JP6783712B2 - Management method and management device - Google Patents

Description

The present invention relates to a management method and a management device in a power management system.

In recent years, a management device that collectively manages a plurality of facilities has been used. The facility is provided with a measuring device that measures the power (for example, the amount of power flow from the power system to the facility). The measurement data of the measuring device is periodically transmitted to the management device via a communication device or the like provided in the facility. The management device periodically collects measurement data from the facility and manages the collected measurement data (for example, Patent Document 1).

By the way, since a malfunction or communication failure of a communication device or the like provided in a facility may occur, a loss period may occur in the measurement data collected by the management device.

As a method for interpolating the missing data in the missing period, the missing data is interpolated using a simple statistical method such as linear interpolation based on the measurement data in the unit time frame (for example, one day) including the missing period. Is possible.

Further, there is known a technique for predicting the future power situation of a certain facility by analyzing the measurement data collected in the past for the facility (for example, Non-Patent Document 1). It is considered that such a prediction technique can be applied to interpolation of missing data.

Japanese Unexamined Patent Publication No. 2014-082703

Noriaki Matsumoto, A Study on Predicting Electricity Demand Patterns for the Next Day in Generations, Proceedings of the 29th Energy System / Economy / Environment Conference, (2013), 223-226

However, when one method is fixedly used as a method for interpolating missing data, there are the following problems.

In the method of interpolating the missing data by using a simple statistical method such as linear interpolation, the interpolation accuracy is lowered when the missing period is long.

Further, the method of interpolating the missing data by using the prediction technique can cope with the case where the missing period is long, but the processing load increases as compared with the case of using the simple statistical method.

Therefore, an object of the present invention is to provide a management method and a management device capable of improving the accuracy of interpolating missing data while suppressing an increase in processing load.

The management method according to one aspect includes step A for managing measurement data periodically collected from a measuring device provided in a facility, step B for detecting a missing period of the measured data, and missing data in the missing period. It includes step C to interpolate. The step C is a step C1 in which the missing data is interpolated by the first method when the missing period is less than the threshold value, and the missing data is obtained by the second method when the missing period is equal to or more than the threshold value. Includes step C2 to interpolate. The first method is a method of interpolating the missing data based on the measurement data in the missing time frame, which is a unit time frame including the missing period. The second method is a method of interpolating the missing data based on the measurement data in the reference time frame which is the unit time frame corresponding to the missing time frame.

The management device according to one aspect includes a management unit that manages measurement data periodically collected from a measurement device provided in a facility, a detection unit that detects a loss period of the measurement data, and a loss data in the loss period. It is provided with an interpolation unit for interpolation. The interpolation unit interpolates the missing data by the first method when the missing period is less than the threshold value, and interpolates the missing data by the second method when the missing period is equal to or more than the threshold value. The first method is a method of interpolating the missing data based on the measurement data in the missing time frame, which is a unit time frame including the missing period. The second method is a method of interpolating the missing data based on the measurement data in the reference time frame which is the unit time frame corresponding to the missing time frame.

According to one aspect, it is possible to provide a management method and a management device that can improve the accuracy of interpolating missing data while suppressing an increase in processing load.

It is a figure which shows the structure of the electric power management system which concerns on embodiment. It is a figure which shows the structure of the power management server which concerns on embodiment. It is a figure which shows the management method which concerns on embodiment. It is a figure for demonstrating group classification which concerns on embodiment. It is a figure which shows the loss period less than the threshold value which concerns on embodiment. It is a figure which shows the loss period which is more than the threshold value which concerns on embodiment. It is a figure which shows the 2nd method which concerns on embodiment. It is a figure which shows the specific example of the 2nd method which concerns on embodiment. It is a figure which shows the transition of the similarity within a group which concerns on embodiment.

The embodiment will be described with reference to the drawings. In the description of the drawings below, the same or similar parts are designated by the same or similar reference numerals.

(Power management system)
The configuration of the power management system according to the embodiment will be described.

