Energy exchange in a set of entities

Abstract

The invention relates to a method for exchanging energy in a set of at least two entities which are configured, respectively, according to at least one energy use profile, characterised in that it implements the following in a current time interval in one of the entities: - receiving (S1) information relating to a value of the price that the energy available in the current time interval costs from the other entity and / or from an energy supplier external to the set, and relating to a value of the amount of energy which, according to the profile of the entity, depends on the amount of energy produced in the current time interval by at least one energy-producing sub-entity associated with said entity, on the amount of energy consumed in the current time interval by at least one energy-consuming sub-entity associated with the entity, and on the amount of energy stored in the current time interval by at least one energy-storing sub-entity associated with the entity; and - on the basis of the value of the amount of energy and the price, selecting (S2, S3a, S3b), where appropriate, one action from among an action of supplying energy to the other entity or to the energy supplier and an action of requesting energy from the other entity or from the energy supplier, said selection being implemented according to a performance criterion (R1i,c; R2i,c) of the energy exchange.

Description

[0001] DESCRIPTION

[0002] Energy exchange in a set of entities

[0003] Field of invention

[0004] The invention relates generally to the field of energy exchange on an energy exchange place, in which several entities which make up this exchange place are capable of implementing energy transactions on this exchange place, according to their corresponding profile which can be for example energy producer and / or energy consumer and / or energy store.

[0005] More specifically, the invention relates to the selection of an optimal energy exchange strategy for each entity, so that each entity of the exchange place implements an energy transaction, in other words requires and / or provides energy while respecting at least one criterion such as the reduction of expenses for requiring energy when the entity consumes energy, the increase in profits when the entity produces energy, the increase / maintenance of the comfort of the user of the entity when the entity consumes energy, the reduction of the carbon footprint when the entity implements an energy transaction, etc.

[0006] Prior art

[0007] There are currently different possible models of energy exchanges. However, the mechanisms implemented to optimize energy exchange can still be improved. Indeed, in some energy exchange modeling works, the calculations are handled by a limited number of entities that make up the exchange, while others are incomplete because they do not take into account the multi-profile nature that an entity can have (consume / produce / store) or they only address one facet of the energy exchange strategy (for example, managing the balance between energy demand and the response to this demand, controlling how to distribute energy needs at the level of each entity, planning the operation of energy-consuming entities, etc.).

[0008] Purpose and summary of the invention One of the aims of the invention is to remedy drawbacks of the aforementioned state of the art by proposing an energy exchange method which makes it possible to take into account, for a given entity of the exchange place, all the possible profiles of use of the energy conferred on this entity, as well as all the possible strategies of energy exchange determined by this entity, this for the benefit of an optimized distribution of the calculations on all the entities of the exchange place, so as to optimize the performances of the energy exchange between the entities.

[0009] To this end, an object of the present invention relates to a method for exchanging energy in a set of at least two entities configured respectively according to at least one energy use profile among a first energy production profile, a second energy consumption profile and a third energy storage profile.

[0010] Such a method is remarkable in that it implements the following in a current time interval, at the level of at least one of the entities:

[0011] - receive information relating to:

[0012] * an energy quantity value which, according to the profile of said at least one entity, depends on:

[0013] -- the quantity of energy produced, in the current time interval, by at least one energy production sub-entity associated with said at least one entity,

[0014] - the quantity of energy consumed, in the current time interval, by at least one energy consumption sub-entity associated with said at least one entity,

[0015] - the quantity of energy stored, in the current time interval, by at least one energy storage sub-entity associated with said at least one entity,

[0016] * a value of the price of the energy available in the current time interval from the other entity and / or from an energy supplier external to said group,

[0017] - from said energy quantity value and the value of said price, select, where appropriate, an action from among an action of supplying energy to said other entity or to the external energy supplier, an action of requesting energy from said other entity or from the external energy supplier, said selection being implemented in accordance with a performance criterion of the energy exchange. The invention advantageously makes it possible to construct an energy exchange place between different entities of the same set which takes into account all the possible energy use profiles of a given entity, namely producing energy and / or consuming energy and / or storing energy, which makes it a particularly complete energy exchange place.Such a set of entities is for example a group of buildings located in the same neighborhood, a set of buildings located in an industrial zone, a fleet of boats in a port, a plurality of base stations respectively serving a plurality of cells of a communication network, etc. The set of entities is not limited to an entity of the same type, for example building, building, boat, base station, etc. Thus, the set of entities can include for example one or more buildings in the same neighborhood and one or more base stations, one or more boats in a port and one or more base stations, etc.

[0018] Such an energy exchange method is not only efficient and precise, at the level of the action that is selected in the current time interval, but it is also adaptable over time, because it is based on a hierarchical structure composed of entities at a higher level that interact with sub-entities at a lower level. Such an interaction is advantageous in the sense that it allows the higher-level entity to select, from the information received from one or more sub-entities associated with this entity, the optimal strategy for using energy in the current time interval, in accordance with an energy performance criterion.

[0019] According to a particular embodiment, the energy performance criterion used minimizes the cost of the energy required by said at least one entity in the current time interval and maximizes the profit of supplying energy to the other entity or to the external energy supplier in the current time interval.

[0020] According to this embodiment, the optimal energy use strategy, in the current time interval, is advantageously based on a compromise of reducing expenditure for the energy required by the entity and maximizing profits for the energy supplied by the entity.

[0021] According to another particular embodiment, the information received further comprises a carbon footprint value determined, in the current time interval, by said at least one energy production or storage sub-entity associated with said at least one entity, and transmitted by said at least one sub-entity to said at least one entity, and in which the selection of an action is further implemented from said carbon footprint value in accordance with a criterion for minimizing the carbon footprint of the energy to be supplied to the other entity or to the external energy supplier.

[0022] This embodiment has the advantage of adding the minimization of the carbon footprint of the energy to be supplied to the other entity to the compromise of reducing expenses for the energy required by the entity and maximizing profits for the energy supplied by the entity. As a result, the energy exchange method according to this embodiment is made more energy-efficient and less polluting.

[0023] According to another particular embodiment, the quantity of energy to be consumed, in the current time interval, by at least one energy consumption sub-entity associated with said at least one entity, is based on the energy consumption calculated at the level of at least one energy consuming device which is associated with said at least one energy consumption sub-entity, in accordance with the minimization of a dissatisfaction criterion of a user of said at least one device, said criterion being linked:

[0024] - either to a shift in the use of said at least one device in a time interval subsequent to the current time interval, if said at least one device has a usage profile that can be shifted over time,

[0025] - either to a reduction in the power level of said at least one device in the current time interval, if said at least one device has a power-shiftable usage profile.

[0026] Such an embodiment allows the energy consumption sub-entity associated with said at least one entity to apply at its level an optimal strategy to determine, in the current time interval, the quantity of energy to be consumed in accordance with an energy performance criterion which is based here on minimizing the dissatisfaction of a user of at least one energy consuming device attached to the energy consumption sub-entity, as a sub-entity of this energy consumption sub-entity.

[0027] According to another particular embodiment, said at least one energy production sub-entity associated with said at least one entity implements the following in the current time interval: - receive information relating to:

[0028] -- a value of the price of the energy available in the current time interval from the other entity and / or from an energy supplier external to said set, -- the quantity of energy stored, in the current time interval, by at least one energy storage sub-entity associated with said at least one entity, if said at least one storage sub-entity is present,

[0029] - based on the price value and, where applicable, the quantity of energy stored, and according to an energy production performance criterion:

[0030] -- select a destination of the energy produced by said at least one energy production sub-entity in the current time interval from among said other entity or the external energy supplier, said at least one energy storage sub-entity associated with said at least one entity, at least one energy consuming device associated with said at least one energy consumption sub-entity,

[0031] -- calculate the quantity of energy produced to be used according to the selected destination. Such an embodiment allows the energy production sub-entity associated with said at least one entity to also apply at its level an optimal strategy to determine, in the current time interval, the quantity of energy to be used in accordance with an energy production performance criterion, depending on the action selected according to this criterion which is either to supply energy to the other entity or to the external energy supplier, to charge an energy storage sub-entity associated with said at least one entity, or to supply energy to at least one energy consuming device attached to said at least one energy consuming entity.

[0032] According to another particular embodiment, said at least one energy storage sub-entity associated with said at least one entity implements the following in the current time interval:

[0033] - receive information relating to at least one value of the price of the energy available in the current time interval from an energy supplier external to said group,

[0034] - from the value of the price, the quantity of energy stored in the current time interval, by said at least one energy storage sub-entity, and according to a performance criterion of the use of the stored energy, select: -- an action relating to the non-recharging or the recharging with energy of said at least one storage sub-entity,

[0035] -- an action relating to the non-discharge or the discharge of energy from said at least one storage sub-entity,

[0036] - calculate the amount of energy required for recharging, respectively unloading, if the action relating to recharging, respectively unloading, is selected.

[0037] Such an embodiment allows the energy storage sub-entity associated with said at least one entity to also apply at its level an optimal strategy to determine, in the current time interval, the quantity of energy necessary for its recharging or discharging in accordance with a performance criterion for the use of the stored energy.

[0038] According to another particular embodiment, the performance criterion for the use of the stored energy maximizes the life cycle duration of said at least one storage sub-entity.

[0039] According to this embodiment, the performance criterion for the use of the stored energy which is used by the energy storage sub-entity advantageously takes into account the maximization of the life cycle duration of the storage sub-entity to determine the energy to be used for its charging or discharging. Thus, said at least one energy storage sub-entity, at its level also, selects its action with a view to saving energy and reducing pollution, which contributes to respecting sustainable development. According to another particular embodiment, the steps implemented by said at least one entity, said energy production, energy consumption, energy storage sub-entities, as well as said at least one energy consuming device, are executed using a learning algorithm.Such a learning algorithm is particularly well suited to the hierarchical structure on which the energy exchange method according to the invention is based, where each action will be learned level by level, that is to say, at the level of the entities of said set, at the level of the sub-entities associated with the entities of said set, and at the level of the energy consuming device(s) associated in particular with said at least one energy consuming sub-entity. Thus, the learning of the actions is advantageously distributed over each of these levels instead of being concentrated solely on the entities of said set, which provides scalability to the energy exchange method according to the invention and accelerates its learning.

[0040] According to another particular embodiment, the learning algorithm is of the reinforcement type, in which:

[0041] - the entities are agents, while the sub-entities and the at least one energy-consuming device are sub-agents associated with the agents,

[0042] - the information received by the agents and sub-agents is representative of an environment in which the energy exchange process is implemented,

[0043] - the selected actions are decisions taken by the agents.

[0044] The advantage of using such a reinforcement learning algorithm is that it is particularly powerful and reliable in the case of an energy exchange method based on a plurality of agents and corresponding sub-agents, such as the energy exchange method according to the invention.

[0045] According to another particular embodiment, for at least one agent considered, the agent transmits at least one objective to at least one sub-agent which is associated with it and which is in a given state, this objective being to be satisfied by the sub-agent and being integrated into the given state of the sub-agent.

[0046] Thus, in this particular embodiment, communication is advantageously established between at least one agent of the exchange place and a sub-agent associated with it, thereby characterizing the hierarchical structure of the exchange place.

[0047] The various embodiments or features mentioned above may be added independently or in combination with each other to the energy exchange method defined above.

[0048] The invention also relates to an entity configured to exchange energy with at least one other entity, said entity and said other entity belonging to a set of entities configured according to at least one energy use profile among a first energy production profile, a second energy consumption profile and a third energy storage profile.

[0049] Such an entity is remarkable in that it implements the following, in a common time interval:

[0050] - receive information relating to: * an energy quantity value which, according to the profile of said at least one entity, depends on:

[0051] -- of the quantity of energy produced, in the current time interval, by at least one energy production sub-entity associated with said at least one entity, -- of the quantity of energy consumed, in the current time interval, by at least one energy consumption sub-entity associated with said at least one entity,

[0052] -- the quantity of energy stored, in the current time interval, by at least one energy storage sub-entity associated with said at least one entity,

[0053] * a value of the price of the energy available in the current time interval from the other entity and / or from an energy supplier external to said group,

[0054] - from the value of the quantity of energy and the value of said price, select, where appropriate, an action from among an action of supplying energy to said other entity or to the external energy supplier, an action of requesting energy from said other entity or from the external energy supplier, said selection being implemented in accordance with a performance criterion of the energy exchange.

[0055] Such an entity is in particular capable of implementing the aforementioned energy exchange method.

[0056] The invention also relates to a computer program comprising instructions for implementing the energy exchange method according to the invention, according to any one of the particular embodiments described previously, when said program is executed by a processor.

[0057] Such instructions can be stored permanently in a non-transitory memory medium of an entity or a sub-entity implementing the energy exchange method according to the invention.

[0058] This program may use any programming language, and may be in the form of source code, object code, or code intermediate between source code and object code, such as in a partially compiled form, or in any other desirable form.

[0059] The invention also relates to a recording medium or information medium readable by a computer, and comprising instructions of a computer program as mentioned above. The recording medium can be any entity or device capable of storing the program. For example, the medium can comprise a storage means, such as a ROM (Read Only Memory), for example a CD ROM (Compact Disc Read-Only Memory) or a microelectronic circuit ROM, or a magnetic recording means, for example a mobile medium, a hard disk or an SSD (Solid State Drive). Furthermore, the recording medium can be a transmissible medium such as an electrical or optical signal, which can be conveyed via an electrical or optical cable, by radio or by other means, so that the computer program it contains is remotely executable.The program according to the invention may in particular be downloaded over a network, for example an Internet-type network. Alternatively, the recording medium may be an integrated circuit in which the program is incorporated, the circuit being adapted to execute or to be used in the execution of the aforementioned energy exchange method.

[0060] According to an exemplary embodiment, the present technique is implemented by means of software and / or hardware components. In this regard, the term "module" can correspond in this document to a software component, a hardware component or a set of hardware and software components.

[0061] Brief description of the drawings

[0062] Other characteristics and advantages will appear on reading particular embodiments of the invention, given as illustrative and non-limiting examples, and the appended drawings, among which:

[0063] [Fig. 1] represents an example of architecture in which the energy exchange method according to the invention is implemented, according to a particular embodiment, [Fig. 2] represents the structure of a set of entities implementing the energy exchange method, in a particular embodiment of the invention, [Fig. 3] represents an example of architecture of an entity implementing the energy exchange method, according to a first particular embodiment of the invention,

[0064] [Fig. 4] represents the main steps implemented by an entity of the set of entities, in the energy exchange method according to the invention, according to a first particular embodiment of the invention, [Fig. 5] represents the main actions implemented by an energy production sub-entity of the set of entities, during the implementation of the energy exchange method, according to a first particular embodiment of the invention, [Fig. 6] represents the main actions implemented by an energy storage sub-entity of the set of entities, during the implementation of the energy exchange method, according to a first particular embodiment of the invention, [Fig. 7] represents the main actions implemented by an electric vehicle sub-entity of the set of entities, during the implementation of the energy exchange method, according to a first particular embodiment of the invention, [Fig.8] represents the main actions implemented by an energy consumption sub-entity of the set of entities during the implementation of the energy exchange method, according to a first particular embodiment of the invention.

