Systems and methods for determining an appropriate model parameter order

1. A method for determining an appropriate parameter order for an energy use model for a building site, the method comprising:

receiving an energy use model for the building site at a communications interface of a building analysis system, the energy use model comprising a weather-related predictor variable;

obtaining, by a processing circuit of the building analysis system, a plurality of data points, each of the data points comprising a value of the weather-related predictor variable and an associated energy consumption value for the building site;

calculating, by the processing circuit, a first regression statistic indicating a fit of the energy use model to the plurality of data points under a null hypothesis that the energy use model has a first parameter order and a second regression statistic indicating a fit of the energy use model to the plurality of data points under an alternative hypothesis that the energy use model has a second parameter order different from the first parameter order;

generating, by the processing circuit, a test statistic that is a ratio comprising an improvement between the first regression statistic and the second regression statistic;

comparing, by the processing circuit, the test statistic to a threshold value to determine whether the improvement warrants rejecting the null hypothesis;

determining, by the processing circuit, an appropriate parameter order for the energy use model based on a result of the comparison; receiving, at the processing circuit, inputs from one or more sensors that measure a variable state or condition of the building site;

using the energy use model with the appropriate parameter order to generate, by the processing circuit, control signals for one or more building control devices as a function of the inputs from the sensors;

and operating the building control devices according to the control signals, wherein operating the building control devices affects the variable state or condition of the building site.

2. The method of claim 1 , wherein obtaining the plurality of data points comprises, for each of the data points:

receiving at least one of: an observed temperature value and an observed enthalpy value; and

calculating the value of the weather-related predictor variable using the observed temperature value or the observed enthalpy value.

3. The method of claim 1 , wherein the energy use model includes a balance point parameter under the alternative hypothesis and does not include a balance point parameter under the null hypothesis.

4. The method of claim 3 , wherein the balance point parameter is at least one of:

a temperature parameter having a temperature value between a minimum and a maximum of a plurality of observed temperature values, and

an enthalpy parameter having an enthalpy value between a minimum and a maximum of a plurality of observed enthalpy values.

5. The method of claim 1 , further comprising analyzing energy consumption data and weather data to identify a balance point parameter for use in the energy use model, wherein:

under the null hypothesis, the balance point parameter is identified by using an extremum of the weather data as the balance point parameter; and

under the alternative hypothesis, the balance point parameter is identified by determining a range of the weather data within which the energy consumption data is a function of the weather data and using an extremum of the determined range as the balance point parameter.

6. The method of claim 3 , wherein the value of the weather-related predictor variable is a function of the balance point parameter.

7. The method of claim 1 , wherein the weather-related predictor variable is at least one of: cooling degree days, heating degree days, cooling energy days, heating energy days, temperature, and enthalpy.

8. The method of claim 1 , wherein the energy use model under the null hypothesis is nested within the energy use model under the alternative hypothesis.

9. The method of claim 1 , wherein the first regression statistic is a sum of squared error under the null hypothesis and the second regression statistic is a sum of squared error under the alternative hypothesis,

wherein the sum of squared error is a function of a difference between an energy consumption of the building site predicted by the energy use model and an actual energy consumption of the building site.

10. The method of claim 1 , wherein the test statistic is a ratio of (a) an improvement between the first regression statistic and the second regression statistic to (b) the second regression statistic divided by a number of degrees of freedom of the second regression statistic.

11. The method of claim 1 , further comprising:

identifying a significance level; and

calculating the threshold value, wherein the threshold value is a function of the identified significance level.

12. The method of claim 11 , wherein the function of the identified significance level is an inverse F-distribution function based on:

the identified significance level;

a difference between a number of degrees of freedom of the first regression statistic and a number of degrees of freedom of the second regression statistic; and

number of degrees of freedom of the second regression statistic.

13. The method of claim 1 , wherein determining the appropriate parameter order for the energy use model comprises:

rejecting the null hypothesis if the result of the comparison reveals that test statistic is not less than the threshold value; and

failing to reject the null hypothesis if the result of the comparison reveals that the test statistic is less than the threshold value.

14. The method of claim 13 , wherein determining the appropriate parameter order for the energy use model further comprises:

determining that a three-parameter model is appropriate in response to rejecting the null hypothesis; and

determining that a two-parameter model is appropriate in response to failing to reject the null hypothesis.

15. The method of claim 13 , wherein determining an appropriate parameter order for the energy use model further comprises:

identifying a range of weather data values used by the energy use model to predict energy consumption; and

determining that the energy consumption predicted by the energy use model is a function of the weather data values for weather data values within the identified range.

16. The method of claim 13 , wherein determining an appropriate parameter order for the energy use model further comprises:

in response to failing to reject the null hypothesis, identifying the building site as at least one of: a building site for which heating is not required and a building site for which cooling is not required.

17. The method of claim 1 , further comprising:

identifying a current parameter order of the energy use model;

comparing the current parameter order with the appropriate parameter order;

updating the energy use model with an energy use model having the appropriate parameter order in response to the current parameter order not matching the appropriate parameter order; and

storing the energy use model for the building site, wherein the stored energy use model has the appropriate parameter order.

