ITRM20100621A1 - METHOD OF OPTIMIZATION OF DOMESTIC LOADS POWERED BY A RENEWABLE ENERGY PLANT, AND ITS OPTIMIZATION SYSTEM. - Google Patents

Description

Method for optimizing domestic loads fed with a renewable energy system, and related optimization system

The present invention relates to a method of optimizing domestic loads fed with a renewable energy system, and a related optimization system.

More precisely, the present invention relates to a method for managing energy savings in individual housing units or more complex buildings powered by photovoltaic or wind energy systems. The invention also relates to a system substantially represented by a home automation aid that is able to activate and deactivate the energy loads through the maximum rationalization of the energy available in the energy system.

There are many devices and methods for controlling the power produced by a photovoltaic system or other renewable source, which is normally subject to large variations in production due to the atmospheric state.

Documents US 2010236253 and WO / 2008/108137 describe a system for controlling the production of a photovoltaic plant based on an expected production value and monitoring the deviation of the present value (or value calculated on the basis of current measurements) from the expected one. .

The document describes a system for controlling the production of a photovoltaic system based on an expected production value and on monitoring the deviation of the current value from the expected one.

Document KR 20050121240 describes a system for controlling the production of a photovoltaic system based on the calculation of solar radiance.

Document WO 2009051870 describes a system for controlling the production of a photovoltaic plant based on the possibility of disconnecting individual modules and detecting maintenance needs.

Document KR 100725755 describes a control system of the type of the documents described above, plus an atmospheric station providing atmospheric data that is used for the control itself.

Documents are also known which indicate how to calculate meteorological indices (see for example EP1324253).

Documents:

- US2007 / 271006 A1;

- US2006 / 276938 A1;

- US2005 / 102068 A1;

- ARMAS J M ET AL: "A Heuristic Technique for Scheduling a Customer Driven Residential Distributed Energy Resource Installation", INTELLIGENT SYSTEM APPLICATIONS TO POWER SYSTEMS, 2009. ISAP '09. 15TH INTERNATIONAL CONFERENCE ON, IEEE, PISCATAWAY, NJ, USA, 8 November 2009 (2009 II 08), pages 17, XP031580112, ISBN: 9781 4244 5097 8;

- US2004 / 117330 A1;

- W02007 / 136456 A2;

disclose further known solutions which suffer from the same problems described above.

There remains the need for a complete system that, in addition to calculating the current and expected values of the generated power, is able to manage this power on the available load, during the period of time in which the alternative energy source is available.

The purpose of the present invention is to provide a method for optimizing loads, in particular domestic loads, fed with a renewable energy system.

A further object of the present invention is a system for optimizing loads, in particular household loads, supplied with a renewable energy system.

The object of the present invention is a real-time optimization method of loads, in particular household loads, powered with the aid of a renewable energy system, characterized by the fact that it uses:

- a microcontroller,

- a weather station close to that installation,

- one or more loads that can be activated and deactivated, · - a remote server for the pre-processing of the data necessary for the microcontroller;

and by understanding the following stages:

A. the microcontroller acquires the general information on said system through a man-machine interface and transmits them to said remote server via the internet;

B. the remote server, on the basis of said general information, takes from a specific database the information on the atmospheric parameters in ideal conditions essential to the calculation of the theoretical power Pt [Fo] of said system as a function of the time slot Fo, and calculates said power theoretical Pt [Fo];

C. the remote server retrieves the weather forecast data at the location of said plant, via the Internet, and organizes them in a matrix called

previsionaldatameteo, and transmits them to the microcontroller;

D. the microcontroller takes the weather data in real time from said local weather station and organizes them in the form of a matrix called

localdatameteo;

E. the microcontroller acquires through a man-machine interface the general settings of the loads in the form of a loadsettings matrix, on the basis of which an activation priority index of said loads is assigned, organized in the form of a matrix

arraypriorityloads [n], with respect to forced activation and / or according to a predetermined time schedule in the form of a set of matrices

timeschedulationloads;

F. the microcontroller calculates, on the basis of forecastedatameteo and localdatameteo, the matrix of maximum available power Peff [Fo] according to the production time bands of said plant;

G. on the basis of the Peff [Fo] matrix, the microcontroller calculates {previsionalbestpriceloadN) a FoN [T] matrix for each load N of said one or more loads, in which the time bands of activation index T of the same load N are indicated , thus indicating the best economic compromise between the maximum energy available and that used by the loads, and the transposed matrix with the data of the temporal activation of the loads

timescheduN [h] for activating loads one at a time;

H. the microcontroller, on the basis of the results of phase G, sends command signals on a serial bus to a transmission peripheral which in turn sends corresponding command signals to a receiving peripheral, which in turn sends corresponding command signals to said loads for their timed activation.

