FR3056790A1 - MASKING A CONSUMPTION CURVE - Google Patents

Abstract

The invention relates to a method of masking the electricity consumption of a residential site (1) equipped with a plurality of electrical equipment (2-4) of different natures. The site comprises at least one counter (5) capable of measuring the electric current consumed by the site. The method comprises the steps of: obtaining (E6), by a masking device (6) connected to the counter (5), an initial general power consumption curve (CICE ') of the site over a first time window; generation (E9) of a masked power consumption curve (CMCE) of the site over the time window. - Restitution (E10) of the masked power consumption curve. The invention also relates to a device (6, 10) for masking the electricity consumption of a residential site (1).

Description

(54) MASKING OF A CONSUMPTION CURVE.

FR 3 056 790 - A1 _ The invention relates to a process for masking the electricity consumption of a residential site (1) equipped with a plurality of electrical equipment (2-4) of different types. The site includes at least one counter (5) capable of measuring the electric current consumed by the site. The process includes the steps of:

- Obtaining (E6), by a masking device (6) connected to the counter (5), of an initial general electrical consumption curve (CICE ') of the site on a first time window;

- generation (E9) of a masked curve of electrical consumption (CMCE) of the site on the time window.

- restitution (E10) of the masked electricity consumption curve.

The invention also relates to a device (6, 10) for masking the electricity consumption of a residential site (1) ·

i

Hiding a consumption curve.

Technical area

The invention relates to respect for privacy in home networks.

The invention relates in particular to an electrical network in the home of a user equipped with an electric meter.

The invention applies to a device connected to the electric meter of the home.

State of the art

We have recently seen the development, in the context of electrical distribution networks, of new generation electric meters known as Advanced Meter Management (AMM) or even Linky defined according to the standards of electricity distributors such as in France, EDF ( French Electricity). This type of equipment makes it possible to observe electrical signals from a home, or residential site, and to deduce therefrom a certain amount of private information concerning the habits and preferences of the people making up the household (the number of people whose composed the fireplace; the electrical equipment switched on at a given time; the number of people present in the fireplace at a given time; the habits of the people making up the fireplace, etc.)

Even with regard to first generation electric meters, it is possible to obtain this information by a simple connection to the electric meter, for example by using an amperometric clamp.

This information, if it can be used wisely by the distributor or by the user, can also be diverted for an intrusive commercial purpose (by the distributor) or for the purpose of a scam (burglary of the home, etc.)

The invention offers a solution which does not have the drawbacks of the state of the art.

The invention

To this end, according to a functional aspect, the invention relates to a process for masking the electricity consumption of a residential site equipped with a plurality of electrical equipment of different natures, comprising at least one meter capable of measure the electric current consumed by the site, the process comprising the steps of:

- obtaining, by a masking device connected to the meter, an initial curve of general electrical consumption of the site over a first time window;

- generation of a masked curve of site electrical consumption over the time window.

- restitution of the hidden electricity consumption curve.

By restitution is meant here the supply of the masked curve to another device which may for example be the electric meter.

Thus, the invention offers the advantage of presenting to the electric meter a modified consumption curve, generated from an initial consumption curve obtained for example downstream from the electrical panel of the home.

The initial curve of the electrical signals of the home being masked, we avoid a malicious use of this consumption curve, since a malicious person cannot deduce from the masked curve private information concerning the habits and preferences of the people composing the household.

According to a particular embodiment of the invention, a method as described above also includes the following steps before the step of generating a masked curve:

- obtaining an initial data structure representative of a set of events related to electrical equipment over the time window, from the initial curve of electrical consumption;

- for at least part of said events, application of a masking operation making it possible to obtain a masked data structure over the time window;

and in that the step of generating the masked power consumption curve of the site is carried out from the masked data structure on the time window

This embodiment of the invention makes it possible, from any electrical curve, representative of a continuous signal, to isolate the various electrical discontinuities and to deduce therefrom a data structure representative of electrical events of the residential site (switching on an appliance, putting it to sleep, switching off, etc.), then modifying this structure to reconstruct a modified electrical curve.

