KR20210043389A - Load power analysis apparatus and load power analysis method by clustering technique - Google Patents

Abstract

Disclosed are a load power analysis device and load power analysis method by a clustering technique. The load power analysis device by a clustering technique of the present invention comprises: a smart power meter unit installed at a power inlet point of the consumer to which at least one load device is connected to measure the power usage of the consumer; and a load decomposition unit analyzing the power usage measured by the smart power meter unit and monitoring the power usage for each load device. The load decomposition unit pre-processes the power usage data measured by the smart power meter unit to obtain active power and apparent power for each time period, and classifies the load devices according to the load power characteristics using the active power and the apparent power.

Description

Load power analysis apparatus and load power analysis method by clustering technique

The present invention relates to a load power analysis apparatus and a load power analysis method, and more particularly, to a load power analysis apparatus and a load power analysis method using a clustering technique based on power characteristics of a load.

Due to environmental issues, policy efforts to lower the rate of generation by fossil energy and increase the rate of generation by renewable energy are continuing. Along with this, a power management policy is in parallel to manage demand-supply in a balanced manner so that the amount of demand for electricity is accurately predicted and the amount of supply is maintained at a certain level accordingly.

In order to realize this energy management policy, an intelligent remote metering infrastructure (AMI) has been installed and operated to accurately predict the demand-side power usage.

AMI metering data is used not only as a standard for calculating electricity bills, but also as a variety of intelligent services in the field of distribution/sales/customer service, such as violations, challenge search, electricity quality (voltage, current) monitoring, and real-time information provision.

However, in order to support the planned/reasonable power consumption and voluntary power saving of customers with existing AMIs installed, real-time information of AMI data must be provided as well as more detailed and useful information.

To do this, it is necessary to analyze by load installed at the customer (by type of electronic product).In order to analyze the load by electronic product in this way, a separate device must be installed at the customer, or meaningful information can be obtained only through a separate analysis equipment. there was.

However, there is a need for the development of a new load power analysis device and a load power analysis method capable of grasping the power usage by load by analyzing the characteristics of a customer's load without additional installation of a separate device other than the already installed AMI.

Patent Document 1: Registered Patent No. 10-1970186 (announcement date: April 18, 2019) Patent Document 2: Registered Patent No. 10-1708709 (announcement date: February 21, 2017) Patent Document 3: Public Patent Publication No. 10-2016-0143456 (Publication date: December 14, 2016)

The present invention conceived by the above-described necessity provides a load power analysis device and a load power analysis method capable of analyzing the power usage and usage pattern of individual loads by grasping the power characteristics of each load of customers and classifying them by a clustering technique. It aims to do.

In order to achieve the above object, a load power analysis device according to a clustering technique according to an embodiment of the present invention is installed at a power entry point of a customer to which at least one load device is connected, and measures the power consumption of the customer. Meter part; And a load decomposition unit that analyzes the electric power consumption measured by the smart power meter unit and monitors the electric power consumption for each load device, wherein the load decomposition unit stores the power consumption data measured by the smart power meter unit. Preprocessing is performed to obtain active power and apparent power for each time slot, and classifies the load device according to load power characteristics using the active power and the apparent power.

In this case, the load decomposition unit clusters the measured power consumption data into a plurality of clusters on a daily or monthly basis.

In this case, the load decomposition unit calculates a power factor using the active power and the apparent power, and calculates the measured power consumption data based on the active power, the power factor, and a meter reading time obtained by measuring the active power. Cluster into multiple clusters.

In this case, the load decomposition unit classifies clusters having a similar change pattern as homogeneous load devices by using the median value of the active power, the median power factor, and the median value of the meter reading time for each of the plurality of clusters.

In this case, the load decomposition unit calculates the total daily operation time by summing the average daily occurrence time of clusters classified as homogeneous load devices, and uses the active power and the total daily operation time to calculate the corresponding load. Classify the device into one of several types of power change.

In this case, the load decomposition unit includes a first type having two power change states, a second type having a plurality of power change states, and a third type having a continuous power change state in a short period of time. And a fourth type having a state of repetitive power change over a long period of time, and classified into any one of the first to fourth types.

