EP3953875A1

EP3953875A1 - Method for evaluating photovoltaic energy production and evaluation and management unit implementing the method

Info

Publication number: EP3953875A1
Application number: EP20717881.5A
Authority: EP
Inventors: Valérick CASSAGNE; Mohammed CHIGUER
Original assignee: Total Renewables SAS
Current assignee: TotalEnergies Renewables SAS
Priority date: 2019-04-11
Filing date: 2020-04-10
Publication date: 2022-02-16
Also published as: CN113950693A; WO2020208237A1; US20220166217A1; JP2022528759A; JP7564124B2; FR3095067A1

Abstract

The invention concerns a method for evaluating photovoltaic energy production for an energy supply facility (11) installed on a site (1), the energy supply facility (11) comprising at least one electricity production unit (12) having at least one photovoltaic module (14); characterised in that the photovoltaic energy production evaluation method comprises the following steps: - determining past or predictive irradiance values for the site (1) on which the energy supply facility (11) is installed, the determination being carried out for time intervals of between one second and two years, and - evaluating the photovoltaic power generated as a function of the determined irradiance values and numerical values prerecorded in a database (30), this database (30) including, for each day of the year and different times of each day, prerecorded numerical values for determining the photovoltaic electrical power generated as a function of the irradiance.

Description

Title of the invention: Method for evaluating the production of photovoltaic energy and unit of evaluation and

management implementing the process

[1] TECHNICAL FIELD

[2] The present invention relates to the field of production evaluation

photovoltaic energy. More particularly, the present invention relates to a method for evaluating the production of photovoltaic energy and to an evaluation and management unit implementing this method.

[3] STATE OF THE ART

[4] Due to the increase in the cost of fossil fuels and the increase in

pollution generated by the consumption of these fossil fuels, we are turning more and more to renewable energy resources and energy consumption in a logic of sustainable development. This trend naturally leads to favoring renewable energies such as solar energy.

It is now conventional to install photovoltaic panels, in particular on the roofs of companies, public buildings, or simply on the roofs of private homes for self-consumption or to supply, for example, all production or only surplus energy to the public network. .

[5] Positive energy neighborhoods are currently being developed.

Such neighborhoods aim to be self-sufficient in electricity by generating the energy they consume in a renewable manner. We can cite, for example, the IssyGrid project which is an experimental project bringing together private homes and offices. For this project, various places were equipped in particular with photovoltaic modules in order to be able to produce electrical energy in order to meet the energy demands of the installations in this district.

[6] The various electrical devices of these installations are for example

interconnected in a network and can also be controlled in order to optimize their operation according to the quantities of electricity produced instantaneously and / or the quantities stored, for example in accumulators, while limiting the additional electricity from the public network to which the district is of course connected.

[7] There are also various control modules making it possible to limit the consumption of various electrical devices, in particular to control the optimized operation of an air conditioning or heating installation (reversible heat pumps), household appliances (washing machine, washing machine). - dishes, water heaters) or lighting.

[8] Thus, the goal of the IssyGrid project is to provide a neighborhood that can manage its

clean energy, and possibly able to inject surplus renewable electrical energy into the public electricity grid.

[9] In order to at least partially achieve such autonomy, it is therefore

necessary to be able to evaluate the photovoltaic power generated in a predictive manner, which makes it possible, in the event of insufficient production of renewable electrical energy in the face of high energy demand, for example to arbitrate between, on the one hand, an additional electrical energy by the public network or by electricity stored in accumulators and on the other hand, simply a deletion of certain consumers considered for example not critical.

[10] Indeed, it may be necessary to activate storage solutions,

warn the operator of the public network or draw on the storage units in the event that the quantity of renewable energy available or produced is insufficient or in overproduction to meet the energy demand of the various consumers in this district or to limit certain identified electricity consumers.

[11] Such a situation can, for example, arise in the case of a severe winter during which the renewable energy production capacities are much lower than the demands or on the contrary in summer with a high photovoltaic production and a summer consumption and weak daytime.

[12] Within the framework of the design of photovoltaic installations, a certain number of software programs are currently known making it possible to calculate a

forecast of renewable electrical energy production. This software is used to determine the dimensioning of photovoltaic installations for a future production site and generally take into account historical meteorological data. [13] These historical meteorological data include, for example, the levels of sunshine at different times of the year in the geographical area planned for this installation.

[14] These software are mainly developed to calculate the return on

investment and therefore the financial viability of such a project.

[15] In more advanced versions for projects with self-consumption, some software also makes it possible to take into account the consumer side and the electrical energy requirements of a more complex installation comprising electrical energy production units, electrical energy storage and electrical energy consumption units.

[16] However, this software is not suitable for reliably or practically predicting the photovoltaic power generated by the production unit for future periods, in particular from day to day and with a fairly precise sampling step in the day. Indeed, the photovoltaic power generated by the electricity production unit is dependent on instantaneous weather conditions. These weather conditions can vary from day to day or even during the same day.

[17] In addition, the calculations performed by this software are quite complex and long, which makes them unsuitable for carrying out a predictive evaluation allowing, for example, to adapt the energy supply or even the consumption of electrical devices in real time.

[18] In addition, known photovoltaic simulation software, initially

designed for an annual performance evaluation, often have a calculation step of the order of one hour and not adjustable, which is much too long for an intra-day evaluation where one needs to evaluate the forecast for a order of magnitude of the minute.

[19] Thus, these software, even if they are very efficient in modeling and in ergonomics, are not suitable for the prediction of the production of photovoltaic energy for short times, are not precise and consume in computing time.

[20] In order to be able to meet the energy demands of the various

installations in these positive energy districts, it is therefore necessary to be able to predict the photovoltaic power generated in real time, and possibly activate the supply or injection of electricity to these installations from the public network, activate storage units or control the reduction in consumption or the start-up of certain consumers or identified electrical devices in the district.