By event, here is meant a characteristic representative of an electrical appliance, for example a washing machine or a radiator, positioned over time.

By initial data structure is meant a structure corresponding to a set of events.

By masking is meant a modification of at least one of these events in the data structure.

By generation of the masked curve is meant the reconstruction of an electrical curve over time, generated from the events of the masked data structure. This curve is notably intelligible by the electric meter.

According to a variant of this particular embodiment of the invention, the step of obtaining the initial data structure of a method as described above also includes the following substeps:

- obtaining a database of signatures relating to the various electrical equipment on the site;

- segmentation of the initial electricity consumption curve into events relating to the various electrical equipment by comparison of the curve with the signatures of the base.

This variant of the invention makes it possible to compare the events which have been isolated in the curve with known signatures of electrical devices and therefore to recognize them. By signature is meant all of the electrical characteristics of a device when it is put into operation. For example, if one of the isolated events during the analysis of the general curve presents a consumption curve close to that of the signing of a washing machine, we can deduce, with a certain margin of error which can be controlled, that the appliance is probably a washing machine.

According to another variant of this particular embodiment of the invention, which can be implemented alternately or cumulatively with the previous one, the masking step also includes the following sub-steps:

- obtaining a database of profiles relating to the site's electrical consumption;

- modification of part of said events, to obtain a masked data structure as a function of at least one profile obtained.

This variant of the invention makes it possible to use known or acquired profiles to modify the consumption curve of the home while retaining a certain realism. For example, if the family consists of two adults and two children, a known profile of consumption of such a type of household makes it possible to generate a “neutral” curve to replace that of the household and thereby prohibit any malicious analysis of the household. in question. This type of profile may have been obtained beforehand from a server with such profiles, in the network, etc. A profile can also be acquired inside the site, to find out the behavior of the household to which the meter and the device are attached. Once this profile has been acquired and stored in the database, it can be reused to, for example, simulate the presence of the inhabitants of the household during a holiday period, a weekend, etc.

According to another variant of this particular mode of implementation of the invention, which can be implemented alternately or cumulatively with the previous ones, the method further comprises, prior to the masking step, the following steps:

- Obtaining, by the masking device, an initial curve of general electrical consumption of the site over a second time window;

- obtaining signatures of electrical appliances;

- segmentation of the initial electricity consumption curve into events relating to the various electrical equipment by comparison of the curve with the signatures obtained;

- modeling of events to obtain at least one behavior profile;

- memorization of said profile in a behavior database.

This variant of the invention makes it possible, by knowing the signatures of electrical devices, to build profiles to modify the consumption curve of the home while retaining a certain realism. For example, learning household habits can simulate the family's presence at home while on vacation, etc.

According to another variant of this particular embodiment of the invention, which can be implemented alternately or cumulatively with the previous ones, the masking operation consists in adding at least one event to the initial data structure.

This variant makes it possible to insert into the consumption curve fictitious consumption “peaks”, corresponding for example to the start-up of an electrical appliance of which the signature is known, in order to mask the real curve.

According to another variant of this particular implementation mode of the invention, which can be implemented alternately or cumulatively with the previous ones, the masking operation consists in removing at least one event from the initial data structure

This variant makes it possible to remove from the consumption curve identified events, corresponding for example to the start-up of an electrical appliance of which the signature is known, in order to mask the real curve. It makes it possible in particular to mask the real curve while providing the electric meter with an identical average power, certain events withdrawn coming to compensate for other added events, while assuring the holder of the subscription an electric billing corresponding to his real use, although this one is hidden.

According to another variant of this particular embodiment of the invention, which can be implemented alternately or cumulatively with the previous ones, the masking operation consists in calculating a transformation of the initial curve representative of several events.

This variant makes it possible to smooth the consumption curve of the identified events, over a certain period of time, in order to mask the real curve while providing the electric meter with identical power, and therefore ensuring the holder of the subscription a corresponding electric billing. to its actual use, although it is hidden.