Meanwhile, the load decomposition unit receives a user's feedback through an application pre-installed in a user terminal previously registered as a user of the smart power meter unit.

On the other hand, the load decomposition unit receives the identification information on the load device acquired by the smart outlet pre-installed in the customer installed the smart power meter unit.

The load power analysis method according to another embodiment of the present invention includes a smart power meter unit installed at a power entry point of a customer to which at least one load device is connected to measure the power consumption of the customer, and the smart power meter unit. In the load power analysis method using a load power analysis device according to a clustering technique including a load decomposition unit that analyzes the generated power consumption and monitors the power consumption of each load device, the load decomposition unit comprises: information on the measured power consumption Preprocessing and calculating a power factor using the active power and the apparent power for each predetermined time slot, and the active power and the apparent power; The load decomposition unit clustering information on the measured power consumption into a plurality of clusters on a daily or monthly basis using the active power, the power factor, and a meter reading time obtained by reading the active power. ; The load decomposition unit may further include classifying clusters having similar patterns of change into homogeneous load devices by using an intermediate value of active power, an intermediate value of power factor, and an intermediate value of meter reading time for each of the plurality of clusters; And the load decomposition unit calculates the total daily operation time by summing the average daily occurrence time of the cluster classified as the homogeneous load device, and uses the active power and the total daily operation time to determine the load device. Classifying into one of a plurality of power change types; Including, wherein the load decomposition unit comprises: a first type having two power change states, a second type having a plurality of power change states, and a short period of time. It is classified as one of a third type having a continuous power change state and a fourth type having a repetitive power change state over a long period of time.

In this case, the load decomposition unit receives a user's feedback through an application pre-installed in a user terminal registered in advance as a user of the smart power meter unit, or a load acquired by a smart outlet pre-installed in the customer where the smart power meter unit is installed. Receiving identification information about the device; And updating classification information on the load device by using the user's feedback or identification information on the load device.

According to various embodiments of the present disclosure, since the power characteristics of the customer's load can be analyzed and classified by load by a clustering technique, the consumer's power consumption can be accurately predicted,

By analyzing the power usage pattern by load and decomposing AMI power usage, customers can reduce costs through unnecessary home appliance usage information and progressive tax notifications.

By using the power consumption pattern of each home appliance, it is possible to detect the failure of the home appliance and prevent the problem caused by the breakdown of the home appliance.

By analyzing the usage patterns of users by household, it predicts emergency situations that occur in elderly households living alone, thereby exerting the effect of preventing accidents.

1 is a block diagram exemplarily illustrating an apparatus for analyzing load power using a clustering technique according to an embodiment of the present invention;
FIG. 2 is a graph of four types of power patterns illustrating power characteristics of a load device;
3 is a 3D simulation diagram according to active power, power factor, and meter reading time for explaining the clustering technique of the present invention;
4 is a graph of data distribution for each power factor group of six clusters according to an embodiment of the present invention;
5 is a graph of distribution of data on holidays and weekdays of six clusters according to an embodiment of the present invention;
6 is a graph of data distribution for each day of the week of six clusters according to an embodiment of the present invention;
7 is a data distribution graph for each power factor group of four clusters according to another embodiment of the present invention.
8 is a data distribution graph of holidays and weekdays of four clusters according to another embodiment of the present invention;
9 is a data distribution graph for each day of the week of four clusters according to another embodiment of the present invention.
10 is a diagram exemplarily illustrating a user interface for receiving feedback from a user terminal according to another embodiment of the present invention; and
11 is a flowchart illustratively illustrating a load power analysis method according to another embodiment of the present invention.

Hereinafter, preferred embodiments of the present invention will be described with reference to the drawings and tables. The embodiments described below are for describing a preferred example, and various design changes may be made within the scope of the technical idea of the present invention.

1 is a block diagram illustrating an apparatus for analyzing load power using a clustering technique according to an embodiment of the present invention. Referring to FIG. 1, the load power analysis apparatus 100 using a clustering technique includes a smart power meter unit 110 and a load decomposition unit 130.