[21] In the case of hybrid electric systems, in particular with

thermal electric generators such as diesel generators, short-term photovoltaic prediction makes it possible to anticipate a drop in solar production due to a cloudy passage for example and to start one or more generators in advance and to avoid a forced cancellation or even a blackout . On the other hand, a reliable forecast helps to minimize the number of running generators, reduce fuel consumption and the need for maintenance while ensuring a reliable supply of electricity.

[22] One of the objects of the invention is therefore to provide a solution to be able to evaluate the photovoltaic production of an installation in real time without resorting to complex calculations.

[23] Another aspect relates to the control of the power generated for diagnostic purposes. Indeed, by evaluating a theoretical production of photovoltaic energy according to the real meteorological conditions in a fairly fine way for the production unit during a past time interval and by comparing this theoretical production with the real production of electricity by the production unit during the same time interval, it is possible to determine an operating state of the production unit and to plan, in the event of too great a deviation, maintenance operations for example which are necessary .

[24] Indeed, if the amount of electrical energy produced is too low compared to the expected power, it can be deduced that the production unit has a technical problem, such as, for example, a chain fault. modules, inverters, deposits of dust or dirt etc.

[25] The present invention therefore aims to overcome, at least partially, the problems of the prior art described above, by proposing a simplified method for evaluating the production of photovoltaic energy allowing monitoring, diagnosis or prediction. of photovoltaic energy production.

[26] DISCLOSURE OF THE INVENTION [27] To this end, the invention relates to a method for evaluating the production of photovoltaic energy for an energy supply installation located on a site, the energy supply installation comprising at least one electricity production unit having at least one photovoltaic module;

characterized in that the energy production evaluation method

photovoltaic comprises the following stages:

- determination of past or predictive irradiance values for the site on which the energy supply installation is located, said determination being carried out for time intervals between one second and two years, and

- evaluation of the photovoltaic power generated based on the determined irradiance values and digital values pre-recorded in a database, this database comprising for each day of the year and different times of each day digital values

pre-recorded to determine the photovoltaic electrical power generated as a function of the irradiance.

[28] The various steps of the process are for example carried out in a

automatic by an evaluation and management unit comprising processing means or a computer system such as a microprocessor or a

microcontroller. The capacity of these processing means may be limited due to the use of the database as will be described in the remainder of the description. In particular, the photovoltaic power values generated are read by the processing means and can be stored and subsequently processed.

[29] The use of a database, for example in the form of a table, in order to determine the photovoltaic power generated by the electricity production unit by the introduction of irradiance makes it possible to limit the powers of calculation necessary to determine this generated photovoltaic power.

[30] Thus, the use of such a database makes it possible to facilitate calculations and therefore allows a simpler and faster determination of the photovoltaic power generated at any time of the day. Furthermore, the use of such a database allows automation of the determination of the photovoltaic power generated by the electricity production unit. [31] The invention may further comprise one or more of the following aspects taken alone or in combination:

[32] According to one aspect, the digital values pre-recorded in the database correspond to photovoltaic electric power values generated as a function of irradiance values.

[33] The database includes, for example, for each instant listed in the database, electrical power values generated as a function of different irradiance levels, the irradiance levels being within a range between 0 and 1300 W / m ² , in particular in steps of 10 W / m ² .

[34] For an irradiance level which is not entered in the database and which is located between two levels entered in the database, the value of the photovoltaic electric power can be determined by linear interpolation of the values of photovoltaic electric power entered at the two irradiance levels entered in the database.

[35] According to another aspect, the digital values pre-recorded in the database correspond to the coefficients of a mathematical function describing the values of electric power generated as a function of

irradiance.

[36] The mathematical function can be a polynomial, in particular of degree 4.

[37] The coefficients of the mathematical function are in particular determined by least-squared regression of the electric power values generated as a function of different irradiance values, the irradiance values being for example within a range between 0 and 1300 W / m ² .

[38] The method may include a step of correcting the power

photovoltaic electricity generated as a function of irradiance, this correction step taking into account predictive, past or measured values of temperature and / or wind speed at the installation site by applying a correction coefficient.

[39] The correction coefficient is in particular a correction coefficient

linear.

[40] The pre-recorded numerical values are for example entered in the database at least between sunrise and sunset of each day with instants of the day spaced apart by a time between 5 min and 2 h, in particular equal to 1 h.

[41] For a moment of the day that is not entered in the database

data and which is between two instants entered in the database, the pre-recorded numerical values can be determined by

linear interpolation of the prerecorded numerical values indicated at the two times indicated in the database.

[42] The evaluation of the production of electrical energy is carried out in particular with a sampling interval less than or equal to five minutes, and in particular every second.

[43] The invention also relates to a method for forecasting the production of photovoltaic energy and for managing electrical devices for a

energy supply installation located on a site equipped with at least one electrical device, the energy supply installation comprising at least one electricity production unit having at least one photovoltaic module (14), comprising the steps :

- execution of a method for evaluating the production of photovoltaic energy as defined above, in which during the step of determining the irradiance values, predictive irradiance values for the site on in which the energy supply installation is located, said determination being carried out for a forecast time interval of between one second and two days, and

- adaptation of the operating mode of at least one electrical device according to the evaluation of the photovoltaic energy production of the energy supply installation.

[44] The invention also relates to a method for monitoring and diagnosing photovoltaic energy production for an energy supply installation located on a site, the energy supply installation comprising at least one production unit of energy. '' electricity having at least one photovoltaic module comprising the steps:

- execution of a method for evaluating the production of photovoltaic energy as defined above, in which during the step of determining the irradiance values, historical irradiance values are determined for the site on which the energy supply installation is installed for a past time interval, and

- comparison of the photovoltaic energy production evaluated with the photovoltaic production measured during the past time interval.