[0065] [Fig. 9A] represents the main actions implemented by an energy consuming device associated with the energy consuming sub-entity of the set of entities during the implementation of the energy exchange method, according to a first particular embodiment of the invention, the operation of the energy consuming device being shiftable over time, [Fig. 9B] represents the main actions implemented by an energy consuming device associated with the energy consuming sub-entity of the set of entities during the implementation of the energy exchange method, according to a first particular embodiment of the invention, the power level of the energy consuming device being variable over time,

[0066] [Fig. 10] represents an example of architecture of an entity or a sub-entity implementing the energy exchange method, according to a second particular embodiment of the invention,

[0067] [Fig. 1 1] represents the main steps implemented by an entity of the set of entities, in the energy exchange method according to the invention, according to a second particular embodiment of the invention,

[0068] [Fig. 12] represents the main actions implemented by an energy production sub-entity of the set of entities, during the implementation of the energy exchange method, according to a second particular embodiment of the invention, [Fig. 13] represents the main actions implemented by an energy storage sub-entity of the set of entities, during the implementation of the energy exchange method, according to a second particular embodiment of the invention,

[0069] [Fig. 14] represents the main actions implemented by an electric vehicle sub-entity of the set of entities, during the implementation of the energy exchange method, according to a second particular embodiment of the invention,

[0070] [Fig. 15] represents the main actions implemented by an energy consuming device associated with the energy consuming sub-entity of the set of entities during the implementation of the energy exchange method, according to a second particular embodiment of the invention, the operation of the energy consuming device being shiftable in time,

[0071] [Fig. 16] represents the main actions implemented by an energy consuming device associated with the energy consuming sub-entity of the set of entities during the implementation of the energy exchange method, according to a second particular embodiment of the invention, the power level of the energy consuming device being variable over time.

[0072] Detailed description of embodiments of the invention

[0073] With reference to Figure 1, an example of architecture is described in which the energy exchange method according to the invention is implemented. According to this architecture, an energy exchange system or Xplace energy exchange place comprises:

[0074] - a set E of at least two entities ENTi and ENT2 configured to exchange energy between them, these at least two entities being located in the set E at a first higher hierarchical level L1,

[0075] - possibly one or more energy suppliers FE, external to said set of entities E, which is / are configured to exchange energy with the entity ENT1 and / or the entity ENT2.

[0076] In this energy exchange place, in the case where an entity produces renewable energy, the climatic conditions CC are also taken into account and are external conditions to the set E. The energy exchanged can be of different types: electricity, gas, fuel, heat, etc.

[0077] According to the invention:

[0078] - at least one sub-entity S_ENTi is attached to the entity ENTi at a second hierarchical level L2 lower than level L1,

[0079] - at least one sub-entity S_ENÏ2 is attached to the entity ENT2 at the second hierarchical level L2.

[0080] According to the invention:

[0081] - at least one SS_ENTi sub-entity can be attached to the S_ENTi sub-entity at a third hierarchical level L3 lower than level L2,

[0082] - at least one sub-entity SS_ENÏ2 can be attached to the sub-entity S_ENÏ2 at the third hierarchical level L3.

[0083] The third level L3 is optional. For this reason, the sub-entities SS_ENTi and SS_ENÏ2 are represented in dotted lines in Figure 1.

[0084] Alternatively, the sub-entities SS_ENTi and SS_ENÏ2 could be located on the second hierarchical level L2.

[0085] Each of the entities ENT1 and ENT2 is advantageously configured according to at least one energy use profile among a first energy production profile PF1, a second energy consumption profile PF2 and a third energy storage profile PF3 or any possible combination of these profiles.

[0086] An entity according to the invention implements an energy transaction in the energy exchange system or place of Figure 1. For this purpose, an entity can be a professional building (factory for example) or a residential building (building for example). An entity can also be a boat, a base station in a communication network. The set E can contain entities of the same nature, such as for example a residential area composed of “buildings” entities, a fishing port composed of “boats” entities, a communication network composed of “base stations” entities. Alternatively, the set E can contain entities of a different nature. Thus, for example, the set E could be a port district of a city which would contain “houses”, “buildings”, “boats”, and / or “base stations” entities.

[0087] An embodiment of the set E is shown in Figure 2, for illustrative purposes only, in which there are “home” and “base station” entities, at a higher hierarchical level L1. In the example shown, fourteen ENTi to ENTu entities are represented, including:

[0088] - the entities ENTi to ENT?, ENTg, ENT10, ENT12 and ENT13 are houses or buildings,

[0089] - the entities ENTs, ENTu and ENTu are base stations.

[0090] Of course, this number of entities may be less or more than 14 depending on the context of the energy transaction to be implemented.

[0091] Each of the entities represented is associated with one or more energy usage profiles. Depending on the usage profile considered, one or more sub-entities are attached to a considered entity, at a hierarchical level L2 or L3 lower than the hierarchical level L1.

[0092] In the example shown in Figure 2:

[0093] - the ENTi entity has a PF2 energy consumption profile: it is thus associated with at least one S_ENTi sub-entity (not shown) which consumes energy, such as for example a refrigerator, one or more radiators, a washing machine, a lifting pump, etc.;

[0094] - the ENT2 entity has a PF1 energy production profile: it is thus associated with at least one S_ENT2 sub-entity which produces energy, for example a photovoltaic panel;

[0095] - the entity ENT3 has the two profiles PF1 and PF2: it is thus associated with at least one sub-entity S_ENTsi (not shown) which consumes energy, of the aforementioned type, as well as at least one sub-entity S_ENT32 which produces energy, for example a photovoltaic panel;

[0096] - the entity ENT4 has the profile PF2 and a profile PF4 for using an electric vehicle, i.e. both an energy storage profile and an energy consumption profile: it is thus associated with at least one sub-entity S_ENT4i (not shown) which consumes energy, of the aforementioned type, and with a sub-entity S_ENT42 which is the electric vehicle;

[0097] - the entity ENTs has the profile PF1 and a profile PF3 for energy storage: it is thus associated with at least one sub-entity S_ENTsi which produces energy, for example a photovoltaic panel, as well as at least one sub-entity S_ENTs2 which stores energy, for example a battery; - the entity ENTe has the three profiles PF1, PF2 and PF4: it is thus associated with at least one sub-entity S_ENTei (not shown) which consumes energy, of the aforementioned type, as well as at least one sub-entity S_ENÏ62 which produces energy, for example a photovoltaic panel and a sub-entity S_ENÏ63 which is an electric vehicle;

[0098] - the entity ENT? has an energy consumption profile PF2: it is thus associated with at least one sub-entity S_ENT?i (not shown) which consumes energy, of the aforementioned type, as well as at least one sub-entity S_ENT?2 which stores energy, for example a battery;

[0099] - the ENTs entity has a PF2 energy consumption profile, both when the base station transmits or receives a signal: it is thus associated with at least one S_ENTsi sub-entity (not shown) which consumes energy;

[0100] - the entity ENT9 has the three profiles PF1, PF2 and PF3: it is thus associated with at least one sub-entity S_ENTgi (not shown) which consumes energy, of the aforementioned type, as well as at least one sub-entity S_ENÏ92 which produces energy, for example a photovoltaic panel and a sub-entity S_ENÏ93 which stores energy, for example a battery;

[0101] - the entity ENT10 has the profile PF3 and is therefore associated with a sub-entity S_ENT i which stores energy, for example a battery;

[0102] - the entity ENT11 has the profiles PF2 and PF3: it is thus associated with at least one sub-entity S_ENTm (not shown) which consumes energy, of the aforementioned type, as well as at least one sub-entity S_ENTn2 which stores energy, for example a battery;

[0103] - the entity ENT12 has the four profiles PF1 to PF4: it is thus associated with at least one sub-entity S_ENTi2i (not shown) which consumes energy, as well as at least one sub-entity S_ENTi22 which produces energy, for example a photovoltaic panel, with a sub-entity S_ENTi23 which stores energy, for example a battery and with a sub-entity S_ENTi24 which is an electric vehicle;

[0104] - the entity ENT13 has the three profiles PF2, PF3, PF4: it is thus associated with at least one sub-entity S_ENTm (not shown) which consumes energy, of the aforementioned type, as well as at least one sub-entity S_ENTi32 which stores energy, for example a battery, and a sub-entity S_ENTi33 which is an electric vehicle; - the entity ENTu has the two profiles PF1 and PF3: it is thus associated with at least one sub-entity S_ENT i which produces energy, for example a photovoltaic panel, as well as at least one sub-entity S_ENT 2 which stores energy, for example a battery.

[0105] Of course, other configurations are possible and depend on the context of the energy transaction to be implemented. In addition, in other examples:

[0106] - an energy production sub-entity could be a wind turbine, a methanization device, etc.,

[0107] - an energy storage sub-entity could be an inertial storage device, a compressed air storage device, a methanation device, etc.,

[0108] - an electric vehicle may include an electric car and / or an electric boat and / or an electric scooter, etc.

[0109] According to the invention, the set E is decomposed in such a way that each of the sub-entities described above is located at a hierarchical level L2 which is lower than the level L1. For the sake of clarity in Figure 2, such a decomposition is only represented for the entity ENT12 for which the sub-entities S_ENTi2i to S_ENTi24 are attached to the entity ENT12 at the lower hierarchical level L2.

[0110] In the example shown, an optional third hierarchical level L3 is represented. Level L3 includes:

[0111] - at least one SS_ENTi2io sub-entity attached to the S_ENTi2i sub-entity and comprising at least one device whose use can be shifted over time, for example an electric radiator, a washing machine, etc., and / or

[0112] - at least one SS_ENTi2n sub-entity attached to the S_ENTi2i sub-entity and comprising at least one device whose operating power level is variable, for example a lighting device, a transmitter of a base station, etc.

[0113] According to the invention, the energy transaction is implemented at the higher hierarchical level L1, by at least two entities ENT1 and ENT2 which compose it. During this transaction, when one of the at least two entities, for example ENT1, exchanges energy with the other entity ENT2 or an energy supplier FE, this action is carried out according to an energy exchange performance criterion. According to a first embodiment, such a criterion is a compromise:

[0114] - reduction of energy expenditure linked to the purchase of energy from the ENT2 entity or the FE energy supplier, and

[0115] - maximizing revenues linked to the sale of energy to the ENT2 entity or to the FE energy supplier.

[0116] According to a second embodiment, such a criterion is a compromise:

[0117] - reduction of energy expenditure linked to the purchase of energy from the ENT2 entity or the FE energy supplier,

[0118] - maximizing revenues linked to the sale of energy to the ENT2 entity or to the FE energy supplier, and

[0119] - minimizing the carbon footprint linked to the energy transaction. According to the invention, at the median hierarchical level L2, the energy distribution is implemented at the level of each of the sub-entities attached to the entity ENT1, respectively ENT2, such energy distribution being controlled by the entity ENT1, respectively ENT2. For this purpose,

[0120] - in the case where, for example, entity ENT1 has a PF2 energy consumption profile, a decision is taken by entity ENT1 as to whether it is better, according to an energy exchange performance criterion, to consume energy from one or more storage sub-entities attached to entity ENT1, if present, and / or from one or more “electric vehicle” sub-entities attached to entity ENT1, if present, and / or

[0121] - in the case where, for example, entity ENT1 has a PF1 energy production profile, a decision is taken by entity ENT1 as to whether it is better, according to an energy exchange performance criterion, to produce energy by the energy production sub-entity (photovoltaic panel, wind turbine, etc.) attached to entity ENT1, and / or

[0122] - whether it is better, according to an energy exchange performance criterion, to require energy (purchase) from an external energy supplier or another entity in the set E of entities, for example ENT2, to supply the energy consumption sub-entity(ies) attached to the entity ENT1, if present, and / or store this purchased energy in the storage sub-entity(ies) attached to the entity ENT1, if present, and / or charge the “electric vehicle” sub-entity(ies) attached to the entity ENT1, if present. According to the invention, at the lower hierarchical level L3, a control of the use of energy is implemented by one or more energy consumption sub-entities, for example the sub-entity S_ENTi2i, at the level of each of the sub-entities attached to the sub-entity S_ENTi2i, such as the sub-entities SS_ENTi2io and SS_ENTi2ii.To this end, it is at this level that any energy consumption sub-entity controls the "demand-response" functionality which allows decisions to be made on how and / or when to use the sub-entity(s) (devices which consume energy) according to their type, namely whose use is shiftable over time and whose operating power level is variable.

[0123] According to the invention, such control is implemented by the energy consumption sub-entity while respecting a performance criterion for the use of the consumed energy which expresses, for example, the minimization of the dissatisfaction of the user of the energy-consuming devices which are present.

[0124] As already explained above, the hierarchical level L3 is optional, the control of energy use can be implemented at the level of the energy consumption sub-entity located at the hierarchical level L2 which then integrates the sub-entities SS_ENTi2io and SS_ENTi2ii.

[0125] Description of a first embodiment of an entity capable of exchanging energy

[0126] Figure 3 shows the simplified structure of an ENTi entity selected from the plurality of ENTi to ENTNG entities of Figure 2, such that 1 <i<Ne, où Ne représente le nombre d’entités dans la place d’échange. Dans l’exemple représenté sur la figure 2, Ne=14. Une telle entité ENTi est configurée pour mettre en oeuvre le procédé d’échange d’énergie qui va être décrit ci-dessous.