The smart power meter unit 110 is installed at a power entry point of a customer to which at least one load device is connected to measure the power consumption of the customer. The smart power meter unit 110 measures power consumption of a plurality of load devices L1 to L5 connected through the smart outlet 300.

According to an embodiment of the present invention, clustering can be performed with the difference between active power and reactive power. In the received data, active power, reactive ground, reactive leading power, and apparent power of one or more home appliances are accumulated for 15 minutes. Therefore, there is a need to perform pre-processing work to fit the structure of the data.

The load decomposition unit 130 preprocesses the power consumption data measured by the smart power meter unit 110. Table 1 below shows an example of power usage data preprocessed by the load decomposition unit 130.

Referring to Table 1 above, METR_NO is the identification number of the smart power meter, MR_YMD is the meter reading date, MR_HHMI is the meter reading time, WHM_NDL_LP is the amount of active power, ONEA_VAR_LP is the amount of reactive ground power, CDNPH_VARNDL_LP is the amount of reactive power, and PH_WHM_NDL_LP means the amount of reactive power. .

At this time, the power factor (factor) means a number that indicates how efficiently the actual voltage and current applied to work. In theory, when the efficiency is 100%, it can be calculated as the ratio of the active power to the total amount of power (apparent power). Power factor = Active power/Apparent power. Add a derived variable that defines the power factor based on the active power and apparent power. Table 2 below is a table that adds the calculated power factor.

The derived variable defining the calculated power factor in Table 2 is defined as factor_check. In addition, the items P, Q1, Q2, S, factor check, Time, holiday, and Day are defined as shown in Table 3 below.

The load decomposition unit 130 acquires active power and apparent power for each time period from the preprocessed power usage data. The load decomposition unit 140 calculates a power factor using the active power and the apparent power.

The load decomposition unit 130 may classify the power factor value according to four criteria and generate a factor group variable item. The factor group item created in this way can be created in the manner shown in Table 4 below.

The types of power factor values are [0.94, 0.93, 1.0, 0.96, 0.99, 0.98, 0.92] for power factor values with a frequency of 95% or more, as shown in Table 5 below, and the types of power factor values with a frequency of 95% or more are 8 based on descending order. It is a kind, and the load decomposition unit 130 extracts a power factor value assuming that the daily product usage frequency is around 95% of the total.

The load decomposition unit 130 pre-processes the measured power consumption data by dividing it on a daily or monthly basis. Table 6 below shows an example of a data set preprocessed on a daily or monthly basis.

The load decomposition unit 130 clusters the measured power usage data into a plurality of clusters based on the active power, the power factor, and the meter reading time obtained by reading the active power. At this time, the load decomposition unit 130 adjusts the number of clusters according to a change in the size (distance) of the clustered group. Preferably, the number of clusters is determined at a critical point where the size (distance) of the clustered group rapidly decreases while increasing the number of clusters. In order to determine the number of clusters in this way, the logic shown in Table 7 below can be used.

In more detail, as shown in Table 8 below, the number of clusters at points that rapidly decrease is selected as the number for clustering by looking at the change in the size of the clustered individual groups formed while increasing the number of clusters.

When the number of clusters is determined in the above-described manner, the load decomposition unit 130 clusters the power consumption data measured according to the determined number of clusters in units of days or months. Table 9 below shows an example of clustering into 6 groups.

Table 9 shows the meter reading time, active power, power factor value, and generation time for each of the six clusters. At this time, the meter reading time, active power, and power factor values of each cluster mean the median value of the power usage data constituting the cluster, and the occurrence time means the average daily occurrence time.

In Table 10, clusters 1, 2, and 4 are classified as homogeneous clusters, and the criterion is the degree of similarity between active power and power factor values. When the average daily occurrence times of clusters 1, 2, and 4 are combined in this way, it takes up 13.2 hours, and a load device showing such a power usage time pattern can be inferred as a refrigerator.

That is, the load decomposition unit 130 classifies clusters having similar patterns of change as homogeneous load devices by using the median value of the active power, the median power factor, and the median value of the meter reading time for each of a plurality of clusters.

