KR102624317B1 - Server, method and computer program for detecting energy consumption of target building - Google Patents

Abstract

The server that detects the energy consumption of the target building includes a zone setting unit that sets a plurality of zones for the target building, a collection unit that collects energy consumption data for the set plurality of zones, and the collected energy consumption data. an energy consumption pattern deriving unit for deriving an energy consumption pattern for each zone based on the derived energy consumption pattern for each zone, a classification unit for classifying the plurality of zones into at least one similar group based on the derived energy consumption pattern for each zone, and the classification unit and an abnormal energy consumption pattern detection unit that detects an abnormal energy consumption pattern based on probability distribution data for energy consumption in at least one zone belonging to the similar group.

Description

Server, method and computer program for detecting energy consumption of target building {SERVER, METHOD AND COMPUTER PROGRAM FOR DETECTING ENERGY CONSUMPTION OF TARGET BUILDING}

The present invention relates to a server, method, and computer program for detecting energy consumption of a target building.

Building Energy Management System (BEMS) is a system that uses IT technology to manage various facilities in a building, such as electricity, air conditioning, crime prevention, and disaster prevention.

The building energy management system collects and analyzes various information related to energy management facilities in the building in real time and can improve energy use efficiency by managing energy usage, facility operation status, indoor environment, and carbon emissions.

In relation to this building energy management system, Korean Patent No. 10-1170743, a prior art, discloses a refrigerator optimal operation system and method through a building energy management system.

Conventional building energy management systems analyze building energy usage patterns on a building-by-building basis and classify buildings showing similar patterns based on the hourly or monthly energy usage patterns of multiple buildings.

However, building energy usage patterns differ depending on the use of each zone set inside the building and the difference between work days and non-work days. When trying to classify energy usage patterns by reflecting the differences in use of each zone and between working days and non-working days, it has the disadvantage that the features that can be used are very limited compared to analyzing energy usage patterns at the building level.

The present invention is intended to solve the above-described problems, and includes a server, method, and computer for setting a plurality of zones for a target building and deriving an energy consumption pattern for each zone based on energy consumption data for the set plurality of zones. We would like to provide a program.

Based on the energy consumption pattern for each zone, a plurality of zones are classified into at least one similar group, and an abnormal energy consumption pattern is determined based on probability distribution data for energy consumption in at least one zone belonging to the classified similar group. It is intended to provide a server, method, and computer program for sensing.

However, the technical challenges that this embodiment aims to achieve are not limited to the technical challenges described above, and other technical challenges may exist.

As a means to achieve the above-described technical problem, an embodiment of the present invention includes a zone setting unit that sets a plurality of zones for a target building, and a collection unit that collects energy consumption data for the set plurality of zones. A unit, an energy consumption pattern deriving unit for deriving an energy consumption pattern for each zone based on the collected energy consumption data, dividing the plurality of zones into at least one similar group based on the derived energy consumption pattern for each zone. A server including a classification unit that classifies and an energy abnormal consumption pattern detection unit that detects an abnormal energy consumption pattern based on probability distribution data for energy consumption in at least one zone belonging to the classified similar group can be provided. .

Another embodiment of the present invention includes setting a plurality of zones for a target building, collecting energy consumption data for the plurality of zones, and assigning energy consumption data to each zone based on the collected energy consumption data. deriving an energy consumption pattern for each zone, classifying the plurality of zones into at least one similar group based on the derived energy consumption pattern for each zone, and A method for detecting energy consumption may be provided, including detecting an abnormal energy consumption pattern based on probability distribution data for energy consumption.

In another embodiment of the present invention, the computer program, when executed by a computing device, sets a plurality of zones for a target building, collects energy consumption data for the set plurality of zones, and collects the collected energy consumption data. An energy consumption pattern for each zone is derived based on energy consumption data, the plurality of zones are classified into at least one similar group based on the derived energy consumption pattern for each zone, and the plurality of zones are classified into at least one similar group. A computer program stored in a medium including a sequence of instructions for detecting an abnormal energy consumption pattern based on probability distribution data for energy consumption in at least one zone to which the device belongs may be provided.