[0127] The ENTi entity comprises according to the invention:

[0128] - a COM communication module adapted to communicate with at least one other entity ENTj of the set E of entities, such as 1 <j<Ne, ou avec au moins un fournisseur d’énergie FE, via un réseau de communication de données (non représenté) qui peut être de type sans fil courte ou moyenne portée, tel que par exemple Bluetooth, NFC, LTE, WiFi, DSRC, C-V2X, etc., de type sans fil longue portée, tel que par exemple 2G, 3G, 4G, 5G, etc., de type filaire tel que ADSL, fibre, etc.., l’objet de la communication pouvant être une requête en énergie auprès de l’entité ENTj ou du fournisseur d’énergie FE ou une proposition de fourniture d’énergie à l’entité ENTj ou au fournisseur d’énergie FE,

[0129] - an energy delivery point (meter, wiring) PTL to receive the energy supplied by the entity ENTj or the energy supplier FE, via an energy distribution network not shown in figure 3, and / or

[0130] - an energy supply point (meter, cabling) PTF to supply energy to the entity ENTj or to the energy supplier FE, via the aforementioned energy distribution network,

[0131] - at least one S_ENTi sub-entity for energy consumption, energy production, energy storage, at least one electric vehicle,

[0132] - possibly at least one SS_ENTi sub-entity such as for example a device whose use is shiftable over time, a device whose operating power level is variable, a device whose use is not shiftable over time and whose operating power level is fixed,

[0133] - a REC module for receiving information relating to the energy context in a current time interval.

[0134] Because the SS_ENTi sub-entity is optional, it is shown in dotted lines in Figure 3.

[0135] According to a particular embodiment of the invention, the actions executed by the entity ENTi, within the framework of the implementation of the energy exchange method in accordance with the present invention, are implemented by instructions of a computer program PG. For this, the entity ENTi comprises a conventional architecture of a computer and notably comprises a memory MEM, a processing unit UTR, equipped for example with a processor PROC, and controlled by the computer program PG stored in memory MEM. The computer program PG comprises instructions for implementing the actions executed by the entity ENTi, when the program is executed by the processor PROC, according to any one of the particular embodiments of the invention.

[0136] At initialization, the code instructions of the computer program PG are for example loaded into a RAM memory (not shown) before being executed by the processor PROC. The processor PROC of the processing unit UTR implements in particular the communication actions via the COM module, the information reception actions via the REC module, the energy supply actions via the energy supply point PTF, the energy request actions via the energy delivery point PTL.

[0137] Description of a first embodiment of an energy exchange method

[0138] We now describe, in relation to figure 4, the progress of an energy exchange process executed by the ENTi entity as illustrated in figure 3.

[0139] Such an energy exchange process takes place as follows at the level of the entity ENTi, in a current time interval IT C .

[0140] In S1, the entity ENTi receives, via the reception module REC in figure 3:

[0141] - from at least one energy production sub-entity S_ENTii (wind turbine, photovoltaic panel, etc.) associated with said at least one entity, if such a sub-entity is present, the quantity QPii, c of energy produced in said current time interval, by said at least one sub-entity S_ENTM,

[0142] - from at least one energy consumption sub-entity S_ENTi2 (radiator, washing machine, lifting pump, etc.) associated with said at least one entity, if such a sub-entity is present, the quantity QCi2,c of energy consumed in said current time interval, by said at least one sub-entity S_ENTi2,

[0143] - from at least one energy storage sub-entity S_ENTi3 (battery for example), associated with said at least one entity, if such a sub-entity is present, the quantity QSi3,c of energy stored in said current time interval, by said at least one sub-entity S_ENTi3,

[0144] - from at least one sub-entity S_ENTi4 electric vehicle, associated with said at least one entity, if such a sub-entity is present, the quantity QSi4,c of energy stored in said current time interval, by said at least one sub-entity S_ENTi4.

[0145] During step S1, a concatenation of this information is implemented in S10 if necessary, so as to obtain a value Qi, c of quantity of energy required by the entity ENTi in the current time interval, with Qi, c = QPh.c + QSi3,c + QSi4,c - QCi2,c (1).

[0146] According to a preferred embodiment, the entity ENTi further receives in S1, via the reception module REC of figure 3:

[0147] - from said at least one energy production sub-entity S_ENTM (wind turbine, photovoltaic panel, etc.) associated with said at least one entity, if such a sub-entity is present, a value of the carbon footprint CarbPii, c which is linked to the quantity of energy to be produced QPri.c, in said current time interval, by said at least one sub-entity S_ENTii,

[0148] - from said at least one energy storage sub-entity S_ENTi3 (battery for example), associated with said at least one entity, if such a sub-entity is present, a value of the carbon footprint CarbSi3,c which is linked to the quantity of energy to be stored QSi3,c, in said current time interval, by said at least one sub-entity S_ENTi3,

[0149] - from said at least one sub-entity S_ENTi4 electric vehicle, associated with said at least one entity, if such a sub-entity is present, a value of the carbon footprint CarbSi4,c which is linked to the quantity of energy to be stored QSi4,c in said current time interval, by said at least one sub-entity S_ENT i4 .

[0150] In step S1, a concatenation of the CarbPii values, c , CarbSi3,c, CarbSi4,c is implemented in S10 if necessary, in order to obtain a carbon footprint value Carbi, c , such as Carbi, c = CarbPii, c + CarbSi3,c + CarbSi4,c (2). According to the invention, the carbon footprint value CarbPii, c is not taken into account in the rest of the energy exchange process, given that it is close to 0 because it is linked to the production of clean, non-polluting energy.

[0151] In S1, the entity ENTi also receives, via the reception module REC in figure 3:

[0152] - from at least one other ENTj entity, a value of the price Pr c ,x p the cost of the energy available in the said current time interval from the other entity ENTj,

[0153] - from at least one FE energy supplier, if available, a value of the Pr price c , g rid that the energy available in the said current time interval costs from the energy supplier FE.

[0154] The Pr Prize c ,gnd of the supplier FE is a fixed price which generally varies according to the season or the time of day (peak hours, off-peak hours). According to the invention, the price Pr c ,x p iace is set prior to the implementation of the energy exchange process. It is a fixed price determined, for example, as being lower than that of the FE supplier. The price Prc ,x p iace can for example be set as being equal to a fraction or a percentage of the price Pr c ,gnd of the supplier. According to a more elaborate strategy, the price Pr c ,x p iace is based on auction theory.

[0155] In S2, from the Qi values, c , Pr c ,x p iace, Pr c ,gnd and Carbi, c , a type of energy transaction is then selected to be implemented in the current time interval.

[0156] The said selection can also be implemented by combining the Qi values, c , Pr c ,x p iace, Pr c ,grid and Carbi, c with respectively values ​​of the same type obtained by learning in previous time intervals, for example using a supervised learning algorithm.

[0157] The selection S2 is implemented in accordance with a first criterion R1 i, c performance of energy exchange in the current time interval.

[0158] In an exemplary embodiment, the criterion R1 i, c minimizes the price of the energy required by the entity ENTi in said current time interval and maximizes the profit from the supply of energy by the entity ENTi to the other entity ENTj or to the external energy supplier FE in said current time interval.

[0159] According to an example, the criterion R1 i, c is expressed in the form Rl iiC = -Pr Ci g rid .Q iiaiC (3) when the selected energy transaction is for example the purchase of energy from the external energy supplier FE or in the form Rl iiC = Pr CiXpiace .Qt iViC(4) when the selected energy transaction is for example the sale of energy to the other entity ENTj or to the external energy supplier FE, with Q i a c and Q iiViC representing the quantities of energy purchased and sold respectively.

[0160] In the preferred embodiment, such selection is implemented in accordance with a second criterion R2i, c performance of energy exchange in the current time interval.

[0161] In an exemplary embodiment, the criterion R2i, c :

[0162] - minimizes the price of the energy required by the entity ENTi in the said current time interval,

[0163] - maximizes the profit from the supply of energy by the entity ENTi to the other entity ENTj or to the external energy supplier FE in the said current time interval,

[0164] - minimizes the Carbi carbon footprint, c .

[0165] The R2i criterion, c is expressed in the form when the selected energy transaction is the purchase of energy from the external energy supplier FE, where a c and (3 C represent weighting coefficients between the cost Pr c ,energy gnd and carbon footprint Carbi, c The value of these weighting coefficients is adjusted by the user, for example via a user interface, and represents preferences of this user.

[0166] The R2i criterion, c is expressed in the form R2 i c = a'c. Pr CiXpiace .Q iiiViC - c.Carb i c (6) when the selected energy transaction is the sale of energy to the other entity ENTj or to the external energy supplier FE, where a'c and [3c represent weighting coefficients between the profit linked to the sale of energy and the carbon footprint Carbi, cThe value of these weighting coefficients is adjusted by the user, for example via a user interface, and represents preferences of this user.

[0167] At the end of said S2 selection, the ENTi entity decides:

[0168] - not to implement any particular action in the following time interval,

[0169] - to implement in the following time interval an “energy purchase” transaction in S3a, i.e. to request energy from the other entity ENTj at the price Pr c ,x p iace and / or the energy supplier FE at the price Pr c ,gnd,

[0170] - to implement in the following time interval an “energy sale” transaction in S3b, i.e. to supply energy to the other entity ENTj and / or to the energy supplier FE at the price Pr c ,x piace or another previously defined price. If the “energy purchase” transaction is implemented in S3a, the entity ENTi determines in S4a the quantity of energy Qreq c to be requested in the current time interval, the energy source Sreq c (FE supplier or other ENTj entity) where to request the quantity of energy Qreq c , as well as the Dreq destination c of quantity of energy Qreq c , that is to say the energy storage sub-entity S_ENTi3, possibly one or more sub-entities associated with the sub-entity S_ENTi2, such as for example a device whose use can be shifted over time, a device whose operating power level is variable, a device whose use cannot be shifted over time and whose operating power level is fixed, as well as the sub-entity S_ENTi4 electric vehicle.

[0171] If the “energy sale” transaction is implemented in S3b, the ENTi entity determines in S4b the quantity of energy Qpro c to be supplied in the current time interval, the energy source Spro c (S_ENTii, S_ENTi3, S_ENTi4) from which the quantity of energy Qpro comes c , as well as possibly the Dpro destination c of the quantity of energy Qpro c to produce, the ENTj entity or the FE supplier in the example shown.

[0172] Thanks to the energy exchange method which has just been described above, the invention advantageously makes it possible to propose an energy exchange location which is more stable, less vulnerable to breakdowns, and energy-efficient.

[0173] We now describe, in relation to figure 5, the different steps carried out at the level of said at least one energy production sub-entity S_ENTii (wind turbine, photovoltaic panel, etc.) which is associated with said at least one entity ENTi, when the energy exchange method is implemented in the current time interval IT C and when such a sub-entity is present.

[0174] In S1 ii, the sub-entity S_ENTii receives, via a reception module identical or similar to that of figure 3:

[0175] - from said at least one energy storage sub-entity S_ENTi3 (battery for example), associated with said at least one entity ENTi, if such a sub-entity S_ENTi3 is present, the quantity of energy QSi3,c stored by the sub-entity S_ENTi3, in said current time interval,

[0176] - from said at least one electric vehicle S_ENTi4 sub-entity, associated with said at least one ENTi entity, if such a S_ENTi4 sub-entity is present, the quantity of energy QSi4,c stored by the S_ENTi4 sub-entity, in said current time interval,

[0177] - from the ENTi entity, the value of the price Pr c ,explains the cost of the energy available in the said current time interval from the other entity ENTj and possibly the value of the price Pr c ,gnd that the energy available in the said current time interval costs from the energy supplier FE.

[0178] In S2ii, from the QPii values, c , QCi2,c, QSi3,c, QSi4,c, Pr c ,xpiace and possibly Pr c ,grid, a type of action is then selected to be implemented in the current time interval by the energy production sub-entity S_ENTii.

[0179] Said selection can also be implemented by combining the QPii values, c , QCi2,c, QSi3,c, QSi4,c, Pr c ,x p iace and possibly Pr c ,gnd with respectively values ​​of the same type obtained by learning in previous time intervals, for example using a supervised learning algorithm. Such a selection S2n is implemented according to a criterion Rii, c performance of the use of the energy produced in the current time interval. In an exemplary implementation, the criterion Rii, c maximizes the profit from supplying the energy produced to the other entity ENTj or to the external energy supplier FE in said current time interval, in the case where the quantity of energy produced by the sub-entity S_ENTii in the current time interval is sold to the other entity ENTj or to the external energy supplier FE.

[0180] The Rii criterion, cis expressed for example in the form Ri1,c=Pr Ci x p iace -QPii,v,c - ( 7 )> °ù QPu, v ,c represents the quantity of energy produced and sold.

[0181] At the end of said selection S2n, the energy production sub-entity S_ENTii decides:

[0182] - in S3aii, to select as recipient of all or part of the energy produced, the other entity ENTj and / or the energy supplier FE,

[0183] - in S3bii, to recharge the storage sub-entity S_ENTi3 with all or part of the energy produced,

[0184] - in S3cii, to recharge the sub-entity S_ENTi4 with all or part of the energy produced,

[0185] - in S3dii, to supply one or more sub-entities associated with the energy consumption sub-entity S_ENTi2, such as for example a device whose use is shiftable over time, a device whose operating power level is variable, a device whose use is not shiftable over time and whose operating power level is fixed.

[0186] In the case where decision S3aii is implemented, the energy production sub-entity S_ENTii calculates in S4aii the quantity QPri.c of energy produced with a view to the sale of this quantity of energy by the entity ENTi in the time interval following ITc+i to the other entity ENTj or to the external energy supplier FE.

[0187] In the case where decision S3bii is implemented, the energy production sub-entity S_ENTii calculates in S4bii the quantity QPri.c of energy produced to be used for recharging the energy storage sub-entity S_ENTi3 in the following time interval.

[0188] In the case where decision S3cii is implemented, the energy production sub-entity S_ENTii calculates in S4cii the quantity QPii, cof energy produced to be used for recharging the electric vehicle sub-entity S_ENTi4 in the following time interval. In the case where decision S3dii is implemented, the energy production sub-entity S_ENTii calculates in S4dii the quantity QPri.c of energy produced to be used to supply, in the following time interval, at least one sub-entity SS_ENTi2i associated with the energy consumption sub-entity S_ENTi2, such as for example a device whose use is shiftable over time, a device whose operating power level is variable, a device whose use is not shiftable over time and whose operating power level is fixed. In S5ii, the quantity QPii, cof calculated produced energy, associated with the selected action, is then transmitted to the entity ENTi, in the following time interval. We now describe, in relation to Figure 6, the different steps carried out at the level of said at least one energy storage sub-entity S_ENTi3 (battery, accumulator, etc.) which is associated with said at least one entity ENTi, when the energy exchange method is implemented in the current time interval IT C and when such a sub-entity is present.

[0189] In S1 i3, the sub-entity S_ENTi3 receives from the entity ENTi, via a reception module identical or similar to that of figure 3, the value of the price Pr c ,gnd that the energy available in the said current time interval costs from the energy supplier FE, possibly the value of the price Pr c ,xpiace.

