Efficient forecasting for hierarchical energy systems

Abstract

Examples of energy forecasting in hierarchical energy systems are provided herein. A global forecast model instance for a hierarchical energy system can be determined through aggregation of energy forecast model data from individual energy smart meters. Energy forecast model data can include values for energy forecast model parameters used by the individual smart meters. The energy smart meters include measurement, forecasting, and calculation capabilities. The smart meters locally determine a forecast model instance used by the smart meter and provide corresponding information to higher levels in the energy system hierarchy. A global forecast model instance is determined based on the provided information.

1 - 20 . (canceled) 
     
     
         21 . An energy smart meter, comprising:
 a communication connection;   a processor; and   one or more computer-readable storage media storing computer-executable modules that, when executed by the processor, cause the energy smart meter to communicate within an energy system hierarchy, the modules comprising:
 a transmission module that, by the communication connection, communicates energy forecast model data to a higher level node in the energy system hierarchy, wherein the energy forecast model data comprises one or more values for one or more parameters of an energy forecast model instance used by the smart meter to forecast energy consumption or production, the one or more parameters being applied to one or more values measured by the smart meter; 
 a usage module that records energy usage measurements by one or more energy consumers or energy prosumers in communication with, and monitored by-the smart meter; 
 a model evaluation engine that assesses accuracy of a current energy forecast model instance using at least some of the energy usage measurements; and 
 an adaptation module that determines an updated forecast model instance having increased accuracy and replaces the current energy forecast model instance with the updated forecast model instance. 
   
     
     
         22 . The energy smart meter of  claim 21 , wherein the modules further comprise an initialization module that estimates an initial forecast model instance based on historical consumption data for the smart meter. 
     
     
         23 . the energy smart meter of  claim 21 , wherein the model evaluation engine asses the accuracy of the current energy forecast model instance by determining a forecast error and comparing the forecast error to a forecast error threshold. 
     
     
         24 . The energy smart meter of  claim 23 , wherein the determining the forecast error and comparing the forecast error to the forecast error threshold is performed at predetermined intervals. 
     
     
         25 . The energy smart meter of  claim 23 , wherein the adaptation module determines the updated forecast model instance when the determined forecast error meets or exceeds the forecast error threshold. 
     
     
         26 . The energy smart meter of  claim 25 , wherein the energy forecast model data communicated by the transmission module comprises one or more parameter values for the updated energy forecast model instance. 
     
     
         27 . The energy smart meter of  claim 25 , wherein the updated energy forecast model instance comprises a same forecast model type as the current energy forecast model instance but a different value for at least one parameter. 
     
     
         28 . The energy smart meter of  claim 21 , wherein the energy smart meter is one of a plurality of energy smart meters within the energy system hierarchy, and wherein the respective smart meters use a same energy forecast model type. 
     
     
         29 . One or more non-transitory computer-readable storage media storing computer-executable instructions that, when executed, cause an energy smart meter to participate in the operation of a global forecast model instance that forecasts energy in a hierarchical energy system, the method comprising:
 recording energy use measurements by one or more energy consumers or energy prosumers in communication with, and monitored by, the energy smart meter;   based at least in part on recorded energy use measurements, evaluating accuracy of a current energy forecast model comprising one or more parameters that are applied to one or more values measured by the energy smart meter;   updating at least one of the one or more parameters based on the evaluating; and   transmitting the updated at least one parameter to a higher level node in the hierarchical energy system.   
     
     
         30 . The one or more non-transitory computer-readable storage media of  claim 29 , the method further comprising:
 generating an initial energy forecast model instance based on historical consumption data measured by the energy smart meter.   
     
     
         31 . The one or more non-transitory computer-readable storage media of  claim 29 , wherein evaluating accuracy of a current energy forecast model comprises determining a forecast error and comparing the forecast error with a forecast error threshold. 
     
     
         32 . The one or more non-transitory computer-readable storage media of  claim 31 , wherein determining the forecast error and comparing the forecast error to the forecast error threshold is performed at predetermined intervals. 
     
     
         33 . The one or more non-transitory computer-readable storage media of  claim 31 , wherein the updating is carried out if it is determined that the determined forecast error meets or exceeds the forecast error threshold. 
     
     
         34 . A method of causing an energy smart meter to participate in the operation of a global forecast model instance that forecasts energy in a hierarchical energy system, the method comprising:
 recording energy use measurements by one or more energy consumers or energy prosumers in communication with, and monitored by, the energy smart meter;   based at least in part on recorded energy use measurements, evaluating accuracy of a current energy forecast model comprising one or more parameters that are applied to one or more values measured by the energy smart meter;   updating at least one of the one or more parameters based on the evaluating; and   transmitting the updated at least one parameter to a higher level node in the hierarchical energy system.   
     
     
         35 . The method of  claim 34 , the method further comprising:
 generating an initial energy forecast model instance based on historical consumption data measured by the energy smart meter.   
     
     
         36 . The method of  claim 34 , wherein evaluating accuracy of a current energy forecast model comprises determining a forecast error and comparing the forecast error with a forecast error threshold. 
     
     
         37 . The method of  claim 36 , wherein determining the forecast error and comparing the forecast error to the forecast error threshold is performed at predetermined intervals. 
     
     
         38 . The method of  claim 36 , wherein the updating is carried out if it is determined that the determined forecast error meets or exceeds the forecast error threshold. 
     
     
         39 . The method of  claim 34 , further comprising:
 receiving a request from a higher level node in the hierarchical energy system for at least one of the one or more parameters; and   transmitting the at least one of the one or more parameters to the higher level node in response to the request.   
     
     
         40 . The method of  claim 39 , wherein the transmitting comprises transmitting a plurality of parameters in a vector.