EP3999861A1

EP3999861A1 - Method for determining electrical losses of an electric line, due to a malfunctioning or tampering of an electricity meter

Info

Publication number: EP3999861A1
Application number: EP20760939.7A
Authority: EP
Inventors: Giacomo VACCARIO; Alessandro TRAMONTI
Original assignee: Liso Srl; AGT SRL
Current assignee: Liso Srl; AGT SRL
Priority date: 2019-07-18
Filing date: 2020-07-17
Publication date: 2022-05-25
Also published as: IT201900012267A1; WO2021009787A1

Abstract

The present invention relates a method for determining electrical losses of at least one electric line (100) due to a malfunctioning or tampering of an electricity meter, reducing the number of false positives. A respective electricity meter is associated to each users and the users to which electrical losses are associated are identified by means of statistical analysis and the processing of measured data, taking into account the power absorbed by a plurality of users (Ui,U2...UN) connected to the electric line (100) and supplied by a transformer room (1) to said electric line (100) measured by a power meter (10), positioned on the electric line (100), between the transformer room (1) and the users of said plurality of users (U₁,U₂...U_N). The present invention also relates to a system for determining said electrical losses.

Description

METHOD FOR DETERMINING ELECTRICAL LOSSES OF AN

ELECTRIC LINE, DUE TO A MALFUNCTIONING OR TAMPERING OF

AN ELECTRICITY METER

The present invention relates to a method for determining electrical losses of an electric line, due to a malfunctioning or tampering of an electricity meter.

In particular, the present invention relates to a method that allows to determine possible frauds to the detriment of an entity responsible for the supply of electrical energy, significantly reducing the number of false positives.

Such frauds may be caused by tampering of the electricity meter.

With reference to the tampering of an electricity meter, in a first example, the electricity meter can be tampered by applying a magnet over it, so that electrical energy continues to be consumed but is not quantified correctly and therefore not paid by the user associated with such electricity meter.

In a second example, the electricity meter can be tampered with by manually changing the internal wiring so that the electricity meter measures a fraction of the power absorbed by the user.

In a further example, the electricity meter can be tampered with by modifying the software of the electricity meter itself, if the latter is an electronic meter .

An electricity meter allows to monitor the trend of flows of electricity consumed during the day, allowing to evaluate a possible tariff change or redistribution of daily electricity consumption.

However, if tampered, the electricity meter is no longer able to provide the actual value of the electric energy consumption.

Electricity meter is defined as an electric meter or an electronic meter by which it is possible to control the energy consumption of a user.

The electricity meter allows at any time to know how much electricity is consumed by a user and to measure the power absorbed by electrical appliances of said user. In fact, although the electricity consumed by a user is displayed by means of an electricity meter, the electricity meter also measures the power absorbed by said user.

Below, reference will be made to the power absorbed by a user as the power absorbed by the electrical equipment of said user.

Prior art

A method of known type for measuring the electrical losses due to a malfunctioning or tampering of an electricity meter is disclosed in the Italian patent application 102015000020863.

However, said method of known type has some drawbacks .

A drawback is due to the fact that the method is efficient when the consumption of electricity by a defrauder connected to the electric line is so high compared to the consumption of the other users connected to the same electric line, that the consumption of such other users is negligible.

Consequently, the greater the number of users connected to the electric line, the greater the difficulty of identifying a defrauder.

In fact, the method of known type is reliable when the number of users connected to the electric line is a reduced number of users, for example equal to twenty or thirty .

It should be taken into account that, in general, the users connected to an electric line are hundreds.

Therefore, it is difficult to identify a user connected to the electric line that consumes a power so high compared to other users connected to the same electric line that it can influence the trend of the electrical losses of the electric line .

In this situation, i.e. in the event that hundreds of users are connected to the same electric line, the method provides an anomalous result regarding the number of possible defrauders.

For example, the number of defrauders can reach a percentage equal to 50-60% of the number of users connected to the electric line, when in reality the number of defrauders is a lower or even zero percentage.

This means that the known method produces a significant number of false positives.

Aim of the invention

Aim of the present invention is to overcome said drawback by providing a method for determining electrical losses of an electric line due to a malfunction of an electricity meter or to a tampering with the same, which allows to significantly reduce the number of false positives so that said method has a higher efficiency than the known method mentioned above.

In particular, the method object of the present invention allows to identify one or more users whose respective electricity meter is malfunctioning or tampered with, where said malfunctioning or tampered electricity meter causes electrical losses of said electric line, with a high accuracy.

A second aim of the present invention is to provide a computer program for carrying out the steps of said method .

A further aim of the present invention is to provide a system for determining said electrical losses

Object of the invention

It is therefore specific object of the invention a method for determining electrical losses of at least one electric line due to a malfunctioning or tampering of an electricity meter, said at least one electric line is connected to a plurality of users and to a transformer room for transmitting electric current, through said at least one electric line, to said plurality of users, and to at least one power meter for measuring the power supplied from said transformer room to said electric line in time units present in a predetermined time period, where said power meter is arranged on said at least one electric line between said transformer room and said users.

