System, controller, and method for predictive control of energy management for a segmented load centre
A system, controller, and method for predictive control of energy management for a segmented load centre. The method including: determining a load prediction for a predetermined prediction timeframe, the load prediction using a machine learning model taking load conditions of the power flow data as input features, the machine learning model trained using historical load conditions; determining a generation prediction for the predetermined prediction timeframe using a renewable generation model, the generation prediction using input features including solar radiation related weather data; determining an energy balance over the prediction timeframe using the load prediction and the generation prediction; modulating a minimum state of charge setpoint based on whether a current value of the setpoint results in curtailed generation in the determined energy balance over the prediction timeframe.
1. A method for predictive energy management for a segmented load centre connected to a mains power supply, the segmented load centre having an energy storage system and renewable energy sources, the method executable on a controller or executed as a model on a computing device, the method comprising:
receiving power flow data from the segmented load centre;
determining a load prediction for a predetermined prediction timeframe, the load prediction using a machine learning model taking load conditions of the power flow data as input features, the machine learning model trained using historical load conditions;
determining a generation prediction for the predetermined prediction timeframe using a renewable generation model representative of the renewable energy sources of the segmented load centre, the generation prediction using input features comprising weather data related to the generation of renewable power by the renewable energy sources of the segmented load centre;
determining an energy balance over the prediction timeframe using the load prediction and the generation prediction in an operation simulation of the segmented load centre;
modulating a minimum state of charge setpoint based on whether a current value of the setpoint results in curtailed generation in the determined energy balance over the prediction timeframe; and
controlling the segmented load centre based on the minimum state of charge setpoint over a future time period, wherein if a state of charge of the energy storage system falls below the minimum state of charge during said future time period, the energy storage system is charged from the mains power supply until the minimum state of charge is reached.
2. The method of claim 1 , wherein modulating the minimum state of charge setpoint comprises iterating the modulation until an end condition is reached.
3. The method of claim 2 , further comprising, where the minimum state of charge setpoint reaches a predetermined minimum threshold before the end condition, modulating a diversion load schedule, repeating the operation simulation of the load centre, and outputting the diversion load schedule.
4. The method of claim 2 , further comprising, where the minimum state of charge setpoint reaches a predetermined maximum threshold, iteratively performing a simulation of outage performance for time periods within the prediction timeframe.
5. The method of claim 2 , further comprising, where an outage results in lost load before a predetermined minimum outage duration threshold, increasing the minimum state of charge setpoint and repeating the simulation of outage performance until an outage end condition is reached.
6. The method of claim 1 , wherein the weather data related to the generation of renewable power comprises time series forecast for one or more of dry bulb temperature, global horizontal irradiation, direct normal irradiation, wind speed, wind direction, and precipitation.
7. The method of claim 1 , wherein the renewable generation model further takes as input installation features or internal system features of a photovoltaic array or a wind turbine array defining one of the renewable energy sources of the segmented load centre.
8. The method of claim 1 , wherein the load prediction comprises predictions for asynchronous load over the prediction timeframe and diversion load over the prediction timeframe.
9. The method of claim 1 , wherein the load conditions comprise loads during certain weather conditions, and loads during certain times of the day, days of the week, or times of the year.
10. The method of claim 1 , wherein the operation simulation comprises:
determining a net load by subtracting effective generation from a current load; and
determining energy storage capacity by:
where the net load is positive, subtracting the determined net load from the energy storage capacity; or
where the net load is negative, charging the energy storage in proportion to the net load or a total remaining capacity the energy storage.
11. A controller for predictive energy management for a segmented load centre connected to a mains power supply, the segmented load centre having an energy storage system and renewable energy sources, the controller comprising one or more processing units in communication with a memory unit comprising executable instructions to:
receive power flow data from the segmented load centre;
determine a load prediction for a predetermined prediction timeframe, the load prediction using a machine learning model taking load conditions of the power flow data as input features, the machine learning model trained using historical load conditions;
determine a generation prediction for the predetermined prediction timeframe using a renewable generation model representative of the renewable energy sources of the segmented load centre, the generation prediction using input features comprising weather data related to the generation of renewable power by the renewable energy sources of the segmented load centre;
determine an energy balance over the prediction timeframe using the load prediction and the generation prediction in an operation simulation of the segmented load centre;
modulate a minimum state of charge setpoint based on whether a current value of the setpoint results in curtailed generation in the determined energy balance over the prediction timeframe; and
control the segmented load centre based on the minimum state of charge setpoint over a future time period, wherein if a state of charge of the energy storage system falls below the minimum state of charge during said future time period, the energy storage system is charged from the mains power supply until the minimum state of charge is reached.
12. The controller of claim 11 , wherein modulating the minimum state of charge setpoint comprises iterating the modulation until an end condition is reached.
13. The controller of claim 12 , where the minimum state of charge setpoint reaches a predetermined minimum threshold before the end condition, the controller further modulates a diversion load schedule, repeating the operation simulation of the load centre, and outputting the diversion load schedule.
14. The controller of claim 12 , where the minimum state of charge setpoint reaches a predetermined maximum threshold, the controller further iteratively performs a simulation of outage performance for time periods within the prediction timeframe.
15. The controller of claim 12 , where an outage results in lost load before a predetermined minimum outage duration threshold, the controller increases the minimum state of charge setpoint and repeating the simulation of outage performance until an outage end condition is reached.
16. The controller of claim 11 , wherein the weather data related to the generation of renewable power comprises time series forecast for one or more of dry bulb temperature, global horizontal irradiation, direct normal irradiation, wind speed, wind direction, and precipitation.
17. The controller of claim 11 , wherein the renewable generation model further takes as input installation features or internal system features of a photovoltaic array or a wind turbine array defining one of the renewable energy sources of the segmented load centre.
18. The controller of claim 11 , wherein the load prediction comprises predictions for asynchronous load over the prediction timeframe and diversion load over the prediction timeframe.
19. The controller of claim 11 , wherein the load conditions comprise loads during certain weather conditions, and loads during certain times of the day, days of the week, or times of the year.
20. The controller of claim 11 , wherein the operation simulation comprises:
determining a net load by subtracting effective generation from a current load; and
determining energy storage capacity by:
where the net load is positive, subtracting the determined net load from the energy storage capacity; or
where the net load is negative, charging the energy storage in proportion to the net load or a total remaining capacity the energy storage.
21. A system for predictive energy management for a segmented load centre, the system comprising the controller of claim 11 in communication with one or more relays, an inverter, a distribution panel, a battery energy storage system, and one or more converters or rectifiers.