Preferably according to the invention, said system is a photovoltaic panel system and said general information includes:

- the geographical position identified by latitude and longitude;

- the physical position of the plant as angle and direction, Azimut;

- V-I characteristic of the panels, with respect to the operating temperature, and irradiation;

- peak power in peak Watts;

- measured power of the photovoltaic system

Pmeasure;

- type of panel

- current date.

Preferably according to the invention, in phase E the microcontroller:

- calculates the electric power curve available in ideal irradiation conditions Pt [Foj per time band Fo.

Preferably according to the invention, in phase G, on the basis of Peff [Fo], the microcontroller:

- considers cloudiness indexes by time slot acquired in phase D and associates each cloudiness index with a replacement value Dp [Fo] relative to the relevant time slot; - creates a matrix of effective power available with respect to the time band Pd [Fo] as:

Peff [Fo] = Dp [Fo]

- obtains for each electric load N-th the time band index matrix FoN [T], adding to each load a minimum absorbed power K as a safety band on the total power absorbed by said one or more loads;

- assigns several time bands, even simultaneously, to different loads,

- carries out the scheduling of the activations of the loads considering the priority of each load acquired in phase E and the available time bands.

Preferably according to the invention, the measured power of the photovoltaic system Pmeasure is measured by data communication between the microcontroller and the inverter of said system and compares it with the power required for each load PowerloadN, according to two alternative methods if Pmeasure <PowerloadN ·.

- the activation of said one or more loads is suspended, postponing the activation to a subsequent time slot in which Pmeasure> PowerloadN, until the completion of the work cycle of the load itself, or

- the activation of said one or more loads is suspended, and activation is postponed as soon as Pmeasure> PowerloadN until the completion of the work cycle of the load itself.

Preferably according to the invention, steps A to H are repeated at regular time intervals.

A further object of the present invention is a real-time optimization system of loads, in particular domestic loads, powered with the aid of a renewable energy system, characterized by the fact that it uses:

- a microcontroller,

- a weather station close to that installation,

- one or more loads that can be activated and deactivated;

- a remote server for the pre-processing of the data necessary for the microcontroller;

- power interfaces that activate and deactivate electrical loads;

and by the fact that the microcontroller comprises a code which, when executed, carries out the phases A, D, E, F, G, H, and the remote server comprises a code which, when executed, carries out the phases B, C according to the method object of the invention.

Preferably according to the invention, the system comprises, connected to the microcontroller:

- an interface for the local weather station;

- a transmission interface for a power line, which sends signals to the relative receiving interface;

- a USB interface;

- an ethernet interface;

- communication bus with the inverter of said plant;

- a GPRS interface for the auxiliary connection to the Internet.

A further object of the present invention is a serving computer, or server, characterized in that it performs steps B, C of the method object of the present invention.

A further object of the present invention is a computer program characterized in that it comprises code means suitable for executing, when operating on a computer, phases A, D, E, F, G, H of the method object of the invention.

The invention will now be described for illustrative but not limitative purposes, with particular reference to the drawings of the attached figures, in which:

Figure 1 shows an illustrative diagram of the system according to the invention;

Figure 2 shows a diagram of the system according to the invention;

- figure 3 shows an example of trends in the available power and the power of the loads, when the system according to the invention is applied;

Figure 4 shows a general flow diagram relating to the management method according to the invention;

- figures 5a, 5b, 5c show, in three parts, a flow chart relating to the management method according to the invention, which details a part of the flow chart of figure 4;

- figure 6 shows an example of the electrical characteristic I-V of a photovoltaic panel as a function of the temperature;

- figure 7 shows an example of I-V electrical characteristic of a photovoltaic panel as a function of solar radiation.

In the following, the structure and operation of the system of the invention will be illustrated with reference to a photovoltaic energy source, but it must be clear from now on that the system according to the invention can be used with any alternative energy source subject to variations, such as wind energy, simply by varying the times of application and related conditions.

Referring to Figure 1, the system according to the invention, hereinafter also called PSM ("Power Solar Manager"), is a hardware and software system 100, based on microcontroller 10, which combines the characteristics of a web server, that is device connected to the internet, with that of a measuring instrument, exercising a measurement and control of the photovoltaic energy available in a domestic system.

The system 100 crosses the data collected, through a remote server 70, from the forecast weather database 60 of the solar irradiation and temperature conditions with those of the local weather station 50 (forecast data correction), and through an appropriate algorithm calculates the maximum available electrical power provided (Peff) with an interval of 4-8 hours of photovoltaic energy, managing the activation ("timeschedulationloads") of the electrical loads that can be deactivated 30, such as: dishwashers, ovens, washing machines, irrigation systems, pumping. All aimed at the best energy compromise, making the most of the photovoltaic system, and creating the best economic compromise net of the energy introduced or taken from the grid ("previsionalbestpriceloadN").

The system according to the invention can be integrated into an intelligent structuring of the electrical distribution network such as Smart Grid 40 (transmission of maximum available power data, or information on the scheduling of electrical loads). This system can also manage systems based on small wind and solar thermal systems (by calculating the availability of the thermal fluid).