Said method comprises the following steps:

A) acquiring a mean value of power s with reference to the power supplied from said transformer room to said electric line and measured by said power meter, so as to obtain a sequence of mean values of power in each time unit present in said predetermined time period ( s ) _k, with k=l , M where M is a positive integer, where k is the index of the number of the time units in the predetermined time period;

B) calculating the mean value s of the mean values of power with reference to the mean values of each sequence of power values (s)_k :

C) acquiring the power absorbed by each user in each time unit present in said predetermined time period, so as to obtain for each user a respective sequence of values of absorbed power (f_i)_k , with i=l , N where N is a positive integer, where i is the index of number of users :

D) calculating the mean value of each sequence of

values of power absorbed by each user: E) determining a mean value of efficiency e associated with said electric line in the predetermined time period with the following formula :

F) comparing said mean value of efficiency e with a predetermined efficiency threshold;

G) if the mean value of said efficiency e is less than said predetermined efficiency threshold, performing the following substeps:

Gl) calculating a sequence of difference values of power t referred to the values of power variation t_j supplied by said transformer room, with j=l , ..., (M-1), where M is a positive integer and j is the number of time units, and where each value of said sequence is obtained by the difference between the mean value of power s supplied by said transformer room in a time unit and the mean value of power s supplied by the transformer room itself in the previous time unit:

where

G2) calculating the mean value t of the difference values of power of the sequence of difference values t_j and the standard deviation d_t :

G3 ) calculating a respective sequence of the difference values of power g_i for each user, referred to the values of power changes (g_i)_j absorbed by each user, obtained by the difference between the value of power (/i)_{k +1} absorbed by a respective user in a time unit and the value of power (f_i)_k absorbed by the same user in the previous time unit:

where

G4 ) calculating the mean value of the difference values of power of each sequence of difference value g_i and the respective standard deviation :

G5 ) calculating a sequence of difference values V_J, where each value is equal to the difference between the value of variation of power supplied by said transformer room in a time unit and the sum of the variations of power absorbed by said user in the same time unit :

G6 ) calculating the mean value V and a standard deviation s_n referred to said sequence of difference values V_j :

G7 ) calculating the statistical correlation coefficient pi associated with each user between each sequence of difference values of power absorbed

by a respective user and said sequence of difference values V_j in M-l time units , with the following formula

H) comparing the statistical correlation coefficient

Pi associated with each user with a predetermined correlation threshold;

I) if one or more statistical correlation coefficients Pi associated to one or more respective users are greater than or equal to said predetermined correlation threshold, reporting said users to an entity responsible for the supply of electricity with reference to electrical losses due to a malfunction or tampering of one or more electricity meters associated with said users .

Preferred embodiments of the method are disclosed in the dependent claims

It is also object of the invention a computer program, comprising code means configured in such a way that, when executed on a computer, perform the steps of the method above mentioned.

Furthermore, it is also object of the invention a system for determining electrical losses of at least one electric line, to which said plurality of users is connected, where said system comprises said at least one electric line, as well as:

- a transformer room for providing electric current to said users of said plurality of users through said at least one electric line,

- at least one electricity meter associated with a respective user of said plurality of users for measuring the power absorbed by said user, said electricity meter measuring the power absorbed by each user of said plurality of users in time units present in a predetermined time period, so as to obtain for each user a respective sequence of values of absorbed power (f_i)_k , with i index of the number of users, i=l , N , with N a positive integer, and k index of the time units , k=l , M, with M a positive integer,

- at least a power meter for measuring the power supplied from said transformer room to said at least one electric line in time units present in a predetermined time period, so as to obtain a sequence of mean values of power ( s ) _k in each of said time units , where said power meter is arranged on said electric line between said transformer room and said users .

Said system comprises a central unit with a computer configured to receive said sequence of values of absorbed power (f_I)_K and said sequence of mean values of power

(s)_k, through a communication channel, and to run the program above mentioned.

Figure list

The present invention will be now described, for illustrative, but not limitative purposes , according to its embodiment, making particular reference to the enclosed figures, wherein:

Figure 1 shows a block diagram of the method according to the invention;

Figure 2 shows an example of application of the method object of the invention;

Figure 3 shows a Cartesian plane on which a logarithmic curve defining a predetermined correlation threshold of logarithmic type and the users connected to the electric line are depicted.

Detailed description of the invention

With reference to Figure 1, a method for determining electrical losses of an electric line, which are not due to the Joule effect.

In fact, it is known that an electric line has electrical losses due to the Joule effect.

However, the method object of the invention is intended to determine that electrical losses due to a malfunctioning of an electricity meter or to a possible fraud, by means of tampering of an electricity meter associated with a user .

In case of tampering of an electricity meter, the bill will show a consumption of electric energy lower than the actual consumption of electric energy.

In the embodiments being described, reference is made to an electric line 100 to which a plurality of users UI,U2...UN are connected (in the specific example described below, eight users Ui, U2...U8 will be considered) to which an electricity meter (not shown) is associated.

Furthermore, a transformer room 1 for transmitting electrical current to said users Ui, U2...U_N feeds the electrical appliances of said users, through said electric line 100, and a power meter 10 to measure the output power to said transformer room 1 (i.e. the power supplied by said transformer room to said electric line ) is arranged on said electric line 100 between said transformer room 1 and said users UI,U2...UN.

