Energy analysis device, energy management system, energy analysis program, and energy analysis method

Abstract

The present invention enables efficient efforts for energy saving by analyzing electric power consumption data of a plurality of electric power supply facilities, the invention comprising: an electric power consumption data acquisition unit that acquires electric power consumption data of each of a plurality of electric power supply facilities for supplying electric power to a plurality of energy demand facilities; an electric power supply pattern model generation unit that performs multivariate analysis on the electric power consumption data, and generates an electric power supply pattern profile in which the plurality of electric power supply facilities are respectively classified into a plurality of electric power fluctuation patterns and / or an electric power supply pattern operation state sequence indicating operation states over time of the electric power supply pattern profile; and an electric power load amount calculation unit that, on the basis of the electric power supply pattern profile and / or the electric power supply pattern operation state sequence, calculates the electric power load amount of a plurality of energy demand facilities or the plurality of electric power supply facilities corresponding to the plurality of electric power fluctuation patterns, or the electric power load amount over time corresponding to the plurality of electric power fluctuation patterns.

Description

Energy analysis device, energy management system, energy analysis program, and energy analysis method

[0001] The present invention relates to an energy analysis device, an energy management system, an energy analysis program, and an energy analysis method.

[0002] BACKGROUND ART Conventionally, energy management systems have been devised for managing energy-consuming facilities such as factory facilities, test facilities, and air conditioning facilities, with the aim of saving energy therein.

[0003] For example, Patent Document 1 discloses an energy usage visualization device that can determine whether energy waste is occurring, identify the causes of the waste, and analyze trends along the production flow of a manufacturing line.

[0004] Japanese Patent Application Laid-Open No. 2013-222256

[0005] Conventional energy conservation measures at research and development sites, which involve the start and stop of energy-consuming equipment, involve creating an equipment operation plan based on the experimental plan and the actual energy consumption of each power supply facility, and the experimenter then determines the schedule to attempt efficient operation.

[0006] However, in actual R&D sites, equipment is updated every time the R&D target changes, and as time passes, the power distribution diagram for the entire building may not be recorded. As a result, the relationship between power supply equipment and each energy demand equipment, as well as the energy consumption of each energy demand equipment, becomes unclear, and there are limitations to manual energy management, making energy conservation efforts inefficient.

[0007] Therefore, the present invention has been made in consideration of the above-mentioned problems, and its main objective is to enable efficient efforts toward energy conservation by analyzing power consumption data of multiple power supply facilities.

[0008] That is, the energy analysis device of the present invention is an energy analysis device that analyzes energy consumption in a plurality of energy demand facilities, and is characterized by comprising: a power consumption data acquisition unit that acquires power consumption data for each of a plurality of power supply facilities that supply power to the plurality of energy demand facilities; a power pattern model generation unit that performs multivariate analysis on the power consumption data to generate a power pattern profile that classifies each of the plurality of power supply facilities into a plurality of power fluctuation patterns and / or a power pattern operating state series that is the operating state of the power pattern profile over time; and a power load calculation unit that calculates the power load of the plurality of energy demand facilities or the plurality of power supply facilities according to the plurality of power fluctuation patterns, or the power load over time according to the plurality of power fluctuation patterns, based on the power pattern profile and / or the power pattern operating state series.

[0009] Such an energy analysis device performs multivariate analysis of the power consumption data of multiple power supply facilities, classifies the multiple power supply facilities into multiple power fluctuation patterns, and calculates the power loads of multiple energy demand facilities or multiple power supply facilities according to the multiple power fluctuation patterns, or the power loads over time according to the multiple power fluctuation patterns.This allows users to consider energy-saving opportunities based on these power loads, thereby enabling efficient efforts toward energy conservation.In addition, it is possible to calculate the standard power consumption of multiple energy demand facilities from past power consumption data.

[0010] The energy analysis device according to the present invention preferably further includes a first work plan data acquisition unit that acquires past work plan data indicating past work plans that used at least some of the plurality of energy demand facilities, and the power supply pattern model generation unit generates correlation data indicating the correlation between power supply facilities for each work based on the power consumption data and the past work plan data, and performs multivariate analysis of the generated correlation data to generate the power supply pattern profile and / or the power supply pattern operating state series.