According to another variant of this particular embodiment of the invention, which can be implemented alternately or cumulatively with the previous ones, the masking operation is followed by a step of dissipation of electrical energy corresponding to the masking.

This variant makes it possible to dissipate the energy which has been introduced by masking (for example by adding a consumption peak corresponding to the start-up of a given device) so as not to unnecessarily charge the holder of the subscription a fictitious consumption. Dissipation can be used to power another electrical device (the simulation of an electric heater can be used to power the water heater).

According to yet another variant of this particular embodiment of the invention, which can be implemented alternately or cumulatively with the previous ones, the masked curve has an average power substantially identical to that of the initial curve, so as to that the overall consumption of the household over the time window considered is substantially equivalent.

This variant allows masking of the consumption curve while ensuring the same power consumed, and therefore identical billing of the household over the time window considered, so as not to penalize either the household (by excessive billing) or the supplier. electricity (by too low billing of the home).

According to a material aspect, the invention also relates to a device for masking the electricity consumption of a residential site equipped with a plurality of electrical equipment of different natures comprising at least one meter capable of measuring the electric current consumed by the site, the masking device being connected to the meter, the masking device comprising the following modules:

- a module for obtaining an initial curve of general electrical consumption of the site over a time window;

- a module for generating a masked curve of the site's electrical consumption over the time window.

- a module for supplying the meter with the masked curve of electrical consumption

The term module can correspond as well to a software component as to a hardware component or a set of hardware and software components, a software component corresponding itself to one or more computer programs or subroutines or more generally to any element of a program capable of implementing a function or a set of functions as described for the modules concerned. In the same way, a hardware component corresponds to any element of a hardware assembly (or hardware) capable of implementing a function or a set of functions for the module concerned (integrated circuit, smart card, memory card, etc. .)

According to another material aspect, the invention also relates to a domestic gateway connected to a device for masking the electricity consumption of a residential site equipped with a plurality of electrical equipment of different natures comprising at least one meter capable of measure the electric current consumed by the site, the masking device being connected to the meter and to the domestic gateway, characterized in that said domestic gateway comprises the following modules:

- a module for obtaining, from the masking device or the counter, an initial curve of general electrical consumption of the site over a time window;

- a module for generating a masked site electrical consumption curve over the time window;

- a module for supplying the device for masking the masked curve of electrical consumption.

According to another material aspect, the invention also relates to a system composed of a domestic gateway as described above and of a masking device comprising:

- a module for supplying the meter with the masked curve of electrical consumption.

a masking module.

According to yet another material aspect, the invention also relates to a system composed of a domestic gateway as described above and of a masking device as described above, further comprising a module for storing and returning electricity.

According to another hardware aspect, the invention also relates to a computer program capable of being implemented in a masking system as defined above, the program comprising code instructions which, when the program is executed by a processor, performs the steps of the masking process.

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

This gateway, this masking device and this computer program have characteristics and advantages similar to those described above in relation to the masking process.

According to yet another material aspect, the invention relates to a recording medium readable by a data processor on which a program is recorded comprising program code instructions for the execution of the steps of the masking method defined above. above. The information medium can be any entity or device capable of storing the program. For example, the support may include a storage means, such as a ROM, for example a CD ROM or a microelectronic circuit ROM, or else a magnetic recording means, for example a floppy disk or a disc. hard. On the other hand, the information medium can be a transmissible medium such as an electrical or optical signal, which can be routed via an electrical or optical cable, by radio or by other means. The program according to the invention can in particular be downloaded from a network of the Internet type. Alternatively, the information medium can 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 process in question.

ίο

The invention will be better understood on reading the description which follows, given by way of example and made with reference to the accompanying drawings.

The figures:

FIG. 1 represents a domestic context for an example of implementation of a masking device according to the invention.

FIG. 2 schematically represents an initial curve of electrical consumption typical of a residential site.