[45] According to a further aspect, the method further comprises a step of generating an alert if the difference between the energy production

PV evaluated and PV production measured during the elapsed time interval is above a predefined threshold.

[46] The invention also relates to an evaluation and management unit comprising means configured for the implementation of a method as defined above.

[47] The invention further relates to a computer program product that can be loaded into the internal memory of an evaluation and management unit comprising portions of software code for executing the steps of a method as defined. above when said computer program is executed by a computer

[48] Other characteristics and advantages of the invention will emerge from the

following description, given by way of example and without limitation, with reference to the appended drawings in which:

[49] [Fig.1] is a schematic perspective representation of a site equipped with a renewable energy supply installation;

[50] [Fig.2] is a schematic synoptic representation of a local electricity network comprising an installation for the supply of electrical energy,

[51] [Fig.3] is an example of an excerpt from a schematic representation of a database for evaluating the photovoltaic power generated as a function of irradiance levels and the time of year for the energy supply installation of the site in figures 1 and 2,

[52] [Fig.4] is an example of a graph showing for several times in the year the photovoltaic power generated as a function of the irradiance for the energy supply installation of the site of Figures 1 and 2,

[53] [Fig.5] is an example of an extract from a schematic representation of a database for evaluating the photovoltaic power generated as a function of the irradiance and the time of year for the supply installation energy of the site of Figures 1 and 2 with coefficients of a function

polynomial,

[54] [Fig.6] is a diagram to illustrate the constitution of the database

evaluation of a photovoltaic power generated by the energy supply system of the site in Figure 1

[55] [Fig. 7] is a flowchart of an exemplary embodiment of a process

evaluation of the photovoltaic electric power generated,

[56] [Fig. 8] is a flowchart of an exemplary embodiment of a method of forecasting photovoltaic energy production and management of electrical devices, and

[57] [Fig.9] is a flowchart of an exemplary embodiment of a photovoltaic energy production monitoring and diagnostic process.

[58] DETAILED DESCRIPTION

[59] In all the figures, the elements having identical functions bear the same reference numbers.

[60] The following embodiments are examples. Although the description is

refers to one or more embodiments, this does not mean

necessarily that each reference relates to the same embodiment, or that the characteristics apply only to a single embodiment. Simple features of different embodiments can also be combined or interchanged to provide other embodiments.

[61] Definitions:

[62] In the following description, the term “photovoltaic module” is understood to mean a most elementary unit for producing electrical energy (direct current), consisting of an assembly of photovoltaic cells interconnected with one another and completely protected from the external environment, that is, as defined by standard IEC-TS61836.

[63] The expression "real time" is understood in the following description to mean an evaluation sampling interval less than or equal to 10 minutes, and in particular less than or equal to 5 minutes, in particular every minute.

[64] The expression “irradiance” or irradiance according to standard ISO 80000-7 §19 in the following description is understood to mean a quantification of a power of electromagnetic radiation striking a unit area. The irradiance corresponds to the surface density of the energy flow arriving at the point considered on a surface. This surface density is expressed in Watt per square meter (W / m ² ) for the entire solar spectrum or a defined part of the spectrum. The irradiance can be in particular the global horizontal irradiance (GHI for "global horizontal irradiance" in English), the irradiance in a fixed or variable defined plane such as the plane of the modules, the diffuse irradiance and / or the direct irradiance normal or a set of this information.

[65] In French the word "temps" can be associated with chronology or with

meteorology. The word "weather" will be used in this description only for the description of temporal elements, the words meteorology or meteorological to describe the elements associated with meteorology.

[66] The term "meteorological parameter" includes any parameter linked to the

meteorology which can have an influence on the operation and in particular the performance of the photovoltaic modules, such as the irradiance value, the temperature and / or the wind speed at the installation site.

[67] "Time interval" means the time between the start and the end of the assessment period or the forecast period. The predicted time interval is the time between a predicted moment in the future and the present moment. If the forecast time interval is 36 h, this allows for example to have at the present moment a forecast of the values of a meteorological parameter between now and in 36 h.

[68] The term “sampling step” is understood to mean the duration between two instants, for example for the determination of irradiance values or the duration between evaluation points of the production of photovoltaic energy over the “time interval”. . Thus for a provisional time interval of 36h and a sampling step of 5 minutes, we calculate for between now and in 36 h for every 5 min the production of photovoltaic energy, i.e. 36x60 / 5 + 1 = 433 production values futures.

[69] Site:

[70] Referring to Figure 1, there is shown a site 1 for example of a commercial nature, such as a service station with in particular a sales building 3 and a fuel distribution structure 5 with a roof 7, for example in the form of a canopy. On this site 1 are also located various electrical devices 10.

[71] The electrical devices 10 are, for example, heat pumps, an air conditioning system, lighting and / or display devices, fuel dispensing pumps, or even vending machines.

[72] On this site 1 is also located an electrical energy supply installation 11 (see Figure 2) which comprises at least one electricity production unit 12 having at least one, preferably a multitude of photovoltaic modules 14 and an evaluation and management unit 16.

[73] Each photovoltaic module 14 of this electricity production unit 12 has known production characteristics, and in particular electrical power generated depending on the sunshine, more

specifically depending on the irradiance as well as for example the temperature. It is possible to use photovoltaic modules 14 of known technologies, for example modules with photovoltaic cells (not shown), for example based on crystalline silicon.

[74] As seen in Figure 1, the photovoltaic modules 14 are for example installed on the roof 7 of the structure 5 of fuel distribution.