[0011] In order to improve the accuracy of the generated power supply pattern profile and / or power supply pattern operating state series, it is desirable that the power supply pattern model generation unit compares the generated power supply pattern profile and / or power supply pattern operating state series with the power consumption data and the past work plan data to confirm the accuracy of the generated power supply pattern profile and / or power supply pattern operating state series.

[0012] The power load calculation unit preferably determines a steady-state range for each task from a statistical index for the variability of the power consumption data based on the power consumption data and the past task plan data, and calculates the power consumption outside the steady-state range for each task. With this configuration, it is possible to consider points where energy can be saved, starting from tasks with high power consumption outside the steady-state range. As a result, it is possible to make efficient efforts toward energy conservation.

[0013] The energy analysis device according to the present invention preferably further includes an equipment management data acquisition unit that acquires equipment management data indicating a correspondence relationship between the energy demanding equipment and the power supply equipment to which the energy demanding equipment is connected, and the power load calculation unit calculates the power loads of the power supply equipment corresponding to the energy demanding equipment or the power loads over time corresponding to the energy demanding equipment based on the power supply pattern operation state series and / or the equipment management data. With this configuration, the power loads of the power supply equipment corresponding to the energy demanding equipment or the power loads over time corresponding to the energy demanding equipment are calculated, allowing a user to consider potential energy savings from the power loads corresponding to the energy demanding equipment. This enables efficient energy conservation efforts.

[0014] As a specific embodiment of the power load calculation unit, it is desirable that the power load calculation unit calculates a frequency distribution of the power load.

[0015] In order to improve usability for the user, it is preferable that the energy analysis device of the present invention further comprises a display unit for displaying the calculation results of the power load obtained by the power load calculation unit.

[0016] Here, the display unit may display the power load for each component in a two-dimensional array, with the vertical axis (rows) representing multiple power fluctuation patterns and the horizontal axis (columns) representing multiple energy demand facilities or multiple power supply facilities. The display unit may also display the power load for each component as a heat map, showing shades of gray according to the numerical value of the power load.

[0017] The display unit may display the power load for each component in a two-dimensional array with the vertical axis (rows) representing multiple power fluctuation patterns and the horizontal axis (columns) representing time. The display unit may also display the power load for each component as a heat map, showing shades of gray according to the numerical value of the power load for each component.

[0018] Furthermore, the display unit may display the power load for each component in a two-dimensional array with the vertical axis (rows) representing multiple energy demand facilities and the horizontal axis (columns) representing multiple power supply facilities or time. The display unit may also display the power load for each component as a heat map, showing shades of gray according to the numerical value of the power load.

[0019] In a specific embodiment, the power consumption data for each of the plurality of power supply facilities is preferably data indicating the power consumption per predetermined time unit for each of the plurality of distribution boards to which the plurality of energy demand facilities are connected. By using the power consumption data for each distribution board in this way, it is possible to easily acquire the power consumption data. Furthermore, it is possible to easily identify the energy demand facilities connected to each distribution board, and it is possible to easily acquire facility management data.

[0020] An energy management system according to the present invention includes the above-described energy analysis device and an energy management device that performs energy management based on the analysis results obtained by the energy analysis device. By performing energy management based on the analysis results obtained by the energy analysis device, this energy management system can suppress large fluctuations in power consumption and enable energy savings.

[0021] It is desirable that the energy management system according to the present invention further includes a plurality of energy demand facilities, and that the energy management device performs energy management of the plurality of energy demand facilities based on the analysis results obtained by the energy analysis device.

[0022] Furthermore, the energy analysis program of the present invention is a program used in an energy analysis device that analyzes energy consumption in a plurality of energy demanding facilities, and is characterized in that it comprises a power consumption data acquisition unit that acquires power consumption data for each of a plurality of power supply facilities that supply power to the plurality of energy demanding facilities, a power pattern model generation unit that performs multivariate analysis on the power consumption data to generate a power pattern profile and / or a power pattern operating state series that is the operating state of the power pattern profile over time, which classifies each of the plurality of power supply facilities into a plurality of power fluctuation patterns, and a power load calculation unit that calculates the power loads of the plurality of energy demanding facilities or the plurality of power supply facilities corresponding to the plurality of power fluctuation patterns, or the power loads over time corresponding to the plurality of power fluctuation patterns, based on the power pattern profile and / or the power pattern operating state series. Note that the energy analysis program may be distributed electronically or may be recorded on a program recording medium such as a CD, DVD, or flash memory.