FIG. 3 represents a curve of electrical consumption after masking by a masking device according to a first embodiment of the invention.

FIG. 4 illustrates a second embodiment of the invention.

FIG. 5 illustrates a third embodiment of the invention, for which the service gateway supports the masking method according to the invention.

FIG. 6 illustrates a flow diagram of the different steps of the method according to an embodiment of the invention.

Detailed description of an embodiment illustrating the invention

FIG. 1 represents a domestic context for an example of implementation of a masking device according to the invention.

The residential site (1) consists for example of a house, or domicile, in which one or more users of electrical equipment live. All of these users are grouped together under the name “foyer”. Alternatively, one could imagine a collective housing building, an office building or an industrial building, etc.

On this site are installed various electrical equipment (2 to 4), for example a lighting lamp (2), a washing machine (3), heating equipment (4), etc. These various pieces of equipment are connected to a private electrical network whose architecture is customary, this private network being itself connected to the electrical network of an energy distributor via a connection station (7), for example a low voltage electrical panel. This electrical panel is linked to the energy operator's counter (5). The device (6) according to an embodiment of the invention, hereinafter called masking device, is for example placed between the electric meter (5) and the electrical panel (7) to which the electrical appliances of the house are connected .

In this embodiment, the residential site is equipped with a service gateway, or “box” (10), controlling a home communications network (LAN) and connected to a broadband network (WAN, not shown). The masking device (6) according to the invention can be connected to this domestic gateway by a wired L link (PLC, Ethernet, etc.) or wireless (WiFi, Bluetooth, etc.). As will be described below, the domestic gateway can, depending on the embodiments and implementation variants, take over all or part of the invention (analysis, learning, masking, etc.).

The masking device (6) appropriates the Initial Curve of Electric Consumption, abbreviated CICE, measured by the meter, representative of the electrical activity of the site during a certain period of time, or time window of observation. The main function of the masking device is to generate a second curve, called the Masked Electricity Consumption Curve, abbreviated as CMCE.

The masking device (6) can be, according to a first embodiment, illustrated in particular in FIGS. 3A and 3B, very simple to implement. It can, for example, randomly generate current “consumption peaks” making it possible to mask the initial curve. According to other embodiments and / or variants, the masking device (6) can be more complex. In particular, in the second embodiment presented in support of FIG. 4, it comprises a module for observing the instant consumption of the home and a module for controlling electrical consumption, responsible for modulating the consumption curve intelligently , based in particular on the electrical devices present in the home and the use made of it by users.

FIG. 2 schematically represents an initial curve of CICE electrical consumption typical of a residential site.

This initial curve of electrical consumption received by the electric meter (5) is masked by the masking device (6) according to a first embodiment of the invention.

The CICE curve is made up of a number of peaks (P1-P4), or current levels, which lie above the quiescent current (the quiescent current can be defined as the observable current value when all the household electrical appliances are turned off). Each of these peaks, or notorious variation of the curve, represents what is called an "event", and is linked to a change in electrical state (switching on, switching off, change of state, of function ...). ) from one of the household electrical appliances. For example :

- at the time of the event Pl, the washing machine is started;

- at the time of event P2, the washing machine begins a spinning phase;

- at the time of event P3, the radiator is on;

- at the time of event P4, the radiator is switched off.

A malicious person can easily, using a device to measure the current (amperometric clamp, or ammeter, or voltmeter / multimeter), connect to the meter (5) located outside the house, and observe these electrical variations for fraudulent purposes.

The purpose of the masking device according to the invention is to avoid this situation.

According to a first embodiment of the invention which will be described in support of FIGS. 3A and 3B, the masking device does not need to observe this curve, nor to store it. He just has to modify it "on the fly" by superimposing fictitious peaks of electrical consumption.

According to other, more complex embodiments of the invention, which will be described in support of FIGS. 4 and 5, the masking device observes this curve and modifies it by taking into account the habits of the inhabitants of the house for regarding their use of electrical appliances.