[0190] In S2i3, from the value of Pr c , gnd and / or Pr c ,xpiace, of the value of the quantity of energy produced QPii, c by the sub-entity S_ENTii, of the value of the quantity of energy consumed QCi2,c by the sub-entity S_ENTi2, of the value of the quantity of energy QSi3,c stored by the sub-entity S_ENTi3, of the value of the quantity of energy QSi4,c stored by the sub-entity S_ENTi4, and possibly of the carbon footprint Carbi3,c of the energy stored by the storage sub-entity S_ENTi3, in the current time interval IT C , a type of action is then selected to be implemented in the current time interval by the energy storage sub-entity S_ENTi3.

[0191] Said selection can also be implemented by combining the aforementioned values ​​with respectively values ​​of the same type obtained by learning in previous time intervals, for example using a supervised learning algorithm.

[0192] Such a selection S2i3 is implemented in accordance with a criterion Ri3,c of performance of the use of the stored energy in the current time interval. In an exemplary embodiment, the criterion Ri3,c maximizes the profit of the supply of the stored energy to the other entity ENTj or to the external energy supplier FE in said current time interval, in the case where the quantity of energy stored by the sub-entity S_ENTi3 in the current time interval is sold to the other entity ENTj or to the external energy supplier FE.

[0193] The criterion Ri3,c is expressed in the form Ri3,c=Pr c ,xpiace-QSi3, v ,c (8), where QS i3iViC represents the amount of stored energy that is sold.

[0194] In another example of implementation, more particularly in the case where the supplier's selling price is different from that of the exchange, the criterion Ri3,c is expressed in the form R i3 ,c=Pr c,grid-QS i3 , v , c (9).

[0195] In a more complex implementation example, the criterion Ri3,c minimizes the cost of purchasing energy, from the energy supplier or another entity of the exchange, which should be stored in the sub-entity ENTi3 in said current time interval.

[0196] The criterion Ri3,c is then expressed in the form R i3iC = -Pr CiXpiace .QS i3 a c (10) or R i3iC = - P r c,grid-Q s i3,a,c (1 1 ), where QS i3 a c represents the amount of energy purchased to be stored.

[0197] Alternatively, the criterion Ri3,c used is a compromise between maximizing the profit of supplying the stored energy to the other entity ENTj or to the external energy supplier FE in said current time interval and minimizing the dissatisfaction of the user of the storage sub-entity S_ENTi3 in said current time interval. Such dissatisfaction is based on the user's fear that the storage sub-entity S_ENTi3 does not have a sufficient charge level to meet the user's needs, in the current time interval.

[0198] The criterion Ri3,c is expressed in the form

[0199] • ai3 and [3i3 are weighting coefficients between the profit linked to the sale of energy and user dissatisfaction,

[0200] • Pi3-(Ei3,max - £i3,c) 2is a factor determining the anxiety of the user of the sub-entity S_ENTi3 of not having enough energy to use this sub-entity, in which E i3 max is the maximum energy consumption of this sub-entity S_ENTi3. In another exemplary embodiment, the criterion Ri3,c used minimizes the dissatisfaction of the user of the sub-entity S_ENTi3 in the said current time interval. It is then expressed in the form

[0201] At the end of said selection S2i3, the energy storage sub-entity S_ENTi3 makes a decision from three decisions D1, D2, D3. Decision D1 relates to the choice of the energy source for recharging the storage sub-entity S_ENTi3. Decision D2 relates to the choice of the destination of the energy discharged from the storage sub-entity S_ENTi3. Decision D3 is not to implement any particular action in the following time interval.

[0202] If decision D1 was selected as the action to be carried out, the storage sub-entity S_ENTi3:

[0203] - either selects in S3ai3, as the source of recharging of the sub-entity S_ENTi3, the other entity ENTj and / or the energy supplier FE, in the following time interval,

[0204] - either selects in S3bi3, as the source of recharging of the sub-entity S_ENTi3, the energy production sub-entity S_ENTii, in the following time interval.

[0205] If decision D2 was selected as the action to be carried out in the following time interval, the storage sub-entity S_ENTi3:

[0206] - either selects in S3ci3, as recipient of all or part of the quantity QSi3,c discharged from the sub-entity S_ENTi3, the other entity ENTj and / or the energy supplier FE,

[0207] - either selects in S3di3, as recipient of all or part of the quantity QSi3,c discharged from the sub-entity S_ENTi3, the energy consumption sub-entity S_ENTi2, such as for example a device whose use is shiftable over time, a device whose operating power level is variable, a device whose use is not shiftable over time and whose operating power level is fixed.

[0208] In the case where the selection S3ai3 is implemented, the energy storage sub-entity S_ENTi3 calculates in S4ai3 the quantity of energy Qi3,c to be received from the entity ENTj or from the energy supplier FE for its recharging in the current time interval. In the case where the selection S3bi3 is implemented, the energy storage sub-entity S_ENTi3 calculates in S4bi3 the quantity of energy Qi3,c to be received from the energy production sub-entity S_ENTii for its recharging in the current time interval.

[0209] In the case where the selection S3ci3 is implemented, the energy storage sub-entity S_ENTi3 calculates in S4ci3 the quantity of energy Qi3,c to be used to supply energy to the other entity ENTj and / or the energy supplier FE.

[0210] In the case where the selection S3di3 is implemented, the energy storage sub-entity S_ENTi3 calculates in S4di3 the quantity of energy Qi3,c of energy to be used to supply energy to said at least one energy consumption sub-entity S_ENT i2 .

[0211] In S5i3, the calculated quantity of energy Qi3,c is then transmitted to the entity ENTi, in the current time interval.

[0212] We now describe, in relation to figure 7, the different steps carried out at the level of said at least one sub-entity S_ENTi4 electric vehicle which is associated with said at least one entity ENTi, when the energy exchange method is implemented in the current time interval IT C and when such a sub-entity is present.

[0213] In S1 i4, the sub-entity S_ENTi4 receives from the entity ENTi, via a reception module identical or similar to that of figure 3, the value of the price Prc ,gnd that the energy available, in the said current time interval, costs from the energy supplier FE and / or the value of the price Pr c ,expiace that the energy available on the exchange place costs, in the said current time interval.

[0214] In S2i4, from the value of Pr c ,gnd and / or Pr c ,xpiace, QSi4,c, QSi3,c, QPii,c, QCi2,c in said current time interval, of the energy Ei4,c consumed by the sub-entity S_ENTi4, in said current time interval, and possibly of the carbon footprint Carbi4,c of the energy stored by the sub-entity S_ENTi4 in the current time interval, a type of action is then selected to be implemented in the current time interval by the sub-entity S_ENTi4 electric vehicle.

[0215] Said selection can also be implemented by combining the aforementioned values, with respectively values ​​of the same type obtained by learning in previous time intervals, for example using a supervised learning algorithm. Such a selection S2i4 is implemented in accordance with a criterion Ri4,c of performance of the use of the energy stored by said sub-entity S_ENTi4, in the current time interval.

[0216] In an exemplary embodiment, the criterion F .c maximizes the profit of supplying the stored energy to the other entity ENTj or to the external energy supplier FE in said current time interval.

[0217] The criterion Ri4,c is expressed in the form Ri4,c=Pr Ci x p iace -Q s i4,v,c (15) or Ri4 :C =Pr c gri(i .QS i4 v c (16), where QS i4 v crepresents the stored quantity that is sold. In a more complex implementation example, the criterion Ri4,c minimizes the cost of purchasing energy, from the energy supplier or another entity in the exchange, which should be stored in the sub-entity ENTi4 in the following time interval.

[0218] The criterion Ri4,c is then expressed in the form R i4iC = -Pr CiXpia this-Q s i4,a,c U 7 ) or R i4iC = - P r c,grid-Q s i4,a,c (18), where QS i4 a c represents the amount of energy purchased to be stored.

[0219] Alternatively, the criterion Ri4,c used is a compromise between maximizing the profit of providing the stored energy to the other entity ENTj or to the external energy supplier FE in said current time interval and minimizing the dissatisfaction of the user of the sub-entity S_ENTi4 electric vehicle in said current time interval. Such dissatisfaction is based on the user's fear that the sub-entity S_ENTi4 electric vehicle does not have enough energy to operate in the current time interval.

[0220] The criterion Ri4,cs is expressed in the form ~ Pi4- (, 'i4,max Pi4,c

[0221] • ai4 and [3i4 are weighting coefficients between the profit linked to the sale of energy and user dissatisfaction,

[0222] • factor determining the anxiety of the user of the sub-entity S_ENTi4 electric vehicle of not having enough energy to use his electric vehicle, in which E i4 max is the maximum energy consumption of this sub-entity S_ENTi4 and Ei4, c is its energy consumption during the current time interval. In another exemplary embodiment, the criterion F .c used minimizes the dissatisfaction of the user of the sub-entity S_ENTi4 electric vehicle in said current time interval. It is then expressed in the form R i4iC = ~ / 3i4.(E i4 max - E i4 c ) 2(21). At the end of said selection S2i4, the electric vehicle sub-entity S_ENTi4 makes a decision from three decisions D1, D2, D3. Decision D1 relates to the choice of the energy source for recharging the sub-entity S_ENTi4. Decision D2 relates to the choice of the destination of the energy discharged from the sub-entity S_ENTi4. Decision D3 is not to implement any particular action in the following time interval, and in this case, the exchange method is iterated from step S1 i4 for the following time interval IT c +i .

[0223] If decision D1 was selected as the action to be carried out, the sub-entity S_ENT i4 :

[0224] - either selects in S3ai4, as the source of recharging of the sub-entity S_ENTi4, the other entity ENTj and / or the energy supplier FE,

[0225] - either selects in S3bi4, as the source of reloading of the sub-entity S_ENTi4, the energy production sub-entity S_ENTii.

[0226] If decision D2 was selected as the action to be carried out, the sub-entity S_ENT i4 :

[0227] - either selects in S3ci4, as recipient of the unloading of the sub-entity S_ENTi4, the other entity ENTj and / or the energy supplier FE,

[0228] - either selects in S3di4, as the recipient of the unloading of the sub-entity S_ENTi4, the energy consumption sub-entity S_ENTi2, such as for example a device whose use is shiftable over time, a device whose operating power level is variable, a device whose use is not shiftable over time and whose operating power level is fixed. In the case where the selection S3ai4 is implemented, the sub-entity S_ENTi4 calculates in S4ai4 the quantity Qi4,c of energy to be received from the entity ENTi or from the energy supplier FE for its reloading in the current time interval.

[0229] In the case where the selection S3bi4 is implemented, the sub-entity S_ENTi4 calculates in S4bi4 the quantity Qi4,c of energy to be received from the energy production sub-entity S_ENTii for its recharging in the current time interval. In the case where the selection S3ci4 is implemented, the sub-entity S_ENTi4 calculates in S4ci4 the quantity Qi4,c of energy to be used to supply energy to the other entity ENTj and / or the energy supplier FE.

[0230] In the case where the selection S3di4 is implemented, the sub-entity S_ENTi4 calculates in S4di4 the quantity Qi4,c of energy to be used to supply energy to said at least one energy consumption sub-entity S_ENTi2.

[0231] In S5i4, the calculated quantity Qi4,c of energy is then transmitted to the entity ENTi, in the current time interval.

[0232] We now describe, in relation to figure 8, the different steps carried out at the level of said at least one energy consumption sub-entity S_ENTi2 which is associated with said at least one entity ENTi, when the energy exchange method is implemented in the current time interval IT C and when such a sub-entity is present. In the example of Figure 8, the architecture of the entity set is on only two levels L1 and L2, the energy-consuming devices associated with the sub-entity S_ENTi2 being all located at level L2, regardless of their type.

[0233] In S1 i2, the sub-entity S_ENTi2 which is located at level L2 receives from the entity ENTi which is located at level L1, via a reception module identical or similar to that of figure 3, the value of the price Pr c,gnd that the energy available costs, in the said current time interval, from the energy supplier FE, as well as the value of the price Pr c ,xpiace that the energy available costs, in the said current time interval, from the other entity ENTj.

[0234] In S2i2, from the value of the price Pr c ,gnd and the price Pr c ,x p iace, of the power consumption êloadi2,c of all the energy-consuming devices corresponding to the sub-entity S_ENTi2, a type of action is then selected to be implemented in the current time interval by the sub-entity S_ENTi2. Said selection can also be implemented by combining the values ​​Pr c ,gnd, Pr c ,x p iace, 5loadi2,c with respectively values ​​of the same type obtained by learning in previous time intervals, for example using a supervised learning algorithm.

[0235] Such a selection S2i2 is implemented in accordance with a criterion Ri2,c of performance of the use of the energy to be consumed in the current time interval. In an exemplary embodiment, the criterion Ri2,c minimizes the dissatisfaction of a user of the energy-consuming device(s), said criterion being linked:

[0236] - either to a shift in the use of all or part of the devices in a time interval subsequent to the current time interval, if this or these devices has or have a usage profile that can be shifted over time,

[0237] - either to a reduction in the power level of all or part of the devices in the current time interval, if this or these devices has or have a power-shiftable usage profile.

[0238] The criterion Ri2,c is expressed in the form Ri2,c=- ' k=i Pk,c (22), where N represents the number of devices whose use is shiftable in time and Pk,c represents the operating power of a device k among N. If this device is switched off during a time interval, Pk,c=0 during this interval.

[0239] In another exemplary embodiment, the criterion Ri2,c is expressed in the form (23), where we use for example an average of the Pr price values c ,xpiace and Pr c , g nd.

[0240] This criterion can also be defined as a weighted sum between the energy consumption (or power) of energy-consuming devices and user dissatisfaction, both of which should be minimized.

[0241] The criterion Ri2,c is expressed in the form Ri2,c= k=i Rk,c (24) if k devices associated with the sub-entity S_ENTi2 remain in operation in the current time interval, represents the sum of the energy performance criteria applied individually for each of the k devices.

[0242] At the end of said selection S2i2, the sub-entity S_ENTi2:

[0243] - either leaves the energy-consuming device(s) associated with it operating S3ai2 during the current time interval,

[0244] - either cuts S3bi2 this or these devices, during the current time interval.

[0245] In the case where the selection S3ai2 is implemented, the sub-entity S_ENTi2 calculates in S4ai2 the quantity QCi2,c of energy to be consumed in the current time interval. In the case where the selection S3bi2 is implemented, in S4bi2, the sub-entity S_ENTi2 updates the quantity QCi2,c of energy to be consumed in the current time interval according to the devices that remain in operation in the current time interval. In S5i2, the sub-entity ENTi2 transmits to the entity ENTi the quantity QCi2,c of energy to be consumed in the current time interval, which was obtained in S3ai2 or S3bi2. We now describe, in relation to Figure 9A, the different steps carried out at the level of at least one energy consumption sub-entity SS_ENTi2o which is associated with said at least one energy consumption sub-entity S_ENTi2, when the energy exchange method is implemented in the current time interval IT Cand that the sub-entity S_ENTi2 has decided to leave in operation the energy-consuming device(s) which make up the sub-entity SS_ENTi2o. To this end, in this decision context, in the example of figure 9A, the architecture of the set of entities is on three levels L1, L2, L3 and the sub-entity SS_ENTi2o which is located at level L3 comprises M energy-consuming devices of the aforementioned type, the use of which can be shifted over time.