[0023] Furthermore, the energy analysis method of the present invention is an energy analysis method for analyzing energy consumption in a plurality of energy demand facilities, and is characterized in that it acquires power consumption data for each of a plurality of power supply facilities that supply power to the plurality of energy demand facilities, performs multivariate analysis on the power consumption data, generates a power supply pattern profile that classifies each of the plurality of power supply facilities into a plurality of power fluctuation patterns and / or a power supply pattern operating state series that is the operating state of the power supply pattern profile over time, and calculates the power load of the plurality of energy demand facilities or the plurality of power supply facilities corresponding to the plurality of power fluctuation patterns, or the power load over time corresponding to the plurality of power fluctuation patterns, based on the power supply pattern profile and / or the power supply pattern operating state series.

[0024] According to the present invention described above, by analyzing the power consumption data of a plurality of power supply facilities, it is possible to make efficient efforts toward energy conservation.

[0025] 1 is an overall configuration diagram of an energy management system according to an embodiment of the present invention. FIG. 2 is a diagram illustrating an energy analysis method according to the embodiment. FIG. 3 is a graph illustrating an example of power consumption data according to the embodiment. FIG. 4 is a table illustrating an example of past work plan data according to the embodiment. FIG. 5 is correlation data illustrating the correlation between power supply equipment for each task according to the embodiment. FIG. 6 is correlation data illustrating the correlation coefficient between power supply equipment for each task according to the embodiment. FIG. 7 is correlation data illustrating the correlation coefficient between power supply equipment for each task according to the embodiment. FIG. 8 is correlation data illustrating the correlation between power supply equipment in a running test according to the embodiment. FIG. 9 is a graph illustrating (a) the reproducibility of data distribution and (b) a comparison result using a confusion matrix according to the embodiment. FIG. 10 is a schematic diagram illustrating a method for creating a confusion matrix according to the embodiment. FIG. 11 is a diagram illustrating (a) a power supply pattern profile and (b) a power supply pattern operating state series according to the embodiment. FIG. 12 is a diagram illustrating an example of display by a display unit according to the embodiment. FIG. 13 is a diagram illustrating the configuration of an energy analysis device according to a modified embodiment. FIG. 14 is a diagram illustrating an example of display by a display unit according to the modified embodiment. FIG. 15 is an overall configuration diagram of an energy management system according to a modified embodiment. FIG. 16 is a flowchart illustrating the operation of the energy management system according to the modified embodiment. FIG. 17 is a table illustrating a steady-state lower limit value, a steady-state upper limit value, total power consumption, and out-of-steady-state range power for each task. 10 is a graph showing the correlation between a parameter (air conditioning fan wind speed) of energy demanding equipment (air conditioner) and power consumption. FIG. 11 is an overall configuration diagram of an energy management system according to a modified embodiment.

[0026] <One embodiment of the present invention> An embodiment of an automotive test bench will be described below as one embodiment of an energy management system using an energy analysis device according to the present invention, with reference to the drawings. Note that all of the drawings shown below are drawn in a schematic manner, with appropriate omissions or exaggerations made, for ease of understanding. Identical components are given the same reference numerals, and descriptions thereof will be omitted where appropriate.

[0027] <Configuration of Energy Management System 100> The energy management system 100 of this embodiment manages energy consumption in a plurality of energy demand facilities X, such as factory facilities, test facilities, or air conditioning facilities.

[0028] Here, the energy demanding facility X includes various types of equipment, such as manufacturing equipment, processing equipment, transport equipment, inspection equipment, testing equipment, analytical equipment, air conditioning equipment, lighting equipment, etc. In this embodiment, the energy consumed by the energy demanding facility X is electricity, but it may be secondary energy other than electricity, such as city gas, propane gas, gasoline, kerosene, or diesel, or primary energy, such as petroleum, coal, wind power, natural gas, hydroelectric power, or geothermal energy.