[75] The electrical power supply installation 11 also comprises

optionally, a storage unit 18 for the electrical energy produced by the electricity production unit 12 and / or a thermal electricity generator 20.

[76] This storage unit 18 can for example be an electric battery or an electric energy accumulator.

[77] The thermal electricity generator 20 can be formed by a group

diesel generator that can be started to provide additional power, for example in the event of failure or absence of the public electricity network or in the event that it is economically more advantageous to start the generator set than to draw on the network public.

[78] Of course, other embodiments are envisaged where the site 1 can be a residential and / or industrial complex comprising individual or collective dwellings, offices, or even industrial buildings. [79] As shown schematically in Figure 2, the electrical energy supply installation 12 and the electrical devices 10 are for example interconnected in a local electrical network 22 controlled by the evaluation and management unit. 16.

[80] As will be detailed later, the evaluation and management unit 16 is configured to analyze the production of electrical energy from the electricity production unit 12, to analyze the electricity requirements of the electrical devices. 10 and for example to control the storage unit 18 and / or one or more thermal electricity generators 20 and the flow of electricity between the various units of the local electrical network 22. This local electrical network 22 is further connected to the public electricity network 24 which can receive the surplus electricity produced by the energy supply installation 12 or makes it possible to complete the supply or to replace it, completely or

partially, to the electricity production unit 12 as appropriate.

[81] To this end, the evaluation and management unit 16 is for example configured to activate / deactivate / control switches, relays and / or converters arranged in this local network 22 and which are not shown, and for manage the various flows of electricity.

[82] Thus, the electrical energy produced by the electricity production unit 12 can be consumed directly by the electrical appliances 10, be stored by the storage unit 18 or be supplied to the public network 24, for example. which is economically the most advantageous for the operator of site 1.

[83] The electrical energy stored by the storage unit 18 can for example be supplied to the electrical devices 10, especially in the event of insufficient electrical energy produced by the production unit 12.

[84] The electrical energy from the public network 24 can for example be supplied to the electrical appliances 10, in particular in the event of insufficient electrical energy produced by the production unit 12 and / or the thermal electricity generator 20 or else the electrical energy available in the storage unit 18.

[85] Finally the electrical energy produced by the local thermal electricity generator such as diesel generators which can be started up to provide additional power, for example in the event of failure or absence of the public electricity network 24, can also be supplied to electrical devices 10.

[86] So that the evaluation and management unit 16 can order such

arbitrations, it is therefore necessary to forecast in real time the production of photovoltaic energy carried out by the electricity production unit 12 which is highly dependent on the weather conditions on site 1, and to take into account the electricity needs of the different devices 10 consuming electricity, but also the conditions for buying back or selling electricity by the public network 24.

[87] In a more advanced variant, the evaluation and management unit 16 is

configured to more finely control the energy consumption of at least some of the electrical devices 10, for example to limit

consumption of certain electrical appliances 10 over a given period.

This makes it possible in particular to increase the energy autonomy of site 1 by limiting consumption on the public network 24 and thus optimize the energy bill of site 1. In the case of air conditioning for example, the evaluation unit and management 16 can be configured to raise the setpoint temperature, for example by one degree to limit consumption.

[88] The predictive aspect may also be necessary to warn the operator of the

public network 24 in order to be authorized to withdraw or inject a certain amount of electrical power from the public network 24 without penalty or simply to be able to start the thermal electricity generator 20 in time.

[89] To this end, the evaluation and management unit 16 is in particular connected by telecommunication means 26 to a meteorological evaluation system 28 configured to communicate predictive or historical meteorological parameters such as values of irradiance, the wind speed or even the temperature at site 1 on which the electricity production unit 12 is located. The meteorological evaluation system 28 is located remotely and for example comprises a remote server.

[90] The evaluation and management unit 16 is for example a computer equipped with memories and one or more processors or microcontrollers as well as communication means configured to communicate and control

the energy supply installation 11 as well as the electrical appliances 10, for communicating with the meteorological evaluation system 28 and also with, for example, the operator of the public network 24. This evaluation and management unit 16 can be installed on site 1 or be installed remotely.

[91] The management evaluation unit 16 is configured to implement a method for evaluating the production of photovoltaic energy of the energy supply installation 11 and in particular of the production unit of electricity 12 using a database 30 comprising, for each day of the year and different times of each day, prerecorded digital values making it possible to determine the photovoltaic electric power generated as a function of the irradiance.

[92] For this purpose, the evaluation and management unit 16 is configured to access a database 30, which can be stored in a memory of the evaluation and management unit 16 (as is schematically represented in FIG. 2) or be located on a remote server.

[93] According to a first embodiment, the digital values

pre-recorded correspond to electrical power values

photovoltaics generated according to the irradiance by the electricity production unit 12.

[94] An example of the presentation and content of this database 30 is shown in Figure 3 in the form of a table.

[95] This database 30 includes for each day of the year and for different times of each day photovoltaic power values generated as a function of irradiance.

[96] More precisely, this database 30 comprises on ordinates 32 in steps of 10 W / m ^2, for example irradiance levels. In the present example, the irradiance values range from 0 to 1300 W / m ² , in this case between 0 and 1100 W / m ² . This range of levels seems sufficient, because in general, the irradiance of a normal surface to direct sunlight is at most around 1000 W / m ² .

[97] On the x-axis 34 are reported times in the year listed by

example by a date and time. In this example, each day of the year is present in the table of the database 30 ^(January 1 (01/01) December 31 (31/12)), for each day, different times listed by one hour are listed, for example twenty-four times ranging from 00: 00h to 23: 00h in intervals of one hour.

[98] Of course, the table of database 30 can be modified without departing from the scope of the present invention. Thus, the time interval between the various instants can be longer or shorter than an hour and we can have more than twenty-four times per day or less. We can also implicitly neglect the night hours when the irradiance value is zero or negligible by listing in this table of this database 30 only times between sunrise and sunset which can easily be determined by calculations. astronomical.