[0246] In S1 i2o, the sub-entity SS_ENTi2o selects, for a kth device among M, a corresponding action ai2o,c among two possible actions ACT1, ACT2 which are the following:

[0247] - ACT1: the kth device remains in operation in the following time interval,

[0248] - ACT2: the kth device stops operating in the following time interval. Step S1 i2o is iterated for each of the M devices of the sub-entity SS_ENTi2o. Such a selection S1120 is implemented on the basis of a prior setting rii2o,c of the user, in accordance with which the user has indicated which, among the M devices detectable over time, are those whose operation must be maintained and those whose operation must be stopped. Such a setting is conventional and can be implemented, for example, via a home automation application installed on a terminal or a domestic central unit, or via a website dedicated to the service offering the energy exchange.

[0249] Such selection is implemented in accordance with an energy performance criterion Ri2o,c.

[0250] In an exemplary embodiment, the criterion Ri2o,c minimizes the dissatisfaction of a user which would be linked to a shift in the use of a kth device shiftable in time, in a time interval subsequent to the current time interval. The criterion Ri2o,c is expressed for example in the following form: • Pt2o,k is la operating power of the kth device,

[0251] • a i2o,c is the action chosen at the end of the current interval IT C , it is a vector representing the decisions taken for all M devices: for example if the sub-entity SS_ENTi2o is representative of M=5 devices and the action is to shift the use of each of them, then a i20 c = [0, 0, 0, 0, 0],

[0252] • £ £2 O is a weighting coefficient,

[0253] • <5 £ 2o,fc rrepresents a user dissatisfaction coefficient for a kth device of the SS_ENTi2o sub-entity.

[0254] In another exemplary embodiment, the criterion Ri2o,c could for example minimize energy consumption in the event of a load peak to avoid purchasing energy at a high price.

[0255] At the end of said selection S1 i2o, each of the M devices of the sub-entity SS_ENTi2o:

[0256] - either remains in operation in S2ai2o, during the following time interval,

[0257] - is cut in S2bi2o, during the following time interval.

[0258] In S3i2o, the sub-entity SS_ENTi2o transmits the criterion Ri2o,c to the sub-entity S_ENTi2. We now describe, in relation to Figure 9B, the different steps carried out at the level of at least one energy consumption sub-entity SS_ENTi2i which is associated with said at least one energy consumption sub-entity S_ENTi2, when the energy exchange method is implemented in the current time interval IT C and that the sub-entity S_ENTi2 has decided to vary the power level of the energy-consuming devices that make up the sub-entity SS_ENTi2i. To this end, in this decision context, in the example of figure 9B, the architecture of the set of entities is on three levels L1, L2, L3 and the sub-entity SS_ENTi2i which is located at level L3 comprises N energy-consuming devices of the aforementioned type, the power level of which is variable over time.

[0259] In S1 i2i, the sub-entity SS_ENTi2i selects, for a kth device among N, an action among three possible actions ACT3, ACT4, ACT5 which are the following:

[0260] - ACT3: the kth device continues to operate in the next time interval, with the same power level as in the current time interval,

[0261] - ACT4: the kth device continues to operate in the next time interval, with a power level higher than that applied in the current time interval, - ACT5: the kth device continues to operate in the next time interval, with a power level lower than that applied in the current time interval.

[0262] Step S1121 is iterated for each of the N devices of the sub-entity SS_ENTi2i. Such a selection S1121 is implemented on the basis of a prior parameterization rii2i,c of the user, in accordance with which the user has indicated which, among the N variable power devices, are those for which the power level remains fixed, those for which the power level can be increased, and those for which the power level can be decreased. Such a parameterization is conventional and can be implemented in a manner similar to the example of FIG. 9A.

[0263] Such a selection is implemented in accordance with an energy performance criterion Ri2i,c.

[0264] In an exemplary embodiment, the criterion Ri2i,c minimizes the dissatisfaction of a user which would be linked to a decrease in the power level of a kth device among N, in the current time interval.

[0265] The criterion Ri2i,c is expressed for example in the following form:

[0266] • a i2i,c is the action chosen at the end of the current interval IT C , has i21 c being a vector of the decisions taken for all N devices of the sub-entity SS_ENTi2i regarding the power level to use among the levels available for each: for example if the sub-entity SS_ENTi2i is representative of 3 devices and the possible values ​​for each are [Pi, P2, P3, P4, Ps] and the action chosen at the end of the current time interval IT C is to use power level Pi for the first two devices and power level P3 for the third device, then the action will be written as follows: a i21 c = [Pi, Pi, P3],

[0267] • £21 is a weighting coefficient,

[0268] • pi2i, max is a vector representing the maximum operating powers for all devices whose power level is variable,

[0269] • 5i2i,k represents a user dissatisfaction coefficient for a kth device of the sub-entity SS_ENTi2i for which the power level is variable. In another exemplary embodiment, the criterion Ri2i,c minimizes the power consumption in the event of a load peak, for example.

[0270] At the end of said selection S1121, the operation of each of the N devices of the sub-entity SS_ENTi2i:

[0271] - is activated in S2ai2i, during the following time interval, with the same power level as that applied in the current time interval,

[0272] - is activated in S2bi2i, during the following time interval, with a higher power level than that applied in the current time interval,

[0273] - is activated in S2ci2i, during the following time interval, with a lower power level than that applied in the current time interval.

[0274] In S3i2i, the sub-entity SS_ENTi2i transmits the criterion Ri2i,c to the sub-entity S_ENTi2.

[0275] Description of a second embodiment of an entity capable of exchanging energy

[0276] We now describe, with reference to figure 10, the simplified structure of an entity ENTi chosen from the plurality of entities ENT1 to ENTK of figure 2, such as for example 1 <K<Ne, selon un deuxième mode de réalisation de l’invention.

[0277] Such an ENTi entity is configured to implement the energy exchange method which will be described below and which is implemented using a reinforcement learning algorithm.

[0278] For this purpose, the ENTi entity is an agent which operates in a multi-agent scenario involving at least one other ENTj agent located at the same level L1.

[0279] In the particular embodiment of figure 10, the agent ENTi comprises at least one sub-agent S_ENTi located at the lower level L2, said at least one sub-agent belonging for example to the following sub-agents:

[0280] - an energy production sub-agent S_ENTM,

[0281] - at least one S_ENTi2o device whose use can be shifted over time,

[0282] - at least one S_ENTi2i device with a variable operating power level,

[0283] - an energy storage sub-agent S_ENTi3,

[0284] - an S_ENTi4 electric vehicle,

[0285] - etc. In the example shown in Figure 10, the agent ENTi performs, in a current time interval IT C , a certain action ai,c belonging to a following action space Ai:

[0286] - do nothing,

[0287] - request energy from an external FE energy supplier or at least one of the K-1 other agents,

[0288] - supply energy to the external energy supplier FE or to at least one of the K-1 other agents,

[0289] - determine the quantity of energy to be required (to buy) or supplied (to sell). The agent ENTi also chooses, in a current time interval ITc, an objective for each of the sub-agents of the lower level L2. These objectives belong to sets of predefined objectives which will be described later.

[0290] Similarly, in this IT time interval C , each of the K-1 other agents defining an energy exchange place, is an agent which performs one of the aforementioned actions and chooses an objective for each of its sub-agents of the lower level L2.

[0291] In order to select the action that optimizes this energy exchange in the current time interval ITc, the agent ENTi explores its environment which is represented by a state belonging to a state space Si which will be described in the rest of this description or exploits the result of its learning and selects the action which has proven to be best so far.

[0292] To this end, the ENTi agent performs different actions, such as those in the aforementioned action space, for a given state Si, c of the state space Si, providing a reward Ri, c which defines a performance criterion for the energy exchange in the exchange place. Ri, c is a signal that defines the reward (or the cost) of having performed the action ai,c while being in the state Si, c. This information is transmitted from the environment to the agent which seeks to optimize it (maximize it in case of reward and minimize it if it is a cost) in order to learn the best actions to take in each state. In the example shown, Ri, c may be representative of reducing expenses and / or maximizing revenues and / or reducing the carbon footprint related to the energy transaction, in the current time interval ITc. Similarly, in this time interval IT C , each of the K-1 other agents explores its corresponding environment, in particular the entity ENTj of figure 10 which carries out different actions, such as those of the action space Aj, for a given state Sj, c of a state space Sj, providing a reward Rj, c which defines the reward (or the cost) of having carried out the action aj, c being in the state Sj,c.

[0293] In accordance with the invention, and as already explained for the first embodiment mentioned above, the agent ENTi, respectively the agent ENTj, is broken down into sub-agents arranged at the hierarchical level L2, of which only one sub-agent S_ENTi, respectively S_ENTj, is represented for the sake of simplification of figure 10.

[0294] At level L2, the sub-agent S_ENTi, respectively S_ENTj, receives information from the agent ENTi, respectively ENTj, which defines a goal to be achieved in order to achieve the optimal energy exchange strategy defined by Ri, c (respectively Rj,c). This information can be added to the state Si, c , respectively Sj, c , of the sub-agent S_ENTi, respectively S_ENTj.

[0295] For this purpose, the sub-agent S_ENTi, respectively S_ENTj, carries out, in the current time interval IT C, the action ai.c, respectively aj,c, which aims to optimally distribute the energy for the agent ENTi, respectively the agent ENTj. When the agent ENTi is decomposed for example into five sub-agents S_ENTii, S_ENTi2o, S_ENTi2i, S_ENTi3, S_ENTi4, each of them carries out, in the current time interval, the respective actions an , c , ai2o,c, ai2i,c, ai3,c, ai4,c which, together, aim to optimally distribute energy for the agent ENTi.

[0296] The action ari,c performed by the sub-agent S_ENTM, in the current time interval ITc, is predefined, such an action being chosen from the following action space:

[0297] - do nothing,

[0298] - select, as recipient of all or part of the energy produced, the other agent ENTj and / or the energy supplier FE,

[0299] - supply energy to the S_ENTi2o sub-agent and / or the S_ENTi2i sub-agent,

[0300] - recharge the energy storage sub-agent S_ENTi3 with all or part of the energy produced,

[0301] - recharge the sub-agent S_ENTi4 with all or part of the energy produced. In order to select the action an , c which optimizes the distribution or use of energy in the current time interval ITc, the sub-agent ENTM explores its environment which is represented by a state belonging to a state space Sii which will be described in the rest of this description. To this end, the sub-agent S_ENTii carries out different possible actions, such as those of the aforementioned action space, for a given state, providing a reward Rii, c which defines a performance criterion for the use of the energy produced by this sub-agent S_ENTii. In the example shown, Rii, c can be representative for example of the maximization of revenues in the current time interval IT C, in the case where the energy produced is supplied to the energy supplier FE or to the other agent ENTj.

[0302] According to this second embodiment, the sub-agent S_ENTi2o carries out, in the current time interval, the action ai2o,c which aims to best adapt the energy consumption for the agent ENTi. Depending on the context of the energy exchange, if a sub-agent S_ENTi2i is present at level L2, the action ai2o,c is implemented jointly with the action ai2i, c implemented by the S_ENTi2i sub-agent in order to best adapt energy consumption for the ENTi agent.

[0303] The action ai2o,c performed by the sub-agent S_ENTi2o, in the current time interval ITc, is predefined, such an action being chosen in the following space of actions Ai2o:

[0304] - operate,

[0305] - not work.

[0306] In order to select the action ai2o,c that optimizes the distribution or use of energy in the current time interval ITc, the sub-agent S_ENTi2o explores its environment which is represented by a state belonging to a state space Si2o which will be described in the remainder of this description. To this end, the sub-agent S_ENTi2o performs different possible actions, such as those of the aforementioned action space Ai2o, for a given state, providing a reward Ri2o,c that defines an energy performance criterion linked to the use of the energy consumed by this sub-agent S_ENTi2o. In the example shown, Ri2o,c can define for example the minimization of user dissatisfaction in the current time interval IT C, in the case of a shift in the use of the sub-agent S_ENTi2o and the minimization of energy consumption in the event of a load peak to avoid purchasing energy at a high price. The action ai2i,c carried out by the sub-agent S_ENTi2i, in the current time interval ITc, is predefined, such an action being chosen from the following action space Ai2i:

[0307] - keep the same power as that applied in the previous time interval,

[0308] - reduce power,

[0309] - increase power.

[0310] In order to select the action ai2i,c which optimizes the energy consumption in the current time interval IT C , in this time interval IT C, the sub-agent S_ENTi2i explores its environment which is represented by a state belonging to the state space Si2i and which will be described in the remainder of this description. For this purpose, the sub-agent S_ENTi2i carries out different possible actions, such as that of the aforementioned action space, for a given state of the state space Si2i , providing a reward Ri2i,c which defines a criterion for optimizing the energy consumption by the sub-agent S_ENTi2i. In the example shown, Ri2i,c can be representative of the minimization of user dissatisfaction linked to the reduction of the operating power level of the sub-agent S_ENTi2i and the minimization of power consumption in the event of a load peak for example, to avoid purchasing energy at a high price.

[0311] According to this second embodiment, the action ai3,c carried out by the sub-agent S_ENTi3, in the current time interval ITc, is predefined, such an action being chosen in the following action space Ais:

[0312] - do nothing, that is to say neither recharge nor discharge energy,

[0313] - sell energy to at least one of the K-1 other agents, ENTj for example,

[0314] - sell energy to the external energy supplier FE,

[0315] - supply energy to the S_ENTi2o sub-agent and / or the S_ENTi2i sub-agent,

[0316] - recharge energy from at least one of the K-1 other agents, ENTj for example,

[0317] - top up energy from the external energy supplier FE,

[0318] - recharge energy from the S_ENTii subagent.

[0319] In order to select the action ai3,c which optimizes the use of the energy stored in the current time interval ITc, the sub-agent ENTi3 explores its environment which is represented by a state belonging to a state space Sis which will be described in the remainder of this description. For this purpose, the sub-agent S_ENTi3 carries out different possible actions, such as those of the aforementioned action space, for a given state, providing a reward Ri3,c which defines an energy performance criterion linked to the use of the energy stored by this sub-agent S_ENTi3. In the example shown, Ri3,c can define the maximization of the duration of the life cycle of said at least one storage sub-agent S_ENTi3.