As shown in FIG. 1, the power management system 100 has a power management server 200 and a plurality of facilities 300. In FIG. 1, facilities 300A, 300B, 300C, ... Are illustrated as a plurality of facilities 300. Each facility 300 is connected to the power system 110. In the following, the flow of electric power from the electric power system 110 to the facility 300 will be referred to as a tidal current. The flow of electric power from the facility 300 to the electric power system 110 is called reverse power flow.

The power management server 200 and the facility 300 are connected to the network 120. The network 120 may provide a line between the power management server 200 and the facility 300 and a line between the power management server 200 and the power operator server 400. The network 120 is, for example, the Internet. The network 120 may provide a dedicated line such as a VPN (Virtual Private Network).

The power management server 200 is a server of a business operator that manages a plurality of facilities 300 in a bundle. The power management server 200 is an example of a management device that manages a plurality of facilities 300. The power management server 200 periodically collects measurement data from each facility 300 and manages the collected measurement data. The measurement data is any one of the amount of power flow power from the power system 110 to the facility 300, the amount of reverse power flow power from the facility 300 to the power system 110, and the amount of power consumed by the facility 300. The power management server 200 predicts the amount of future power (the amount of power flow, the amount of reverse power flow, or the amount of power consumed) in the facility 300 based on the measurement data collected from the facility 300. May be good.

Facility 300 is, for example, a house such as a detached house. The facility 300 may be a store, a building, or a factory. Facility 300 has a measuring device 310. In the embodiment, the measuring device 310 is a meter having a communication function (for example, a smart meter). The measuring device 310 measures any one of the amount of power flow power from the power system 110 to the facility 300, the amount of reverse power flow power from the facility 300 to the power system 110, and the amount of power consumed by the facility 300. The measuring device 310 periodically (for example, every minute) transmits the measurement data to the power management server 200.

Facility 300 may further have a load. A load is a device that operates using electric power. Facility 300 may further have a distributed power source. A distributed power source is a device having at least one of a function of outputting electric power and a function of storing electric power. Facility 300 may further have a local controller. The local control device is a device (EMS; Energy Management System) that manages the electric power of the facility 300. The measuring device 310 may transmit the measurement data to the power management server 200 via the local control device.

(Management device)
The power management server 200, which is a management device according to the embodiment, will be described.

As shown in FIG. 2, the power management server 200 includes a communication unit 210, a storage unit 220, and a control unit 230. The communication unit 210 is composed of a communication module. The communication unit 210 communicates with the facility 300 (specifically, the measuring device 310 and / or the local control device) via the network 120. The storage unit 220 is composed of a non-volatile memory and / or a storage medium such as an HDD (Hard Disk Drive). The storage unit 220 stores information used for processing and control by the control unit 230. In an embodiment, the storage unit 220 stores and stores measurement data collected via the communication unit 210. The control unit 230 is composed of a memory, a CPU (Central Processing Unit), and the like. The control unit 230 controls each configuration provided in the power management server 200. The control unit 230 includes a management unit 231, a classification unit 232, a detection unit 233, and an interpolation unit 234.

The management unit 231 manages the measurement data periodically collected from the measuring device 310 provided in the facility 300. The management unit 231 stores the measurement data in the storage unit 220 for each facility 300 and each measurement time, and periodically updates the measurement data in the storage unit 220.

The classification unit 232 classifies the plurality of facilities 300 into groups based on the similarity of the measurement data. The classification unit 232 may periodically execute a process of classifying the plurality of facilities 300 into groups. The information on the correspondence between the facility 300 and the group is used in the second method of data interpolation described later.

The detection unit 233 detects the missing period of the measurement data. For example, the detection unit 233 monitors the measurement data in the storage unit 220 for each facility 300, recognizes that the measurement data could not be obtained, and detects the period during which the measurement data was not obtained as the missing period.

The interpolation unit 234 interpolates the missing data in the missing period detected by the detecting unit 233. The interpolation unit 234 compares the missing period with the threshold value, and when the missing period is less than the threshold value, interpolates the missing data by the first method. The first method is a method of interpolating the missing data based on the measurement data in the missing time frame, which is a unit time frame including the missing period. The first method may be a method using a simple statistical method such as linear interpolation or mean value interpolation. The unit time frame is a time unit in which the management unit 231 manages the measurement data. In an embodiment, the unit time frame is one day (ie, 24 hours). The unit time frame is not limited to one day, but may be half a day (that is, 12 hours) or the like.