The above-described means for solving the problem are merely illustrative and should not be construed as limiting the present invention. In addition to the exemplary embodiments described above, there may be additional embodiments described in the drawings and detailed description of the invention.

According to one of the means for solving the problem of the present invention described above, the target building is set into a plurality of zones, the energy consumption pattern of each zone is derived, and the plurality of zones are classified into at least one similar group. , a server, method, and computer program for detecting abnormal energy consumption patterns based on probability distribution data for energy consumption in at least one zone belonging to a classified similar group can be provided.

Servers, methods, and computer programs can be provided that allow building operations managers to focus on zones with high energy savings potential, rather than focusing on zones with low energy savings potential.

Servers, methods, and computer programs can be provided to reduce energy costs by blocking energy wasted due to incorrect building operation in advance.

1 is a configuration diagram of a server according to an embodiment of the present invention.
2A and 2B are exemplary diagrams for explaining a process of classifying a plurality of zones into at least one similar group according to an embodiment of the present invention.
Figure 3 is an exemplary diagram for explaining a process for detecting an abnormal energy consumption pattern according to an embodiment of the present invention.
Figure 4 is an exemplary diagram showing energy consumption patterns for a plurality of zones classified into at least one similar group in a manager terminal according to an embodiment of the present invention.
Figure 5 is a flowchart of a method for detecting energy consumption of a target building in a server according to an embodiment of the present invention.

Below, with reference to the attached drawings, embodiments of the present invention will be described in detail so that those skilled in the art can easily implement the present invention. However, the present invention may be implemented in many different forms and is not limited to the embodiments described herein. In order to clearly explain the present invention in the drawings, parts that are not related to the description are omitted, and similar parts are given similar reference numerals throughout the specification.

Throughout the specification, when a part is said to be "connected" to another part, this includes not only the case where it is "directly connected," but also the case where it is "electrically connected" with another element in between. . In addition, when a part is said to "include" a certain component, this does not mean excluding other components unless specifically stated to the contrary, but may further include other components, and one or more other features. It should be understood that it does not exclude in advance the presence or addition of numbers, steps, operations, components, parts, or combinations thereof.

In this specification, 'part' includes a unit realized by hardware, a unit realized by software, and a unit realized using both. Additionally, one unit may be realized using two or more pieces of hardware, and two or more units may be realized using one piece of hardware.

In this specification, some of the operations or functions described as being performed by a terminal or device may instead be performed on a server connected to the terminal or device. Likewise, some of the operations or functions described as being performed by the server may also be performed on a terminal or device connected to the server.

Hereinafter, an embodiment of the present invention will be described in detail with reference to the attached drawings.

1 is a configuration diagram of a server according to an embodiment of the present invention. Referring to FIG. 1, the server 100 includes a zone setting unit 110, a collection unit 120, an energy consumption pattern derivation unit 130, a classification unit 140, an abnormal energy consumption pattern detection unit 150, and a notification unit. It may include unit 160.

The zone setting unit 110 may set a plurality of zones for the target building. For example, the zone setting unit 110 may classify the space of each floor of the target building into a plurality of zones.

The collection unit 120 may collect energy consumption data for a plurality of set zones. For example, the collection unit 120 may collect first energy consumption data consumed on working days and second energy consumption data consumed on non-working days for each zone.

The energy consumption pattern deriving unit 130 may derive the energy consumption pattern for each zone based on the collected energy consumption data. This is because energy consumption patterns may vary depending on the use of each zone.

The classification unit 140 may classify a plurality of zones into at least one similar group based on the derived energy consumption pattern for each zone. The process of classifying a plurality of zones into at least one similar group will be described with reference to FIGS. 2A and 2B.

2A and 2B are exemplary diagrams for explaining a process of classifying a plurality of zones into at least one similar group according to an embodiment of the present invention.