By analyzing this clustered data set, if it shows a similar pattern, it can be inferred as a homogeneous load device. Referring to Table 11, since the active power and power factor values of the clusters 8 and 10 are similar, the load decomposing unit 130 can be classified as a load device of the same type.

The load decomposing unit 130 may classify the corresponding cluster into four types corresponding to the corresponding power change pattern by using the change pattern of the active power and the power factor value.

The load decomposition unit 130 calculates the total operating time per day by summing the average daily occurrence times of clusters classified as load devices of the same homogeneity. The load decomposition unit 130 classifies the load device into one of a plurality of power change types using the active power and the total operating time per day.

The load decomposition unit 130 includes a first type having two power change states, a second type having a plurality of power change states, a third type having a continuous power change state for a short period of time, and a long period of the plurality of power change types. It is classified into a fourth type having a repetitive power change state, and classified into one of the first to fourth types. Referring to Table 12 below, the four types can be distinguished by power change patterns, and additional power change patterns in addition to the representative four change patterns can be interpreted as much as possible.

The load decomposition unit 130 receives a user's feedback through an application preinstalled in the user terminal 200 registered in advance as a user of the smart power meter unit 110. The load decomposition unit 130 receives identification information about the load device acquired by the smart outlet 300 pre-installed in the customer in which the smart power meter unit 110 is installed.

2 is a graph of four types of power patterns illustrating power characteristics of a load device. Referring to FIG. 2, the four types are divided into first to fourth types.

The first type is a type having two power change patterns consisting of a section in which the power is turned off and a section in which the power is turned on (see Fig. 2(a)),

The second type is a type having various power change patterns in an operation period (refer to FIG. 2(b)), for example, a first period in which the power is turned off, a second period operating at a first state level value, and a second There are various types of power change patterns as a third section operating as a state level value and a fourth section operating as a third state level value.

The third type is a type showing a continuous power change pattern in a short-term operation period (see Fig. 2(c)). For example, a period in which power is turned off, a period in which the power is raised to a maximum value, and a period in which the power is dropped to a lower level than the maximum value period is a type in which a power change pattern occurs in a short period of time. Here, the short term means tens of minutes or hours.

The fourth type is a type having a repetitive power change pattern in a long-term operation period (see Fig. 2(d)). For example, while showing a continuous power usage pattern, it shows a repetitive pattern at two power levels: operating power and standby power. In the case of this fourth type, a pattern lasting from tens to hundreds of hours is shown.

3 is a 3D simulation diagram according to active power, power factor, and meter reading time for explaining the clustering technique of the present invention. Referring to FIG. 3, the load decomposition unit 130 clusters the measured power usage data into 10 clusters.

At this time, the standard elements of clustering are active power, meter reading time, and power factor value, and if the characteristics of active power, power factor value, and meter reading time of 10 clusters are grouped together, the cluster is a power use section by the same load device. I can interpret it.

4 is a graph of data distribution for each power factor group of six clusters according to an embodiment of the present invention. Referring to FIG. 4, data distribution of 6 groups for 4 power factor groups can be confirmed.

In the case of the first cluster, about 600 power usage data are concentrated in the third power factor group. In the case of the second cluster, the first power factor group and the second power factor group are distributed as 980 and 1100 data items, respectively. In the case of the third cluster, data of the third power factor group and the fourth power factor group are distributed. In the case of the fourth cluster, data of the first power factor group and the second power factor group are distributed. In the case of the fifth cluster, data of the first power factor group, the second power factor group, and the third power factor group are distributed. In the case of the sixth cluster, data of the third power factor group and the fourth power factor group are distributed.

As shown in FIG. 4, this is to accurately analyze the power usage of the load device even when various load devices are included in the same cluster through data distribution analysis for each power factor group for a plurality of clusters.

5 is a graph of distribution of data on holidays and weekdays of six clusters according to an embodiment of the present invention. Referring to FIG. 5, distribution of power use data on public holidays and weekdays can be confirmed.