[0029] Specifically, as shown in Fig. 1, energy management system 100 includes an energy analysis device 10 that analyzes energy consumption in a plurality of energy demanding facilities X, and an energy management device 20 that performs energy management based on the analysis results obtained by energy analysis device 10. Based on the analysis results, energy management device 20 can calculate power consumption for an input work plan (including an operation plan, maintenance, calibration, etc.) for energy demanding facilities X and / or propose an energy-saving work plan.

[0030] The energy analysis device 10 or the energy management device 20 is configured by a computer having a CPU, memory, an input / output interface, an AD converter, a display device such as a display, etc. The energy analysis device 10 performs the functions described below by the CPU and peripheral devices working together based on an energy analysis program or an energy management program stored in the memory. The energy analysis device 10 and the energy management device 20 may be configured by a physically integrated computer, or may be configured by physically separate computers.

[0031] 1, the energy analysis device 10 includes a power consumption data acquisition unit 21, a first work plan data acquisition unit 22, a power supply pattern model generation unit 3, and a power load calculation unit 4. The power consumption data acquisition unit 21 and the first work plan data acquisition unit 22 may be a common data acquisition unit.

[0032] The power consumption data acquiring unit 21 acquires power consumption data (see FIG. 3 ) for each of a plurality of power supply facilities S that supply power to a plurality of energy demanding facilities X. This power consumption data acquiring unit 21 acquires past power consumption data for each of the plurality of power supply facilities S. The power consumption data acquiring unit 21 may acquire and store power consumption data from the plurality of power supply facilities S, or may measure the power consumption data of the plurality of power supply facilities S with a separate meter and acquire the measured data.

[0033] This past power consumption data is data indicating the power consumption of each of the multiple power supply facilities S for each predetermined time unit. Here, the predetermined time unit may be, for example, a predetermined time unit such as one hour, or a predetermined number of days such as one day. In this embodiment, the multiple power supply facilities S are multiple distribution boards to which multiple energy demand facilities X are connected. Note that the past power consumption data may indicate the power consumption of each power supply board, or may be the total power consumption data of two or more distribution boards. Additionally, the power supply facilities S may be outlets to which power is supplied from a distribution board, or may be a distribution board that supplies power to the distribution board.

[0034] The first work plan data acquisition unit 22 acquires past plan result data (see FIG. 4 ) that indicates past work plans that used at least some of the multiple energy demanding facilities X. This first work plan data acquisition unit 22 acquires past work plan data that corresponds to power consumption data. The past work plan data may be data that indicates actual results (results) based on past work plans. Furthermore, the past work plan data is data that indicates work content, including preparation, testing, and / or production, for each predetermined time unit in the multiple energy demanding facilities X for which power consumption data has been acquired. Here, the predetermined time unit may be, for example, a predetermined time unit such as one hour, or a predetermined number of days such as one day.

[0035] The power supply pattern model generation unit 3 generates a power supply pattern profile and / or a power supply pattern operating state sequence by using a statistical method from the power consumption data acquired by the power consumption data acquisition unit 21 and / or the past work plan data acquired by the first work plan data acquisition unit 22. Here, the power supply pattern model generation unit 3 uses multivariate analysis as the statistical method, and the multivariate analysis may use, for example, unsupervised learning such as nonnegative matrix factorization (NMF), cluster analysis, principal component analysis, or self-organizing map (SOM).

[0036] Here, as shown in FIG. 2, the power supply pattern profile classifies each of a plurality of power supply equipment S (S1 to S6) into a plurality of power fluctuation patterns (here, P1 to P4). Each of the power fluctuation patterns P1 to P4 indicates a combination of the power consumption of each of the plurality of power supply equipment S (S1 to S6). For example, the power fluctuation pattern P1 shown in FIG. 2 is a pattern in which all of the power supply equipment S1 to S6 consumes power, and the power fluctuation pattern P4 is a pattern in which only the power supply equipment S1, S3, and S4 consume power. Furthermore, as shown in FIG. 2, the power supply pattern operating state series indicates the operating state (time change) of each of the plurality of power fluctuation patterns P1 to P4 over time.