PSM HARDWARE STRUCTURE DESCRIPTION

The PSM hardware system is shown in figure 2. The structure includes the following components.

The heart of the PSM device is the microcontroller (10), which can be interfaced to the Ethernet network with remarkable reliability and modularity characteristics.

The program is inserted in the microcontroller (10) according to the software structure described below.

The following are connected to the microcontroller (10):

- an interface (51) for the local weather station; - an interface (31) for the power line;

- a Usb interface;

- an interface to communicate with the photovoltaic system inverter;

- an Ethernet interface, for connection to the Internet;

- a GPRS interface for connection to the Internet, if the Internet service is not available.

DESCRIPTION OF MANAGEMENT METHOD

The functions of the power management system according to the invention are described below from the software point of view according to the schematization of Figure 4.

Web server function and peripheral communication The system according to the invention is based on a hardware software structure such as Web Server (as described previously). The information to be processed (general system characteristics, forecast weather data, local station weather data, general load settings, data preprocessing from remote Server 70) and to be displayed are based on a web structure.

The system collects (man-machine interface, for example Web interface) and processes the general information of the system and data, through the protocols Tcp-Ip Server Http, Server FTP (proprietary libraries of the microcontroller).

The microcontroller 10 communicates with the local barometric station peripherals, power interface "Power line Interface Tx", with the photovoltaic system inverter to collect data, including the measured power of the Pmeasure system, through serial protocols such as Spi, I2c, owners of the microcontroller (10). The electric load activation and deactivation function is performed by the "Power line interface TX" (31) and Receiver (32) electric load activation signal transmission device via communication protocols, for example PLC (Power Line Communication, substantially based on data system / conveyed wave energy).

General plant information

Through the Web interface, the fundamental information of the solar photovoltaic system can be loaded, essential for the calculation of the irradiation to which it is subjected according to:

- the geographical position identified by latitude (Lat) and longitude (Long);

- the physical position of the plant as panel inclination angle (angleP) and direction (azimuthP);

- peak power in peak Watts (Wp);

- type of panel (Type, for example polycrystalline, monochitalline, amorphous, thin film);

- Current Date (Date), current time (Time).

This information will initially be transmitted from the microcontroller (10) to the remote server (70), this information will be used to:

- trace the V-l characteristic of the panels, with respect to the operating temperature (t) and irradiation;

- take from a specific database (PVGIS) the information on the maximum daily irradiation (for time bands Fo) expected in ideal conditions (for example table 13).

This database will be used for the calculation of the theoretical available electric power Pt [FO] through the "fit curve available power theoretical" of the PD algorithm (figure 5a), and for the preparation of the information suitable for the creation of the available power curve (Peff) of the 'plant.

Weather forecast data

This module incorporates the forecast data (matrix of "previsionaldatameteo" values, table 1) at 8 hours of cloud cover (clouds), or of irradiation (Irragg) and forecasted ground temperature (tempP).

This part of the method is performed by the remote Server; it collects weather forecast data from the temporal instant Ta (Sunrise time) to the temporal instant Tt (Sunset time) of the current day, taking the information on the forecasted cloud conditions and creating an index (Clouds) that varies from S (clear ) at MN (very cloudy) and predicted temperature (TempP).

Irragg Clouds TempP band

hourly

6-9

9-12

12-15

15-18

18-21

Table 1

The Clouds index can assume the following values: S clear

F haze

Partly cloudy PN

N Cloudy

MN = very cloudy

Each day there will be a band of solar irradiation that goes from sunrise to sunset, which is variable, with the respective times Ta, Tt.

This data will be used to calculate the effective maximum available power Peff using the PD algorithm. The forecast weather data will be sent to the microcontroller 10 via the FTP client of the remote Server, in a suitable format (for example .txt).

Local weather data

The real-time weather data (matrix of "localdatameteo" values, table 2) is taken from the local barometric station, atmospheric pressure rate (barometric) current temperature (temp), direction and intensity of the wind (wind) for the calculation of the temperature effective operation of the panels. These data will be used to calculate the effective maximum available power Peff [Fo] using the PD algorithm "fit curve avalaible power correct".

Real data of measured temperature (temp) and cloud tendency (CloudsCorr) are collected. The CloudsCorr indices are identical to and have the same meaning as those of Table 1 (Clouds).

The IrraggM irradiation is measured by the specific sensor.

Barome Irragg Clouds intensity W Direc Temp tric M Corr ind tion

wind

Val

hours

Table 2

Dynamic web page - web interface and "loadsettings" setting

The system may include the use of a Web page with information that is easy and usable to consult. From the Web interface {or other man-machine interface) it is possible to establish the "loadsettings", the index of priority of the loads "priorityloads") organized in the matrix "Arraypriorityload", with respect to the forced activation (manual), or according to the time scheduling of the electrical load established by the device (auto), maximum load power (powerloadN), activation mode of the electrical load scheduling (setloadN), number of electrical loads managed (numberloadP), and minimum power margin K that the system will yield to the external electrical network with respect to the load, to have information on the activation status of the loads (statusloadN). In addition, all the information and general characteristics of the system.