18. The method of claim 17 , further comprising:

using the stored energy use model to perform a peer analysis of energy use model parameters for a class of buildings;

calculating a difference between an energy use model parameter of the stored energy use model and a mean of the energy use model parameters for the class of buildings; and

detecting an outlier model parameter based on a result of the calculation.

19. The method of claim 17 , further comprising:

monitoring changes to one or more energy use model parameters in the stored energy use model;

detecting the existence of a fault condition using a monitored change to the energy use model parameters; and

determining a change to an energy consumption that results from the fault condition based on the change to the energy use model parameters.

20. The method of claim 1 , further comprising:

updating the energy use model with an energy use model that has the appropriate parameter order;

applying inputs to the updated energy use model;

conducting a performance analysis using the updated energy use model; and

providing an output using a result of the performance analysis.

21. A system for determining an appropriate parameter order for an energy use model for a building site, the system comprising:

one or more sensors that measure a variable state or condition of the building site;

a communications interface that receives an energy use model for the building site, the energy use model comprising a weather-related predictor variable;

a processing circuit comprising a processor and memory, wherein the processing circuit:

obtains a plurality of data points, each of the data points comprising a value of the weather-related predictor variable and an associated energy consumption value for the building site;

calculates a first regression statistic indicating a fit of the energy use model to the plurality of data points under a null hypothesis that that energy use model has a first parameter order and a second regression statistic indicating a fit of the energy use model to the plurality of data points under an alternative hypothesis that the energy use model has a second parameter order different from the first parameter order;

generates a test statistic that is a ratio comprising an improvement between the first regression statistic and the second regression statistic;

compares a test statistic to a threshold value to determine whether the improvement warrants rejecting the null hypothesis;

determines an appropriate parameter order for the energy use model based on a result of the comparison;

receives inputs from the one or more sensors that measure the variable state or condition of the building site; and

uses the energy use model with the appropriate parameter order to generate control signals as a function of the inputs from the sensors; and

building control devices that operate according to the control signals and affect the variable state or condition of the building site.

22. The system of claim 21 , wherein obtaining the plurality of data points comprises, for each of the data points:

receiving at least one of: an observed temperature value and an observed enthalpy value; and

calculating the value of the weather-related predictor variable using the observed temperature value or the observed enthalpy value.

23. The system of claim 21 , wherein the energy use model includes a balance point parameter under the alternative hypothesis and does not include a balance point parameter under the null hypothesis.

24. The system of claim 23 , wherein the balance point parameter is at least one of:

a temperature parameter having a temperature value between a minimum and a maximum of a plurality of observed temperature values, and

an enthalpy parameter having an enthalpy value between a minimum and a maximum of a plurality of observed enthalpy values.

25. The system of claim 23 , wherein the value of the weather-related predictor variable is a function of the balance point parameter.

26. The system of claim 21 , wherein the first regression statistic is a sum of squared error under the null hypothesis and the second regression statistic is a sum of squared error under the alternative hypothesis,

wherein the sum of squared error is a function of a difference between an energy consumption of the building site predicted by the energy use model and an actual energy consumption of the building site.

27. The system of claim 21 , wherein the processing circuit:

identifies a significance level; and

calculates the threshold value, wherein the threshold value is a function of the identified significance level.

28. The system of claim 21 , wherein determining the appropriate parameter order for the energy use model comprises:

rejecting the null hypothesis if the result of the comparison reveals that test statistic is not less than the threshold value; and

failing to reject the null hypothesis if the result of the comparison reveals that the test statistic is less than the threshold value.

29. The system of claim 28 , wherein determining the appropriate parameter order for the energy use model further comprises:

determining that a three-parameter model is appropriate in response to rejecting the null hypothesis; and

determining that a two-parameter model is appropriate in response to failing to reject the null hypothesis.

30. The system of claim 27 , wherein determining an appropriate parameter order for the energy use model further comprises:

identifying a range of weather data values used by the energy use model to predict energy consumption; and

determining that the energy consumption predicted by the energy use model is a function of the weather data values for weather data values within the identified range.

31. The system of claim 28 , wherein determining an appropriate parameter order for the energy use model further comprises:

in response to failing to reject the null hypothesis, identifying the building site as at least one of: a building site for which heating is not required and a building site for which cooling is not required.

32. The system of claim 21 , wherein the processing circuit is further:

identifies a current parameter order of the energy use model;

compares the current parameter order with the appropriate parameter order;

updates the energy use model with an energy use model having the appropriate parameter order in response to the current parameter order not matching the appropriate parameter order; and

stores the energy use model for the building site, wherein the stored energy use model has the appropriate parameter order.

33. The system of claim 21 , wherein the processing circuit:

updates the energy use model with an energy use model having the appropriate order;

applies inputs to the updated energy use model;

conducts a performance analysis using the updated energy use model; and

provides an output using a result of the performance analysis.