Central plant control system with building energy load estimation
View Patent ↗Predictor variables that affect a building energy load are sampled at a plurality of times within a time period and aggregated to generate an aggregated value for each predictor variable over the time period. A model is generated which estimates the building energy load in terms of the predictor variables. A regression process is performed to generate values for a plurality of regression coefficients in the model based on a cumulative building energy load for the time period and the aggregated values. The sampled values of the predictor variables are then applied as inputs to the model to estimate building energy loads at each of the plurality of times. The estimated building energy loads may be used as inputs to a controller that optimizes operation of a central plant.
1. A system for optimizing operation of a central plant configured to serve building energy loads comprising:
a load estimator circuit configured to:
identify a predicted cumulative building energy load for a future time period and sample values of a plurality of predictor variables that affect a building energy load at a plurality of times within the future time period;
aggregate the values of the plurality of predictor variables to generate an aggregated value for each of the plurality of predictor variables over the future time period;
generate a model configured to estimate the building energy load in terms of the predictor variables;
perform a regression process to generate values for a plurality of regression coefficients in the model based on the predicted cumulative building energy load and the aggregated value for each of the plurality of predictor variables;
input, to the model, the values of the plurality of predictor variables sampled at each of the plurality of times within the future time period and further configured to estimate building energy loads at each of the plurality of times within the future time period as a function of the predictor variables sampled at each of the plurality of times within the first time period and the plurality of regression coefficients, wherein the plurality of times within the future time period at which the building energy loads are estimated occur more frequently than a duration of the future time period such that the predicted cumulative building energy load for the future time period is disaggregated into the plurality of building energy loads at each of the plurality of times within the future time period; and
a central plant controller configured to control a distribution of energy loads in the central plant by operating equipment of the central plant using the estimated amount of building energy loads at each of the plurality of times.
2. The system of claim 1 , wherein the load estimator circuit is further configured to:
determine a number of the plurality of times that occur within the future time period; and
sum each of the values of the plurality of predictor variables sampled during the number of the plurality of times to aggregate the plurality of predictor variable values.
3. The system of claim 1 , wherein the load estimator circuit is further configured to select the predictor variables based on an effect of each of the predictor variables on the building energy load.
4. The system of claim 1 , wherein the load estimator circuit is further configured to generate a regression model based on the predicted cumulative building energy load identified.
5. The system of claim 1 , wherein the building energy load is a heating load.
6. The system of claim 5 , wherein the time period is one month.
7. The system of claim 1 , wherein the building energy load is a cooling load.
8. The system of claim 7 , wherein the future time period is one day.
9. A method for optimizing operation of a central plant configured to serve building energy loads, the method comprising:
identifying a predicted cumulative building energy load for a future time period;
sampling values of a plurality of predictor variables that affect a building energy load at a plurality of times within the future time period;
aggregating the values of the plurality of predictor variables to generate an aggregated value for each of the plurality of predictor variables over the future time period;
generating a model that estimates the building energy load in terms of the predictor variables;
performing a regression process to generate values for a plurality of regression coefficients in the model based on predicted cumulative building energy loads and the aggregated value for each of the plurality of predictor variables;
inputting, to the model, the values of the plurality of predictor variables sampled at each of the plurality of times within the future time period;
estimating building energy loads at each of the plurality of times within the future time period as a function of the predictor variables sampled at each of the plurality of times within the future time period and the plurality of regression coefficients, wherein the plurality of times within the future time period at which the building energy loads are estimated occur more frequently than a duration of the future time period such that the predicted cumulative building energy load for the future time period is disaggregated into the plurality of building energy loads at each of the plurality of times within the future time period; and
controlling a distribution of energy loads in the central plant by operating equipment of the central plant using the estimated building energy loads at each of the plurality of times.
10. The method of claim 9 , wherein aggregating the values of the plurality of predictor variables comprises:
determining a number of the plurality of times that occur within the future time period; and
summing each of the values of the plurality of predictor variables sampled during the number of the plurality of times.
11. The method of claim 9 , further comprising selecting the predictor variables based on an effect of each of the predictor variables on the building energy load.
12. The method of claim 9 , wherein performing the regression process comprises generating a model based on the predicted cumulative building energy load identified.
13. The method of claim 9 , wherein the building energy load is a heating load.
14. The method of claim 13 , wherein the future time period is one month.
15. The method of claim 9 , wherein the building energy load is a cooling load.
16. The method of claim 15 , wherein the future time period is one day.
17. A load estimator circuit for a central plant configured to serve building energy loads, the load estimator circuit configured to:
identify a predicted cumulative building energy load for a future time period and sample values of a plurality of predictor variables that affect a building energy load at a plurality of times within the future time period;
aggregate the values of the plurality of predictor variables to generate an aggregated value for each of the plurality of predictor variables over the future time period;
generate a model configured to estimate the building energy load in terms of the predictor variables;
perform a regression process to generate values for a plurality of regression coefficients in the model based on the predicted cumulative building energy load and the aggregated value for each of the plurality of predictor variables;
input, to the model, the values of the plurality of predictor variables sampled at each of the plurality of times within the future time period and further configured to estimate building energy loads at each of the plurality of times within the future time period as a function of the predictor variables sampled at each of the plurality of times within the future time period and the plurality of regression coefficients, wherein the plurality of times within the future time period at which the building energy loads are estimated occur more frequently than a duration of the future time period such that the predicted cumulative building energy load for the future time period is disaggregated into the plurality of building energy loads at each of the plurality of times within the future time period; and
control a distribution of energy loads in the central plant by operating equipment of the central plant using the estimated building energy loads.
18. The load estimator circuit of claim 17 , wherein the load estimator circuit is further configured to:
determine a number of the plurality of times that occur within the future time period; and
sum each of the values of the plurality of predictor variables sampled during the number of the plurality of times to aggregate the plurality of predictor variable values.
19. The load estimator circuit of claim 17 , wherein the load estimator circuit is further configured to select the predictor variables based on an effect of each of the predictor variables on the building energy load.
20. The load estimator circuit of claim 17 , wherein the load estimator circuit is further configured to generate a regression model based on the predicted cumulative building energy load identified.