FIG. 2A is an exemplary diagram illustrating a process for deriving an energy consumption pattern for each zone according to an embodiment of the present invention. Referring to FIG. 2A, the energy consumption pattern deriving unit 130 performs cluster analysis on the first energy consumption data to derive the first energy consumption pattern, and performs cluster analysis on the second energy consumption data to derive the second energy consumption pattern. Energy consumption patterns can be derived. This is because energy consumption patterns may vary depending on the use of each zone, and energy consumption patterns may vary depending on work days and non-work days.

For example, the energy consumption pattern derivation unit 130 sets the cluster to 1, and divides the plurality of dependencies 200 into each zone 200, including the first energy consumption data 210 consumed on a working day and the first energy consumption data 210 consumed on a non-working day. The second energy consumption data 220 may be classified, and a cluster centroid may be derived using the k-means algorithm for the first energy consumption data 210 and the second energy consumption data 220. At this time, the energy consumption pattern deriving unit 130 may perform two cluster analyzes and derive a first energy consumption pattern representing a working day and a second energy consumption pattern representing a non-working day through each cluster analysis. .

The energy consumption pattern derivation unit 130 may determine the difference between the first energy consumption pattern corresponding to a working day and the second energy consumption pattern corresponding to a non-working day for each zone.

Conventionally, energy consumption distribution was simply derived by collecting annual or monthly energy consumption of a large number of buildings, but the present invention has the advantage of being able to derive energy consumption patterns over a continuous flow of time.

FIG. 2B is an exemplary diagram illustrating a process of classifying a plurality of zones into at least one similar group according to an embodiment of the present invention. Referring to FIGS. 2A and 2B, the classification unit 140 determines whether there is a difference in consumption patterns between the first energy consumption pattern and the second energy consumption pattern based on the Two Sample K-S test (240). You can. Here, the two-sample K-S test is a statistical technique that verifies whether two data are extracted from the same population using the difference in cumulative probability of the sampling distribution of the two groups. The two-sample K-S test (240) is used to determine the first energy consumption pattern and It is possible to determine the difference between the second energy consumption patterns.

The classification unit 140 may classify the plurality of zones 200 into one of the first similar group 250 and the second similar group 260 based on the presence or absence of the determined consumption pattern difference.

For example, if there is no difference in consumption patterns (231, 232, 233), the classification unit 140 may classify the zone into the first similar group 250. At this time, the classification unit 140 may classify at least one zone belonging to the first similar group 250 into one of a plurality of lower similar groups based on the degree of change in energy consumption over time. For example, based on the degree of change in energy consumption over time, the classification unit 140 divides at least one zone belonging to the first similar group 250 into a first sub-similar group 271 in which the change in hourly energy consumption is constant. It can be classified into a second sub-similar group 272 in which there is a change in energy consumption.

For another example, if there is a difference in consumption patterns (234, 235, 236), the classification unit 140 may classify the corresponding zone into a second similar group 260. At this time, based on the time at which energy consumption occurs (eg, time corresponding to work hours), at least one zone belonging to the second similar group 260 may be classified into one of a plurality of lower similar groups. For example, based on the time at which energy consumption occurs, the classification unit 140 divides at least one zone belonging to the second similar group 260 into a first sub-similar group 281 that consumes energy only during work hours, and a zone belonging to the second similar group 260. In addition to the time zone, it can be classified into a second sub-similar group (282) that consumes energy.

This is to prevent the problem of providing incorrect information when comparing zones with different energy use patterns on work days and non-work days on the same line.

In other words, the present invention manages zones with similar energy consumption patterns together, thereby comparing zones in which energy is used intermittently and zones in which energy is used continuously, so that zones in which energy is used intermittently consume less energy, and zones in which energy is used intermittently consume less energy. Zones that are used continuously are judged to be zones that use a lot of energy, and through these classification results, it is possible to check the energy consumption ranking or consumption growth rate ranking for each zone with similar energy consumption patterns, and compare similar energy use patterns. This can reduce the risk of providing inaccurate benchmarking results to building operations managers. Additionally, it has the advantage of being able to provide building operation managers or building owners with more objective and scientific results to reduce energy use.

Returning to FIG. 1 , the abnormal energy consumption pattern detection unit 150 may detect an abnormal energy consumption pattern based on probability distribution data for energy consumption in at least one zone belonging to a classified similar group.