Transformer room is defined as any electrical unit capable of transmitting electric current which, via the electric line, reaches the users connected to said electric line .

Moreover, the measurement of power by said power meter 10 is performed in a plurality of time units present in a predetermined period of time.

Although not described, in the case of more transformer rooms and more electric lines, each of which is connected to a respective transformer room, a respective electric power meter is present for each electric line .

Said method comprises the following steps: A) acquiring a mean value of power s with reference to the power supplied from said transformer room 1 to said electric line 100 and measured by said power meter 10, so as to obtain a sequence of mean values of power in each time unit present in said predetermined time period ( s ) _k, with k=l , M where M is a positive integer, where k is the index of the number of the time units in the predetermined time period;

C) acquiring the power absorbed by each user UI,U2...U_N in each time unit present in said predetermined time period, so as to obtain for each user UI,U2...U_N a respective sequence of values of absorbed power (f_i)_k , with 1=1 , N where N is a positive integer, where i is the index of the number of users:

D) calculating the mean value of each sequence of

values of power absorbed by each user U₁,U₂...U_N:

E) determining a mean value of efficiency e associated with said electric line 100 in the predetermined time period with the following formula :

F) comparing said mean value of efficiency e with a predetermined first threshold or predetermined efficiency threshold;

Gl) calculating a sequence of difference values of power t referred to the values of power variation t_j supplied by said transformer room 1, with j=l , (M- 1), where M is a positive integer and j is the number of time units, and where each value of said sequence is obtained by the difference between the mean value of power s supplied by said transformer room 1 in a time unit and the mean value of power s supplied by the transformer room 1 in the previous time unit:

where

G2 ) calculating the mean value of the difference values of power of the sequence of difference values t_j and the standard deviation d_t :

G3 ) calculating a respective sequence of the difference values of power g_i for each user U₁, U₂...U_N, referred to the values of power variation absorbed by each user U_i, U₂...U_N, obtained by the difference between the value of power 1 absorbed

by a respective user in a time unit and the value of power absorbed by the same user in the previous time unit :

where

G4 ) calculating the mean value of the difference

values of power of each sequence of difference value gi and the respective standard deviation :

G5 ) calculating a sequence of difference values V_j, where each value is equal to the difference between the value of variation of power supplied by said transformer room 1 in a time unit , and the sum of the variations of power absorbed by said user U₁,U₂...U_N in the same time unit :

G6 ) calculating the mean value V and a standard deviation s_n referred to said sequence of difference values V; :

G7 ) calculating the statistical correlation coefficient pi associated with each user U₁,U₂...U_N between each sequence of difference values of power absorbed by a respective user U₁,U₂...U_N and said

sequence of difference values V_j in M-l time units, with the following formula:

H) comparing the statistical correlation coefficient r_i associated with each user U₁,U₂...U_N with a predetermined second threshold or predetermined correlation threshold; I) if one or more statistical correlation coefficients

Pi associated to one or more respective users UI,U2...UN are greater than or equal to said predetermined correlation threshold, reporting said users to an entity responsible for the supply of electricity with reference to electrical losses due to a malfunction or tampering of one or more electricity meters associated with said users .

With reference to steps A and B the power to which reference is made is the output power to the transformer room 1 which is measured by the power meter 10, placed between the transformer room 1 and the users U_I,U2...U_N connected to said electric line 100.

With reference to steps C and D, the power to which reference is made is the power absorbed by a respective user U₁,U₂...U_N, i.e. the power that is measured by the electricity meter associated with the respective user, proportional, according to a multiplicative factor, to the energy consumed by the user itself which is indicated by said electricity meter. On the basis of the energy consumed, the amount to be paid indicated in the bill is established.

The power absorbed by each user UI,U2...UN is also acquired in the same time units in which the power supplied by the transformer room 1 to the electric line 100 is acquired and measured by the power meter 10.

With reference to step E, the mean value of the efficiency associated with the electric line 100 is the first reference parameter for determining the presence of electric losses on the electric line due to a malfunction or tampering of an electricity meter.

With reference to step F concerning the comparison of the mean value of efficiency with a predetermined first threshold or predetermined efficiency threshold, it is preferable that said predetermined efficiency threshold is set equal to 0,9.

It should be taken into account that generally the electrical losses of an electric line not due to a possible fraud, such as electrical losses due to the Joule effect, are in percentage terms equal to approximately 7% of the power supplied by the transformer room to the electric line itself. In terms of efficiency, this means that the electric line has an efficiency of around 93%. Therefore, choosing a predetermined efficiency threshold equal to 0,9 means analysing electrical losses greater than 10%, i.e. electrical losses due to possible fraud.

In a first case, when the mean value of efficiency e is associated with the electric line is greater than or equal to a predetermined efficiency threshold, no electrical loss is reported to an entity responsible for the supply of electricity.

This means that no electrical loss has been determined due to a malfunction or tampering of an electricity meter associated with a user.

In a second case, when the mean value of said efficiency e is less than said predetermined efficiency threshold, step G comprises a plurality of substeps.