FIG. 3B illustrates an example of a masked curve of electrical consumption as generated by the masking device (6) according to a first embodiment of the invention.

The masking device according to this embodiment can be very simple. It is preferably connected to a base of "signatures" (BS) which it can use randomly or not, each of these signatures simulating a change of state of an electrical appliance. Note that such a device can be an existing device in the house, or correspond to the simulation of any device, existing or not in the house, whose signature is obtained for example from a database on the network to which the service gateway (10) is connected. In other words, this base includes signatures characteristic of domestic electrical equipment corresponding to various current events in a household: switching on a curling iron, a radiator, switching off a television, switching on a washing machine from a "washing" state to a "spin" state (which consumes more current), etc.

According to this embodiment, the device is composed of a charging module (MC), responsible for generating the current curve after masking. The load module is able in particular to insert signatures (PO and P5 in FIGS. 3A and 3B) in the initial CICE curve at random or chosen moments, thus resulting in a new masked curve, CMCE, comprising “peaks” of current at the instants at which signatures are inserted. Ideally, the instant consumption curve is enriched by consumption peaks generated according to a model close to reality, that is to say corresponding to possible signatures at probable times. It is recalled that a signature became an event when it is positioned over time. According to the example in Figure 3A / 3B:

- at the time TO of the event PO, a new electrical consumption is simulated by the device, corresponding to the starting of a curling iron. The PO event corresponds to the insertion of the signature of the curling iron at the time TO, the most likely time (for example, at 8 am in the morning) for switching on such a device.

- at the instant T5 of the event P5, a new electrical consumption is simulated by the device, corresponding to restarting of the heating, most probable instant (for example, at 8 pm in the evening) for the activation of such a device.

According to the example shown in FIG. 3B, the initial curve is preserved and the simulated events are added to it at selected times.

According to a variant, the selection of the events can be random over time and / or in the choice of the event. The masking curve generated in this case corresponds to a random superposition of the initial curve and of the new events.

An observer using an amperometric clamp at the output (or alternatively at the input) of the counter (5) would therefore observe the composite, masked curve (CMCE), presented in FIG. 3B. This consumption at the meter output is therefore unusable for an individual or a malicious entity who would try to appropriate it.

Note that the current peaks added to the initial curve to mask it can disturb the customer's electrical billing. To this end, it will be possible, according to a variant, to select events so as to keep the same signal power over a given period of time, so as to keep the same average electrical billing. In another variant, it is possible to introduce a dissipative element, for example into the masking device, to absorb the added peaks. According to yet another variant, such a dissipative element can be used to supply an electrical device of the house such as for example a hot water tank, a radiator, etc.

FIG. 4 illustrates a second embodiment of the invention.

In this second embodiment, the masking device (6) has intelligent functions which allow it in particular to know the electrical appliances of the house, as well as their signatures, but also the behavior of the users of the household.

According to this embodiment, the device is composed of the following functional modules:

- a module for acquiring the electrical curve of the fireplace, denoted ACQ.

- a module for storing signatures of household electrical appliances, denoted BS; this module is identical to that described above in support of FIG. 3.

- an observation and learning module (MOA) responsible for learning and memorizing the behaviors of the home; this module observes the initial electricity consumption curve (CICE) over a certain period of time, for example a few days, or a few weeks, and deduces therefrom a certain number of events taking place over time with a certain recurrence; for example, if the “television on” event is detected several evenings at 8 p.m. and the dryer every Wednesday at 5 p.m., the module enters this information in the behavior database.

- a behavior database (BC) to record household behaviors that have been learned by the MOA module, so that they can be used again later

- the load module (MC), responsible for generating the current curve after masking. As in the previous embodiment, the instantaneous consumption curve can only be observed by the device (6), the latter being responsible for generating a masked consumption curve (CMCE). In this embodiment, the load module is based, thanks to learning, on realistic modeling; it is capable of generating and / or eliminating peaks in current consumption so as to simulate the "electrical behavior" of a real hearth.