For example, the abnormal energy consumption pattern detection unit 150 derives an hourly probability distribution for energy consumption based on past energy consumption data for each zone, and derives a probability value indicating abnormal energy consumption for each zone based on the hourly probability distribution. can do. Probability distribution by time has the advantage of minimizing management points by applying a technique (for example, GMM) that can express various consumption patterns as a single probability distribution.

At this time, the abnormal energy consumption pattern detection unit 150 may use a Gaussian Mixture Model (GMM) to derive a probability value representing abnormal energy consumption for each zone. A Gaussian mixture model can be created by performing data preprocessing on training data and performing data filtering.

To build a Gaussian mixture model, energy consumption data for each zone over a specific past period (for example, 1 month or 6 months) can be used. This is because the probability distribution may have two or more peaks depending on the zone's energy consumption pattern. For example, when using one month's worth of energy consumption data, a Gaussian mixture model for energy usage at a specific time can be built based on 30 pieces of energy consumption data.

In a conventional building energy management system, abnormal energy consumption is detected on a building-by-building basis. In this case, an energy prediction model is created using climate data, heat source facility operation logs, etc., and energy consumption abnormalities are detected based on this (actual value vs. predicted value). This method was mainly used.

However, when detecting abnormal energy consumption on a zone basis, it is difficult to find common variables that can create an energy usage prediction model. Therefore, in order to detect abnormal energy consumption patterns, the present invention uses a method of deriving a probability value representing abnormal energy consumption for each zone based on a probability distribution by time.

The collection unit 120 can collect current energy consumption data for each zone in real time.

The abnormal energy consumption pattern detection unit 150 may detect an abnormality in the current energy consumption data by determining whether the current energy consumption data corresponds to a probability value indicating abnormal energy consumption. The process of detecting anomalies in current energy consumption data will be explained with reference to FIG. 3.

Figure 3 is an exemplary diagram for explaining a process for detecting an abnormal energy consumption pattern according to an embodiment of the present invention. Referring to FIG. 3, the first figure 300 is a diagram showing electricity consumption for a specific zone for one week. Here, the horizontal axis (x-axis) of the first image 300 represents the day of the week (Monday to Sunday), the vertical axis (y-axis) represents the time (0 to 23:00), and the color in the first image 300 represents the corresponding Indicates electricity consumption for each day/time. The general method of diagnosing abnormal energy consumption using the consumption pattern in a conventional building energy management system is performed based on the consumption pattern of a daily unit (0 to 23:00), but the present invention has the advantage of using the consumption pattern of the same time period. there is.

Accordingly, the abnormal energy consumption pattern detection unit 150 can acquire horizontal data as shown in the red dotted box.

The second figure 310 is a diagram showing the distribution of energy consumption data extracted in the horizontal direction. The four graphs represent the distribution of electricity consumption in different time zones, with the horizontal axis representing electricity consumption and the vertical axis representing frequency.

For example, if working days and non-working days are relatively clearly distinguished in the target building, the distribution may have a form with two peaks. Here, the left bud may represent a non-working day, and the right bud may represent a working day.

In the third figure 320, the probability distribution can be estimated using the distribution of electricity consumption in the same time period. For example, when using electricity consumption in the same time period without distinction between work and non-work, two or more buds (321, 322) may appear depending on the energy consumption pattern of the target building, and a Gaussian mixture model is applied to form one peak. It can be expressed as a probability distribution.

When using the probability distribution obtained through a Gaussian mixture model, it is possible to determine the probability that a specific amount of electricity consumption will occur, and when using a Gaussian mixture model, if it is below a certain probability value, it is possible to determine whether the amount of electricity consumption is specifically used.

When the notification unit 160 detects an abnormality in the current energy consumption data, it can notify the abnormality detection to the manager terminal of the target building. For example, based on the probability of occurrence of a specific amount of electricity consumption, if the amount of electricity consumption is less than a certain probability value (for example, 5%), abnormality detection may be notified to the manager terminal of the target building.

