Central energy system with expert subplant models
A central plant includes a plurality of subplants and circuitry. The circuitry is configured to obtain an expert model associated with a first subplant of the plurality of subplants. The expert model includes a map of values of independent variables to one or more dependent variables. The circuitry is also configured to receive input data indicating input values for the independent variables and estimate outputs values for the one or more dependent variables associated with the input variables by determining a spatial relationship between the input data and the expert model, reducing the expert model to a reduced model based on the spatial relationship, and performing an interpolation using the reduced model to generate the output values for the one or more dependent variables. The circuitry is also configured to initiate an automated action using the output values.
1. A central plant comprising:
a plurality of subplants;
circuitry configured to:
obtain an expert model associated with a first subplant of the plurality of subplants, wherein the expert model comprises a map of values of one or more independent variables characterizing operation of the first subplant to one or more dependent variables characterizing operation of the first subplant;
receive input data indicating input values for the independent variables;
estimate outputs values for the one or more dependent variables associated with the input variables by:
determining a spatial relationship between the input data and the expert model;
reducing the expert model to a reduced model based on the spatial relationship; and
performing an interpolation using the reduced model to generate the output values for the one or more dependent variables; and
initiate an automated action using the output values.
2. The central plant of claim 1 , wherein determining the spatial relationship between the input data and the expert model comprises:
defining a plurality of n-dimensional directions in coordinate system of the expert model, wherein n is a count of the independent variables; and
determining a set of closest points of the expert model to the input data in the plurality of n-dimensional directions.
3. The central plant of claim 2 , wherein the plurality of n-dimensional directions consists of 2 n n-dimensional directions.
4. The central plant of claim 2 , wherein reduced model is defined by the set of the closest points of the expert model to the input data in the plurality of n-dimensional directions.
5. The central plant of claim 1 , further comprising generating a setpoint for the first subplant by performing an optimization of a predictive objective function using the output values.
6. The central plant of claim 1 , wherein the circuitry is configured to initiate the automated action by executing a validation procedure comprising collecting measurements while the first subplant operates in accordance with at least one of the input data and the output data and generating a report using the measurements, the input data, and the output data.
7. The central plant of claim 6 , further comprising a user display device, wherein the circuitry is further configured to cause the user device to display the report.
8. The central plant of claim 1 , wherein performing the interpolation comprises a step configured to minimize the output values relative to other possible outcomes of the interpolation.
9. A method, comprising:
obtaining an expert model associated with a first subplant of a central plant, wherein the expert model comprises a map of values of independent variables to one or more dependent variables;
receiving input data indicating input values for the independent variables;
estimating outputs values for the one or more dependent variables associated with the input variables by:
determining a spatial relationship between the input data and the expert model;
reducing the expert model to a reduced model based on the spatial relationship; and
performing an interpolation using the reduced model to generate the output values for the one or more dependent variables; and
initiating an automated action using the output values.
10. The method of claim 9 , wherein determining the spatial relationship between the input data and the expert model comprises:
defining a plurality of n-dimensional directions in coordinate system of the expert model, wherein n is a count of the independent variables; and
determining a set of closest points of the expert model to the input data in the plurality of n-dimensional directions.
11. The method of claim 10 , wherein the plurality of n-dimensional directions consists of 2 n n-dimensional directions.
12. The method of claim 10 , wherein reduced model is defined by the set of the closest points of the expert model to the input data in the plurality of n-dimensional directions.
13. The method of claim 9 , wherein initiating the automated action comprises generating a setpoint for the first subplant by performing an optimization of a predictive objective function using the output values.
14. The method of claim 9 , wherein initiating the automated action comprises executing a validation procedure comprising collecting measurements while the first subplant operates in accordance with at least one of the input data and the output data and generating a report using the measurements, the input data, and the output data.
15. The method of claim 9 , wherein performing the interpolation comprises a step configured to minimize the output values relative to other possible outcomes of the interpolation.
16. One or more non-transitory computer-readable media storing program instructions that, when executed by one or more processors, cause the one or more processors to perform operations comprising:
obtaining an expert model associated with a first subplant of a central plant, wherein the expert model comprises a map of values of independent variables to one or more dependent variables;
receiving input data indicating input values for the independent variables;
estimating outputs values for the one or more dependent variables associated with the input variables by:
determining a spatial relationship between the input data and the expert model;
reducing the expert model to a reduced model based on the spatial relationship; and
performing an interpolation using the reduced model to generate the output values for the one or more dependent variables; and
initiating an automated action using the output values.
17. The non-transitory computer-readable media of claim 16 , wherein determining the spatial relationship between the input data and the expert model comprises:
defining a plurality of n-dimensional directions in coordinate system of the expert model, wherein n is a count of the independent variables; and
determining a set of closest points of the expert model to the input data in the plurality of n-dimensional directions.
18. The non-transitory computer-readable media of claim 17 , wherein the plurality of n-dimensional directions consists of 2 n n-dimensional directions.
19. The non-transitory computer-readable media of claim 18 , wherein reduced model is defined by the set of the closest points of the expert model to the input data in the plurality of n-dimensional directions.
20. The non-transitory computer-readable media of claim 16 , wherein the operations further comprise generating a setpoint for the first subplant using the output values.