The interpolation unit 234 interpolates the missing data by the second method when the missing period detected by the detecting unit 233 is equal to or longer than the threshold value. The second method is a method of interpolating the missing data based on the measurement data in the reference time frame, which is the unit time frame corresponding to the missing time frame. In the embodiment, the reference time frame is a unit time frame belonging to a reference facility which is a facility other than the target facility in which the missing data should be interpolated. In other words, in the second method, the interpolation unit 234 interpolates the missing data based on the measurement data of a facility different from the facility (target facility) where the measurement data is lost. In this case, the interpolation unit 234 may specify a facility included in the group to which the target facility belongs as a reference facility based on the information obtained by the classification unit 232. By narrowing the range of searching for the reference facility within the group, the interpolation unit 234 can quickly identify the reference facility.

Alternatively, the reference time frame is a unit time frame belonging to the target facility in which the missing data should be interpolated, and may be a unit time frame before the missing time frame. In other words, in the second method, the interpolation unit 234 interpolates the missing data based on the measurement data past the missing time frame in the facility (target facility) where the measurement data is lost. However, when the past measurement data in the target facility is used for interpolation, the power management server 200 stores the measurement data in association with various conditions (weather conditions, etc.), and the past measurement data that matches the condition of the missing time frame. Need to be identified. On the other hand, when interpolating the missing data based on the measurement data of the reference facility different from the target facility, it is not necessary to consider the weather conditions and the like.

As described above, the power management server 200 according to the embodiment uses the first method and the second method properly according to the length of the loss period when interpolating the missing data.

Since the first method is a method in which the measurement data in the missing time frame, which is the unit time frame including the missing period, is used for interpolation, it is not necessary to refer to the unit time frame other than the missing time frame. Therefore, the first method has an advantage that the processing load is small. However, the first method has a demerit that the interpolation accuracy is lowered when the loss period is long.

The second method is a method in which the measurement data in the reference time frame, which is a unit time frame other than the loss time frame, is used for interpolation. It has the disadvantage of being large.

According to the embodiment, the merits of the first method and the merits of the second method can be compatible with each other by properly using the first method and the second method according to the length of the loss period of the measurement data. Therefore, it is possible to improve the accuracy of interpolating the missing data while suppressing the increase in the processing load.

(Management method)
The management method according to the embodiment will be described.

As shown in FIG. 3, in step S1, the classification unit 232 classifies the plurality of facilities 300 into a plurality of groups. As shown in FIG. 4, the classification unit 232 defines a group consisting of facilities 300 having similar transition patterns of measurement data in a unit time frame (for example, one day). In the example of FIG. 4, the classification unit 232 defines eight groups. The transition pattern of the measurement data of each group shows the average value of the measurement data of the facilities 300 in the group. After the groups are defined in this way, the classification unit 232 calculates the similarity between the transition pattern of the measurement data acquired from the facility 300 and the transition pattern of the measurement data of each group, and the group having the highest similarity. The facility 300 is classified into. The classification unit 232 stores the results of classification into groups in the storage unit 220.

In step S2, the detection unit 233 detects the missing period of the measurement data of the target facility. The deficiency period is detected, for example, as follows. First, the detection unit 233 acquires measurement data belonging to the target facility from the storage unit 220. This measurement data is, for example, one day's worth of measurement data collected by the management unit 231 from the measurement device 310 of the target facility. Second, the detection unit 233 performs preprocessing such as changing the sampling time and converting the acquired measurement data into a pivot graph. Third, the detection unit 230 detects the period in which the preprocessed measurement data does not exist as the missing period.

In step S3, the interpolation unit 234 compares the loss period with the threshold value. The threshold value is a predetermined value (for example, 2 hours). The threshold may be determined according to the characteristics of the specific statistical method used as the first method.