Through this process, the Gaussian mixture model and the unusual usage boundary are constructed for each zone and each time, so that abnormal use can be checked for each zone by referring to the specific probability value every hour.

This server 100 may be executed by a computer program stored on a medium containing a sequence of instructions for detecting energy consumption of a target building. When executed by a computing device, the computer program sets a plurality of zones for the target building, collects energy consumption data for the plurality of zones, and calculates energy consumption for each zone based on the collected energy consumption data. Deriving a consumption pattern, classifying a plurality of zones into at least one similar group based on the derived energy consumption pattern for each zone, and probability distribution data for energy consumption in at least one zone belonging to the classified similar group It may include a sequence of instructions to detect abnormal energy consumption patterns based on .

Figure 4 is an exemplary diagram showing energy consumption patterns for a plurality of zones classified into at least one similar group in a manager terminal according to an embodiment of the present invention. Referring to FIG. 4, the manager terminal may display a screen 400 related to energy consumption of the target building.

The administrator terminal displays a plurality of zones classified into similar group A (410), similar group B (420), similar group C (430), and similar group D (440) through the screen 400 related to the energy consumption of the target building ( 450) can display the energy consumption pattern.

For example, the manager terminal may group zones of a target building into zones with similar electricity consumption and display an electricity consumption ranking for each group. At this time, the manager terminal may highlight (e.g., display with a red box) a zone in which the current electricity consumption is diagnosed as abnormal.

Through this, the risk of providing inaccurate benchmarking results to managers can be reduced through comparison between patterns of similar groups.

Figure 5 is a flowchart of a method for detecting energy consumption of a target building in a server according to an embodiment of the present invention. The method of detecting energy consumption of a target building in the server 100 shown in FIG. 5 includes steps processed in time series by the server 100 according to the embodiments shown in FIGS. 1 to 4. Therefore, even if the content is omitted below, it also applies to the method of detecting energy consumption of a target building by the server 100 according to the embodiment shown in FIGS. 1 to 4.

In step S510, the server 100 may set a plurality of zones for the target building.

In step S520, the server 100 may collect energy consumption data for a plurality of set zones.

In step S530, the server 100 may derive an energy consumption pattern for each zone based on the collected energy consumption data.

In step S540, the server 100 may classify a plurality of zones into at least one similar group based on the derived energy consumption pattern for each zone.

In step S550, the server 100 may detect an abnormal energy consumption pattern based on probability distribution data on energy consumption in at least one zone belonging to the classified similar group.

In the above description, steps S510 to S550 may be further divided into additional steps or combined into fewer steps, depending on the implementation of the present invention. Additionally, some steps may be omitted or the order between steps may be switched as needed.

The method of detecting the energy consumption of a target building in a server described with reference to FIGS. 1 to 5 can also be implemented in the form of a computer program stored on a medium executed by a computer or a recording medium containing instructions executable by a computer. there is. Additionally, the method of detecting energy consumption of a target building in a server described with reference to FIGS. 1 to 5 may also be implemented in the form of a computer program stored in a medium executed by a computer.

Computer-readable media can be any available media that can be accessed by a computer and includes both volatile and non-volatile media, removable and non-removable media. Additionally, computer-readable media may include computer storage media. Computer storage media includes both volatile and non-volatile, removable and non-removable media implemented in any method or technology for storage of information such as computer-readable instructions, data structures, program modules or other data.

The description of the present invention described above is for illustrative purposes, and those skilled in the art will understand that the present invention can be easily modified into other specific forms without changing the technical idea or essential features of the present invention. will be. Therefore, the embodiments described above should be understood in all respects as illustrative and not restrictive. For example, each component described as unitary may be implemented in a distributed manner, and similarly, components described as distributed may also be implemented in a combined form.

The scope of the present invention is indicated by the claims described below rather than the detailed description above, and all changes or modified forms derived from the meaning and scope of the claims and their equivalent concepts should be construed as being included in the scope of the present invention. do.