In the case of the first to sixth clusters, the distribution of power usage data on holidays and weekdays can be distinguished, and thus the data distribution pattern can be analyzed to distinguish which type of load device is used.

For example, in the case of the sixth cluster shown in FIG. 5, since the distribution of power use data on weekdays is much larger than the distribution of power use data on holidays, it is estimated as a load device that is frequently used on weekdays, and the accuracy of this estimation is improved. For this purpose, if information on the characteristics of each load device is received from the user as feedback information and reflected, the corresponding load device can be inferred more precisely. If you analyze the distribution of power use data by day of the week more precisely than analyzing the distribution of power use data on weekdays or holidays, the classification accuracy of load devices can be further improved.

6 is a data distribution graph for each day of the week of six clusters according to an embodiment of the present invention. Referring to FIG. 6, in the case of the second cluster, since the power usage data distribution is concentrated on Saturday, it is inferred as a load device with a high frequency of use on Saturday. The load device can be classified by receiving identification information of the operating load device from equipment that can identify the type of the load device, such as an outlet.

Although the power usage data pattern shown in FIG. 6 is exemplary, when analyzing the power usage data pattern for each day of the week of the second cluster, it is evenly used on weekdays, and power due to the use of load devices (e.g., washing machines) with a high frequency of use on Saturdays. Indicate the pattern of change.

7 is a data distribution graph for each power factor group of four clusters according to another embodiment of the present invention. Referring to FIG. 7, distribution characteristics for each power factor group for the first to fourth clusters can be confirmed. In the case of the third cluster, it can be seen that the distribution of the power usage data of the first power factor group is shown, and the distribution of the power usage data of the second power factor group is shown in Kyungwoong of the remaining clusters.

8 is a data distribution graph of public holidays and weekdays of four clusters according to another embodiment of the present invention. Referring to FIG. 8, a specific cluster may be classified as a cluster representing the operating characteristics of a specific load device through the power usage distribution pattern of the first to fourth clusters on holidays and weekdays.

For example, the fourth cluster may be classified as a load device having a characteristic that is frequently used on weekdays, and the second cluster may be classified as a load device having a characteristic that is used on a public holiday. If feedback information is received from the user in order to grasp the characteristics of the load device, it is possible to more accurately identify the type of the main load device used in the cluster by using the power usage data distribution and the feedback information of the cluster.

9 is a graph of data distribution for each day of the week of four clusters according to another embodiment of the present invention. Referring to FIG. 9, since data distribution characteristics can be checked for each day of the week for four clusters, in the case of the 10th cluster, the frequency of use tends to be high on Wednesday, so the characteristics of load devices with high frequency used on Wednesday are shown. It can be analyzed as a cluster.

10 is a diagram exemplarily illustrating a user interface for receiving feedback from a user terminal according to another embodiment of the present invention. Referring to FIG. 10, if feedback information on the use cycle, use time, and day of use for each load device is received from the user terminal and referenced when analyzing data, the power change pattern in each cluster is analyzed after clustering and the load device is classified by type. It can be disassembled and analyzed for power consumption.

11 is a flowchart illustrating a method of analyzing load power according to another embodiment of the present invention. Referring to FIG. 11, the load power analysis method pre-processes the measured power usage information and calculates a power factor using the active power and apparent power for each predetermined time slot, and the active power and the apparent power (S1101). Clustering information on the power consumption measured using the power, power factor, and the meter reading time of the corresponding active power into a plurality of clusters in daily or monthly units (S1103), effective for each of a plurality of clusters Classifying clusters showing similar patterns of change as homogeneous load devices using the median power, median power factor, and median reading time (S1105), the daily average occurrence time of clusters classified as homogeneous load devices And calculating the total daily operation time by adding them, and classifying the load device into one of a plurality of power change types by using the active power and the total daily operation time (S1107).

At this time, before and after executing step (S1107), the user's feedback is received through an application pre-installed in the user terminal 200 registered in advance as a user of the smart power meter unit 110, or the smart power meter unit 110 is installed. The step (S1108) of receiving identification information on the load device acquired by the smart outlet 300 previously installed in the customer may be further performed. In addition, after step S1108, a step S1109 of updating or correcting classification information for a corresponding load device may be further performed by using the user's feedback or identification information about the load device.