[0320] According to this second embodiment, the action ai4,c carried out by the sub-agent S_ENTi4, in the current time interval IT C , is predefined, such an action being chosen from the following action space Ai4:

[0321] - do nothing, that is to say neither recharge nor discharge energy,

[0322] - sell energy to at least one of the K-1 other agents, ENTj for example,

[0323] - sell energy to the external energy supplier FE,

[0324] - supply energy to the S_ENTi2o sub-agent and / or the S_ENTi2i sub-agent,

[0325] - recharge energy from at least one of the K-1 other agents, ENTj for example,

[0326] - top up energy from the external energy supplier FE,

[0327] - recharge energy from the S_ENTii subagent.

[0328] In order to select the action ai4,c which optimizes the use of the energy stored in the current time interval IT C, the sub-agent S_ENTi4 explores its environment which is represented by a state belonging to a state space Si4 which will be described in the remainder of this description. For this purpose, the sub-agent S_ENTi4 carries out different possible actions, such as those of the aforementioned action space, for a given state, providing a reward Ri4,c which defines an energy performance criterion linked to the use of the energy stored by this sub-agent S_ENTi4. In the example shown, Ri4,c can define the user's dissatisfaction based on the fear that the sub-agent S_ENTi4 electric vehicle does not have enough energy to operate in the current time interval. Alternatively, Ri4,c can be representative of the maximization of the duration of the life cycle of said at least one sub-agent S_ENTi4.

[0329] The embodiment described in connection with Figure 10 is particularly advantageous for the following reasons:

[0330] - it uses a reinforcement learning algorithm that is particularly effective for modeling a sequential decision-making system or a multi-agent energy exchange place with complex state spaces and action spaces. Furthermore, it is well suited for hierarchical decomposition in accordance with the invention of the energy exchange place, where on several levels, for example two levels L1, L2 in the example shown in Figure 10, sequential decisions are implemented by the corresponding agents or sub-agents,

[0331] - it is based on a hierarchical decomposition of an agent into different sub-agents linked to a specific energy use profile (consumption, storage, energy production, electric vehicle, etc.), such a decomposition making it possible to reduce action spaces and state spaces while preserving the scalability of the place or the energy exchange system,

[0332] - it allows to operate on various time scales: for example, the energy exchange strategy at level L1 can be defined per day, while the energy use strategies in the lower level L2 can be defined over shorter times, for example one or more hours, one or more minutes, etc. Such hierarchical operation allows an acceleration of reinforcement learning for the place or the energy exchange system,

[0333] - the modularity of the energy exchange system on different levels L1, L2 greatly facilitates the transfer of learning between agents with the same characteristics, which also contributes to this acceleration of reinforcement learning. Thus, reinforcement learning makes it possible to optimize the efficiency of energy exchange on the place or the energy exchange system,

[0334] - the decomposition of the energy exchange system into several hierarchical levels allows better protection of the user's personal data concerning the sub-agents of the lower L2 levels of the place or of the energy exchange system, little information linked to these lower levels being transmitted to the entities or agents of the higher level.

[0335] Description of a second embodiment of an energy exchange method

[0336] We now describe, in relation to figure 11, the progress of an energy exchange process executed by the agent ENTi as illustrated in figure 10. Such an energy exchange process takes place as follows at the level of the agent ENTi, in a current time interval IT C . In S'1, a given state of the energy exchange system is initialized in the current time interval IT C .

[0337] In a preferred embodiment, the state space Si configured for the system is for example the following:

[0338] Si=Prgrid X PrXplace X Qi1 ,v X Qcons XHX Ui3 X Ui4 (27), where:

[0339] • Prgrid and Prxpiace are the set of possible values ​​for prices from the traditional FE provider and the exchange, respectively; they can be defined as two intervals between a minimum price and a maximum price (which are to be defined) with values ​​that are either continuous or discrete in between,

[0340] • Qh,v is the quantity of energy produced by the sub-agent S_ENTii and sold to the traditional supplier FE or to the exchange place;

[0341] • Qcons is the set defining the possible values ​​for the amount of energy consumed during a predefined time interval; Qcons can be defined as an interval between a minimum and a maximum consumption value, such as Q cons = [Qcons, min , Qcons, max ], with continuous or discrete values ​​in between;

[0342] • H is the set defining the time: H = [0, 23];

[0343] • Ui3 (resp. Ui4) is a vector composed of the quantity of energy to be sold by the sub-agent S_ENTi3 (resp. S_ENTi4) and the carbon footprint Carbi3 (resp. Carbi4) which is the set of possible values ​​for the carbon footprints in the sub-agent S_ENTi3 (resp. S_ENTi4), where Carbi3 (resp. Carbi4) can be between a predefined minimum and maximum value.

[0344] In S'2, the ENTi agent selects an action according to a compromise of exploitation of the learning result and exploration of the action space, with a given probability (for example 0.9 for exploitation and 0.1 for exploration).

[0345] In a preferred embodiment, the action is selected from an action space Ai which is for example the following:

[0346] Ai= Ti x Q x Sbi x Ssi v (28), where:

[0347] • Ti represents the vector defining possible energy transactions. It is a three-valued vector, Ti = {-1, 1, 0}, where -1 represents an energy purchase from the FE supplier or at least one of the K-1 other agents, 1 represents an energy sale to the FE supplier or at least one of the K-1 other agents, and 0 represents no energy transaction action,

[0348] • Qti is the set defining the possible values ​​for the amount of energy to be bought or sold during a predefined time interval; Qti can be defined as an interval between 0 and a maximum consumption value, such as Qti = [0, Qti, max], with continuous or discrete values ​​in between,

[0349] • Sbi is the set defining the source of the energy purchased by the agent ENTi in a given time interval, Sbi being a vector with two values, for example Sbi = {1, 2}, where 1 is representative of the energy supplier FE and 2 is representative of the exchange place,

[0350] • Ssi is the set defining the source of the energy sold by the agent ENTi in a given time interval, Ssi being a vector with three values, for example Ssi = {1, 2, 3}, where 1 is representative of the energy producing sub-agent S_ENTii, 2 is representative of the energy storage sub-agent S_ENTi3, 3 is representative of the electric vehicle sub-agent S_ENTi4,

[0351] • Di is the set defining the destination of the energy purchased by the agent ENTi in a given time interval, Di being a vector with four values, for example Di = {1, 2, 3}, where 1 is representative of the energy storage sub-agent S_ENTi3, 2 is representative of the electric vehicle sub-agent S_ENTi4, 3 is representative of the energy consuming devices,

[0352] • Pn,v is a set defining the possible values ​​of the selling price of the energy proposed by the agent ENTi, it can be defined as being an interval between a minimum price and a maximum price (which are to be defined) with values ​​which are either continuous or discrete between the two.

[0353] In S'3, the ENTi agent selects an objective for each sub-entity according to a compromise of exploitation of the learning result and exploration of the action space, with a given probability (for example 0.9 for exploitation and 0.1 for exploration).

[0354] In a preferred embodiment, each objective is selected from an objective space Gi,c which is for example the following:

[0355] Gi=Gii x Gi2ox Gi2i x G i3 x Gi4 (29), where:

[0356] • Gii is the objective space concerning the photovoltaic panel sub-agent S_ENTii, such that Gii = {0,1}, where: ■ 0 means to stop the photovoltaic panel S_ENTii (this can be useful for energy balancing: input = output) or to continue not to use it if it is already switched off,

[0357] ■ 1 means to turn it on / or keep it on if it is already functional;

[0358] • Gi2o is the objective space concerning the S_ENTi2o devices detectable in time, such that Gi2o = {0,1}, where:

[0359] ■ 0 means do not use the device running time adjustment feature, 1 means enable this feature;

[0360] • Gi2i is the objective space concerning the devices S_ENTi2i whose power level is adjustable over time, such that Gi2i = {0,1}, where:

[0361] ■ 0 means not to use the power level adjustment feature of the devices,

[0362] ■ 1 means enable this feature;

[0363] • Gi3 is the objective space concerning the energy storage sub-agent S_ENTi3, such that Gis = {0, 1, 2} where:

[0364] ■ 0 means do nothing,

[0365] ■ 1 means discharge the energy storage subagent S_ENTi3,

[0366] ■ 2 means load this subagent;

[0367] • Gi4 is the objective space concerning the electric vehicle sub-agent S_ENTi4, such that: {0, 1, 2} where:

[0368] ■ 0 means do nothing,

[0369] ■ 1 means unload the vehicle,

[0370] ■ 2 means charge it.

[0371] At the end of this selection, a gi objective, c , at a current time interval IT C, can be a combination of the different possible values ​​belonging to the sets Gii, Gi2o, Gi2i, Gi3, Gi4, for example:

[0372] • gi, c = {0, 0, 0, 1 , 1} means that the ENTi agent of level L1 transmits an instruction to each of the sub-agents S_ENTii, S_ENTi2o, S_ENTi2i, S_ENTi3, S_ENTi4 of lower level L2, to stop the operation or continue the shutdown, if it is already the case, of the sub-agents S_ENTii, to deactivate the functionality of adjusting the use of the sub-agents S_ENTi2o and S_ENTi2i, and to unload the sub-agents S_ENTi3 and S_ENTi4, • or gi, c= {1 , 1 , 1 , 1 , 1} means that the ENTi agent of level L1 transmits an instruction to each of the sub-agents S_ENTii, S_ENTi2o, S_ENTi2i, S_ENTi3, S_ENTi4 of lower level L2, to respectively control the operation or the continuation of the operation of the sub-agent S_ENTii, activate the functionality of adjusting the use of the sub-agent S_ENTi2o, activate the functionality of adjusting the power of the sub-agent S_ENTi2i, activate the unloading of the sub-agents S_ENTi3 and S_ENTi4.

[0373] In S'4, each of the five values ​​of the objective gi, c is sent respectively to each of the corresponding sub-agents S_ENTii, S_ENTi2o, S_ENTi2i, S_ENTi3, S_ENTi4 and will then be part of their corresponding state.

[0374] In S'5, the agent ENTi receives, from its environment, a reward signal Ri, c which is for example the following in a preferred embodiment: Ri, c=m,c.ti, c. profit - [3i,c.carbon_footprint (30), where:

[0375] • you, c is the type of action chosen at the end of the current interval IT C for the next time interval, ti, c e Ti, Sbi and qt i c e Qt £ ,

[0376] • carbon_footprint is a carbon footprint factor that corresponds to the carbon footprint transmitted by the energy supplier FE or at least one of the K-1 other agents from which the agent ENTi purchased the energy,

[0377] • have, c and Pi,c are two weighting coefficients between profit and carbon footprint.

[0378] Alternatively, Profit=pr i V c .qt i c (33) if ti,c=1 and Ri, c =0 if ti,c=O, where pr i v c e Pr i v and represents a sale price that is not fixed in advance, for example a price deducted following an auction.

[0379] Steps S'2 to S'5 are iterated in S'6 until a stopping criterion is reached, so as to select for each given state an optimal action a opt i, c among Ai, that is, an optimal vector value for ti, c , qti.c, sbi.c, ssi, c said, c , pn,v,c for given values ​​of Pre, grid, Prc,Xplace, Qi1 ,v, Qi3,v, Qi4,v, Qcons, H, Ui3, Ui4.

[0380] Such optimal action has opt i, c is recorded in S7 in a dedicated memory. In a particular embodiment, the optimal action has opt i, cis stored so as to be possibly selected in the following time interval in response to either an exploitation or an exploration. In another particular embodiment, steps S'1 to S'7 can be carried out prior to the real-time operation of the exchange place, in a phase of simulation of the operation of this exchange place, so as to obtain in S'8 a mapping between each possible state and the corresponding optimal action and store in S'9 this mapping, for example in the form of a correspondence table TC. Thus, when the exchange place operates in real time, in S'10 the agent ENTi observes for example its state, then selects in S'11 directly from the table TC the action which has proven to be optimal.We now describe, in relation to figure 12, the progress of an energy distribution method executed by the sub-agent S_ENTii as illustrated in figure 10, within the framework of the energy exchange method of figure 11.

[0381] Such an energy distribution process takes place as follows at the level of the sub-agent S_ENTii, in a current time interval IT C .

[0382] In S'1 ii, a given state of the energy exchange system is initialized in the current time interval IT C .

[0383] In a preferred embodiment, the state space Sii configured for the system is for example the following:

[0384] Si1 = Prgrid X Prxplace X Qi1 ,pr X CHi3 xCHi4 X Qcons X Gil XH (34), where:

[0385] - Pr g rid, Prxpiace, Gu and H are as mentioned above,

[0386] - Qn.prest the set defining the quantity of energy that can be produced by the sub-agent S_ENTn and which can be defined for example as follows: Qii, pr = [0, Qn.max], where Qn.max is the maximum amount of energy that can be produced during a predefined time interval, e.g., one hour, one day, etc.,

[0387] - CHi3 = [0, 100] defines the load state of the sub-agent S_ENTi3, in the current time interval IT C ,

[0388] - CHi4 = [0, 100] defines the load state of the sub-agent S_ENTi4, in the current time interval IT C ,

[0389] - Qcons is the set defining the possible values ​​for the amount of energy consumed during a predefined time interval; Qcons can be defined as an interval between a minimum and a maximum consumption value, such that Q cons = [Qcons, min , Qcons, max ], with continuous or discrete values ​​in between. In S'2ii, the sub-agent S_ENTii selects an action according to a compromise of exploiting the learning result and exploring the action space, with a given probability (e.g. 0.9 for exploitation and 0.1 for exploration). In a preferred embodiment, the action is selected from an action space An which is for example the following:

[0390] Aii = Dii x Qh.utx lii (35), where:

[0391] • Dii is the set of possible destinations of the energy produced by the sub-agent S_ENTii in the current time interval IT C, and is expressed as follows: Dii = {1, 2, 3, 4, 5}, where 1 is representative of the energy exchange location, 2 is representative of the external energy supplier FE, 3 is representative of the storage sub-agent S_ENTi3, 4 is representative of the electric vehicle sub-agent S_ENTi4, 5 is representative of the energy consuming devices,

[0392] • Qii ,ut is the set of quantities of energy produced by the sub-agent S_ENTii which can be used in the current time interval IT C , and being defined for example as follows: Qii.ut = [0, Qii ,max],

[0393] • lii={0,1}, where 0 means that the sub-agent S_ENTii is not put into operation and 1 means that the sub-agent S_ENTii is put into operation or continues its operation if it was in operation in the previous time interval.