100: server
110: Zone setting government
120: Collection department
130: Energy consumption pattern derivation unit
140: Classification department
150: Abnormal energy consumption pattern detection unit
160: Notification unit

Claims (19)

In the server that detects the energy consumption of the target building,
a zone setting unit that sets a plurality of zones for the target building;
a collection unit that collects energy consumption data for the plurality of zones set;
an energy consumption pattern deriving unit that derives an energy consumption pattern on working days and an energy consumption pattern on non-working days for each zone based on the collected energy consumption data;
The plurality of zones are classified into at least one similar group based on the consumption pattern difference between the energy consumption pattern on working days and the energy consumption pattern on non-working days for each zone, and the zones without the consumption pattern difference are classified into hourly energy consumption. a classification unit that classifies zones with differences in consumption patterns into lower similar groups based on the time at which energy consumption occurs, based on the degree of change; and
For each of the classified zones, it includes an abnormal energy consumption pattern detection unit that detects an abnormal energy consumption pattern based on probability distribution data for energy consumption in at least one zone belonging to a similar group,
A server wherein energy consumption patterns are displayed for each similar group by an administrator terminal, and zones in which the abnormal energy consumption patterns are detected are highlighted.
According to claim 1,
The server wherein the collection unit distinguishes and collects first energy consumption data consumed on working days and second energy consumption data consumed on non-working days for each zone.
According to claim 2,
The energy consumption pattern deriving unit performs cluster analysis on the first energy consumption data to derive a first energy consumption pattern, and performs cluster analysis on the second energy consumption data to derive a second energy consumption pattern. In, server.
According to claim 3,
The classification unit determines the presence or absence of a consumption pattern difference between the first energy consumption pattern and the second energy consumption pattern based on the Two Sample KS test, and determines the presence or absence of a consumption pattern difference based on the determined consumption pattern difference. A server that classifies the plurality of zones into one of a first similarity group and a second similarity group.
According to claim 4,
If the consumption pattern difference does not exist, the classification unit classifies the zone into a first similar group, and classifies at least one zone belonging to the first similar group into a plurality of sub-similar groups based on the degree of change in energy consumption over time. Server, which is classified into one of the groups.
According to claim 4,
If there is a difference in the consumption pattern, the classification unit classifies the zone into a second similar group, and divides at least one zone belonging to the second similar group into a plurality of lower similar groups based on the time at which the energy consumption occurs. It is classified as one of the servers.
In the server that detects the energy consumption of the target building,
a zone setting unit that sets a plurality of zones for the target building;
a collection unit that collects energy consumption data for the plurality of zones set;
an energy consumption pattern deriving unit that derives an energy consumption pattern for each zone based on the collected energy consumption data;
Based on the derived energy consumption pattern for each zone, the plurality of zones are classified into at least one similar group, and in the zones where there is no difference in consumption pattern, the difference in consumption pattern is based on the degree of change in energy consumption over time. a classification unit that classifies the zones into lower similar groups based on the time at which energy consumption occurs; and
For each of the classified zones, it includes an abnormal energy consumption pattern detection unit that detects an abnormal energy consumption pattern based on probability distribution data for energy consumption in at least one zone belonging to a similar group,
The abnormal energy consumption pattern detection unit derives a time-dependent probability distribution for the energy consumption based on past energy consumption data for each zone,
A server that derives a probability value indicating abnormal energy consumption for each zone based on the time-specific probability distribution.
According to claim 7,
The collection unit collects current energy consumption data for each zone in real time,
The abnormal energy consumption pattern detection unit detects an abnormality in the current energy consumption data by determining whether the current energy consumption data corresponds to a probability value indicating the abnormal energy consumption.
According to claim 8,
When detecting an abnormality in the current energy consumption data, the server further includes a notification unit notifying the abnormality detection to a manager terminal of the target building.
In a method for detecting energy consumption of a target building in a server,
Setting a plurality of zones for the target building;
collecting energy consumption data for the plurality of zones set;
Deriving energy consumption patterns on working days and energy consumption patterns on non-working days for each zone based on the collected energy consumption data;
