PREDICTION OF ELECTRICAL POWER SYSTEM BEHAVIOR, AND RELATED SYSTEMS, APPARATUSES, AND METHODS

1 . A controller of an electrical power system, the controller comprising:

a data storage device configured to store model data indicating, for time points of a time period of operation of the electrical power system, a model load power for one or more loads of the electrical power system; and

a processor operably coupled to the data storage device and configured to:

determine, based on information received from one or more sensors of the electrical power system, current data including a current load power for the one or more loads for time points of a current time period of operation of the electrical power system, the current time period corresponding to an early portion of the time period of the model data stored by the data storage device;

fit the model data to the current data to produce predicted data, a future portion of the predicted data corresponding to time points occurring after the early portion of the time period of the model data; and

determine a set of control values to effectuate a change to operation of the electrical power system based, at least in part, on the future portion of the predicted data.

2 . The controller of claim 1 , wherein the processor is configured to fit the model data to the current data to produce the predicted data by performing a least squares regression to determine a scale and offset of the model data that minimize a sum of squares of error between the model data and the current data.

3 . The controller of claim 1 , wherein the least squares regression comprises a weighted least squares regression that favors more recent samples of the current data than older samples of the current data.

4 . The controller of claim 1 , wherein the processor is configured to interpolate the model data to have a same time step length between samples as the current data.

5 . The controller of claim 4 , wherein the processor is configured to interpolate the model data using a linear interpolation.

6 . The controller of claim 4 , wherein the processor is configured to interpolate the model data using a nonlinear interpolation.

7 . The controller of claim 1 , wherein an amount of time corresponding to the early portion of the time period is about three (3) hours to about eighteen (18) hours.

8 . The controller of claim 1 , wherein:

the model data indicates a plurality of different load power profiles of power for the one or more loads of the electrical power system; and

the processor is configured to select one of the different load power profiles that fits the current data better than the others of the different load power profiles to fit to the current data to produce the predicted data.

9 . The controller of claim 8 , wherein one of the different load power profiles corresponds to a weekday and another of the different load power profiles corresponds to a weekend day.

10 . The controller of claim 1 , wherein:

the model data also indicates a model generator power provided by one or more generators of the electrical power system for the time points of the time period of operation of the electrical power system; and

the processor is configured to determine, based on the information received from the one or more sensors of the electrical power system, current generator data of the current data including a current generator power provided by the one or more generators for the time points of the current time period of operation of the electrical power system.

11 . The controller of claim 1 , wherein the time period of operation of the electrical power system corresponding to the model data spans twenty-four (24) hours.

12 . An electrical power system, comprising:

one or more loads;

one or more sensors operably coupled to the one or more loads and configured to measure power for the one or more loads; and

a controller operably coupled to the one or more sensors, the controller configured to:

store model data indicating, for time points of a time period of operation of the electrical power system, a model load power of the one or more loads;

determine, based on information received from the one or more sensors, current data including a current load power consumed by the one or more loads for time points of a current time period of operation of the electrical power system, the current time period corresponding to an early portion of the time period of the model data stored by the data storage device;

fit the model data to the current data to produce predicted data, a future portion of the predicted data corresponding to time points occurring after the early portion of the time period of the model data; and

determine a set of control values to effectuate a change to operation of the electrical power system based, at least in part, on the future portion of the predicted data.

13 . The electrical power system of claim 12 , further comprising:

one or more generators operably coupled to the one or more sensors and configured to provide power;

wherein:

the one or more sensors are configured to measure power provided by the one or more loads;

the model data also indicates, for the time points of the time period, a model generator power provided by the one or more generators; and

the current data also indicates a current generator power provided by the one or more generators for the time points of the current time period.

14 . The electrical power system of claim 13 , wherein the one or more generators include one or more of a solar photovoltaic (PV) system, a wind generator, a combined heat and power (CHP) system, or a diesel generator.

15 . The electrical power system of claim 12 , wherein the one or more loads include one or more of an air conditioning system, a motor, or an electric heater.

16 . A method of operating an electrical power system, the method comprising:

storing current data indicating power for the one or more loads at a first plurality of different points of time over a previous number of hours of operation;

interpolating model data indicating a historic average power for the one or more loads at a second plurality of different points of time over a time span to have an interval of discretization that is the same as that of the current data, a subset of the time span corresponding to the previous number of hours of operation;

determining a scale and an offset of the interpolated model data that fits a subset of the interpolated model data corresponding to the subset of the time span to the current data;

scaling and offsetting the interpolated model data to produce predicted data, a future portion of the predicted data comprising future data; and

determining a set of control values to effectuate a change to operation of the electrical power system based, at least in part, on the future data.

17 . The method of claim 16 , wherein storing current data indicating power for the one or more loads at a first plurality of different points of time over a previous number of hours of operation comprises storing the current data indicating the power for the one or more loads over a period of about three (3) to eighteen (18) hours.

18 . The method of claim 16 , wherein interpolating model data includes interpolating the model data from having the interval of discretization between five (5) minutes and 120 minutes.

19 . The method of claim 16 , wherein interpolating model data includes linearly interpolating the model data.

20 . The method of claim 16 , wherein determining a scale and an offset of the interpolated model data comprises performing a weighted least squares regression that favors more recent observations of the current data.