[99] In the database 30 of FIG. 3 are therefore entered the values of photovoltaic power 36 generated as a function of the irradiance 32, at least between sunrise and sunset of each day with times of the day. spaced at intervals of time, preferably regular, comprised between 30 min and 2 h, in particular equal to 1 h.

[100] More specifically, each day of the year can be broken down into hours, ie 8760 time steps for a common year.

[101] Over a range of 0 to 1100 W / m ² and in steps of 10 W / m ² , for each level of irradiance and for each moment of the year listed in the database 30, a value is pre-recorded of power 36 corresponding to a photovoltaic power generated by the electricity production unit 12.

[102] Consequently, if we receive for example a global horizontal irradiance of 470 W / m ² on 01/01 / at 12:00, we can evaluate the photovoltaic power generated by the electricity production unit 12 to 3 305,573 kW for example. At 4 pm, this same irradiance produces a photovoltaic power of 2,325,012 kW.

[103] We can therefore understand that from an irradiance value, which can be a predictive or historical value, we can very simply assess the photovoltaic power generated by the electricity production unit 12.

[104] This database 30 therefore takes into account, depending on the date and the area

predefined geographical, for astronomical reasons, the angle formed between the direct light rays of the sun and the surface of the photovoltaic modules 14 which varies according to the seasons and the time of day. [105] If the value of irradiance 32 determined by the meteorological evaluation system 28 is not exactly one of the irradiance levels

prerecorded from the database 30, it is possible for example to perform a linear correction in order to approximate, for example by interpolation, the photovoltaic power 36 generated under these particular conditions.

[106] Let us take for example a predicted irradiance 32 of 468 W / m ² at a given hour of a day, such as for example at 10 am on July 20. In such a case, we can perform the following calculation:

[107] [Math. 1]

[109] In which:

[110] - P1 corresponds to the photovoltaic power 36 generated for a

irradiance 32 predicted at a given instant T; and

[111] - Ee corresponds to a predicted value of irradiance 32.

[112] According to the data in figure 3, on July 20 at 10 a.m.,

[113] Pl (E _e = 468 W / m ² ) = 3003057

[114] On the other hand, if the time chosen for the evaluation of the power is not an exact hour entered in the database 30 of FIG. 3, it is also possible to carry out a linear correction in order to determine the photovoltaic power 36 generated at this precise moment.

[115] For example, it is possible that the prediction of irradiance 32 will be made on July 20 at 10:15 a.m. In such a case, and in order to predict the photovoltaic power 36 generated at this precise time on July 20 and for a predetermined time interval, the following calculation can be carried out:

[116] [Math. 2]

[117]

[118] In which:

[119] - P2 corresponds to the photovoltaic power generated at a time

particular of the day for a given irradiance. [120] According to the data in figure 3, on July 20 for 468 W / m ² ,

[121] For 10:00 am:

[122] P2 (10 / i00) = Pl (E _e = 468 / m ² ) = 3003057 + ^ (3140604 - 3003057) =

3113094 / d / K

[123] For 11 a.m.:

[124] P2 (ll i00) = Pl (P _e = 468 W / m ² ) = 2947088

3051926 / d / K

and therefore for 10:15

15

P2 (10 / il5) = 3113094 + - (3051926 - 3113094) = 3097802 kW

60

[125] Since the corrections in this case are linear, they can be made first at the level of irradiance and then at the intra-interval time or in the reverse order.

[126] According to a second exemplary embodiment which makes it possible to further reduce the size of the database 30, the pre-recorded digital values correspond to coefficients of a mathematical function, in particular a polynomial function, for example of degree 4, describing values of electric power generated as a function of the irradiance in a range of irradiance values between 0 and 1300 W / m ² .

[127] For example, if we plot for different times in the year the photovoltaic power 36 values as a function of the irradiance levels 32, we obtain graphs according to Figure 4.

[128] Thus the curves 41 (triangles), 43 (diamonds), 45 (squares), and 47 (circles) respectively show photovoltaic power values 36 as a function of the irradiance levels 32 for 01/01 to 12: 00, 01/01 at 4:00 p.m., 07/20 at 10:00 a.m. and 07/20 at 11:00 a.m.

[129] By using for example an interpolation by a polynomial function of degree 4 as a mathematical function by a least squared regression, one can determine coefficients which will be pre-recorded as numerical values in the database 30 to calculate the photovoltaic electric power generated. according to the irradiance by the electricity production unit 12. [130] More specifically, the values of the modeling calculations described in the following paragraph illustrated graphically in FIG. 4, we can model the power 36 vs irradiance 32 relation by a polynomial of degree 4 by least squares regression. This polynomial is written

[131] P (Ee) = A x Ee ⁴ + B x Ee ³ + C x Ee ² + D x Ee + E

[132] We deduce from this regression the 5 coefficients A, B, C, D, and E which will be recorded in the database 30 for the moment of the year defined as illustrated in FIG. 5 showing an example of presentation and content of this database 30 in the form of a table with the different

coefficients.

[133] Also for this embodiment, this database 30 comprises for each day of the year and for different times of each day a set of coefficients, for example A, B, C, D, E making it possible to calculate PV power values generated as a function of irradiance using a mathematical function.