When the loss period is less than the threshold value (step S3: Yes), the interpolation unit 234 selects the first method of data interpolation and proceeds to the process in step S4. In step S4, the interpolation unit 234 interpolates the missing data by the first method. As described above, the first method is a method of interpolating the missing data based on the measurement data in the missing time frame, which is the unit time frame including the missing period. As shown in FIG. 5, it is assumed that the loss period is from 14:00 to 15:00 and the threshold value is 2 hours. In this case, since the loss period (1 hour) is less than the threshold value, the interpolation unit 234 uses the measurement data in the loss time frame (for example, the measurement data immediately before and after the loss period) at 14:00 by a statistical method. Interpolate the missing data from ~ 15:00.

On the other hand, when the loss period is equal to or greater than the threshold value (step S3: No), the interpolation unit 234 selects the second method of data interpolation and proceeds to the process in step S5. In step S5, the interpolation unit 234 interpolates the missing data by the second method. As described above, the second method is a method of interpolating the missing data based on the measurement data in the reference time frame, which is the unit time frame corresponding to the missing time frame. As shown in FIG. 6, it is assumed that the deficiency period is from 8:00 to 16:00 and the threshold value is 2 hours. In this case, since the missing period (8 hours) is equal to or longer than the threshold value, the interpolation unit 234 interpolates the missing data from 8:00 to 16:00 based on the measurement data in the reference time frame. The reference time frame is a unit time frame belonging to a reference facility that is a facility other than the target facility in which the missing data should be interpolated.

Next, a second method of data interpolation will be described.

As shown in FIG. 7, in step S51, the interpolation unit 234 identifies the group to which the target facility belongs based on the group classification result in step S1.

In step S52, the interpolation unit 234 identifies the reference facility from the group to which the target facility belongs. As shown in FIG. 8, it is assumed that the target facility is facility 300A, the group to which facility 300A belongs includes six facilities from facility 300A to facility 300F, and the unit time frame is one day (24 hours). To do. In this case, the interpolation unit 234 specifies the reference facility as follows, for example. First, the interpolation unit 234 acquires the measurement data of each of the facilities 300A to 300F before the loss time frame (X month N day). For example, the interpolation unit 234 acquires the measurement data of the facilities 300A to the facility 300F on the previous day (N-1 day of the X month) and the day before the previous day (N-2 days of the X month) from the storage unit 220. Secondly, the interpolation unit 234 calculates the degree of similarity between the measurement data of the facility 300A and the measurement data of the facilities 300B to 300F. As a method for calculating the similarity of measured data, a general calculation method for calculating the similarity of data time-series data that changes over time can be used. As an example, a calculation method using the Euclidean distance is used. Specifically, the interpolation unit 234 calculates the difference for each measurement time with respect to the measurement data of the facility 300A and the measurement data of the facilities 300B to 300F. Thirdly, the interpolation unit 234 specifies the facility having the smallest integrated value of the difference from the measurement data of the facility 300A as the reference facility. Here, the explanation will proceed on the assumption that the facility 300B is specified as the reference facility. As a method for calculating the similarity of measurement data, a method called Dynamic Time Warping (DTW) may be used.

In step S53, the interpolation unit 234 interpolates the missing data of the facility 300A based on the measurement data in the reference time frame belonging to the specified reference facility (facility 300B). As shown in FIG. 8, the reference time frame may be a unit time frame having the same date as the missing time frame. The reference time frame may be a unit time frame of a date earlier than the missing time frame. The interpolation unit 234 acquires measurement data in the loss period (8:00 to 17:00) in the reference time frame of the facility 300B. The interpolation unit 234 may use the measurement data acquired from the reference time frame as it is instead of the missing data of the facility 300A. The interpolation unit 234 may modify the measurement data acquired from the reference time frame and use the modified measurement data in place of the missing data of the facility 300A. For example, the interpolation unit 234 may modify the measurement data acquired from the reference time frame so as to ensure continuity with the measurement data (non-defective data) in the loss time frame.

(Other embodiments)
Although the present invention has been described by embodiment, the statements and drawings that form part of this disclosure should not be understood as limiting the invention. Various alternative embodiments, examples and operational techniques will be apparent to those skilled in the art from this disclosure.