[0394] In S'3ii, the sub-agent S_ENTii receives a reward signal Rii, cwhich is for example the following in a preferred embodiment:

[0395] • Oii is a predefined weighting coefficient,

[0396] • Ru,ext,c defines an extrinsic reward from the environment in response to the action taken by the subagent S_ENTii. It can be defined in a current time interval as follows: il,ext,c = PTc,Xplace X qi1,ut,c If di1,c = 1 (37) and

[0397] Ril,ext,c = Pfc, grid X qi1,ut,c If dil.c = 2 (38) and n,ext,c = 0 otherwise, where: • pfc.Xplace G Prxpiace and pr c , grid G ​​Prgrid are respectively the prices of energy in the exchange and of the energy supplier in the current time interval IT Cconsidered and qii.ut.c G Qii ,ut and dii,c G DH respectively designate the decision made at the end of this time interval considered concerning the quantity of energy to be used and for which destination, this for the following time interval,

[0398] • R u,int,c defines an intrinsic reward that is received by the sub-agent S_ENTii, which is consistent with the objectives transmitted by the agent ENTi in S'4 (figure 1 1 ), this intrinsic reward can be defined in the current time interval as:

[0399] Rn,int,c = "I if iii.c = gri.c and R^int.c = 0 otherwise, where iii, c G hi and gii, c G Gii respectively designate the action chosen by the sub-agent S_ENTii to either operate or not and the objective transmitted by the agent ENTi in S'4 (figure 1 1 ), both for the following time interval IT C .

[0400] Steps S'2n to S'3ii are iterated in S'4ii up to a stopping criterion so as to select for each given state an optimal action a opt ii, c among Aii , i.e. an optimal vector value for dii, c , qii ,ut,c and iii, c for given values ​​of pre, grid, prc,Xplace, qi1 ,pr,c, Chi3,c, Chi4,c, qcons,c, Çi1 ,c and hc.

[0401] Such optimal action has opt ii, c is recorded in S'5ii in a dedicated memory so as to be possibly selected in the following time interval in response to either an exploitation or an exploration or to be used for the implementation of real-time steps S'6ii to S'9ii of the same type as the aforementioned steps S'8 to S'11.

[0402] We now describe, in relation to figure 13, the progress of an energy distribution method executed by the storage sub-agent S_ENTi3 as illustrated in figure 10, within the framework of the energy exchange method of figure 11.

[0403] Such an energy distribution process takes place as follows at the level of the sub-agent S_ENTi3, in a current time interval IT C .

[0404] In S'1 i3, a given state of the energy exchange system is initialized in the current time interval IT C .

[0405] In a preferred embodiment, the state space Sis configured for the system is for example as follows:

[0406] Si3= Prgrid x Prxpiace x Carbi3 x CHi3 xCHi4 x Qcons x Gi3 XH (39), where:

[0407] Prgrid, Prxpiace, Carbi3, Qcons, Gis and H are as described above, • CHi3 defines the load percentage of the sub-agent S_ENTi3, in the current time interval IT C , and is expressed as follows CHi3= [0,100].

[0408] In S'2i3, the sub-agent S_ENTi3 selects an action according to a compromise of exploitation of the learning result and exploration of the action space, with a given probability (for example 0.9 for exploitation and 0.1 for exploration). In a preferred embodiment, the action is selected in an action space Ais which is for example the following: Ai3 =Ci3 x Di3 x Qi3 (40), where:

[0409] • Ci3 is a set defining the possible source of energy recharged in the sub-agent S_ENTi3 in the current time interval IT Cand is expressed for example as follows: Ci3 = {0,1,2,3}, where 0 is representative of the absence of recharging, 1 is representative of the energy exchange place, 2 is representative of the external energy supplier FE, 3 is representative of the sub-agent S_ENTii in the case it produces excess energy,

[0410] • Di3 is a set defining the possible destination of the energy discharged from the sub-agent S_ENTi3 in the current time interval IT C , and is expressed for example as follows: Dis = {0,1,2,3}, where 0 is representative of the absence of unloading, 1 is representative of the energy exchange place, 2 is representative of the external energy supplier FE, 3 is representative of the energy consuming devices,

[0411] • Qi3 is a set defining the percentage of reloading / unloading of the sub-agent S_ENTi3 in the current time interval IT C , and is expressed as follows: Qi3 = [0,100].

[0412] The action selected by the sub-agent S_ENTi3 thus consists of choosing to recharge or discharge, with what quantity of energy (or otherwise what percentage of its capacity), as well as the source of this charge / destination of this discharge. If the sub-agent S_ENTi3 chooses to discharge by selling in the exchange place or to the supplier, this information is then transmitted to the higher-level ENTi agent L1 to be taken into account in the exchange strategy. This information bears the reference Ui, c in figure 10, such that here Ui, c = Ui3,c. As mentioned above, Ui3,c takes as value the quantity of energy to be sold in the exchange or to the supplier and the carbon footprint Carbi3, and 0 otherwise. It will subsequently enter the state of the agent ENTi. In Figure 10, Ui, cis represented by dotted lines because it is not transmitted by all the other sub-agents considered, notably S_ENTii, S_ENTi2o, S_ENTi2i.

[0413] In S'4i3, the sub-agent S_ENTi3 receives a reward signal Ri3,c which is for example the following in a preferred embodiment:

[0414] • ai3 is a predefined weighting coefficient,

[0415] • Rn.ext.c defines an extrinsic reward from the environment in response to the action taken by the S_ENTi3 subagent. It can be defined in a current time interval as follows:

[0416] R i3,ext,c = &3. profit, c + (1 - £i3).storagei3,c (42), where:

[0417] . prof i ti3,c = pr c ,Xplace (qi3,cQ max) / 100 (43) if di3,c = 1 and

[0418] • profiti3,c = pTc, grid (qi3,c.Qmax) / 100 (44) If di3,c = 2 and

[0419] • prof iti3,c = - prc,Xplace (qi3,c.Qmax) / 100 (45) if Ci3,c = 1 and

[0420] • profiti3,c = - pTc, grid (qi3,c.Qmax) / 100 (46) If Ci3,c = 2 and

[0421] . profit = 0 otherwise and

[0422] . storagei3,c = qi3,c - |clte,c + qi3,c - Chi3,max| (47) if Ci3,c 0 (loading) and

[0423] . storagei3,c = qi3,c - |clte,c - qi3,c - CH i3,min| (48) if di3,c 0 (unloading), and |x| symbolizing the absolute value of a real number x, where:

[0424] • pr c ,expiace and pr c ,grid are respectively the energy prices in the exchange and the energy supplier in the current time interval IT Cconsidered and qi3,ce Qis, di3,ce Dis and clte,ce CHi3 respectively designate the decision made at the end of this time interval considered concerning the quantity of energy to be used and for which destination, this for the following time interval and the state of charge of the sub-agent S_ENTi3 at the end of the current time interval, .storagei3,c is a function that aims to maximize the quantity to be loaded and minimize the distance between the current load of the S_ENTi3 subagent, incremented by the amount of energy to be loaded, and the maximum load value of the S_ENTi3 subagent in order not to move away from this maximum value in the case of a reload and in the case of an unload storagei3,c aims to maximize the quantity of energy to be unloaded and minimize the distance between the current load of the S_ENTi3 subagent, decremented by the amount of energy to be unloaded, and the minimum load of the S_ENTi3 subagent, this being in order not to exceed too much the minimum load required by the S_ENTi3 subagent.

[0425] . Eis is a weighting coefficient,

[0426] • Ri3,int,c defines an intrinsic reward which is received in coherence with the objectives transmitted by the agent ENTi in S'4 (figure 1 1 ), this reward being able to be defined in a current time interval as follows:

[0427] • Ri3,int,c = 1 if Ci3,c = di3,c = gi3,c = 0 (do nothing) or if di3,c 0 and gi3,c = 1 or if Ci3,c + 0 and gi3,c = 2 and Ri3,int,c = 0 otherwise, where Ci3,c G Ci3 and gi3,c G Gi3 respectively designate the action chosen by the sub-agent S_ENTi3 to choose the energy source for recharging and the objective transmitted by the agent ENTi in S'4 (figure 11), both at the end of the current time interval. The steps S'2i3 to S'3is are iterated in S'4i3 until a stopping criterion so as to select for each given state an optimal action a opt i3,c among Ais, i.e. an optimal vector value for Ci3,c, di3,c, qi3,c for given values ​​of pre, grid, preface, Carbi3,c, chi3,c, chi4,c, qcons,c, gis.c and hc.

[0428] Such optimal action has opt i3,c is recorded in S'5i3 in a dedicated memory so as to be possibly selected in the following time interval IT c +i in response to either exploitation or exploration or be used for the implementation of real-time steps S'6i3 to S'9i3 of the same type as the aforementioned steps S'8 to S'1 1.

[0429] We now describe, in relation to figure 14, the progress of an energy distribution method executed by the sub-agent S_ENTi4 electric vehicle as illustrated in figure 10, within the framework of the energy exchange method of figure 11. Such an energy distribution method takes place as follows at the level of the sub-agent S_ENTi4, in a current time interval IT C .

[0430] In S'1 i4, a given state of the energy exchange system is initialized in the current time interval IT C .

[0431] In a preferred embodiment, the state space Si4,c configured for the system is for example the following:

[0432] Si4= Pr g rid x Prxpiace x Carbi4 x CHi4 x Qcons x Gi4 x Ei4 x H (49), where:

[0433] • Pr c ,grid, Pr c ,x p iace, Carbi4,, Qcons, Gi4 and H are as described above,

[0434] • CHi4 defines the load percentage of the sub-agent S_ENTi4, in the current time interval IT C , and is expressed as follows: CHi4= [0,100],

[0435] • Ei4 is the set of possible values ​​for the amount of energy that can be consumed by the sub-agent S_ENTi4 during a given time interval, between 0 and a predefined maximum value.

[0436] In S'2i4, the sub-agent S_ENTi4 selects an action according to a compromise of exploitation of the learning result and exploration of the action space, with a given probability (for example 0.9 for exploitation and 0.1 for exploration). In a preferred embodiment, the action is selected in an action space Ai4 which is for example the following:

[0437] Ai4 =Ci4 x Di4 x Qi4 (50), where:

[0438] • Ci4 is a set defining the possible source of energy used for recharging the sub-agent S_ENTi4 in the current time interval IT C and is expressed for example as follows: Ci4= {0,1,2,3,4}, where 0 is representative of the absence of recharging, 1 is representative of the energy exchange place, 2 is representative of the external energy supplier FE, 3 is representative of the sub-agent S_ENTii in the case it produces excess energy, 4 is representative of the storage sub-agent S_ENTi3,

[0439] • Di4 is a set defining the possible destination of the energy discharged from the sub-agent S_ENTi4 in the current time interval IT C and is expressed for example as follows: Di4= {0,1,2,3}, where 0 is representative of the absence of unloading, 1 is representative of the energy exchange place, 2 is representative of the external energy supplier FE, 3 is representative of the energy consuming devices, • Qi4 is a set defining the percentage of recharging / unloading of the sub-agent S_ENTi4 in the current time interval IT C , and is expressed as follows: Qi4= [0,100].

[0440] The action selected by the sub-agent S_ENTi4 thus consists of choosing to recharge or discharge, with what quantity of energy (or otherwise what percentage of its capacity), as well as the source of this charge / destination of this discharge. If the sub-agent S_ENTi4 chooses to discharge by selling in the exchange place or to the supplier, this information is then transmitted to the higher-level ENTi agent L1 to be taken into account in the exchange strategy. This information bears the reference Ui, c in figure 10, such that here Ui, c = Ui4,c. As mentioned above, Ui4,c takes as value the quantity of energy to be sold in the exchange or to the supplier and the carbon footprint Carbi4, and 0 otherwise. It will subsequently enter the state of the agent ENTi.

[0441] In S'4i4, the sub-agent S_ENTi4 receives a reward signal Ri4,c which is for example the following in a preferred embodiment:

[0442] • ai4 is a predefined weighting coefficient,

[0443] • Ri4,ext,c defines an extrinsic reward from the environment in response to the action taken by the subagent S_ENTi4. It can be defined in a current time interval as follows:

[0444] Ri4,ext,c = Ei,i4.profiti4,c + S2,i4.storagei4,c + £3,i4.disconforti4,c (52), where: . prof i ti4,c = pr c ,Xplace x (qi4,cQ max) / 100 (53) if di4,c = 1 and

[0445] • prof iti4,c = pTc, grid X (qi4,c.Qmax) / 100 (54) If di4,c = 2 and

[0446] . prof i ti4,c = - pr c ,Xplace x (qi4,cQ max) / 100 (55) if Ci4,c = 1 and

[0447] • profiti4,c = - pTc, grid X (qi4,c.Qmax) / 100 (56) If Ci4,c = 2 and

[0448] . profiti4,c = 0 otherwise and storagei4,c = qi4,c - |chi4,c + qi4,c - CHi4,max| (57) if Ci4,c 0 (reloading) and

[0449] . storagei4,c = qi4,c - |chi4,c - qi4,c - CHi4,min| (58) if cc 0 (unloading) and

[0450] • discomforti4,c= (Ei4,max - ei4,c) 2 (59), and £i,i4, £2,i4, £3 4 are predefined weighting coefficients, and where:

[0451] • pr c ,x P iace and pr c ,grid are respectively the energy prices in the exchange and the energy supplier in the current time interval IT C considered and qi4,ce Qi4, di4,ce Di4 and chi4,ce CHi4 respectively designate the decision made at the end of this time interval considered concerning the quantity of energy to be used and for which destination, this for the following time interval and the state of charge of the sub-agent S_ENTi4 at the end of the current time interval,

[0452] • Ei4,max and ei4,c are respectively the maximum energy consumption during a predefined time interval and ei4, cis the energy consumption of the sub-agent S_ENTi4 during the current time interval,

[0453] • disconforti4, c is a factor determining the anxiety of the user of the subagent S_ENTi4 of not having enough energy to use it,

[0454] • storagei4,c is a function that aims to maximize the quantity to be loaded and minimize the distance between the current load of the S_ENTi4 sub-agent, incremented by the quantity of energy to be loaded, and the maximum load value of the S_ENTi4 sub-agent in order not to move away from this maximum value in the case of a reload and in the case of an unload storagei4,c aims to maximize the quantity of energy to be unloaded and minimize the distance between the current load of the S_ENTi4 sub-agent, decremented by the quantity of energy to be unloaded, and the minimum load of the S_ENTi4 sub-agent, this being in order not to exceed too much the minimum load required by the S_ENTi4 sub-agent,

[0455] • R i4,int,c defines an intrinsic reward which is received in coherence with the objectives transmitted by the agent ENTi in S'4 (figure 1 1 ), this reward being able to be defined in a current time interval as follows:

[0456] • Ri4,int,c = 1 if Ci4,c = di4,c = gi4,c = 0 (do nothing) or if di4,c 0 and gi4,c = 1 or if Ci4,c + 0 and gi4,c = 2 and Ri4,int,c = 0 otherwise, where Ci4,c G Ci4 and gi4,c G Gi4 respectively designate the action chosen by the sub-agent S_ENTi4 to choose the energy source for recharging and the objective transmitted by the agent ENTi in S'4 (figure 11), both at the end of the current time interval. Steps S'2i4 to S'3i4 are iterated in S'4i4 until a stopping criterion is reached so as to select for each given state an optimal action a opti4,c among Ai4, i.e. an optimal vector value for Ci4,c, di4,c, qi4,c for given values ​​of pre, grid, prc,xpiace, Carbi4,c, chi4,c, qcons,c, gi4,c, 6i4,c and hc.