- memories articulated around a processor (CPU). The memories can be of type ROM (from English Read Only Memory) or RAM (from English Random Access Memory), removable or fixed, etc. The random access memory makes it possible to store in a non-durable manner calculation data used during the implementation of a method according to embodiments. The non-volatile memory (for example of the EEPROM type) makes it possible to store for example a computer program according to an embodiment for its execution by a processor, as well as the BS and BC databases.

- a communication module (COM) which allows it to communicate with the service gateway (10), wirelessly via a radio channel of Wi-Fi, Bluetooth, NFC, Zigbee type (not shown), etc. or with wires via, for example, an Ethernet link or by carrier current (PLC).

- all of these modules classically communicate with each other via a data bus (B).

Thanks to this device, the learned consumption behaviors of the household are stored in the behavior database (BC) and reused in due time to generate a realistic masked consumption curve. Many variations of this principle are possible in this context.

In a variant of this embodiment, the masking device (6) also has a current storage module, which can for example be an inverter, or a set of batteries, denoted BA, able to store and then restore electric current, controlled by the charging module. Unlike the previously described modes, in which the masking device could only modify the initial curve received by the charging module by the insertion of additional events, this mode with storage module makes it possible to modify the curve received by the module more elaborate charge. It is in particular possible, thanks to this module, to:

- remove events from the initial curve; it suffices to "absorb" them, that is to say, to draw from the reserves of the inverter the amount of energy equivalent to the events to be removed, then to recharge the inverter over a sufficiently long period that 'there is no resulting consumption peak;

- transform the curve by transformations well known to those skilled in the art (average, integration, derivation, transformation of fast fourrier - FFT, etc.). In this way the inverter can in particular absorb the current peaks and smooth the curve of electrical consumption, transforming for example the graph of instantaneous consumption into a sinusoid which completely masks the initial curve without modifying certain characteristics (power, energy, etc. ) influencing in particular the invoicing of the household. The greater the capacity of the inverter device, and the longer the sinusoid can have, the more the curve is masked and therefore unusable for an intruder.

According to another variant, the masking device is equipped with a security element (for example a SIM card - Subscriber Identification Module, or a security software element of HCE type - Host Card Emulation) to protect the bases of behavior and signatures; in such a case, the databases can be saved in a wide area network (WAN), external to the local network, and accessed in a secure manner.

According to this embodiment and its variants, the different modules are part of the masking device (6).

FIG. 5 illustrates a third embodiment of the invention, for which the service gateway (10) supports part of the masking method according to the invention and of the corresponding modules.

The domestic gateway hosts, as before, the learning modules (MOA), signature storage (BS) and behavior (BC), as well as the part of the load module responsible for generating the masking curve (MC ' ). Only a hardware part of the charging module, able in particular to open / close the output of the inverter (relay R) is kept on the device, thus advantageously reducing the masking device (6) to its simplest expression. In FIG. 5, the module for acquiring the electric curve of the hearth, denoted ACQ, is positioned on the masking device but it could also, according to a variant, be located on the domestic gateway connected to the table (7).

Advantageously, moreover, if these modules are located on the domestic gateway (10), an interface can moreover be associated with it in order to propose different behavior profiles to the user, by means of signatures and behaviors obtained in the wide area network, via a module marked MR. The user could for example enter a profile corresponding to the composition of the household (number of adults, children, etc.), to the devices existing in the household, etc.

FIG. 6 illustrates a flow diagram of the various stages of the method according to the second or the third embodiments of the invention.

According to this example, the method links two distinct phases: a first learning phase, during which the two bases BS and BC are created and / or updated by the masking device; then a second masking phase proper, during which the initial curve CICE is transformed into a masked curve CMCE intended for the counter.

Learning phase (PO)

The initial step E1 corresponds to the construction of the base of signatures of the electrical devices (BS), real or simulated, of the household. In the real case, this step consists in recognizing the electrical appliances of the house and memorizing their signature in the base. In the simulated case, this step consists in appropriating the signatures of devices that could be present in the house, by accessing for example an external database of signatures, or descriptive manuals of electrical devices, etc. The signatures are stored in the signature database (BS).