[0037] The number of power fluctuation patterns in the power supply pattern profile can be determined by the user using facility information such as a facility management table that indicates the correspondence between the energy demanding facility X and the power supply facility S to which the energy demanding facility X is connected. In this embodiment, the number of power fluctuation patterns is eight, which matches the number of power supply facilities S. As another method for determining the number of power fluctuation patterns, it is also possible to determine the number of patterns using an evaluation index based on an information criterion such as AIC (Akaike Information Criterion).

[0038] The power supply pattern model generating unit 3 will be described in detail below.

[0039] The power supply pattern model generation unit 3 generates correlation data showing the correlation between power supply equipment S for each task, as shown in Figures 5 to 7, based on the power consumption data (see Figure 3) and past work plan data (see Figure 4).

[0040] For example, as shown in FIG. 5, the correlation between power supply equipment S for each task can be qualitatively confirmed based on a scatter diagram showing the correlation between power supply equipment S for each task. Furthermore, details of the correlation coefficients between power supply equipment S for each task are shown in FIGS. 6a and 6b. From FIGS. 6a and 6b, the differences in the correlation between power supply equipment S for each task can be quantitatively understood based on the correlation coefficients between power supply equipment S for each task. Specifically, from FIG. 6a, a tendency for high correlation between power supply equipment S1 and power supply equipment S4, S7, and S8 is observed for tasks other than dynamo warm-up and analyzer calibration test. Furthermore, for example, the correlation can be quantitatively understood based on the correlation coefficients in a form such as FIG. 7, which focuses on a specific task compared to FIGS. 6a and 6b. Specifically, in the "driving test," power supply equipment S5 has a higher correlation with power supply equipment S3 than the other power supply equipment, demonstrating a distinctive tendency.

[0041] Furthermore, the power supply pattern model generation unit 3 performs multivariate analysis on the generated correlation data to generate a power supply pattern profile (see FIG. 8( a)) and / or a power supply pattern operation state sequence (see FIG. 8( b)). Because correlations are observed between power supply equipment and because the energy consumption is a non-negative value, the power supply pattern model generation unit 3 of this embodiment generates the power supply pattern profile and / or the power supply pattern operation state sequence using non-negative matrix factorization, which is a multivariate analysis model with non-negative value constraints. Furthermore, the power consumption data of this embodiment includes information on the energy consumption distribution of eight power supply equipment, and the power supply pattern model generation unit 3 generates power supply pattern profiles classified into eight power fluctuation patterns.

[0042] Furthermore, the power supply pattern model generation unit 3 compares the generated power supply pattern profile and / or power supply pattern operating state series with power consumption data and past work plan data to confirm the accuracy of the generated power supply pattern profile and / or power supply pattern operating state series.

[0043] Specifically, the power supply pattern model generating unit 3 checks the accuracy of the power supply pattern profile and / or the power supply pattern operating state sequence by the following methods (1) and / or (2).

[0044] (1) As shown in Figure 9(a), the power supply pattern model generation unit 3 calculates the energy consumption (actual measurement value) for each actual task from the power consumption data and past task plan data. The power supply pattern model generation unit 3 also estimates the energy consumption (calculated value) for each task from the generated power supply pattern profile and / or power supply pattern operation state sequence. The power supply pattern model generation unit 3 then compares the actual measurement value and the calculated value of the energy consumption and calculates the degree of agreement between them (reproducibility of the actual data distribution), thereby confirming the accuracy of the power supply pattern profile and / or power supply pattern operation state sequence.

[0045] (2) As shown in FIG. 9B, the power supply pattern model generation unit 3 performs comparison using a confusion matrix using the power supply pattern operation state sequence. Specifically, as shown in FIG. 10, the power supply pattern model generation unit 3 divides the power supply pattern operation state sequence (including multiple power fluctuation patterns) into tasks and calculates a representative power fluctuation pattern for each task using a statistical method such as averaging for each power fluctuation pattern. The power supply pattern model generation unit 3 then calculates the similarity (e.g., cosine similarity) between the calculated representative power fluctuation pattern for each task and each task included in the power supply pattern operation state sequence. The power supply pattern model generation unit 3 also determines the task with the highest similarity to the task at each time included in the power supply pattern operation state sequence, and estimates the task by creating, for example, a similarity table. Next, the power supply pattern model generation unit 3 compares the obtained estimation result (estimated task) with the actual task using, for example, cosine similarity, counts the number of matches for each task, and creates a confusion matrix. Finally, the power supply pattern model generating unit 3 checks the accuracy of the power supply pattern profile and / or the power supply pattern operating state sequence from the degree of agreement between the estimated work and the actual work obtained from this mixing matrix.