The first substep G1 is calculating a sequence of difference values of power t referred to the values of power variation t_j supplied by the transformer room 1

Each value is obtained by the difference of the mean value of power s supplied by the transformer room 1 in a time unit and the mean value of power s supplied by the same transformer room 1 in the previous time unit.

The second substep G2 is calculating the mean value t of the sequence of difference values of power t and its standard deviation d_t :

The third substep G3 is calculating a respective sequence of difference values of power g_i for each user U_I,U2...U_N referred to the values of power variation (,g_t)_j absorbed by each user U₁,U₂...U_N.

With reference to each sequence (the number of sequences is equal to the number of users), each value of each sequence . is obtained by the difference between the value of power absorbed by a respective

user in a time unit ad the value of power absorbed by the same user in the previous time unit.

The fourth substep G4 is calculating the mean value of each sequence and the respective standard deviation

The fifth substep G5 is calculating a sequence of difference values V· .

Each value of said sequence j is equal to the difference between the value of power variation supplied by the transformer room 1 in a time unit and the sum of the power variations absorbed by said user U₁, U₂...U_N in the same time unit :

The sixth substep G6 is calculating the mean value V and the standard deviation s_n of said sequence.

The seventh substep G7 is calculating the statistical correlation coefficient r_i associated with each user

U₁,U₂...U_N.

For each user U₁,U₂...U_N a respective statistical correlation coefficient will be obtained:

for the user Ui

With reference to the step H, the predetermined correlation threshold is the second reference parameter for determining the presence of electrical losses on the electric line due to a malfunction or tampering of an electricity meter.

The user to which the statistical correlation coefficient Pi greater than the predetermined correlation threshold is associated will be reported to the entity responsible for the supply of electricity .

With reference to the predetermined correlation threshold, said predetermined correlation threshold can be a correlation threshold of linear type or a correlation threshold of logarithmic type.

In the case of a predetermined correlation threshold of linear type, said predetermined correlation threshold can be chosen in a first range of values , from a value greater than or equal to 0,4 to a value equal to 1, or in a second range of values, from a value greater than or equal to 0,2 to a value less than 0,4, or in a third range of values, from a value greater than or equal to 0,1 to a value less than 0,2.

In the case of a predetermined correlation threshold of logarithmic type, said predetermined correlation threshold is defined by a logarithmic curve Y = a . log(x) + b, where a and b are predetermined constants chosen on the basis of the number of said users U₁,U₂...U_N in such a way that the slope associated with said logarithmic curve decreases when the number of said users U₁,U₂...U_N increases .

In particular, the parameter a is a constant between 0,02 and 0,2, preferably between 0,05 and 0,15, and parameter b is a constant between 0,1 and 0,3, preferably between 0,15 and 0,25.

The values of parameters a and b were obtained on the basis of a series of experiments with reference to an electric line to which a number of users between two users and three hundred users are connected.

The maximum number of users (i.e. three hundred users ) was chosen taking into account that on average the number of users connected to an electric line is equal to seventy-five users.

In this specific case, the method comprises the step of :

- calculating for each user a respective first parameter Q_i equal to the percentage value of the ratio between the mean value of each sequence of values of power absorbed by each user U₁, U₂...U_N and the mean value s with the following formula:

Furthermore, the step H comprises the following substeps :

- drawing on a Cartesian plane x, y a point for each user Ui, U2...UN, where the position of each user U₁,U₂...U_N on said Cartesian plane x, y is identified through a value of the x axis equal to the value of the respective first parameter Q_i and a value on the y axis equal to the value of the respective statistical correlation coefficient

Pi ^;

- representing on said Cartesian plane x, y said predetermined correlation threshold,

Consequently, with reference to step I, the users reported to an entity responsible for the supply of electricity are those to which respective statistical correlation coefficients greater than or equal to said predetermined correlation threshold are associated.

Notwithstanding the foregoing ( i . e . the predetermined correlation threshold can be of linear or logarithmic type), in a variant, the method according to the invention can provide both the application of a predetermined correlation threshold of linear type and the application of a predetermined correlation threshold of logarithmic type.

As a result , the method performs two comparisons:

• a first comparison between a statistical correlation coefficient (associated with a user ) with a predetermined correlation threshold of linear or logarithmic type, and

• a second comparison between the same statistical correlation coefficient with a predetermined correlation threshold of logarithmic type (when a predetermined correlation threshold of linear type was used in the first comparison) or between the same statistical correlation coefficient with a further predetermined correlation threshold of logarithmic type ( when a predetermined correlation threshold of logarithmic type was used in the first comparison) .

The second comparison allows to obtain more accurate results, reducing the error in determining the users to which electrical losses due to malfunctions or tampering of electricity meters are associated.