The plurality of zones are classified into at least one similar group based on the consumption pattern difference between the energy consumption pattern on working days and the energy consumption pattern on non-working days for each zone, and the zones without the consumption pattern difference are classified into hourly energy consumption. Based on the degree of change, reclassifying the zones with differences in consumption patterns into lower similar groups based on the time at which energy consumption occurs; and
A method for detecting energy consumption, comprising the step of detecting an abnormal energy consumption pattern based on probability distribution data for energy consumption in at least one zone belonging to a similar group for each of the classified zones.
According to claim 10,
The step of collecting the energy consumption data is,
A method for detecting energy consumption, comprising the step of collecting and distinguishing first energy consumption data consumed on working days and second energy consumption data consumed on non-working days for each zone.
According to claim 11,
The step of deriving the energy consumption pattern for each zone is,
deriving a first energy consumption pattern by performing cluster analysis on the first energy consumption data; and
A method for detecting energy consumption, comprising deriving a second energy consumption pattern by performing cluster analysis on the second energy consumption data.
According to claim 12,
The step of classifying into at least one similar group is,
determining whether there is a consumption pattern difference between the first energy consumption pattern and the second energy consumption pattern based on a Two Sample KS test; and
A method for detecting energy consumption, comprising classifying the plurality of zones into one of a first similar group and a second similar group based on the presence or absence of the determined consumption pattern difference.
According to claim 13,
The step of classifying into at least one similar group is,
If there is no difference in the consumption pattern, classifying the corresponding zone into a first similar group; and
A method for detecting energy consumption, comprising classifying at least one zone belonging to the first similarity group into one of a plurality of lower similarity groups based on the degree of change in energy consumption over time.
According to claim 13,
The step of classifying into at least one similar group is,
If there is a difference in the consumption pattern, classifying the corresponding zone into a second similar group; and
A method for detecting energy consumption, comprising classifying at least one zone belonging to the second similarity group into one of a plurality of lower similarity groups based on the time at which energy consumption occurs.
In a method for detecting energy consumption of a target building in a server,
Setting a plurality of zones for the target building;
collecting energy consumption data for the plurality of zones set;
Deriving an energy consumption pattern for each zone based on the collected energy consumption data;
Based on the derived energy consumption pattern for each zone, the plurality of zones are classified into at least one similar group, and in the zones where there is no difference in consumption pattern, the difference in consumption pattern is based on the degree of change in energy consumption over time. reclassifying the existing zones into lower similar groups based on the time at which energy consumption occurs; and
For each classified zone, detecting an abnormal energy consumption pattern based on probability distribution data for energy consumption in at least one zone belonging to a similar group,
The step of detecting the abnormal energy consumption pattern is,
Deriving a probability distribution for the energy consumption by time based on the past energy consumption data for each zone; and
A method for detecting energy consumption, comprising deriving a probability value indicating abnormal energy consumption for each zone based on the time-dependent probability distribution.
According to claim 16,
The step of collecting the energy consumption data is,
Comprising the step of collecting current energy consumption data for each zone in real time,
The step of detecting the abnormal energy consumption pattern is,
A method for detecting energy consumption, comprising the step of detecting an abnormality in the current energy consumption data by determining whether the current energy consumption data corresponds to a probability value indicating abnormal energy consumption.
According to claim 17,
When detecting an abnormality in the current energy consumption data, the method further includes notifying the abnormality detection to a manager terminal of the target building.
A computer program stored on a medium containing a sequence of instructions for detecting energy consumption of a target building, comprising:
When the computer program is executed by a computing device,
Set a plurality of zones for the target building,
Collect energy consumption data for the plurality of zones set above,
Based on the collected energy consumption data, derive energy consumption patterns on working days and energy consumption patterns on non-working days for each zone,
The plurality of zones are classified into at least one similar group based on the consumption pattern difference between the energy consumption pattern on working days and the energy consumption pattern on non-working days for each zone, and the zones without the consumption pattern difference are classified into hourly energy consumption. Based on the degree of change, zones with differences in consumption patterns are reclassified into lower similar groups based on the time when energy consumption occurs,
A computer program stored in a medium, comprising a sequence of instructions for detecting an abnormal energy consumption pattern based on probability distribution data for energy consumption in at least one zone belonging to a similar group for each classified zone.