[134] To evaluate the photovoltaic power of the electricity production unit 12, for an irradiance 32 resulting from measurement or forecasting, it will suffice to recover the coefficients A to E corresponding to the time of the evaluation in the database. data 30 and calculate the power 36 with the polynomial using these coefficients. For example, to determine the power 36 on July 20 at 10:00 a.m., the coefficients A to E for this time of year extracted from database 30 are respectively: -4.6074x10 ⁶ , 3.1851x10 ³ , 3.4227x10 °, 4.7805 x10 ³ , 0x10 °. The photovoltaic power 36 for a GHI irradiance of 468 W / m ² will be:

[135] P (468) =: -4.6074x10 ⁶ x 468 ⁴ + 3.1851 x10 ³ x 468 ³ + 3.4227x10 ° x 468 ² + 4.7805x10 ³ x 468 + 0x10 ° = 3,092,387 kW

[136] Compared to the first embodiment, this method does not require an irradiance interpolation calculation to process a precise level of irradiance. On the other hand, it will always be necessary to make an interpolation to evaluate the power at a time between 2 time steps of the database as that was described above. In the example discussed above, to evaluate the power at 10:15 a.m. under a GHI irradiance of 468 W / m ² ,

[137] P2 (1 OhOO) = 3,092,387 kW [138] P2 (11 hOO) = 3,048,581 kW

[139] P2 (10 / il5) = 3092387

[140] That is to say only a variation of 0.5% compared to the first embodiment.

[141] This second embodiment of the database 30 describing the

behavior of the power generation unit 12 as a function of irradiance and time is more compact with only five pre-recorded digital values per time step instead of 111 data in the case of the base with a 10 W step / m ² between 0 and 1100 W / m ^2, i.e. a storage requirement of approximately 22 times.

[142] Other mathematical models than polynomials of degrees 4

can be used to describe the behavior of the power 36 as a function of the irradiance 32 using other mathematical functions or other regressions or other coefficient determination methods. The methods chosen will influence the size of the database and the computational capacity requirements.

[143] Thus, the photovoltaic production of the power generation unit 12 can be evaluated at any time of the year and for all irradiance conditions with precision.

[144] In order for this assessment to be as precise as possible, a

Particular attention has been paid to the calculation and the constitution of the database 30, which will be detailed below.

[145] Creation of database 30 and use of it:

[146] Figure 6 shows in a simplified and schematic way the blocks

functional making it possible to constitute the database 30. Block B1 represents the modeling of the energy supply installation 11 and in particular the electricity production unit 12 as a whole beforehand and block 2 represents a program of dimensioning or simulation known from the prior art such as the commercial products PVSYST, PVSOL or academic SAM or collaborative LADYBUG. Such a model takes into account a multitude of parameters such as in particular the characteristics of the photovoltaic modules 14, their location / orientation / inclination on site 1 and their number and efficiency. [147] The modeling also includes the description of the wiring of the various photovoltaic modules 14 together, or the description of the power conversion equipment (inverters, controllers, transformers, etc.) used for example.

[148] It can also take into account the environment such as far or near shading, self-shading.

[149] Once this modeling has been carried out, numerical values are calculated and recorded in the database 30 making it possible to determine the photovoltaic power generated by the electricity production unit 12 thus modeled for all the predefined moments of the year. (which is represented by block B3) and at the time step defined by the user or by the software and for irradiance values, for example the GHI, notably between 0 and 1100 W / m ² in steps of 10 W / m ² (exemplary embodiment of FIG. 3) or the coefficients of the mathematical function (exemplary embodiment of FIG. 5) which is represented by block B4, which makes it possible to obtain for example the tables of the database 30 of Figure 3 or 5 explained above.

[150] This calculation is therefore carried out in particular with software known from

sizing and modeling of photovoltaic installations taking into account the specific configuration of the electricity production unit 12 on site 1.

[151] These calculations, which are relatively consuming in terms of computing power and specialized software, are only carried out once to constitute the database 30. As described above, subsequent evaluations are carried out by simple reading or calculations. interpolation of values in said database 30. In addition, the constitution of such a database 30 is facilitated by the fact that the modeling of the energy supply installation 11 is often carried out at least once before its deployment and assembly on site 1, in particular within the framework of profitability studies.

[152] One of the advantages of using such a database 30 lies in the fact that the predictive or historical irradiance value directly takes into account certain meteorological phenomena such as, for example, the cloud layer.

[153] PV energy production evaluation process: [154] FIG. 7 shows an embodiment of a method for evaluating the production of photovoltaic energy for the energy supply installation 11 located on site 1.

[155] According to a step E 1, past irradiance values or

predictive for the site 1 on which the energy supply installation 11 is located, this for moments of the day spaced at intervals of time between one second and two days, or even two years. This step E1 is carried out in particular by supplying the irradiance values by the meteorological evaluation system 28 to the evaluation and management unit 16 using telecommunication means 26.

[156] As part of a predictive evaluation, the determination of irradiance and other meteorological parameters is carried out for forecast time intervals of between one second and two days, in particular less than 5 min. It is clear that the shorter the forecast time interval, the more reliable the forecast of irradiance values. Then the step between two forecasts of irradiance values can be chosen as required, for example a forecast every 10 s in the case of the management of a local network 22 as will be explained below.

[157] As part of a historical assessment, the meteorological parameters collected can span intervals ranging from a few minutes to a few days.

[158] The determination (prediction or collection) of meteorological parameters over a given forecast time interval is defined over a time step less than or equal to the forecast time interval typically given between one second and a few hours.

[159] Then during a step E2, the photovoltaic power generated is evaluated according to the determined irradiance values, the time point in the year of this evaluation and the generated power values pre-recorded 36 in the database 30. by simple reading or by calculation using the mathematical function and the coefficients as has been detailed above for the two embodiments, as the case may be by carrying out linear interpolations for the irradiance 32 or the instant in the year 34. [160] According to an optional step E3, a step is carried out for correcting the prerecorded values 36 in the database 30 of the photovoltaic power generated by taking account of predictive, past or measured values of the temperature (in particular temperature of the modules 14 or ambient temperature on site 1) and / or the wind speed on site 1 of installation by applying a correction coefficient, in particular linear.