The steps E2 to E4 which follow it correspond to the construction of the behavioral base (BC) of the home. Step E2 begins with the acquisition of the general electricity consumption curve of the subscriber's facilities, denoted CICE, in order to generate the basis of behaviors. It is carried out for example over a period of several days.

During a successive step E3, the CICE curve obtained is segmented by the observation module. This module recognizes the electrical uses (use of devices over time) of the home by comparison with the base of signatures previously constructed; this step requires the sub-steps well known to those skilled in the art of sampling the signal of the curve, followed by a step of discriminating each of the events relating to the electrical devices by comparison with the signatures available in the base (BS) . Such a process is described for example in French patent application FR2917161A1 from the company EDF. This step allows, for example, to result in the discrimination of events PI, P2, P3, P4 in the diagram of FIG. 2, and incidentally to recognize that an event corresponds to a particular electrical appliance.

During a step E4, starting from the segmentation into events, learning makes it possible to model behaviors of the members of the household on the electrical devices over time, that is to say to establish the consumption profile of the household. For example, an observation over several days of a segmented curve would lead to a profile reflecting all the behaviors of the users of a household, of the type:

- the washing machine is started every Wednesday at 12 noon;

- the television is switched on every evening at 8 p.m. and off at around 11 p.m .;

- the curling iron is used every morning between 8h and 8hl5;

- etc.

It will be noted that, alternatively, steps E2 to E4 can be replaced by the acquisition of a standard curve, comparable to the consumption curve of the household. For example, if the family consists of three adults, two children and has a certain income, we can construct a fictitious curve of electric consumption which can approach the real curve, and replace all or part of the real curve by this curve acquired.

Then during a step E5, these profiles are stored in the base of behaviors BC. Steps E1 to E5 can be repeated as often as necessary, thus making it possible to generate and consolidate the BS and BC bases over time.

Masking phase fPl)

The masking phase of the curve itself may or may not be distinct from the learning phase. In the embodiment presented in FIG. 6, the two phases are successive for reasons of simplicity. However, the two phases could be nested, and the masking step E9 could alternatively be inserted after step E5 directly.

Step E6 corresponds to the acquisition of an initial CICE 'curve over a certain period. Contrary to what has been described previously for step E2, this acquisition period can be short because it is not a question here of learning but of a masking which can be carried out on the fly.

Then during a step E8, a segmentation step similar to the step

E3 can be performed in order to isolate the different events of the CICE 'curve.

During a step E9, the electrical consumption curve is masked, in transient mode, over this period of time. This step can be preceded by a step E7 of acquiring the masking options. Indeed, in accordance with what has been described previously, it is possible to mask the initial CICE curve in different ways, which can be specified during this configuration:

- by simple addition of events corresponding to consumption peaks generated from the signature base and inserted in time randomly or calculated;

- by deleting events, if there is a storage and current restitution module (BA inverter);

- by transformation of one or more events, or even of the entire curve, over time (smoothing, average, Fourrier transformation, etc.)

- etc.

At the end of this step E9, there is a masked curve CMCE of the electrical behavior of the house. This curve can become the input and / or output curve of the electric meter during a restitution step E10.

It is possible to repeat steps E6 to E10 as often as necessary to perform the masking over a certain period of time.

It goes without saying that the embodiment which has been described above has been given for purely indicative and in no way limiting, and that numerous modifications can be easily made by those skilled in the art without departing from the scope. of the invention.