In the case in which the method provides a second comparison and in said second comparison a predetermined correlation threshold of logarithmic type is used (when in the first comparison a predetermined correlation threshold of linear type was used) or a further predetermined correlation threshold of logarithmic type (when a predetermined correlation threshold of logarithmic type was used in the first comparison) , the method comprises the following step:

Furthermore, said step H comprises the following substeps :

comparing the statistical correlation coefficient

Pi associated with each user U₁, U₂...U_N with a predetermined third threshold which can be a predetermined correlation threshold (when a predetermined correlation threshold of linear type was used in the first comparison), where said predetermined correlation threshold is defined by a logarithmic curve Y = a ^. log(x) + b , or a predetermined further correlation threshold (when a predetermined correlation threshold of logarithmic type was used in the first comparison) , where said predetermined further correlation threshold is defined by a further logarithmic curve Y' = a' . log(x) + b' , ,

- drawing on a Cartesian plane x, y a point for each user Ui, U2...UN, where the position of each user U₁,U₂...U_N on said Cartesian plane x, y is identified by a value on the x axis equal to the value of the respective first parameter and gives a value on the y axis equal to the value of the respective statistical correlation coefficient Pi ,

- representing on said Cartesian plane x, y said predetermined correlation threshold or said predetermined further correlation threshold.

The parameters a and b that define the logarithmic curve and the parameters a' and b' that define the further logarithmic curve are predetermined constants chosen on the basis of the number of said users U₁, U₂...U_N in such a way that the slope associated with said logarithmic curve or said further logarithmic curve decreases when the number of said utilities U₁,U₂...U_N increases.

Consequently, with reference to step I, the users reported to an entity responsible for the supply of electricity are those to which respective statistical correlation coefficients greater than or equal to said predetermined correlation threshold and to said predetermined further correlation threshold are associated.

With reference to the logarithmic curve (which defines the predetermined correlation threshold) , parameters a and b have already been described.

With reference to the further logarithmic curve (which defines the predetermined further correlation threshold) , at least one of the two parameters a' and b' is different from the parameter a and b respectively.

In any case, the two parameters a' and b' are predetermined constants with a value included in a respective range of values. In particular, the parameter a' is a constant between 0,02 and 0,2, preferably between 0,05 and 0,15, and parameter b' is a constant between 0,1 and 0,3, preferably between 0,15 and 0,25.

Consequently, although it is possible to choose the parameter a' and the parameter b' in the same ranges of values associated respectively with the parameter a and the parameter b , the parameters a' and b' are chosen in such a way that one of the two parameters is different respectively from the parameter a and from the parameter b or that both parameters a' and b' are different respectively from the parameter a and from the parameter b , so that the further logarithmic curve defined by parameters a' and b' is different from the logarithmic curve defined by parameters a and b .

In other words, one of the following conditions may occur :

i) a' is a constant different from a and is between

0,02 and 0,2, preferably between 0,05 and 0,15, and b' is a constant equal to b and is between 0,1 and 0,3, preferably between 0,15 and 0,25,

or

ii) a' is a constant equal to a and is between 0,02 and

0,2, preferably between 0,05 and 0,15, and b' is a constant different from b and is between 0,1 and 0,3, preferably between 0,15 and 0,25,

or

ill) a' is a constant different from a and is between

0,02 and 0,2, preferably between 0,05 and 0,15, and b' is a constant different from b and is between 0,1 and

0,3, preferably between 0,15 and 0,25.

Also with reference to said further logarithmic curve, the values of parameters a' and b' have been obtained on the basis of a series of experiments with reference to an electric line to which a number of users between two users and three hundred users are connected.

The maximum number of users (i.e. three hundred users) was chosen taking into account that the on average the number of users connected to an electric line is equal to seventy-five users.

When the standard deviation of one or more sequences of values of power absorbed { ( fi ) i, ( fi ) 2... ( f1 ) M; (f₂) 1, (f₂) 2... (f₂)M,·... ( f_N) 1, ( f_N) 2... ( f_N) M} from a respective user U₁, U₂...U_N is equal to zero, it is not possible to calculate the statistical correlation coefficient r_i of said one or more sequences of values of the absorbed power and the respective user U₁, U₂...U_N is ignored or reported to said entity responsible for the supply of electricity.

Furthermore, the present invention relates to a computer program, comprising code means configured in such a way that, when performed on a computer, they perform the steps of the method above mentioned.

Furthermore, the present invention relates to a system for determining electrical losses of at least one electric line 100, to which a plurality of users U₁, U₂...U_N is connected.

Said system comprises said at least one electric line, as well as: - a transformer room 1 for providing electric current to said users of said plurality of users UI,U2...U_N through said at least one electric line 100,

- at least one electricity meter associated with a respective user of said plurality of users UI,U2...U_N for measuring the power absorbed by said user, where said electricity meter measures the power absorbed by each user of said plurality of users UI,U2...U_N in time units present in a predetermined time period, so as to obtain for each user UI,U2...U_N a respective sequence of values of absorbed power(f_i)_k , with i index of the number of users, i=l,...,N , with N a positive integer, and k index of the time units, k=l,...,M, with M a positive integer,

- at least a power meter 10 for measuring the power supplied from said transformer room 1 to said at least one electric line 100 in time units present in a predetermined time period, so as to obtain a sequence of mean values of power (s) _k in each of said time units, said power meter 10 being arranged on said electric line 100 between said transformer room 1 and said users Ui, U₂...UN ·

Furthermore, said system comprises a central unit with a computer configured to receive said sequence of values of absorbed power(f_i)_k and said sequence of mean values of power (s)_k, through a communication channel, and to run the program above mentioned.