Maximum power available Peff [Fo]

Using the information relating to the forecast weather data, and to those of the local weather station, and to the general settings of the load, the PD algorithm calculates the maximum power curve available at 4-8 hours. This will serve to establish the best economic compromise (matrixes of "previsionalbestpriceloadN" values) and the temporal activation of the loads (matrixes of "timeschedulationloads" values).

Best economic compromise expected

Using the data provided by the general load settings modules, "loadsettings", and Peff [Fo], the method of the invention attributes "previsionalbestprice" activation matrices of FoN [T] and TimescheduN [h] values which determines the best compromise between the available energy Peff [Fo] and that used by the electrical loads, always supplying the surplus to the external electricity grid (80) connected to the photovoltaic system. This allows to make the most of the energy that the plant (20) produces and to make management economical, trying not to buy energy from the grid, creating the advantageous economic condition.

Hourly_ scheduling_ activation_ loads "timeschedulationloads"

Through the data provided by "loadsettings", Peff, "previsionalbestpriceloadN", the temporal activations of the loads are established according to the maximum available power foreseen and the best economic and energy compromise, with respect to the minimum / maximum work cycle of a load (washing machine, etc.) . The Arraypriorityload [n] matrix together with that created in "timeschedulationloads", TimescheduN [h], determines the signals that the microcontroller (10) sends as input to the peripheral (31), and then to the receiver (32) for activation and deactivation loads (30).

Real cost analysis "realbestprice"

This procedure can be envisaged within the method according to the invention and serves to establish whether the best economic energy compromise has actually occurred, by calculating the "realbestprice" and "data collection matrix" values (it is a matrix for the collection of all data relating to the method according to the invention).

The statistical and real-time data of the power used by the loads (powerM) and Peff are thus saved in a file that the microcontroller will send to the remote Server via its communication protocol via internet (for example Email), together with the scheduling matrix M which collects all the information deriving from "Timeschedulationloads", and "loadsettings" usable by future protocols for Smart Grid 40 (Intelligent Electric Grids).

Communication with the power interface

The method according to the invention comprises the use of serial type protocols suitable for communicating with the power interface ("power line interface TX") type Spi, I2c (software libraries and protocols, proprietary to the microcontroller).

The above specified the modules of the algorithm according to the invention. Below, with reference to figures 5a, 5b, 5c, we enter in greater detail the methodology for creating the available power curve of a system, for example photovoltaic, that is, it specifies the work of the modules starting from the remote Server block and from the subsequent one. Peff's calculation in figure 4.

Remote server "Remote server"

The remote server (70) prepares, on the basis of the general plant information (20), the information relating to:

1. Electric power curve available in ideal irradiation conditions Pt [Fo], weather forecast on cloud cover (Clouds); this information is processed daily and transmitted to the microcontroller (10) of the system according to the invention via the Ftp protocol;

2. During the self-calibration phase, DpP power rates are established by means of the Clouds cloudiness indexes (table 5), corresponding to the measured irradiation Irragg; these rates will correct the Pt [Fo] trend.

Theoretical available electric power curve under ideal irradiation conditions Pt [Fo]

From the general system data and the PVIGS database relating to irradiation in ideal conditions according to the daily distribution in time bands Fo (Gc ideal irradiation with a clean sky. Table 13), the expected temperature, the type of panel and the V-I characteristic (temperature) and V-I (solar irradiation) (figures 6 - 7), from the predicted temperature according to the weather forecast data tempP, the theoretical electric power is obtained, throughout the day, that the photovoltaic panel can supply, in theoretical conditions of irradiation at the expected temperature. The remote server will send to the microcontroller 10 an array that expresses the trend of the theoretical available power Pt with respect to the time band Fo, represented by the matrix Pt [Fo]:

Fo Ta Tt

Table 3

The creation of the theoretical available power curve Pt [Fo] is done considering the data of the hourly distribution of the theoretical clean sky irradiation Gc (table 13) with those of the plant's peak power Wp, through a "for" cycle that starts from Ta to Tt increasing every 30 '(time band Fo amplitude). The cycle restarts every new day.

Power rate DpP [Fo] compared to the weather forecast ("fit differential previsionai")

Evaluation of the power DpP [Fo] which will correct the theoretical available power Pt [Fo]. The evaluation must be made considering the Clouds cloudiness indexes (weather forecast data) to which a DpP power rate corresponds (as per Auto calibration) Through a "switch case" cycle (in C), a value is associated to each time slot which effectively replaces the Pt [Fo] with DpP [Fo] in relation to the time band of competence Fo. except for clear (S), in fact in the clear condition the original rate is re-established for that time slot, Pt [Fo]. The matrix with the DpP [Fo] (Table) 4 is constructed from Ta to Tt. The cycle restarts every new day.