[0046] Then, if the power supply pattern profile and / or the power supply pattern operating state sequence satisfy a predetermined accuracy by the above (1) and / or (2), the power supply pattern model generation unit 3 outputs the power supply pattern profile and / or the power supply pattern operating state sequence. On the other hand, if the power supply pattern profile and / or the power supply pattern operating state sequence does not satisfy the predetermined accuracy, the power supply pattern model generation unit 3 selects, executes, and evaluates a multivariate analysis method again to generate a power supply pattern profile and / or a power supply pattern operating state sequence.

[0047] The power load calculation unit 4 calculates the power loads of the plurality of energy demand facilities X or the plurality of power supply facilities S according to the plurality of power fluctuation patterns, or the power loads over time according to the plurality of power fluctuation patterns, based on the power supply pattern profile and / or the power supply pattern operating state series. The calculated power loads are for a predetermined period such as daily or monthly. Here, the power loads are the amounts of power consumption converted from power consumption data from the plurality of power supply facilities S.

[0048] The energy analysis device 10 of this embodiment may further include a display unit 5 that displays the calculation results of the power load obtained by the power load calculation unit 4.

[0049] As shown in Fig. 11 , the display unit 5 can display the calculation results of the power load amount for a predetermined period, such as daily or monthly, on the display as a heat map. Specifically, as shown in Fig. 11( a), the display unit 5 can display the power load amount for each component in a two-dimensional array in which the vertical axis (rows) represents multiple power fluctuation patterns and the horizontal axis (columns) represents multiple energy demand facilities or multiple power supply facilities. As shown in Fig. 11( b), the display unit 5 can display the power load amount for each component in a two-dimensional array in which the vertical axis (rows) represents multiple power fluctuation patterns and the horizontal axis (columns) represents time.

[0050] Additionally, the display unit 5 can display power consumption data (see FIG. 3), past work plan data (see FIG. 4), or various results calculated by the power supply pattern model generation unit 3. Examples of the various results calculated by the power supply pattern model generation unit 3 include correlation data showing the correlation between power supply facilities S for each work (see FIGS. 5 to 7), a power supply pattern profile (see FIG. 8(a)), a power supply pattern operation state series (see FIG. 8(b)), a comparison graph between actual measured values ​​and calculated values ​​of energy consumption for each work (see FIG. 9(a)), or a comparison result using a confusion matrix (see FIG. 9(b)).

[0051] <Effects of the Present Embodiment> According to the energy management system 100 of the present embodiment configured as described above, multivariate analysis is performed on the power consumption data of each of the plurality of power supply facilities S, the plurality of power supply facilities S are classified into a plurality of power fluctuation patterns, and the power loads of the plurality of energy demand facilities X or the plurality of power supply facilities S according to the plurality of power fluctuation patterns, or the power loads over time according to the plurality of power fluctuation patterns, are calculated. Therefore, the user can consider points where energy can be saved (e.g., facilities corresponding to a pattern and their operating times) from these power loads. As a result, efficient efforts toward energy conservation can be made. Furthermore, the standard power consumption of the plurality of energy demand facilities can also be calculated from past power consumption data.

[0052] 12, energy analysis device 10 may further include an equipment management data acquisition unit 23. This equipment management data acquisition unit 23 acquires equipment management data (equipment management table) indicating the correspondence between energy demanding equipment X and power-source equipment S to which the energy demanding equipment X is connected.

[0053] Then, the power load calculation unit 4 calculates the power load of multiple power supply facilities S corresponding to multiple energy demand facilities X, or the power load over time corresponding to multiple energy demand facilities X, based on the power supply pattern profile and the facility management data.

