Efficient forecasting for hierarchical energy systems
원문보기
IPC분류정보
국가/구분
United States(US) Patent
등록
국제특허분류(IPC7판)
G06G-007/54
G06F-017/50
G06Q-050/06
출원번호
US-0170354
(2014-01-31)
등록번호
US-9977847
(2018-05-22)
발명자
/ 주소
Dannecker, Lars
Lorenz, Robert
Roesch, Philipp
출원인 / 주소
SAP SE
대리인 / 주소
Klarquist Sparkman, LLP
인용정보
피인용 횟수 :
0인용 특허 :
2
초록▼
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
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.
대표청구항▼
1. One or more non-transitory computer-readable storage media storing computer-executable instructions that, when executed, cause a computing system operating a global forecast model instance to perform a method of forecasting energy for a hierarchical energy system, the method comprising: receiving
1. One or more non-transitory computer-readable storage media storing computer-executable instructions that, when executed, cause a computing system operating a global forecast model instance to perform a method of forecasting energy for a hierarchical energy system, the method comprising: receiving energy forecast model data from a plurality of energy smart meters, the plurality of energy meters being at a lower level of the hierarchical energy system than the computer system operating the global forecast model instance, wherein the energy forecast model data includes values for a plurality of forecast model parameters that are applied to measured values to provide a forecast, the plurality of forecast model parameters comprising at least a first parameter type and at least a second parameter type, used by the energy smart meters to forecast an amount of energy produced or consumed by one or more consumers, producers, prosumers, or a combination thereof, in communication with, and monitored by, a respective energy smart meter of the plurality of energy smart meters, and wherein the values for the forecast model parameters are determined by the energy smart meters and are used by the energy smart meters to adjust an energy forecast produced by a respective energy smart meter of the plurality of energy smart meters;separately aggregating the respective forecast model parameters of the first parameter type determined by, and received from, the plurality of energy smart meters;separately aggregating the respective forecast model parameters of the second parameter type determined by, and received from, the plurality of energy smart meters;determining the global energy forecast model instance for the energy system based at least in part on the separately aggregated parameters of the first type and the separately aggregated parameters of the second type;forecasting energy consumption and production for the hierarchical energy system using the global energy forecast model instance; andadjusting a level of energy production based at least in part on the forecasted energy consumption and production. 2. The one or more non-transitory computer-readable storage media of claim 1, wherein the energy forecast model data for an individual smart meter further comprises at least one of energy usage data for an entity associated with the smart meter, temperature data, or weather data. 3. The one or more non-transitory computer-readable storage media of claim 1, wherein the global energy forecast model instance comprises a plurality of global components each having at least one corresponding global energy forecast model parameter, and wherein the separately aggregated forecast model parameters correspond to the respective global components. 4. The one or more non-transitory computer-readable storage media of claim 3, wherein the plurality of global components are determined using a weighted linear combination. 5. The one or more non-transitory computer-readable storage media of claim 3, wherein the aggregating further comprises: calculating a weighting factor for each of at least some of the plurality of energy smart meters, the weighting factor based on a current share of energy use for the smart meter out of a total energy use for the energy system; andfor the plurality of global components, determining the global components using the energy forecast model data for the respective individual smart meters and the weighting factors for the respective individual smart meters. 6. The one or more non-transitory computer-readable storage media of claim 1, wherein the aggregating further comprises approximating the current share of energy use for a smart meter as an average historical share for the smart meter. 7. The one or more non-transitory computer-readable storage media of claim 1, wherein the method further comprises: receiving updated energy forecast model data from one or more of the plurality of energy smart meters; andupdating the global energy forecast model instance for the energy system to reflect the received updated energy forecast model data. 8. The one or more non-transitory computer-readable storage media of claim 7, wherein the updated energy forecast model data represents an updated energy forecast model instance for the corresponding smart meter that was updated when a forecast error for the smart meter exceeded a forecast error threshold. 9. A method of providing a global energy forecast for a hierarchical energy system, the method comprising: receiving energy forecast model data for a plurality of energy smart meters, the plurality of energy smart meters being located lower in the hierarchical energy system than one or more computing devices providing the global energy forecast, and the energy forecast model data comprising, for respective smart meters of the plurality of energy smart meters, at least a first parameter type and at least a second parameter type, wherein a respective smart meter forecasts energy use associated with the respective smart meter using an energy forecast model instance having a plurality of energy forecast model parameters that are applied to values measured by the smart meter to provide a forecast, comprising the at least a first parameter type and the at least a second parameter type, and wherein the forecast model data for a respective smart meter comprises values for the plurality of energy forecast model parameters that are used by the respective smart meter to forecast an amount of energy produced or consumed by one or more consumers, producers, prosumers, or a combination thereof, in communication with, and monitored by, a respective energy smart meter of the plurality of energy smart meters;using the values for the plurality of energy forecast model parameters of the respective smart meters of the at least first type and the at least a second type, determining values for a plurality of global energy forecast model parameters, wherein the value for a respective global energy forecast model parameter is based on contributions from values of corresponding energy forecast model parameters of the respective smart meters in the energy forecast model data;determining a global energy forecast model instance for the energy system based on the plurality of global energy forecast model parameter values;forecasting energy consumption and production for the hierarchical energy system using the global energy forecast model instance; andadjusting a level of energy production based at least in part on the forecasted energy consumption and production. 10. The method of claim 9, wherein at least some of the global energy forecast model parameter values are determined as weighted linear combinations of the corresponding energy forecast model parameter values of the respective smart meters, and wherein the weighting for a respective smart meter corresponds to a current share of energy use measured by the smart meter out of a total energy use for the energy system. 11. The method of claim 9, further comprising: receiving updated forecast model data for one or more of the plurality of energy smart meters; andupdating the global forecast model instance for the energy system to reflect the received updated forecast model data. 12. The method of claim 11, wherein the updated forecast model data is generated after the one or more of the plurality of smart meters determines that a forecast error for the smart meter exceeded a forecast error threshold.
연구과제 타임라인
LOADING...
LOADING...
LOADING...
LOADING...
LOADING...
이 특허에 인용된 특허 (2)
Barnes, David L.; Juarez, Ruben; Wade, John, Forecasting an energy output of a wind farm.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.