Claims (14)

Claims 1. A method of masking the electricity consumption of a residential site (1) equipped with a plurality of electrical equipment (2-4) of different natures, comprising at least one meter (5) capable of 5 measure the electric current consumed by the site, the process comprising the steps of: - Obtaining (E6), by a masking device (6) connected to the counter (5), of an initial general electrical consumption curve (CICE ') of the site on a first time window; 10 - generation (E9) of a masked curve of electrical consumption (CMCE) of the site on the time window. - restitution (E10) of the masked electricity consumption curve. 2. Masking method according to claim 1, characterized in that it further comprises the following steps before the step of generating a 15 hidden curve: - obtaining (E8) an initial data structure (P0-P5) representative of a set of events related to electrical equipment over the time window, from the initial curve of electrical consumption (CICE ⁷ ); 20 - for at least part of said events (P0-P5), application of a masking operation (E9) making it possible to obtain a masked data structure (ΡΌ-ΡΊ) on the time window; and in that the generation step (E10) of the masked electrical consumption curve (CMCE) of the site is carried out from the 25 structure of masked data on the time window. 3. Masking method according to claim 2, characterized in that the step of obtaining (E8) the initial data structure (P0-P5) comprises the following substeps: - obtaining a signature base (BS) relating to the various electrical equipment on the site; - segmentation of the initial electricity consumption curve (CICE ') into events (P1-P4) relating to the different equipment 5 electric by comparison of the curve with the base signatures (BS). 4. Masking method according to claim 2, characterized in that the masking operation (E9) comprises the following sub-steps: - obtaining a database of profiles (BC) relating to consumption 10 electrical site; - modification (E9) of part of said events (P0-P5), to obtain a masked data structure (ΡΌ-P'i) as a function of at least one profile obtained. 5. Masking method according to claim 2, characterized in that it 15 further comprises, prior to the masking step, the following steps: - Obtaining (E2), by the masking device (6), of an initial general electrical consumption curve (CICE) of the site over a second time window; 20 - obtaining (E3) signatures of electrical devices (El); - segmentation of the initial electricity consumption curve into events (P1-P4) relating to the different electrical equipment by comparison of the curve with the signatures obtained (BS); 25 - modeling (E4) of events to obtain at least one behavior profile; - memorization (E5) of said profile in a behavior base (BC). 6. Masking method according to claim 2, characterized in that the masking operation (E9) consists in adding at least one event (PO, P5) in the initial data structure. 7. Masking method according to claim 2, characterized in that 5 the masking operation (E9) consists in removing at least one event from the initial data structure. 8. Masking method according to claim 2, characterized in that the masking operation (E 9) consists in calculating a transformation (FFT) of the initial curve representative of several events. 9. Masking method according to claim 2, characterized in that the masking operation (E9) is followed by a step of dissipation of the electrical energy corresponding to the masking. 10. Masking method according to claim 2, characterized in that the masked curve (CMCE) has an average power substantially 15 identical to that of the initial curve (CICE ′), so that the overall consumption of the focus over the time window considered is substantially equivalent. 11. Device for masking (6) the electricity consumption of a residential site equipped with a plurality of electrical equipment of kinds 20 different comprising at least one counter (5) capable of measuring the electric current consumed by the site, the masking device being connected to the counter, the masking device comprising the following modules: - a module for obtaining (ACQ) an initial consumption curve 25 general electrical (CICE, CICE ') of the site over a time window; - a generation module (MC) of a masked curve of electrical consumption (CMCE) of the site on the time window. - a supply module (MC, R) to the counter (5) of the masked electricity consumption curve. 12. System composed of a domestic walkway (10) and a masking device (6): - the masking device comprising: i. a module for obtaining (ACQ) an initial curve of general electrical consumption (CICE, CICE ') of the site over a time window; ii. a masking module (R). iii. a supply module (R) to the counter (5) of the masked electricity consumption curve. - the gangway comprising: i. an obtaining module (E2), coming from the masking device (5) or from the counter (5), of an initial general electrical consumption curve (CICE) of the site over a time window; ii. a module for generating (MC) a masked curve of electrical consumption (CMCE) of the site over the time window. iii. a supply module (MC) to the masking device (5) of the masked power consumption curve. 13. The system of claim 12 further comprising a module for storage and return of electricity (BA). 14. Computer program comprising code instructions for the implementation of the masking method according to claim 1, when the latter is executed by a processor. 1/3