DpP [Fo]:

Fo Ta Tt

DpP

Table 4

So ultimately DpP [Fo] represents the trend of the available power of the photovoltaic system considering the weather forecast data

Self-calibration

The self-calibration cycle takes place semi-automatically in the sense that through the interface (man-machine interface) during the system testing phase, the correspondences between the cloudiness indices and the irradiation measured by the local weather station are established. 'DpP power rate for each clouds index is evaluated by the Remote Server starting from the V-I curves (figures 6-7) of the photovoltaic panels, and then retransmitted to the microcontroller device (10) as general system data, table 4 represents the structure of the DpP with respect to the cloudiness index and corresponding irradiation, measured by the local weather station (50), also with respect to the temperature tempP predicted by the weather forecast. The correspondence between the measured irradiation and the expected cloudiness index and temperature is performed only in the testing phase. This operation determines the same and identical rates for the CloudCorr index at the temp temperature measured by the local weather station (50).

Cloud index Irradiation DpP at Forecast from weather measured temperature Clouds / CloudCorr (IrraggM) (tempP)

S I (S) DpP (S)

F 1 (F) DpP (F)

PN Ip (PN) DpP (PN)

N I (N) DpP (N)

MN I (MN) DpP (MN)

Table 5

Power rate Dp [Fo] compared to the data of the local weather station "Fit Differential previsionai correct".

The Dp [Fo] (table 6) is constructed considering the trend of the cloud index (cloudCorr) that the local weather station perceives (by evaluating the pressure rate, and with what intensity it varies, it is possible to establish the variations of the on-site cloudiness and the accuracy with which the predicted trend is valid), the construction on the time bands takes place with a shift of 3 hours ahead of the currentTime starting from Ta, up to Tt-2h. Starting from DpP [Fo], through a cycle (Table 5) "switch case", the DpP rates are replaced, but referring to the cloud index provided by the local weather station (cloudCorr), relative to the time slot Fo , except for clear (S), in the clear condition the original rate is re-established for that time slot, Pt [Fo].

The replacement takes place in real time for all the useful time (from currentTime onwards) that is from Ta + 3h to Tt-2h. The cycle restarts every new day.

Fo Ta + 3h Tt-2h

Dp

Table 6

Therefore Dp [Fo] represents the trend of the available power of the system considering the trends of the local weather parameters.

Example:

If the conditions are of clear sky (S) all day (from Ta to Tt) no correction Dp [fo] takes place, the correct predicted available power Peff [Fo] coincides with the theoretical one Pt [Fo].

Creation of the correct available power curve ("Fit curve avalaible power theoretical - correct") Peff [Fo].

Creation of the corrected available power curve Peff [Fo], considering the forecast cloud conditions according to a time band, with the indices specified in the weather forecast data module, goes from Ta to Tt, updating the actual power available matrix with respect to Peff [Fo] time slot:

Peff [Fo] = Dp [Fo]

Peff [Fo] array representation:

Fo Ta Tt

Pef f

Table 7

The curve is updated in real time, because the real-time substitutions of the Dp rates due to the data coming from the local weather station take place.

Allocation of time bands "PredictionalbestpriceLoadN" FoN [T]

The best time band for activating the loads (LoadN) depends on the characteristics of the loads and their settings (number of loads that can be managed by the device numberloadP, priority (priorityN)). From the table relating to the trend of the available power corrected with respect to the daily hourly distribution Peff [Fo], the time bands of maximum energy availability for that electrical load according to its power PowerloadN (maximum power absorbed by load N) are obtained:

PowerloadN K <Peff [Fo] (condition 1)

with K indicating the minimum margin of power transferred to the electricity grid (80) with respect to the load.

Therefore, this value will correspond to a time band index T, equal to the time band corresponding to Fo increased every 30 '(time band width Fo), where it is possible to power the electrical load N IoadN, according to condition 1.

Correspondence of time band FoN [T], with time band index T:

T 1 2 3 4 5 6 7 8 9 10 11 12 13 Fo Ta Ta + 30 'Ta + 60' Tt- Tt 30 '

The indexing of the time bands has been carried out to create the correspondence between the time band Fo which starts from Ta and arrives at Tt, and the incremental interval of 30 '(time band width), this width can be varied, and therefore different from 30 '. The maximum number of T indices depends on Ta, Tt and the time band width.

A matrix FoN [T] is thus constructed for each load (N) with PowerloadN power, in which the electrical load N (IoadN) is associated with each time band index T. The cycle starts from Ta to Tt and is updated every time the Peff [Fo] changes, in fact every 30 'because the weather conditions are updated.

For example, Fol [T] corresponds to the time band matrix Fo of index T, load 1 (Loadl).