[0054] Furthermore, the display unit 5 can display the calculation results of the power load amount for a predetermined period, such as daily or monthly, as a heat map, as shown in Fig. 13. Specifically, as shown in Fig. 13(a), the display unit 5 can display the power load amount for each component in a two-dimensional array in which the vertical axis (rows) represents multiple energy demand facilities and the horizontal axis (columns) represents multiple power supply facilities. As shown in Fig. 13(b), the display unit 5 can display the power load amount for each component in a two-dimensional array in which the vertical axis (rows) represents multiple energy demand facilities and the horizontal axis (columns) represents time.

[0055] 14 , the energy management system 100 may be configured such that the energy analysis unit (energy analysis device) 10 of the above embodiment has a power consumption data acquisition unit 21, a first work plan data acquisition unit 22, and a power supply pattern model generation unit 3, and the energy management unit (energy management device) 20 has a second work plan data acquisition unit 6, an energy demand prediction unit 7, an equipment control unit 8, and an energy saving amount calculation unit 9. The operation of this energy management system 100 is as shown in FIG. 15 .

[0056] The power consumption data acquisition unit 21, the first work plan data acquisition unit 22, and the power supply pattern model generation unit 3 have the same configurations as those in the above embodiment.

[0057] The second work plan data acquisition unit 6 acquires work plan data indicating, for example, a work plan to be carried out. This work plan data can be input by the user.

[0058] The energy demand prediction unit 7 calculates the power consumption predicted for the work plan based on the power pattern profile and / or power pattern operating state series generated by the power pattern model generation unit 3 and the work plan data acquired by the second work plan data acquisition unit 6.

[0059] The equipment control unit 8 controls the plurality of energy demanding facilities X based on the work plan data acquired by the second work plan data acquisition unit 6. Here, it is considered that the equipment control unit 8 adaptively controls the plurality of energy demanding facilities X.

[0060] The energy-saving amount calculation unit 9 acquires the power consumption of the multiple energy demanding facilities X as a result of the control by the equipment control unit 8, and calculates the amount of energy saving based on the power consumption predicted by the energy demand prediction unit 7. The energy-saving amount calculation unit 9 then determines whether the calculated amount of energy saving is equal to or greater than a target value. If the amount of energy saving is less than the target value, the user can review the work plan.

[0061] In addition, the power load calculation unit 4 can determine a steady-state range for each task from a statistical index for the variability in the power consumption data based on the power consumption data and past work plan data, and calculate the power consumption that is not included in the steady-state range for each task.

[0062] Specifically, the power load calculation unit 4 calculates a statistical index, such as a Z-score, for the variability of the power consumption data for each task based on the power consumption data and past task plan data. The power load calculation unit 4 then determines a steady-state range for each task based on the statistical index (Z-score value) and calculates the power consumption outside the steady-state range for each task. The steady-state range is determined based on the average value and standard deviation of the power consumption data for each task, and can be set, for example, within ±1σ to ±2σ of the average value. The steady-state range can also be adjusted depending on the number of samples (number of data points). Specifically, as shown in FIG. 16 , the power load calculation unit 4 determines a steady-state lower limit and a steady-state upper limit for each task, and calculates the total power consumption [kWh] and the power outside the steady-state range [kWh] for each task. The power load calculation unit 4 can also calculate the deviation (e.g., power outside the steady-state range [kWh] / total power consumption [kWh]) for each task from the total power consumption [kWh] and the power outside the steady-state range [kWh].

[0063] Here, the display unit 5 displays the out-of-steady-state range power, for example, as a bar graph, so that the tasks with large deviations can be visually grasped (see FIG. 16 ). The display unit 5 may also visually display the steady-state range on a graph that displays the power consumption data for each task.

[0064] Furthermore, as shown in Figure 17, if there is a correlation between a parameter of energy demanding equipment X and power consumption, a change in that parameter may be the cause of the deviation from the steady-state range. Conversely, a deviation from the steady-state range indicates that the parameter of energy demanding equipment X has been changed. Note that Figure 17 shows the power consumption of an air conditioner when the parameter is the wind speed of an air conditioning fan. Furthermore, by formulating the correlation between the parameter of energy demanding equipment X and power consumption, it is possible to quantify the deviation in power consumption due to a change in the parameter.

[0065] In the above embodiment, the energy analysis device 10 and the energy management device 20 are combined together. However, the energy analysis device 10 may be configured as a standalone device without being combined with the energy management device 20 .

