Constructing and calibrating enthalpy based predictive model for building energy consumption

Patent metadata

Field	Value
Publication number	US-9292888-B2
Application number	US-201313930536-A
Country	US
Kind code	B2
Filing date	Jun 28, 2013
Priority date	Jun 28, 2013
Publication date	Mar 22, 2016
Grant date	Mar 22, 2016

Abstract

A predictive model for building energy consumption may be constructed and run to predict energy consumption in a building or to detect anomaly in energy consumption in a building or combinations thereof. Historic energy consumption data associated with energy consumed in a building may be received. Enthalpy of air outside the building may be determined. An energy consumption model may be calibrated based on the historic energy consumption data and the enthalpy of air outside the building. The energy consumption model incorporates enthalpy difference between a balance enthalpy and the enthalpy of air outside the building.

First claim

We claim:

1. A method of constructing a predictive model for building energy consumption, comprising:
receiving historic energy consumption data associated with energy consumed in a building;
determining enthalpy of air outside the building; and
calibrating, by a processor, an energy consumption model that predicts the building energy consumption, the energy consumption model incorporating enthalpy difference between a balance enthalpy and the enthalpy of air outside the building, the calibrating performed based on the historic energy consumption data and the enthalpy of air outside the building,
wherein the energy consumption model implements energy consumption measured per time period per sensor in the building in terms of a base enthalpy load over the time period, heating enthalpy load over the time period and cooling enthalpy load over the time period,
the heating enthalpy load over the time period determined as a sum over a number of time intervals during the time period, of maximum of a zero value and a difference between the balance enthalpy associated with the sensor and an enthalpy of outside air at a respective time interval,
the cooling enthalpy load over the time period determined as a sum, over the number of time intervals during the time period, of maximum of a a zero value and a difference between the enthalpy of outside air at the respective time interval and the balance enthalpy associated with the sensor,
wherein the calibrating determines at least a first factor associated with the base enthalpy load over the time period, a second factor associated with the heating enthalpy load over the time period and a third factor associated with the cooling enthalpy load over the time period,
the method further comprising providing a user interface allowing a user to interact with the processor to execute the predictive model and to present on a display device a graph display of one or more predicted results of the building energy consumption.

2. The method of claim 1 , wherein the determining enthalpy of air outside the building comprises receiving historic weather data, and computing the enthalpy of air outside the building based on the historic weather data.

3. The method of claim 1 , wherein the calibrating determines values associated with balance enthalpy for heating, balance enthalpy for cooling, coefficient for base enthalpy, coefficient for heating enthalpy, coefficient for cooling enthalpy terms incorporated in the energy consumption model.

4. The method of claim 1 , wherein the energy consumption model is calibrated for different buildings using historic energy consumption data and historic weather data respectively associated with the different buildings.

5. The method of claim 1 , wherein the energy consumption model is run to predict energy consumption associated with the building in a future time, based on forecasted weather of the future time.

6. The method of claim 1 , wherein the energy consumption model is run to detect anomaly in energy consumption associated with the building.

7. The method of claim 6 , wherein the anomaly is detected by one or more of:
identifying energy consumption outside of a control bound; or
identifying drifting trend of energy consumption by cumulative sum of difference analysis; or
combinations thereof.

8. A method of constructing a predictive model for building energy consumption, comprising:
receiving historic energy consumption data associated with energy consumed in a building;
determining enthalpy of air outside the building; and
calibrating, by a processor, an energy consumption model that predicts the building energy consumption, the energy consumption model incorporating enthalpy difference between a balance enthalpy and the enthalpy of air outside the building, the calibrating performed based on the historic energy consumption data and the enthalpy of air outside the building,
wherein the energy consumption model comprises an enthalpy based regression model in a mathematical form of

E
ij

=

b
j

+

h
j

·

HEL
ij

⁡

(

h

h
,
bal
,
j

)

+

c
j

·

CEL
ij

⁡

(

h

c
,
bal
,
j

)

+

ɛ
ij

HEL
ij

⁡

(

h

h
,
bal
,
j

)

=

∑

k
=
1

N
j

⁢

max
⁡

(

h

h
,
bal
,
j

-

h
k

,
0

)

,

⁢

CEL
ij

⁡

Assignees

• Globalfoundries Inc

Inventors

• An Lianjun
• Chae Young T
• Lee Young Min
• Liu Fei
• Zhang Rui

Classifications

• G06Q50/06 Primary

  Energy or water supply · CPC title

• G06Q10/04

  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