Example in the case of 2 loads 13 time bands Fo:

Fol [T]:

T 1 2 3 4 5 6 7 8 9 10 11 12 13

Load 1 1 1 1 1

Table 8

Fo2 [T]:

T 1 2 3 4 5 6 7 8 9 10 11 12 13

Load 2 2 2 2 2

Table 9

The algorithm assigns time bands (of T indices), even simultaneously to different loads, then it will be the activation priorities of ArraypriorityLoad [n] in "timeschedulationloads" to establish the actual activation time band (Fo) of the electrical load N.

From FoN [T], we obtain the correspondence load index of activation time band TimescheduN [h], (transposed matrix of FoN [T]), of index h, the maximum number of h à ̈ to the number of activation time bands, attributed to that electrical load N. For example:

Electric load 1 with 5 indices of activation time bands T (load 1):

Timeschedul [h] array representation:

h 1 2 3 4 5

Timeschedul 3 5 6 9 10

Table 10

The time bands Fo available are progressing those represented by the indexes T 3,5,6,9,10.

The time band-electric load assignment varies if the available power curve Pef f [fo] varies.

If there are no available time slots, the service is not available, and a message is sent to the man-machine interface. The cycle returns to the initial condition.

The priority of the loads is organized in a matrix of priorities "ArraypriorityLoad [n]", where n is represented by the priority index of activation of the load, progressive, the maximum size of the matrix is the number of programmed loads (numberloadP) . For example:

Priority (n) 1 2 3 4

Load (N) Load (3) Load (1) Load (2) Load (4)

Table the

"Timeschedulationloads" scheduling The scheduling (activation of electrical loads according to priority) must be done considering the "loadsettings", the priority of the electrical load N (priorityN), then the matrix (ArrayprioriyLoad [n]). The cycle begins by assigning the first value of this initial index matrix n = l (which represents the highest priority), the load number N (IoadN) to be activated first, which will correspond to the time bands (index T) available (TimescheduN [h]). When the current time (CurrentTime) is in the corresponding time band Fo (of index T, established in "previsionalbestpriceloadN"), the electrical load is activated or deactivated according to the modalities of a specific "Cycle work IoadN" routine, until the completion of the duty cycle of the electrical load (CountdownN = 0).

All of the fundamental relative data on the characteristics of the loads and on the scheduling are collected in a scheduling matrix M:

nume powerloa priorit Countdow statusloa setloa ti ro dN yN nN dN dN me cari se co he du N

No.

Table 12

The status of the load N (statusIoadN) varies from wait / run / make depending on whether the load is waiting for activation / pause, execution (run), or work cycle performed (make), reporting the status to the man interface- machine.

The load scheduling cycle is governed by the priority matrix Arraypriorityloand [n], the cycle ends when all the loads have been scheduled (n = numberloadP). It restarts every new day.

Mode and cycle of activation of electric load N "Cycle work IoadN"

The mode of the work cycle and therefore activation of the electrical load N, can be established through the man-machine interface (setloadN) and depends on the priorities (priorityN) and the work time (countdownN), these modes are two FO and O (SetLoandN):

FO mode "Cycle work IoadN mode O"

In FO mode (setloanN = FO) the activation cycle takes place in the time band (Fo) of index T established in "Timeschedulationload", if there is a decrease in the measured irradiation rate (dlrragg / dt <0), determining a decrease in the power actually supplied by the photovoltaic system (DPmeasure / dt <0), and the measured available power (Pmeasure) is less than that necessary for the load (PowerloadN), the activation is suspended (statusloadN = wait) for that time band (Fo) and resumption in the activation time band subsequently available given by (TimescheduN [h]), increasing the h index. The load is activated if the current time (currentTime) is in the time band (Fo) of index T. The cycle is repeated until the complete execution of the work cycle of the electrical load N (CountdownN = 0, statusloadN = make) and then the index n of Arraypriorityload [n] is increased, passing to the scheduling of the next lower priority electrical load.

Mode O, "Cycle work IoadN mode O"

In mode 0 (setloadN = 0), the load N is activated in the set time band Fo of index T as in "Timeschedulationloads", if there is a decrease in the irradiation rate (dlrragg / dt <0), causing a decrease in the power actually supplied by the photovoltaic system (DPmeasure / dt <0), and the measured available power (Pmeasure) is less than that necessary for the load (PowerloadN), the activation is suspended (statusloadN = wait), and recovery (statusloadN = run) when the irradiation rate measured by the local weather station increases (dlrragg / dt> 0) (increase in the power that can be received from the photovoltaic system) if (Pmeasure) is greater than that required for the load (PowerloadN), the activation is resumed, until the complete work cycle of the load (CountdownN = 0, statusloadN = make) in this case the index n of Arraypriorityload [n] is increased, passing to the scheduling of the next electrical load minus priori tario. The activation of the electrical load is subject to the real availability of energy by the photovoltaic system 20.

The work cycle of each load ends when each countdownN = 0, the entire load activation cycle ends when the list of priorities has been completed and n = numberloadP, when all the loads have been scheduled.