100: load power analysis device
110: smart power meter unit
130: load disassembly part
200: user terminal
300: smart outlet

Claims (10)

In the load power analysis apparatus using a clustering technique,
A smart power meter unit installed at a power entry point of a customer to which at least one load device is connected to measure the power consumption of the customer; And
Including; a load decomposition unit for monitoring the power consumption of each load device by analyzing the power consumption measured by the smart power meter unit,
The load decomposition unit pre-processes the power consumption data measured by the smart power meter unit to obtain active power and apparent power for each time period, and uses the active power and the apparent power to determine the load device according to load power characteristics. An apparatus for analyzing load power using a clustering technique, characterized in that for classifying.
The method of claim 1,
The load decomposition unit clusters the measured power usage data into a plurality of clusters on a daily or monthly basis.
The method of claim 2,
The load decomposition unit calculates a power factor using the active power and the apparent power, and converts the measured power consumption data into a plurality of clusters based on the active power, the power factor, and a meter reading time of the active power. An apparatus for analyzing load power using a clustering technique, characterized in that clustering is performed.
The method of claim 3,
The load decomposition unit classifies clusters having similar patterns of change into homogeneous load devices by using an intermediate value of active power, an intermediate value of power factor, and an intermediate value of meter reading time for each of the plurality of clusters. By load power analysis device.
The method of claim 4,
The load decomposition unit calculates the total daily operation time by summing the average daily occurrence time of clusters classified as homogeneous load devices, and uses the active power and the total daily operation time to determine a plurality of corresponding load devices. Load power analysis apparatus using a clustering technique, characterized in that classified as one of the power change types.
The method of claim 5,
The load decomposition unit includes a first type having two power change states, a second type having a plurality of power change states, a third type having a continuous power change state in a short period, and repetitively in a long period of time. A load power analysis apparatus according to a clustering technique, characterized in that it is classified into a fourth type having a power change state and classified into any one of the first to fourth types.
The method of claim 1,
The load decomposition unit receives a user's feedback through an application preinstalled in a user terminal registered in advance as a user of the smart power meter unit.
The method of claim 1,
The load decomposition unit receives the identification information on the load device obtained by the smart outlet pre-installed in the customer in which the smart power meter unit is installed, the load power analysis device according to the clustering technique.
A smart power meter unit installed at the power entry point of the customer to which at least one load device is connected to measure the power consumption of the customer, and the power consumption for each load device by analyzing the power consumption measured by the smart power meter unit. In the load power analysis method using a load power analysis device by a clustering technique including a load decomposition unit to monitor,
The load decomposition unit may pre-process the measured power usage information to calculate a power factor using active power and apparent power for each predetermined time slot, and the active power and the apparent power;
The load decomposition unit clustering information on the measured power consumption into a plurality of clusters on a daily or monthly basis by using the active power, the power factor, and a meter reading time obtained by reading the active power. ;
The load decomposition unit comprises the steps of classifying clusters having similar patterns of change into homogeneous load devices by using an intermediate value of active power, an intermediate value of power factor, and an intermediate value of meter reading time for each of the plurality of clusters; And
The load decomposition unit calculates the total daily operation time by adding the average daily occurrence time of the clusters classified as the homogeneous load devices, and multiplexes the corresponding load devices using active power and the total daily operation time. Including; classifying as one of the power change types
The load decomposition unit includes a first type having two power change states, a second type having a plurality of power change states, a third type having a continuous power change state in a short period, and repetitively in a long period of time. Load power analysis method, characterized in that classified into any one of the fourth type having a power change state.
The method of claim 9,
The load disassembly unit receives user feedback through an application pre-installed in a user terminal that is registered as a user of the smart power meter unit, or identifies a load device acquired by a smart outlet pre-installed at the customer in which the smart power meter unit is installed. Receiving information; And
And updating classification information for a corresponding load device by using the user's feedback or identification information about the load device.

Images