Estimation of a deleted fluid consumption
A device for estimating a deleted fluid consumption during a deletion phase, where said device comprises: a collection module configured to collect: a) first consumption data comprising information about the consumption of fluid from n fluid meters coming from a first group, and b) second consumption data comprising information about the consumption of fluid from m fluid meters coming from a second group, a computer analysis module which is configured for calculating, as a function of the first and second consumption data weighting coefficients β i minimizing the distance between the fluid consumptions of the first and second groups, where i is a positive integer included between 1 and m.
1. A method for estimation of a deleted fluid consumption during a deletion phase, said method implemented by computer means comprising:
during a learning phase:
a first collection step which consists of collecting first consumption data comprising information about the consumption of fluid from n fluid meters coming from a first group, where n is a positive integer and where said n fluid meters from the first group have subscribed to a deletion service;
a second collection step which consists of collecting second consumption data comprising information about the consumption of fluid from m fluid meters coming from a second group, where m is a positive integer and where said m fluid meters from the second group have not subscribed to a deletion service;
a step of analysis of the first and second collected consumption data during which weighting coefficients β i are calculated as a function of these first and second consumption data so as to minimize the distance between the fluid consumption of the first and second groups, where i is a positive integer included between 1 and m; and
a determination step during which a control group is determined from the second collected consumption data coming from m fluid meters from the second group and as a function of the weighting coefficients β i calculated during the analysis step; and
during the deletion phase:
an estimation step consisting of estimating the deleted fluid consumption during a deletion phase by calculating the difference between the fluid consumption of the control group and the average fluid consumption of the n fluid meters from the first group.
2. The method according to claim 1 , wherein the weighting coefficients are calculated during the analysis step so as to solve the following equation:
arg
min
β
1
…β
m
(
dist
(
P
_
1
(
t
)
-
∑
j
=
1
m
β
j
P
2
,
j
(
t
)
)
)
in which:
P 2,j (t) is the individual fluid consumption during the learning phase of the fluid meter j belonging to the second group; and
P 1 (t) is the average fluid consumption during the learning phase of the n fluid meters from the first group.
3. The method according to claim 1 , wherein m is strictly greater than n.
4. The method according to claim 1 , wherein, during the analysis step, the calculation of weighting coefficients comprises a sequential selection of fluid meters from the second group minimizing the distance between the fluid consumptions of the first and second groups.
5. The method according to claim 4 , wherein, during the analysis step, a value of “1” is assigned to the weighting coefficient β k if the individual fluid consumption of the fluid meter k from the second group is selected and a value of “0” is assigned otherwise, where k is a positive integer included between 1 and m.
6. The method according to claim 4 , wherein the step of determination of the control group comprises the calculation of the average fluid consumption P 2 (t) of the second group as a function of the weighting coefficients, where such a calculation is preferably done according to the following mathematical formula:
P
_
2
(
t
)
=
1
∑
j
=
1
m
β
j
∑
j
=
1
m
β
j
P
2
,
j
(
t
)
in which P 2,j (t) is the individual fluid consumption during the learning phase of the fluid meter j belonging to the second group.
7. The method according to claim 1 , wherein, during the analysis step, the calculation of the weighting coefficients β i comprises a linear regression of the average fluid consumption of the n fluid meters from the first group over the individual fluid consumption of each of the m fluid meters from the second group.
8. The method according to claim 7 , wherein the linear regression is a constrained linear regression.
9. The method according to claim 7 , wherein the linear regression is a Ridge type constrained linear regression.
10. The method according to claim 7 , wherein the linear regression is a Lasso type constrained linear regression.
11. The method according to claim 7 , wherein the linear regression is an elastic net type constrained linear regression.
12. The method according to claim 7 , wherein the linear regression is of PLS or PCR type.
13. The method according to claim 7 , wherein the step of determination of the control group comprises the calculation of the average fluid consumption P 2 (t) of the second group as a function of the weighting coefficients, where such a calculation is preferably done according to the following mathematical formula:
P
2
_
(
t
)
=
∑
j
=
1
m
β
j
P
2
,
j
(
t
)
in which P 2,j (t) is the individual fluid consumption during the learning phase of the fluid meter j belonging to the second group.
14. The method according to claim 1 , wherein the fluid consumption is an electric power representative of the consumed electric power.
15. A non-transmissible computer-readable storage medium comprising a computer program comprising instructions for executing the steps of the estimation method according to claim 1 , when said computer program is executed by a computer.
16. A device for estimation of the deleted fluid consumption during the deletion phase, where said device comprises:
a collection module configured to collect:
a) first consumption data comprising information about the consumption of fluid from n fluid meters coming from a first group, where n is a positive integer and where said n fluid meters from the first group have subscribed to a deletion service; and
b) second consumption data comprising information about the consumption of fluid from m fluid meters coming from a second group, where m is a positive integer and where said m fluid meters from the second group have not subscribed to a deletion service; and
a computer analysis module which is configured for calculating, as a function of the first and second consumption data, weighting coefficients β i minimizing the distance between the fluid consumptions of the first and second groups, where i is a positive integer included between 1 and m;
a processor configured to determine a control group from the second consumption data collected coming from m fluid meters from the second group, and as a function of said weighting coefficients β i calculated by the computer analysis module; and
an estimation computer module configured for estimating the deleted fluid consumption during a deletion phase by calculating the difference between the fluid consumption of the control group and the average fluid consumption of the n fluid meters from the first group.
17. The device according to claim 16 , further comprising computer means configured for implementation of the method for estimation of a deleted fluid consumption during a deletion phase, said method implemented by computer means comprising:
during a learning phase:
a first collection step which consists of collecting first consumption data comprising information about the consumption of fluid from n fluid meters coming from a first group, where n is a positive integer and where said n fluid meters from the first group have subscribed to a deletion service;
a second collection step which consists of collecting second consumption data comprising information about the consumption of fluid from m fluid meters coming from a second group, where m is a positive integer and where said m fluid meters from the second group have not subscribed to a deletion service;
a step of analysis of the first and second collected consumption data during which weighting coefficients β i are calculated as a function of these first and second consumption data so as to minimize the distance between the fluid consumption of the first and second groups, where i is a positive integer included between 1 and m; and
a determination step during which a control group is determined from the second collected consumption data coming from m fluid meters from the second group and as a function of the weighting coefficients β i calculated during the analysis step; and
during the deletion phase:
an estimation step consisting of estimating the deleted fluid consumption during a deletion phase by calculating the difference between the fluid consumption of the control group and the average fluid consumption of the n fluid meters from the first group.