Some examples of the results obtainable with the method according to the invention are provided below:

Example 1

Theoretical irradiation table and its hourly distribution:

Latitude: 40 ° 29'29 "North,

Longitude: 15 ° 59'45 "East

Results for January

Tilt of the plane: 35 degrees

Orientation (Azimuth) of the plane: 0 degrees

Time G Gd Gc A Ad Ac

(Fo)

07:37 34 34 21 18 15 11

07:52 45 44 28 24 21 15

08:07 54 54 34 31 27 19

: 22 63 62 40 37 33 23: 37 233 99 442 356 109 774: 52 264 108 507 381 116 826: 07 293 116 568 402 122 869: 22 320 124 625 419 128 904: 37 344 130 677 434 132 933: 52 366 135 725 447 136 957: 07 385 14 0 768 457 139 977: 22 403 144 806 466 142 994: 37 417 147 839 473 144 1010: 52 430 150 867 479 14 6 1020: 07 440 152 889 484 147 1030: 22 447 154 906 487 14 8 1030: 37 452 155 917 489 149 1040: 52 454 155 923 490 150 1040: 07 454 155 923 490 150 1040: 22 452 155 917 489 149 1040: 37 447 154 906 487 148 1030: 52 440 152 889 484 147 1030: 07 430 150 867 479 146 1020: 22 417 147 839 473 144 1010: 37 403 144 806 466 142 994: 52 385 14 0 768 457 139 977: 07 366 135 725 447 136 957: 22 344 130 677 434 132 933: 37 320 124 625 419 128 904: 52 293 116 568 402 122 869: 07 264 108 507 381 116 826: 22 233 99 442 356 109 774 15: 37 201 89 374 326 101 711

15: 52 54 54 34 31 27 19

16: 07 45 44 28 24 21 15

16: 22 34 34 21 18 15 11 16:37 22 22 14 11 9 7

Table 13

Where is it :

- G is the total irradiance of a fixed plane (w / m2); - Gd is the diffuse irradiance on a fixed plane;

(W / m2)?

- Gc is the total irradiance with clear weather on a fixed plane (W / m2);

- A is the global irradiance on a two-axis stakeout plane (W / m2);

- Ad is the diffused irradiance on a two-axis tracking plane (W / m2);

- Ac is the total diffuse irradiance on a two-axis stakeout plane (W / m2).

Example 2

Peff calculation on photovoltaic panel system Wp = 3Kw, with relative loadl and load2 scheduling matrix, panel temperature 18 ° C:

T Fo Gc Pt Peff

1 7.37 21 63 63

2 8 .07 34 102 102

3 8.37 442 1326 800

4 9.07 568 1704 900

5 9.37 677 2031 1300

6 10.07 768 2304 2000

7 10.37 839 2517 2200

8 11.07 889 2667 1500

9 11.37 917 2751 1100

10 12.07 923 2769 1500

11 12.37 906 2718 500

12 13.07 867 2601 1200

13 13.37 806 2418 2418

14 14.07 725 2175 2175

15 14.37 625 1875 1200

16 15.07 507 1521 90

17 15.37 374 1122 1000

18 16.07 28 84 84

19 16.37 14 42 42

Relative scheduling matrix M electrical loads 1 and 2, matrix M:

N powerIoadN priorityN SetloadN CountdownN etatualoadN timesched car uN ico

1 1,5 kw 1 FO Ih 40 'Run / make / [from wait T4 to 2 1,8 kW 2 0 40' Run / make / T8 / wait from T12 to T14] Example 3

With a plant of 6 KWp (peak power), the system predicts a trend for the maximum power available Peff, processing the data establishes that "load 1" (3kw washing machine), compared to the "best price forecast", or the best placement , is activated from 10:00 to 11:00. Therefore the curve of the loads PI is always below that provided by the system (Figure 3) where PowerM (Pmeasure) is the electrical power actually measured by the photovoltaic system and Pi is the electrical power used by the loads according to a average trend. Therefore, only the energy that the photovoltaic system provides is used, not taking energy from the grid (giving it a certain amount), with a safety margin K (minimum energy transferred to the grid with respect to the load).

Example 4

In a condominium of 5 apartments, an installation of 15KWp (peak power) is normally expected; by introducing the system according to the invention, it is possible to foresee an installed power of lOKWp, organizing the activation of loads using the method according to the invention (scheduling of loads with respect to priorities) with a considerable saving on the cost of the plant investment.

With the method and the system according to the invention, an optimized exploitation of the energy produced with a domestic renewable energy system is obtained, and therefore a considerable economic saving thanks to the lesser use of energy obtainable from the electricity grid.

In the foregoing the preferred embodiments have been described and variants of the present invention have been suggested, but it is to be understood that those skilled in the art will be able to make modifications and changes without thereby departing from the relative scope of protection, as defined by claims attached.