In the embodiment, an example in which the interpolation unit 234 identifies the reference facility after the loss period of the measurement data is detected has been described. However, the classification unit 232 may pre-identify the reference facility before the period of loss of measurement data is detected. Specifically, the classification unit 232 specifies the reference facility for each facility in the same group in advance. In addition, the classification unit 232 periodically updates the facility used as the reference facility. As shown in FIG. 9, from February 25 to February 28, the facility 300B has the highest degree of similarity to the facility 300A in the group to which the facility 300A belongs. From March 5th to March 8th, the facility 300C has the highest degree of similarity to the facility 300A in the group to which the facility 300A belongs. From March 25th to March 28th, the facility having the highest degree of similarity to the facility 300A in the group to which the facility 300A belongs is the facility 300F. In this way, the similarity within the group changes over time. Therefore, it is preferable to calculate the similarity within the group on a regular basis and update the facility used as a reference facility on a regular basis.

In the embodiment, when the number of facilities 300 managed by the power management server 200 is less than a predetermined number, the group classification process by the classification unit 232 may be omitted. In other words, the group classification process by the classification unit 232 may be executed only when the number of facilities 300 managed by the power management server 200 is larger than the predetermined number.

A program that causes a computer to execute the management method according to the embodiment may be provided. The program may also be recorded on a computer-readable medium. Computer-readable media can be used to install programs on a computer. Here, the computer-readable medium on which the program is recorded may be a non-transient recording medium. The non-transient recording medium is not particularly limited, but may be, for example, a recording medium such as a CD-ROM or a DVD-ROM.

100 ... Power management system, 110 ... Power system, 120 ... Network, 200 ... Power management server, 210 ... Communication unit, 220 ... Storage unit, 230 ... Control unit, 231 ... Management unit, 232 ... Classification unit, 233 ... Detection unit , 234 ... Interpolator, 300 ... Facility, 310 ... Measuring device

Claims (9)

Step A to manage the measurement data that is regularly collected from the measuring device installed in the facility,
Step B for detecting the loss period of the measurement data and
A step C for interpolating the missing data in the missing period is provided.
The step C is
In step C1 of interpolating the missing data by the first method when the missing period is less than the threshold value,
Including step C2 of interpolating the missing data by the second method when the missing period is equal to or greater than the threshold value.
The first method is a method of interpolating the missing data based on the measurement data in the missing time frame, which is a unit time frame including the missing period.
The second method is a management method in which the missing data is interpolated based on the measurement data in the reference time frame which is the unit time frame corresponding to the missing time frame. The management method according to claim 1, wherein the reference time frame is the unit time frame belonging to the target facility where the missing data should be interpolated, and is the unit time frame before the missing time frame. The management method according to claim 1, wherein the reference time frame is the unit time frame belonging to a reference facility that is a facility other than the target facility in which the missing data should be interpolated. The management method according to claim 3, wherein step C2 includes a step of identifying the reference facility based on the similarity to the measurement data belonging to the target facility. Prior to step B, step D is provided for classifying a plurality of facilities into groups based on the similarity of the measurement data.
The management method according to claim 3 or 4, wherein step C2 includes a step of identifying the facility included in the group to which the target facility belongs as the reference facility. The management method according to claim 5, wherein step D is periodically executed. In step C2, the measurement data in the reference time frame is modified based on the comparison result between the measurement data in the loss time frame and the measurement data in the reference time frame, and the measurement data in the reference time frame is modified based on the corrected measurement data. The management method according to any one of claims 3 to 5, which includes a step of interpolating the missing data. The measurement data includes data indicating the amount of power flow from the power system to the facility, data indicating the amount of reverse power flow from the facility to the power system, and data indicating the amount of power consumed by the facility. The management method according to any one of claims 1 to 6, which is at least one of them. A management department that manages measurement data that is regularly collected from measuring equipment installed in the facility,
A detection unit that detects the missing period of the measurement data,
It is provided with an interpolation unit that interpolates the missing data in the missing period.
The interpolation unit is
When the missing period is less than the threshold value, the missing data is interpolated by the first method.
When the missing period is equal to or greater than the threshold value, the missing data is interpolated by the second method.
The first method is a method of interpolating the missing data based on the measurement data in the missing time frame, which is a unit time frame including the missing period.
The second method is a method of interpolating the missing data based on the measurement data in the reference time frame which is the unit time frame corresponding to the missing time frame.

Images