[0457] Such optimal action has opt i4,c is recorded in S'5i4 in a dedicated memory so as to be possibly selected in the following time interval IT c +i in response to either exploitation or exploration or be used for the implementation of real-time steps S'6i4 to S'9i4 of the same type as the aforementioned steps S'8 to S'1 1.

[0458] We now describe, in relation to figure 15, the progress of an optimal energy consumption method, as executed by the sub-agent S_ENTi2o illustrated in figure 10, within the framework of the energy exchange method of figure 11.

[0459] The S_ENTi2o sub-agent designates an energy-consuming device, the use of which can be shifted over time.

[0460] Such an energy consumption process takes place as follows at the level of the energy consumption sub-agent S_ENTi2o, in a current time interval IT C .

[0461] In S'1 i2o, a given state of the energy exchange system is initialized in the current time interval IT C .

[0462] In a preferred embodiment, the Si2o state space configured for the system is for example as follows:

[0463] Si2o= 5i2o x Gi2o x H (60), where Gi2o and H is as defined above and 5i2o = {5i2o,k}i <k<M, où ôi2o,k représente un coefficient d’insatisfaction de l’utilisateur pour un kième sous-agent S_ENTi2o et M représente le nombre d’appareils dont l’utilisation est décalable dans le temps.

[0464] In S'2i2o, the sub-agent S_ENTi2o selects an action ai2o,c according to a compromise of exploitation of the learning result and exploration of the state space, with a given probability (for example 0.9 for exploitation and 0.1 for exploration). In a preferred embodiment, the action is selected in an action space Ai2o which is for example the following:

[0465] Ai2o ={ak}i <k<M, où ak G {0,1} où pour un kième appareil, 0 signifie que son utilisation est décalée dans un autre intervalle de temps et 1 signifie qu’il reste en fonctionnement.

[0466] In S'3i2o, the sub-agent S_ENTi2o receives a reward signal Ri2o,c which is for example the following in a preferred embodiment:

[0467] • ai2o is a predefined weighting coefficient,

[0468] • Ri2o,ext,c defines an extrinsic reward from the environment in response to the action taken by the subagent S_ENTi2o. It can be defined in a current time interval as follows:

[0469] • p i2o,k is the operating power of the kth device,

[0470] • a i2o,c is the action chosen at the end of the current interval IT C , it is a vector representing the decisions taken for all M devices: for example if the sub-agent S_ENTi2o is representative of M=5 devices and the action is to shift the use of each of them, then a i20 c = [0, 0, 0, 0, 0],

[0471] • £ £2 O is a weighting coefficient,

[0472] • Ri2o,int,c represents an intrinsic reward which can be defined as follows: and R i20, int.c = 0 otherwise, where gi2o,ce Gi2o is the objective transmitted by the ENTi agent at the end of the current time interval.

[0473] Steps S'2i2o to S'3i2o are iterated in S'4i2o until a stopping criterion is reached so as to select for each given state an optimal action a opt i2o,c among Ai2o, that is, an optimal vector value for ai2o,c of the M devices of the sub-agent S_ENTi2o, for given values ​​of ôi2o, gi2o,c, h c .

[0474] Such optimal action has opt i2o,c is recorded in S'6i2o in a dedicated memory so as to be possibly selected in the following time interval ITc+i in response to either an exploitation or an exploration or to be used for the implementation of real-time steps S'6i2o to S'9i2o of the same type as the aforementioned steps S'8 to S'1 1.

[0475] We now describe, in relation to figure 16, the progress of an optimal energy consumption method, as executed by the sub-agent S_ENTi2i illustrated in figure 10, within the framework of the energy exchange method of figure 11.

[0476] The sub-agent S_ENTi2i designates an energy-consuming device, the operating power level of which can be made variable over time. Such an energy consumption process takes place as follows at the level of the energy consumption sub-agent S_ENTi2i, in a current time interval IT C .

[0477] In S'1 i2i, a given state of the energy exchange system is initialized in the current time interval IT C .

[0478] In a preferred embodiment, the state space Si2i configured for the system is for example as follows:

[0479] Si2i= 5i2i x Gi2i x H (63), where Gi2i and H is as defined above and 6121 = {êi2i,k}i <k<N, où 5i2i,k représente un coefficient d’insatisfaction de l’utilisateur pour un kième sous-agent S_ENTi2i décalable en puissance et N représente le nombre d’appareils dont la puissance est variable dans le temps.

[0480] In S'2i2i, the sub-agent S_ENTi2i selects an action ai2i,c according to a compromise of exploitation of the learning result and exploration of the action space, with a given probability (for example 0.9 for exploitation and 0.1 for exploration). In a preferred embodiment, the action is selected in an action space Ai2i which is for example the following:

[0481] Ai2i ={Pk}i <k<N, où Pkest un vecteur qui désigne les valeurs de puissances possibles pour le fonctionnement d’un kième appareil du sous-agent S_ENTi2i.

[0482] In S'3i2i, the sub-agent S_ENTi2i receives a reward signal Ri2i,c which is for example the following in a preferred embodiment:

[0483] - Oi2i is a predefined weighting coefficient,

[0484] ■ Ri2i,ext,c defines an extrinsic reward from the environment in response to the action taken by the subagent S_ENTi2i. It can be defined in a current time interval as follows: Or :

[0485] • a i2i,c is the action chosen at the end of the current interval IT C , has i21 c being a vector of the decisions taken for all N devices of the sub-agent S_ENTi2i regarding the power level to use among the levels available for each: for example if the sub-agent S_ENTi2i is representative of 3 devices and the possible values ​​for each are [Pi, P2, P3, P4, Ps] and the action chosen at the end of the current time interval ITC is to use power level Pi for the first two devices and power level P3 for the third device, then the action will be written as follows: a i21 c = [Pi, Pi, P3],

[0486] • £ £2 I is a weighting coefficient,

[0487] • Pi2i,max is a vector representing the maximum operating powers for all devices whose power level can be adjusted,

[0488] Pi2i,int,c represents an intrinsic reward which can be defined as follows: and Ri2i, int.c = 0 Otherwise, where gi2i,ce Gi2i is the objective transmitted by the agent ENTi at the end of the current time interval.

[0489] Steps S'2i2i to S'3i2i are iterated in S'4i2i until a stopping criterion is reached so as to select for each given state an optimal action a opti2i,c among Ai2i , that is, an optimal vector value for ai2i,c of the N devices of the sub-agent S_ENTi2i, for given values ​​of 6121, gi2i,c, h c .

[0490] Such optimal action has opt i2i,c is recorded in S'6i2i in a dedicated memory so as to be possibly selected in the following time interval ITc+i in response to either an exploitation or an exploration or to be used for the implementation of real-time steps S'6i2i to S'9i2i of the same type as the aforementioned steps S'8 to S'1 1.

[0491] In another embodiment, one can consider an additional central entity located in the first hierarchical level L1 which is responsible for managing the correspondence between the offers and requests deposited in the Xplace exchange. It should be noted that in the aforementioned mathematical equations (1) to (65), all the terms are normalized.

[0492] It goes without saying that the embodiments described above have been given for purely indicative purposes and are in no way limiting, and that numerous modifications can easily be made by those skilled in the art without departing from the scope of the invention.

Claims

DEMANDS

1. A method for exchanging energy in a set of at least two entities communicating with each other via a communication network and being configured respectively according to at least one energy use profile from among a first energy production profile, a second energy consumption profile and a third energy storage profile, the method being characterized in that it implements the following in a current time interval, at the level of at least one of said entities: - receive (S1; S'1), via a receiving module of at least one of said entities, information relating to: * an energy quantity value which, according to the profile of said at least one entity, depends, within said interval, on the quantity of energy produced and / or consumed and / or stored, respectively, by at least one energy production and / or consumption and / or storage sub-entity associated with said at least one entity, * a value representing the price of energy available from the other entity and / or an external energy supplier audited together, - from said values, select (S2, S3a, S3b; S'2) where applicable an action from among an energy supply to said other entity or to the external energy supplier, via an energy supply point of said at least one of said entities, an energy request to said other entity or to the external energy supplier, via an energy delivery point of said at least one of said entities, said selection being implemented in accordance with a performance criterion (R1 i, c ; R2i, c ; Ri,c) of energy exchange.

2. Energy exchange method according to claim 1, wherein said energy exchange performance criterion (R1 i, c ) used minimizes the cost of energy required by said at least one entity in said current time interval and maximizes the profit of supplying energy to the other entity or external energy supplier in said current time interval.

3. An energy exchange method according to claim 1, wherein the information received further includes a carbon footprint value determined, within said current time interval, by said at least one energy production or storage sub-entity associated with said at least one entity, and transmitted by said at least one sub-entity to said at least one entity, and in which said selection of an action is further implemented from said carbon footprint value in accordance with a criterion of minimizing the carbon footprint of the energy to be supplied to the other entity or external energy supplier.

4. A method of energy exchange according to any one of claims 1 to 3, wherein the amount of energy to be consumed, in said current time interval, by at least one energy-consuming sub-entity associated with said at least one entity, is based on the energy consumption calculated at the level of at least one energy-consuming device that is associated with said at least one energy-consuming sub-entity, in accordance with the minimization of a dissatisfaction criterion of a user of said at least one device, said criterion being linked to: - either to a shift in the use of said at least one device to a time interval subsequent to the current time interval, if said at least one device has a time-shiftable usage profile, - or to a decrease in the power level of said at least one device in the current time interval, if said at least one device has a power-shiftable usage profile.

5. A method of energy exchange according to any one of claims 1 to 4, wherein said at least one energy-producing sub-entity associated with said at least one entity implements the following in said current time interval: - receive (S1 ii; S'1 ii), via a receiving module of said at least one energy production sub-entity, information relating to: -- a value of the price that energy available in the said current time interval costs from the other entity and / or an external energy supplier to said entity, -- the amount of energy stored, in the said current time interval, by at least one energy storage sub-entity associated with said at least one entity, if said at least one storage sub-entity is present, - based on the aforementioned price value and, where applicable, the aforementioned quantity of stored energy, and according to an energy production performance criterion: -- select (S3aii-S3dii ; S'2ii) a destination for the energy produced by said at least one energy production sub-entity in the current time interval from among said other entity or the external energy supplier, said at least one energy storage sub-entity associated with said at least one entity, at least one energy-consuming device associated with said at least one energy-consuming sub-entity, -- calculate (S3aii -S3dii ; S'2ii ) the amount of energy produced to be used according to the selected destination.

6. An energy exchange method according to any one of claims 1 to 5, wherein said at least one energy storage sub-entity associated with said at least one entity implements the following in said current time interval: - receive (S1 i3; S1 i4; S'2i3; S'2i4), via a receiving module of said at least one energy storage sub-entity, information relating to at least one value of the price of the energy available in said current time interval from an external energy supplier to said assembly, - from said price value, said quantity of energy stored in said current time interval, by said at least one energy storage sub-entity, and according to a performance criterion of the use of stored energy, select (S3ai3-S3di3 ; S3ai4-S3di4 ; S'2i3- S'4i3 ; S'2i4- S'4i4): -- an action relating to the non-recharging or recharging with energy of said at least one storage sub-entity, -- an action relating to the non-discharge or discharge of energy from said at least one storage sub-entity, - calculate (S4ai3-S4di3 ; S4ai4-S4di4 ; S'2i3- S'4i3 ; S'2i4- S'4i4) the amount of energy required for recharging, respectively unloading, if the action relating to recharging, respectively unloading, is selected.

7. Energy exchange method according to claim 6, wherein said performance criterion of the use of stored energy maximizes the life cycle of said at least one storage sub-entity.

8. A method for exchanging energy according to any one of claims 1 to 7, wherein the steps implemented by said at least one entity, said energy production sub-entities, energy consumption, energy storage, as well as said at least one energy-consuming device, are executed using a learning algorithm.

9. An energy exchange method according to claim 8, wherein the learning algorithm is of the reinforcement type, wherein: - said entities are agents, while said sub-entities and said at least one energy-consuming device are sub-agents associated with the agents, - The information received by the agents and sub-agents is representative of an environment in which the energy exchange process is implemented. - the aforementioned selected actions are decisions taken by the aforementioned agents.

10. Energy exchange method according to claim 9, wherein for at least one agent considered, said agent transmits (S'4) at least one objective to at least one sub-agent associated with it and which is in a given state, said objective being to be satisfied by said sub-agent and being integrated into said given state.

11. Entity (ENTi) configured to exchange energy with at least one other entity (ENTj), said entity and said other entity communicating with each other via a communication network and belonging to a set (E; Xplace) of entities configured according to at least one energy use profile from among a first energy production profile, a second energy consumption profile and a third energy storage profile, said entity being characterized in that it implements the following, in a current time interval: - to receive, via a receiving module of said entity, information relating to: * an energy quantity value which, according to the profile of said at least one entity, depends, within said interval, on the quantity of energy produced and / or consumed and / or stored, respectively, by at least one energy production and / or consumption and / or storage sub-entity associated with said at least one entity, * a value of the price that energy costs from the other entity and / or an external energy supplier audited together, - from said values, select where appropriate an action from among an energy supply to said other entity or to the external energy supplier, via an energy supply point of said at least one of said entities, an energy request to said other entity or to the external energy supplier, said selection being implemented in accordance with a performance criterion of energy exchange.

12. A computer program comprising program code instructions for implementing the energy exchange method according to any one of claims 1 to 10, when executed on a computer.

13. A computer-readable information carrier comprising instructions for a computer program according to claim 12.

Images