Example of method

Figure 2 shows an example of application of the method above described with reference to an electric line 100 to which eight users are connected: Ui, U₂,U₃,U₄,U₅,U₆,U₇,U₈.

A power meter 10 is positioned between a transformer room 1 which supplies electrical energy to said electrical line 100, to which said users U₁,

U₂, U₃, U₄, U₅, U₆, U₇, Us are connected.

The predetermined time period taken into account is a time period equal to 5 consecutive days and each time unit belonging to said predetermined time period has a duration of 15 minutes.

The power meter 10 acquires in said predetermined time period the mean value of the power supplied by said transformer room 1 in each time unit .

Mean values can be stored in a database or recorded in a format that allows said data to be reused.

During the same days, in the time units of the predetermined time period, the mean values of the power absorbed by each user are recorded, where said power is measured by the electricity meter associated with a respective user U₁, U₂, U₃, U₄, U₅, U₆, U₇, U₈.

The mean values of power absorbed by each user are stored in a database or recorded in a format that allows said data to be reused.

As mentioned above, choosing a predetermined time period of 5 consecutive days and a time unit of 15 minutes, the index k corresponding to the time units ranges from 1 to 480.

This is due to the fact that 96 time units of 15 minutes are present in one day and the result of the multiplication of 96 units of time for 5 days is equal to 480. With reference to steps A and C, in order not to make the reading heavier, a table containing all data used to implement the method above described is given below .

The first column of the table shows the mean values of power referred to the output power from the transformer room 1 measured by the power meter 10.

As already said, said power is the power supplied by the transformer room 1 to the electric line 100, to which the eight utilities U₁, U₂, U₃, U₄, U₅, U₆, U₇, U₈ are connected.

Each of said mean values of power was acquired in a respective time unit of 15 minutes.

From the second column to the ninth column of the table the values of each sequence of absorbed power values (f_i)_k associated with a respective user

U₁, U₂, U₃, U₄, U₅, U₆, U₇, Us are reported.

The power values of each sequence of power values are the values measured by the electricity meter associated with a respective user.

All the power values shown in the table are expressed in the same unit of measurement.

In the example being described, the unit of measurement is kilowatts (kW) . However, it is possible to use any unit of measurement, for example Joule, provided that all power values are expressed with the same unit of measurement.

The method will be described below with reference to its steps.

With reference to step A, the first column of the table shows a mean value of power s with reference to the power supplied by the transformer room 1 to the electric line 100 and measured by the power meter 10, so as to obtain a sequence of mean values of power (s)_k ( with the index k ranging from 1 to 480) in each time unit of 15 minutes present in the predetermined time period of 5 days . With reference to step B, the mean value s of the mean values of power referred to the mean values of each sequence of power values (s)_k is calculated.

Below the obtained result .

With reference to step C, from the second column to the ninth column of the table, the powers absorbed by a respective user and measured by the electricity meters associated with the respective user are reported.

The powers absorbed by each user are the powers absorbed in each time unit of 15 minutes present in the predetermined time period of 5 consecutive days, so as to obtain for each user U₁, U₂, U₃, U₄, U₅, U₅, U₇, U₈ a respective sequence of values of power absorbed by the same user:

With reference to step D, the mean value of each

sequence of values of power absorbed by each user U₁, U₂, U₃, U₄, U₅, U₆, U₇, U₈ is calculated

Below the obtained results.

With reference to step E, the mean value of efficiency is associated with said electric line 100 in the predetermined time period .

Below the obtained result .

With reference to step F, said mean value of efficiency must be compared with a predetermined efficiency threshold.

In this example, reference is made to an electric line 100 having an efficiency equal to about 93% in the absence of fraud. In the example being described, the predetermined efficiency threshold is equal to 0,9.

With reference to step G, in the example being described, the mean value of efficiency e is less than said predetermined efficiency threshold, therefore the following substeps are performed.

With reference to steps G1, G3 and G5, concerning the definition of the sequences of difference values, a table containing all the data calculated according to the method above described is given below.

With reference to step G2, the mean value t of the power difference values of the sequence of difference values t_j and its standard deviation d_t are calculated.

Below the obtained results.

With reference to step G4, the mean value of the

power difference values of the sequence of difference values g_i and its standard deviation are calculated.

Below the obtained results.

With reference to step G6, the mean value V and a standard deviation referring to said sequence of difference values _J are calculated.

Below the obtained results .

With reference to step G7, the statistical correlation coefficient r_i associated with each user

U₁, U₂, U₃, U₄, U₅, U₆, U₇, U₈ is calculated.

Below the obtained result.

With reference to step H, it was chosen to compare the statistical correlation coefficient with a predetermined correlation threshold of logarithmic type . Consequently, in said example, the method provides only one comparison and the predetermined correlation threshold is of logarithmic type.

In particular, the predetermined correlation threshold is defined by the following logarithmic curve: In this specific case, in an example in which only eight users are connected to the electric line 100, the parameters a and b of the logarithmic curve have been chosen so that the slope of the curve is steep.