Claims (9)

CLAIMS 1) Real-time optimization method of loads, in particular domestic loads, powered with the aid of a renewable energy system, characterized by the use of: - a microcontroller (10), - a meteorological station (50) close to that installation, - one or more loads that can be activated and deactivated (30); - a remote server (70) for the pre-processing of the data necessary for the microcontroller (10); and by understanding the following steps: A. the microcontroller (10) acquires the general information on said plant through a man-machine interface and transmits them to said remote server via the internet; B. the remote server, on the basis of said general information, takes from a specific database the information on the atmospheric parameters in ideal conditions essential to the calculation of the theoretical power Pt [Fo] of said system as a function of the time slot Fo, and calculates said power theoretical Pt [Fo]; C. the remote server retrieves the weather forecast data in correspondence with the position of said plant, via the Internet, and organizes them in a matrix called weather forecast, and transmits them to the microcontroller (10); D. the microcontroller (10) takes the weather data in real time from said local weather station and organizes them in the form of a matrix called localdatameteo; E. the microcontroller (10) acquires through a man-machine interface the general settings of the loads in the form of a loadsettings matrix, on the basis of which an activation priority index of said loads is assigned, organized in the form of an arraypriorityloads matrix [ n], with respect to forced activation and / or according to a predetermined time schedule in the form of a set of timeschedulationloads matrices; F. the microcontroller (10) calculates, on the basis of forecastedatameteo and localdatameteo, the maximum power available matrix Peff [Fo] according to the production time bands of said plant; G. on the basis of the Peff [Fo] matrix, the microcontroller calculates (previsionalbestpriceloadN) a FoN [T] matrix for each load N of said one or more loads (30), in which the time bands of the T activation index are indicated same load N, thus indicating the best economic compromise between the maximum energy available and that used by the loads, and the corresponding matrix transposed with the data of the temporal activation of the load N timescheduN [h], for the activation of loads one at a time; H. the microcontroller (10), on the basis of the results of phase G, sends command signals on a serial bus to a transmission peripheral (31) which in turn sends corresponding command signals to a receiving peripheral (32), the which in turn sends corresponding final control signals to said loads (30) for their timed activation. 2) Method according to claim 1, characterized in that said system is a photovoltaic panel system and said general information includes: - the geographical position identified by latitude and longitude; the physical position of the plant as angle and direction, Azimuth; - V-I characteristic of the panels, with respect to the operating temperature, and irradiation, - peak power in peak Watts; - measured power of the Pmeasure photovoltaic plant; - type of panel; - current date. 3) Method according to claim 2, characterized in that in phase E the microcontroller (10): - calculates the electric power curve available in ideal irradiation conditions Pt [Fo] for time band Fo. 4) Method according to any one of claims 2 to 3, characterized in that, in phase G, on the basis of Peff [Fo], the microcontroller (10): - considers cloudiness indexes by time slot acquired in phase D and associates each cloudiness index with a replacement value Dp [Fo] relative to the relevant time slot; - creates a matrix of effective power available with respect to the time slot Pd [Fo] as: Peff [Fo] = Dp [Fo] - obtains for each N-th electrical load the time band index matrix FoN [T], adding to each load (30) a minimum absorbed power K as a safety band on the total power absorbed by said one or more loads (30 ); - assigns several time bands, even simultaneously, to different loads, - carries out the scheduling of the activations of the loads (30) considering the priority of each load acquired in phase E and the available time bands. 5) Method according to any one of claims 2 to 4, characterized in that the measured power of the photovoltaic system (20) is measured by means of data communication between the microcontroller (10) and the inverter of said system Pmeasure and compares it with the power required for each load PowerloadN (30), according to two alternative modes if Pmeasure <PowerloadN: the activation of said one or more loads is suspended, postponing the activation to a subsequent time band (FO) in which Pmeasure> PowerloadN, until the completion of the work cycle of the load itself, or - the activation of said one or more loads is suspended, and activation is postponed as soon as Pmeasure> PowerloadN (0) until the completion of the work cycle of the load itself. 6) Method according to any one of claims 1 to 5, characterized in that steps A to H are repeated at regular time intervals. 7) Real-time optimization system of loads, in particular domestic loads, powered with the aid of a renewable energy system, characterized by the use of: - a microcontroller (10), - a meteorological station (50) close to that installation, - one or more loads that can be activated and deactivated (30); - a remote server (70) for the pre-processing of the data necessary for the microcontroller (10); - power interfaces that activate and deactivate electrical loads; and by the fact that the microcontroller comprises a code which, when executed, carries out the phases A, D, E, F, G, H, and the remote server comprises a code which, when executed, carries out the phases B, C according to any one of claims 1 to 6. 8) System according to the preceding claim, characterized in that it comprises, connected to the microcontroller (10): - an interface (51) for the local weather station; - a transmission interface (31) for a power line, which sends signals to the relative receiving interface (32); - a usb interface; - an ethernet interface; - communication bus with the inverter of said plant; - a GPRS interface for the auxiliary connection to the Internet. 9) Computer program characterized in that it comprises code means suitable for executing, when operating on a computer, steps A, D, E, F, G, H of the method according to any one of claims 1 to 6.