Intelligent source switching in microgrid connected households

1 . A method of controlling an energy supply system comprising a central power grid and at least one local power grid configured to supply at least one entity with energy, comprising:

acquiring information indicating, for a past time period, amount of energy having been supplied to the at least one entity by the central power grid and the at least one local power grid, energy supply capacity of the central power grid and the at least one local power grid, and information indicating conditions affecting energy supply and energy capacity during said past time period;

training a machine learning, ML, model with the acquired information;

supplying the trained ML model with information indicating an upcoming time period;

computing, by the trained ML model, expected energy delivery capacity of the central power grid and the at least one local power grid and expected energy demand of the at least one entity during the upcoming time period; and

controlling selection of amount of energy supplied from the central power grid and from the at least one local power grid according to an energy criteria to be satisfied such that the expected energy demand of the at least one entity is met during the upcoming time period based on the computation of the trained ML model,

wherein the ML model is trained by applying reinforcement learning, RL, and the trained ML model is an RL agent used for controlling the selection of one or more energy sources supplying the at least one entity with energy.

2 . The method of claim 1 , wherein the controlling of selection of amount of energy supplied further comprises:

controlling feedback of energy produced by the at last one local power grid to the central power grid.

3 . The method of claim 1 , the at least one local power grid comprising one or more renewable energy sources supplying the at least one entity with energy.

4 . The method of claim 3 , the one or more renewable energy sources comprising one or more of solar energy sources, wind energy sources, geothermal energy sources and hydropower energy sources.

5 . The method of claim 3 , the at least one local power grid comprising one or more energy storage devices capable of storing energy produced by the one or more renewable energy sources.

6 . The method of claim 1 , the information indicating conditions affecting the energy supply and the energy capacity comprising one or more of weather conditions, load on the central power grid, time of day, time of week, time of year, occurred central power grid blackouts and/or maintenance.

7 . The method of claim 1 , the computing further comprising taking into account expected weather conditions and/or planned power grid maintenance during the upcoming time period.

8 . The method of claim 1 , wherein a plurality of entities are to be supplied with energy, each entity being supplied with energy from the central power grid and a respective local power grid, the controlling of the selection of amount of energy supplied comprises controlling selection of amount of energy supplied from the central power grid and from the respective local power grid according to an energy criteria to be satisfied such that the expected energy demand of each entity is met.

9 . A non-transitory computer program product comprising a non-transitory computer readable medium storing a computer program comprising computer-executable instructions for causing a device to perform the method of claim 1 when the computer-executable instructions are executed on processing circuitry included in the device.

10 . A device configured to control an energy supply system comprising a central power grid and at least one local power grid configured to supply at least one entity with energy, said device comprising a processing circuitry and a memory, said memory containing instructions executable by said processing circuitry, whereby the device is operative to:

acquire information indicating, for a past time period, amount of energy having been supplied to the at least one entity by the central power grid and the at least one local power grid, energy supply capacity of the central power grid and the at least one local power grid, and information indicating conditions affecting energy supply and energy capacity during said past time period;

train a machine learning, ML, model with the acquired information;

supply the trained ML model with information indicating an upcoming time period;

compute, by the trained ML model, expected energy delivery capacity of the central power grid and the at least one local power grid and expected energy demand of the at least one entity during the upcoming time period; and

control selection of amount of energy supplied from the central power grid and from the at least one local power grid according to an energy criteria to be satisfied such that the expected energy demand of the at least one entity is met during the upcoming time period based on the computation of the trained ML model,

wherein the ML model is trained by applying reinforcement learning, RL, and the trained ML model is an RL agent used for controlling selection of one or more energy sources supplying the at least one entity with energy.

11 . The device of claim 10 , further being operative to, when controlling selection of amount of energy supplied:

control feedback of energy produced by the at least one local power grid to the central power grid.

12 . The device of claim 10 , the at least one local power grid comprising one or more renewable energy sources supplying the at least one entity with energy.

13 . The device of claim 12 , the one or more renewable energy sources comprising one or more of solar energy sources, wind energy sources, geothermal energy sources and hydropower energy sources.

14 . The device of claim 12 , the at least one local power grid comprising one or more energy storage devices capable of storing energy produced by the one or more renewable energy sources.

15 . The device of claim 10 , the information being configured to indicate conditions affecting the energy supply and the energy capacity comprising one or more of weather conditions, load on the central power grid, time of day, time of week, time of year, occurred central power grid blackouts and/or maintenance.

16 . The device of claim 10 , further being operative to, when computing the expected energy delivery capacity, taking into account expected weather conditions and/or planned power grid maintenance during the upcoming time period.

17 . The device of claim 10 , the ML model being trained by applying the reinforcement learning, RL.