Furthermore, below a respective first parameter Q_i equal to the percentage value of the ratio between the average value of each sequence of power values absorbed

by each user and the mean value is calculated for each user .

Below the obtained results.

Figure 3 shows a Cartesian plane x, y on which the logarithmic curve that defines the predetermined correlation threshold and the users are represented.

Each user is represented on the Cartesian plane by a respective point.

The position of each user on said Cartesian plane x, y is identified by a value on the x axis equal to the respective value Qi and by a value on the y axis equal to the value of the respective statistical correlation coefficient p_i .

With reference to step I, the users reported to an entity responsible for the supply of electricity are those to which respective statistical correlation coefficients having a value greater than or equal to said predetermined logarithmic correlation threshold are associated, i.e. in the specific case, user Ui and user

U₂.

Advantages

Advantageously, according to the method object of the invention, it is possible to determine electrical losses due to a malfunction or tampering of an electricity meter.

The present invention has been described for illustrative, but not limitative purposes with reference to a preferred embodiment, but it well evident that one skilled in the art can introduce modifications to the same without departing from the relevant scope, as defined in the enclosed claims.

Claims

1. Method for determining electrical losses of at least one electric line (100) due to a malfunctioning or tampering of an electricity meter, said at least one electric line (100) being connected to a plurality of users (Ui, U₂...U_N) , a transformer room (1) for transmitting electric current, through said at least one electric line (100) , to said plurality of users (Ui, U2...UN) , and to at least one power meter (10) for measuring the power supplied from said transformer room (1) to said electric line (100) in a plurality of time units present in a predetermined time period, where said power meter (10) is arranged on said at least one electric line (100) between said transformer room (1) and said users

(U₁, U₂...U_N) , said method comprising the following steps:

A) acquiring a mean value of power s with reference to the power supplied from said transformer room (1) to said electric line (100) and measured by said power meter (10), so as to obtain a sequence of mean values of power in each time unit present in said predetermined time period ( s ) k, with k=l , M where M is a positive integer, where k is the index of the number of the time units in the predetermined time period;

B) calculating the mean value of the mean values of

power with reference to the mean values of each sequence of power values (s)_k :

C) acquiring the power absorbed by each user (UI,U2...U_N) in each time unit present in said predetermined time period, so as to obtain for each user (UI,U2...UN) a respective sequence of values of absorbed power (f_i)_k , with i=l , N where N is a positive integer, where i is the index of the number of users:

D) calculating the mean value of each sequence of

values of power absorbed by each user (U₁,U₂...U_N) :

E) determining a mean value of efficiency e associated with said electric line (100) in the predetermined time period with the following formula:

Gl) calculating a sequence of difference values of power t referred to the values of power variation t_j supplied by said transformer room (1), with j=1 , (M- 1), where M is a positive integer and j is the number of time units, and where each value of said sequence is obtained by the difference between the mean value of power s supplied by said transformer room (1) in a time unit and the mean value of power s supplied by the transformer room (1) itself in the previous time unit :

where

G3) calculating a respective sequence of the difference values of power g_i for each user (U₁,U₂...U_N) , referred to the values of power variation (g_i)j absorbed by each user (U₁,U₂...U_N) , obtained by the difference between the value of power absorbed

G4 ) calculating the mean value of the difference

values of power of each sequence of difference value g_i and the respective standard deviation :

5

G5) calculating a sequence of difference values V_j, where each value is equal to the difference between the value of variation of power supplied by said transformer room (1) in a time unit , and the sum of the variations of power absorbed by said user (U₁,U₂...U_N) in the same time unit :

G6) calculating the mean value and a standard deviation s_n referred to said sequence of difference values

G7 ) calculating the statistical correlation coefficient r_i associated with each user (U₁,U₂...U_N) between each sequence of difference values of power absorbed by a respective user (U₁,U₂...U_N) and said

sequence of difference values V_j in M-1 time units, with the following formula:

H) comparing the statistical correlation coefficient

Pi associated with each user (U₁,U₂...U_N) with a predetermined correlation threshold;

I) if one or more statistical correlation coefficients

Pi associated to one or more respective users (UI,U2...UN) are greater than or equal to said predetermined correlation threshold, reporting said users to an entity responsible for the supply of electricity with reference to electrical losses due to a malfunction or tampering of one or more electricity meters associated with said users. 2. Method according to claim 1, characterized in that said predetermined correlation threshold is chosen in a first range of values, from a value greater than or equal to 0,4 to a value equal to 1, or in a second range of values, from a value greater than or equal to 0,2 to a value lower than 0,4, or in a third range of values, from a value greater than or equal to 0,1 to a value lower than 0,

2.