[161] Indeed, the photovoltaic modules 14 and particularly the cells which compose them are the components whose conversion efficiency

Photovoltaic is linked almost linearly to the operating temperature of the cells. Secondly, ambient temperature can limit the conversion efficiency of energy converters such as inverters.

[162] The temperature can therefore for example be determined using a temperature sensor attached to a module 14 and connected to the evaluation and management unit 16 or be the result of weather forecast calculations transmitted by the meteorological rating system 28.

[163] By performing steps E1 and E2 or E1 to E3 in a loop with a step

of sampling less than or equal to five minutes, and especially every second, it is therefore possible to carry out precise and rapid photovoltaic energy production assessments.

[164] In the case where the sampling interval of the received values of irradiance, temperature or wind speed is greater than that of the

sampling evaluation calculations (step between two determinations of irradiance values longer than step between two evaluations of electricity production), linear interpolation can also be used as a first approximation to refine the calculations.

[165] The use of a database 30, for example in the form of a table, in order to determine the past or future theoretical photovoltaic power 36 generated by the electricity production unit 12 by simple determination ( predictive or historical collection) of irradiance makes it possible to limit the necessary computing powers to the strict minimum. Thus, the use of such a database 30 makes it possible to facilitate calculations and therefore allows a simpler and faster determination of the photovoltaic power 36 generated by the energy supply installation 11 at any time of the day. and the year. Use of this database 30 allows an almost instantaneous evaluation of the electrical power generated, which is not possible with the sizing programs known from the prior art.

[166] In addition, the use of such a database 30 allows a

automation of the determination of the photovoltaic power theoretically generated by the energy supply installation 11 with a step

fine sampling less than or equal to 5 minutes, and in particular every minute.

[167] The valuation calculations can also be done on a physical machine or a virtual machine on the cloud computing which is not always possible with for example commercial photovoltaic sizing software.

[168] These evaluations can be used for the management of the local electricity network 22, for example, or for an operational check of the electricity production unit 12.

[169] Management of a local electricity network 22:

[170] FIG. 8 shows an embodiment of a method for forecasting the production of photovoltaic energy and for managing electrical appliances which is for example implemented by the evaluation and management unit 16 for management of the local electricity network 22.

According to a step F1 which is similar to step E1 in FIG. 7, predictive meteorological parameters (irradiance value, temperature on site 1, wind speed on site 1 for example) are collected for site 1 where the energy supply installation 11 is located. Parameter values

predictive weather are for example provided by the system

meteorological evaluations 28. Steps E2 and E3 follow (E3 being optional) as described above in relation to FIG. 5.

Finally, during a step F4, the evaluation and management unit 16 is configured to control the local electrical network 22 according to the forecast of the photovoltaic energy production of the energy supply installation 11. .

[171] This control of the local electrical network may include one or more of the following actions in the non-exhaustive list, for example:

- adaptation of the consumption of one or more electrical devices 10, - storage in from the storage unit 18 and / or the injection of electric electricity from the storage unit 18,

- starting or stopping one or more thermal electricity generators 20,

- withdrawal or injection of electricity from or to the public network 24.

[172] Process for monitoring and diagnosing energy production:

[173] The agility, speed and low consumption of computing power also make it possible to use the evaluation process as part of a

monitoring and diagnosis of photovoltaic energy production for an energy supply installation 11 located on a site 1 as shown schematically according to an example in FIG. 9. The speed and frugality of calculation of the method makes it possible to monitor d 'a large number of sites with high frequencies with limited IT resources. According to a step G1 which is similar to step E1 in FIG. 7, historical meteorological parameters are collected (irradiance value, temperature on site 1, wind speed on site 1 for example) for site 1 where the energy supply installation 11 is located. Parameter values

historical meteorological data are for example provided by the system

28 weather evaluations working from satellite images.

According to a variant not shown, these historical values can be provided by sensors of site 1 and recorded for example in a memory of the evaluation and management unit 16.

Then steps E2 and E3 follow (E3 being optional) as described above in relation to FIG. 7. These steps E2 and E3 therefore make it possible to provide a reference value that the energy supply installation 11 would have. had to be produced depending on the irradiance and possibly the temperature.

Then, in a step G4, the evaluation and management unit 16 is configured to compare the photovoltaic energy production evaluated on the basis of the historical meteorological values with the actual photovoltaic production measured during a past time interval.

[174] Finally, according to a step G5, the evaluation and management unit 16 is configured to generate an alert if the difference between the energy production

PV evaluated and PV production measured during the elapsed time interval is above a predefined threshold. [175] Indeed, if the measured value is significantly lower than the theoretical value (for example 10%), it can be deduced that the energy supply installation 11 is subject to a malfunction such as the presence of dust or dirt on the photovoltaic modules 14, a fault in the chain of modules 14 etc. requiring maintenance.

[176] Historical meteorological parameters can be collected with great precision and / or with a large amount of data. A more detailed evaluation of the theoretical power over time or else an evaluation over a longer time interval is therefore possible with the same computing power.

[177] It is therefore understood that one of the determining aspects of the present invention lies in the use of a database 30 and in the choice of how this database 30 is organized in order to make it possible to evaluate the production of electricity with pre-recorded digital values making it possible to determine the specific photovoltaic power of installation 11 on site 1 and from a few variables such as a time of year for example (date + time making it possible to take into account the aspects astronomical and geometric) and irradiance (allowing meteorological and astronomical aspects to be taken into account) or coefficients as explained above.

[178] Moreover, in view of the iteration frequency of the photovoltaic energy production prediction method, even short disturbances

weather, such as the passage of clouds which would cause the irradiance to fall on the photovoltaic modules 14 of the production unit

of electricity 12 can be apprehended for the prediction of the photovoltaic power generated by the energy supply installation 11.

