
Grant date	—

Abstract

A method for predicting consumer power demand uses power consumption data measured over a long term and a power usage pattern immediately before a target time and for controlling ESS charge/discharge of an ESS based on the predicted power demand. A power demand prediction apparatus using the method includes respective components for collecting weather data and data on power used by the consumer; selecting data points according to a preset condition from among data collected by the data collector based on a specific time span; generating long term prediction data for the power demand in the specific time span based on the selected data points; analyzing a power usage pattern immediately before the specific time span and comparing the power usage pattern with the long-term prediction data, to determine whether prediction data correction is required; and correcting the prediction data based on the power usage pattern when correction is required.

First claim

What is claimed is:

1 . A power demand prediction apparatus for predicting power demand of a consumer, the apparatus comprising:
a data collector for collecting weather data and data on power used by the consumer; a data selector for selecting data points according to a preset condition from among data collected by the data collector based on a specific time span for which power demand is to be predicted; a long-term prediction data generator for generating long term prediction data for the power demand in the specific time span based on the selected data points; a power usage pattern analyzer for analyzing a power usage pattern immediately before the specific time span and comparing the power usage pattern with the generated long-term prediction data, to determine whether correction of the long-term prediction data is required; and a prediction data corrector for correcting the long-term prediction data based on the analyzed power usage pattern when the power usage pattern analyzer determines that the correction is required.

2 . The apparatus according to claim 1 , wherein the long-term prediction data generator includes:
a weight setting unit for setting a weight for each selected data point; and a data generating unit for generating the long-term prediction data using a weighted average based on the selected data, in which the weighted average is obtained by dividing a sum of values, obtained by multiplying each data point and the weight set for each data point on all the data, by a sum of the weights set for all data.

3 . The apparatus according to claim 2 , wherein the long-term prediction data generator further includes a data arranging unit for calculating a difference between a temperature at the specific time span and a temperature at the time of collecting each selected data point and arranging the selected data points based on the difference,
the weight setting unit sets the weight only for a predetermined number of selected data points having the smallest temperature difference from the temperature at the specific time span among the arranged data points, and the data generating unit generates the long-term prediction data by obtaining a weighted average of the predetermined number of selected data points having the smallest temperature difference from the temperature at the specific time span.

4 . The apparatus according to claim 3 , wherein the weight setting unit sets the weight for each selected data point based on a difference between a year corresponding to the specific time span and a year when the selected data is collected.

5 . The apparatus according to claim 4 , wherein the weight setting unit sets the weight as 1 when the difference is zero, sets the weight as 0.9 when the difference is one year, sets the weight as 0.8 when the difference is two years, sets the weight as 0.7 when the difference is three years, sets the weight as 0.6 when the difference is four years, and sets the weight as 0.5 when the difference is five years.

6 . The apparatus according to claim 1 , wherein the power usage pattern analyzer determines that correction of the long-term prediction data is unnecessary when a line connecting two used power amounts collected at two points of time earlier than the specific time span and a line connecting two long-term prediction power demand amounts at the same point of time generated by the long-term prediction data generation unit cross each other, and determines that correction of the long-term prediction data is necessary when the lines do not cross each other.

7 . The apparatus according to claim 1 , wherein the data collector collects temperature data and used power data at fifteen-minute intervals.

8 . The apparatus according to claim 1 , wherein the preset condition is based on a past date, the past date being at least one of
a date falling during a time span of fifteen days before and after a date corresponding to the specific time span, and a date similar in type to a date corresponding to the specific time span, in which the type is one of a holiday, a weekend, a weekday, or a date falling between a holiday and another non-working day.

9 . A charge/discharge control apparatus for controlling charge/discharge of a battery of an energy storage system, the apparatus comprising:
a power demand prediction unit for predicting power demand required by a consumer in at least one time span; a charge/discharge time setting unit for setting a charge time range when the battery is charged and a discharge time range when the battery is discharged; a charge scheduling unit for scheduling an amount of power to be supplied for charging the battery for each time span of the charge time range; a discharge scheduling unit for scheduling an amount of power to be discharged from the battery for each time span of the discharge time range; and a control unit for respectively controlling the charging and discharging of the battery as scheduled by the charge scheduling unit and the discharge scheduling unit.

10 . The apparatus according to claim 9 , wherein the amount of power to be supplied for charging the battery for each time span of the charge time range is constant.

11 . The apparatus according to claim 9 , wherein the discharge scheduling unit is provided to discharge all the power charged into the battery in the discharge time range, and discharge an amount of power obtained by subtracting a certain amount of power supplied from a system from the power demand predicted by the power demand prediction unit for each time span to allow an amount of power supplied from the system to be constant.

12 . The apparatus according to claim 11 , wherein the discharge scheduling unit obtains a value of P L satisfying
P B =Σ( P a ( n )− P L )

where P B is an amount of power charged into the battery, P a (n) is the predicted power demand at each time span, and P L is an amount of power input from the system, and
wherein a discharge power P o (n) at each time span is obtained using
P o ( n )=max(0, P a ( n )− P L )

where the max(A, B) is a function of selecting a larger value of A and B.

13 . A charge/discharge control apparatus, the apparatus comprising:
a power demand prediction unit for predicting power demand of a consumer within a specific period based on past power usage data of the consumer; a charge/discharge time setting unit for establishing a charge/discharge operation schedule of an energy storage apparatus for a day according to the predicted power demand and manages the energy storage apparatus according to the charge/discharge operation schedule; a power demand correcting unit for comparing the predicted power demand with an actual load of the consumer and corrects the predicted power demand when there is a difference of a predefined size; and an operation schedule update unit for updating the charge/discharge operation schedule based on the corrected power demand and a charge/discharge state of the energy storage apparatus.

14 . The apparatus according to claim 13 , wherein the power demand prediction unit includes:
an artificial intelligence engine; and a power demand prediction unit for using the artificial intelligence engine to predict the power demand of the consumer within the specific period via at least one of a prediction algorithm, a statistical technique, a Delphi technique of an artificial intelligence based on the past power usage data, a weather forecast pattern, and a consumer characteristic pattern.

15 . The apparatus according to claim 13 , wherein the charge/discharge time s

Assignees

• Doosan Heavy Ind & Construction Co Ltd

Inventors

• Youn Ju Young
• Park Sung Ku
• Rhie Young Hoon

Classifications

• G05B15/02

  electric · CPC title

• G06F1/263

  Arrangements for using multiple switchable power supplies, e.g. battery and AC (G06F1/30 takes precedence) · CPC title

• G06N99/005

  Physics · mapped topic

• G06F1/28 Primary

  Supervision thereof, e.g. detecting power-supply failure by out of limits supervision · CPC title

• G06F1/3212 Primary

  Monitoring battery levels, e.g. power saving mode being initiated when battery voltage goes below a certain level · CPC title