3. Method according to the claim 1, characterized in that ,

said method comprises the following step:

- calculating for each user a respective first parameter Qi equal to the percentage value of the ratio between the mean value fi of each sequence of values of power absorbed by each user (U₁,U₂...U_N) and the mean value s with the following formula: in that

said step H comprises the following substeps:

- drawing on a Cartesian plane x, y a point for each user (U₁,U₂...U_N) , where the position of each user (U₁,U₂...U_N) on said Cartesian plane x, y is identified through a value of the X axis equal to the value of the respective first parameter Qi and a value on the y axis equal to the value of the respective statistical correlation coefficient pi ;

- represent on said Cartesian plane x, y said predetermined correlation threshold, where said predetermined correlation threshold is defined by a logarithmic curve Y = a . log(x) + b , where a and b are predetermined constants chosen on the basis of the number of said users (U₁,U₂...U_N) in such a way that the slope associated with said logarithmic curve decreases when the number of said users (U₁,U₂...U_N) increases.

4. Method according to the previous claim, characterized in that , with reference to the logarithmic curve, a is a constant between 0,02 and 0,2, preferably between 0,05 and 0,15, and b is a constant between 0,1 and 0,3, preferably between 0,15 e 0,25.

5. Method according to the claim 1 or 2, characterized in that ,

said method comprises the following step:

- calculating for each user a respective first parameter Q_i equal to the percentage value of the ratio between the mean value of each sequence of values of power absorbed

by each user (U₁,U₂...U_N) and the mean value s with the following formula:

in that

said step H comprises the following substeps:

- comparing the statistical correlation coefficient pi associated with each user (UI,U2...UN) with a predetermined further threshold correlation;

- drawing on a Cartesian plane x, y a point for each user (U₁,U₂...U_N) , where the position of each user (U₁,U₂...U_N) on said Cartesian plane x, y is identified through a value of the X axis equal to the value of the respective first parameter Qi and a value on the y axis equal to the value of the respective statistical correlation coefficient r_i ;

- representing on said Cartesian plane x, y said predetermined further correlation threshold, where said predetermined further correlation threshold is defined by a further logarithmic curve Y' = a' ^. log(x) + b ' , where a' and b' are predetermined constants chosen on the basis of the number of said users (U₁,U₂...U_N) in such a way that the slope associated with said further logarithmic curve decreases when the number of said users (U₁,U₂...U_N) increases, at least one between a' and b' being respectively different from a and b ,

in that

with reference to the step I, the users reported to an entity responsible for the supply of electricity are the users to which respective statistical correlation coefficients pi greater than or equal to said predetermined correlation threshold and to said predetermined further correlation threshold are associated.

6. Method according to any one of claims 3-5, characterized in that , with reference to the further logarithmic curve,

a' is a constant different from a and between 0,02 and 0,2, preferably between 0,05 and 0,15, and b' is a constant equal to b and between 0,1 and 0,3, preferably between

0,15 and 0,25,

or

a' is a constant equal to a and between 0,02 and 0,2, preferably between 0,05 and 0,15, and b' is a constant different from b and between 0,1 and 0,3, preferably between 0,15 and 0,25,

or

a' is a constant different from a and between 0,02 and

0,2, preferably 0,05 and 0,15, and b' is a constant different from b and between 0,1 and 0,3, preferably between 0,15 and 0,25.

7. Method according to any one of the previous claims, characterized in that , when the standard deviation of one or more sequences of values of absorbed power

{ (f₁)1, (f₁)₂...(f₁)_M; (f₂)1, (f₂)2...(f₂)_M,·... (f_N)₁, ( f_N) 2... ( fN) _M } by a respective user (UI,U₂...UN) is equal to zero, it is not possible to calculate the statistical correlation coefficient Pr of said one or more sequences of values of absorbed power (f_i)_k, and the respective user (U₁,U₂...U_N) is ignored or reported to said entity responsible for supplying electricity.

8. Method according to any one of the previous claims, characterized in that , with reference to the step F, if the mean value of said efficiency e is greater than or equal to said predetermined efficiency threshold, no electrical loss due to a malfunctioning or tampering with an electricity meter associated with a respective user (UI,U₂...UN) is reported to an entity responsible for the supply of electricity.

9. Computer program, comprising code means configured in such a way that, when executed on a computer, perform the steps of the method according to any of previous claim .

10. System for determining electrical losses of at least one electric line (100) , said at least one electric line (100) being connected to a plurality of users (Ui, U₂...U_N) , said system comprising said at least one electric line (100) , as well as:

- a transformer room (1) for providing electric current to said users of said plurality of users (UI,U₂...UN) through said at least one electric line (100) ,

at least one electricity meter associated with a respective user of said plurality of users (UI,U2...U ) for measuring the power absorbed by said user, said electricity meter measuring the power absorbed by each user of said plurality of users (UI,U2...U ) in time units present in a predetermined time period, so as to obtain for each user (UI,U2...U ) a respective sequence of values of absorbed power (f_i)_k , with i index of the number of users, 1=1 , N , with N a positive integer, and k index of the time units, k=l, ... _/ M, with M a positive integer,

- at least a power meter (10) for measuring the power supplied from said transformer room (1) to said at least one electric line (100) in time units present in a predetermined time period, so as to obtain a sequence of mean values of power (s)_k in each of said time units, said power meter (10) being arranged on said electric line (100) between said transformer room (1) and said users

(U₁, U₂...U_N) ,

characterized in that

said system comprises a central unit with a computer configured to receive said sequence of values of absorbed power and said sequence of mean values of power ( s ) _k through a communication channel, and to run the program according to claim 9.