[179] Likewise, short meteorological disturbances will disturb less

evaluation of theoretical power with parameter readings

sufficiently frequent and close weather conditions during the time interval studied thanks to a short calculation time.

[180] Therefore, the evaluation calculations can be done on a physical machine or a virtual machine on the cloud. [181] The embodiments given above are given by way of illustration and are not restrictive.

Claims

[Claim 1] Method for evaluating energy production

photovoltaic for an energy supply installation (11) located on a site (1), the energy supply installation (11) comprising at least one electricity production unit (12) having at least one module

photovoltaic (14);

characterized in that the photovoltaic energy production evaluation method comprises the following steps:

- determination (E1, F1, G1) of past or predictive irradiance values for the site (1) on which the energy supply installation (11) is located, said determination being carried out for time intervals between one second and two years, and

- evaluation of the photovoltaic power generated (E2) as a function of the determined irradiance values and of digital values pre-recorded in a database (30), this database (30) comprising for each day of the year and different times prerecorded digital values for each day to determine the photovoltaic electrical power generated as a function of irradiance.

[Claim 2] Method for evaluating energy production

photovoltaic according to the preceding claim, characterized in that the digital values pre-recorded in the database (30)

correspond to photovoltaic electric power values generated as a function of irradiance values.

[Claim 3] Method for evaluating energy production

photovoltaic according to any one of the preceding claims, characterized in that the database (30) comprises for each instant listed in the database (30) values of electrical power generated as a function of different levels of irradiance, the levels

irradiance being within a range between 0 and 1300 W / m ² , in particular in steps of 10 W / m ² .

[Claim 4] Method for evaluating energy production

photovoltaic according to Claim 3, characterized in that for a level of irradiance which is not entered in the database (30) and which is located between two levels entered in the database (30), the value of the photovoltaic electric power is determined by linear interpolation of the power values electric photovoltaic system entered at the two irradiance levels entered in the database (30).

[Claim 5] Method for evaluating energy production

Photovoltaic according to Claim 1, characterized in that the digital values pre-recorded in the database (30) correspond to coefficients of a mathematical function describing values of electrical power generated as a function of the irradiance.

[Claim 6] Method for evaluating energy production

Photovoltaic according to Claim 5, characterized in that the mathematical function is a polynomial, in particular of degree 4.

[Claim 7] Method for evaluating energy production

Photovoltaic according to Claim 6, characterized in that the coefficients of the mathematical function are determined by least-squared regression of values of electrical power generated as a function of different irradiance values, the irradiance values being included in a range between 0 and 1300 W / m ² .

[Claim 8] Method for evaluating energy production

photovoltaic according to any one of the preceding claims, characterized in that it comprises a step of correcting the photovoltaic electric power generated as a function of the irradiance, this correction step taking account of predictive, past or measured values of temperature and / or wind speed on the installation site (1) by applying a correction coefficient.

[Claim 9] Method for evaluating energy production

photovoltaic according to the preceding claim, characterized in that the correction coefficient is linear.

[Claim 10] Method for evaluating energy production

photovoltaic according to any one of the preceding claims, characterized in that the prerecorded digital values are entered in the database (30) at least between lifting and sunset each day with moments of the day spaced between 5 min and 2 h, in particular equal to 1 h.

[Claim 11] Method for evaluating energy production

Photovoltaic according to Claim 10, characterized in that for an instant of the day which is not entered in the database (30) and which is situated between two instants entered in the database (30), the digital values prerecorded values are determined by linear interpolation of the prerecorded digital values entered at the two instants entered in the database (30).

[Claim 12] Method for evaluating energy production

photovoltaic according to any one of the preceding claims, characterized in that the evaluation of the production of electrical energy is carried out with a sampling interval of less than or equal to five minutes, and in particular every second.

[Claim 13] A method of predicting energy production

photovoltaic and electrical appliance management for an energy supply installation (11) located on a site (1) equipped with at least one electrical appliance (10), the energy supply installation (11) comprising at least one electricity production unit (12) having at least one photovoltaic module (14),

comprising the steps:

- execution of an evaluation process for energy production

Photovoltaic according to any one of claims 1 to 12, in which during the step of determining the irradiance values, predictive irradiance values are determined (F1) for the site (1) on which the installation of energy supply (11) is implemented, said determination being carried out for a forecast time interval of between one second and two days, and

- adaptation of the operating mode of at least one electrical device (10) based on the evaluation of the photovoltaic energy production of the energy supply installation (11).

[Claim 14] A method of monitoring and diagnosing photovoltaic energy production for an energy supply installation (11) located on a site (1), the energy supply installation (11) comprising at least one electricity production unit (12) having at least one photovoltaic module (14) comprising the steps:

- execution of an evaluation process for energy production

Photovoltaic according to any one of claims 1 to 12, in which during the step of determining the irradiance values, historical irradiance values are determined (G1) for the site (1) on which the installation of power supply (11) is implanted for a past time interval, and

- comparison (G4) of the photovoltaic energy production evaluated with the photovoltaic production measured during the past time interval.

[Claim 15] A method for monitoring and diagnosing photovoltaic energy production according to the preceding claim, characterized in that it further comprises a step of generating an alert (G5) if the difference between the energy production PV evaluated and PV production measured during the elapsed time interval is above a predefined threshold.

[Claim 16] Evaluation and management unit (16) comprising means configured for the implementation of a method according to a

any one of claims 1 to 15.

[Claim 17] Product loadable computer program in the

internal memory of an evaluation and management unit (16) comprising portions of software code for executing the steps of a method according to one of claims 1 to 15 when said computer program is executed by a computer.