[0066] Furthermore, the energy management system 100 is not limited to multivariate analysis, and may also predict the power consumption of each of multiple power supply facilities S from a work plan using multiple energy demand facilities X, using a learning model that shows the correlation between the operating history of multiple energy demand facilities X and the power consumption of multiple power supply facilities S in those operating history, as shown in Figure 18.

[0067] Here, the learning model is generated by a machine learning unit (not shown) by machine learning training data including work records (e.g., work plans such as test schedules that have already been carried out) using a plurality of energy demand facilities X and power consumption data indicating the power consumption per unit time of a plurality of power supply facilities S in the work records. The machine learning unit may be included in the energy management system 100, or may be included in a computer separate from the energy management system 100. This learning model is stored in the learning model storage unit 11.

[0068] The power consumption prediction unit 12 of the energy management system 100 predicts the power consumption per unit time of each of the plurality of power supply facilities S, using the work plan using the plurality of energy demand facilities input by the user and the learning model stored in the learning model storage unit 11. The power consumption prediction unit 12 can also calculate the total power consumption by adding up the power consumption of each of the plurality of power supply facilities S.

[0069] In addition, various modifications and combinations of the embodiments may be made as long as they do not go against the spirit of the present invention.

[0070] According to the present invention, by analyzing power consumption data of a plurality of power supply facilities, it is possible to make efficient efforts toward energy conservation.

[0071] REFERENCE SIGNS LIST 100: Energy management system X: Energy demand facility 10: Energy analysis unit (energy analysis device) 21: Power consumption data acquisition unit 22: First work plan data acquisition unit 3: Power supply pattern model generation unit 4: Power load calculation unit 5: Display unit 23: Equipment management data acquisition unit 20: Energy management unit (energy management device)

Claims

1. An energy analysis device that analyzes energy consumption in a plurality of energy demand facilities, comprising: a power consumption data acquisition unit that acquires power consumption data for each of a plurality of power supply facilities that supply power to the plurality of energy demand facilities; a power pattern model generation unit that performs multivariate analysis on the power consumption data to generate a power pattern profile that classifies each of the plurality of power supply facilities into a plurality of power fluctuation patterns and / or a power pattern operating state series that is the operating state of the power pattern profile over time; and a power load calculation unit that calculates the power load of the plurality of energy demand facilities or the plurality of power supply facilities in accordance with the plurality of power fluctuation patterns, or the power load over time in accordance with the plurality of power fluctuation patterns, based on the power pattern profile and / or the power pattern operating state series.

2. The energy analysis device of claim 1, further comprising a first work plan data acquisition unit that acquires past work plan data indicating past work plans using at least some of the plurality of energy demand facilities, wherein the power supply pattern model generation unit generates correlation data indicating the correlation between power supply facilities for each work based on the power consumption data and the past work plan data, and performs multivariate analysis of the generated correlation data to generate the power supply pattern profile and / or the power supply pattern operating state series.

3. The energy analysis device of claim 2, wherein the power supply pattern model generation unit compares the generated power supply pattern profile and / or the power supply pattern operating state series with the power consumption data and the past work plan data to confirm the accuracy of the generated power supply pattern profile and / or the power supply pattern operating state series.

4. The energy analysis device according to claim 2 or 3, wherein the power load calculation unit determines a steady-state range for each task from a statistical index for the variability of the power consumption data based on the power consumption data and the past work plan data, and calculates the power consumption that does not fall within the steady-state range for each task.

5. An energy analysis device according to any one of claims 1 to 4, further comprising an equipment management data acquisition unit that acquires equipment management data indicating the correspondence between the energy demand equipment and the power supply equipment to which the energy demand equipment is connected, wherein the power load calculation unit calculates the power load of the plurality of power supply equipment corresponding to the plurality of energy demand equipment, or the power load over time corresponding to the plurality of energy demand equipment, based on the power supply pattern profile and / or the power supply pattern operating state series and the equipment management data.

6. An energy analysis device according to any one of claims 1 to 5, further comprising a display unit that displays the calculation results of the power load obtained by the power load calculation unit.

7. The energy analysis device according to claim 6, wherein the display unit displays the calculation results of the amount of power load as a heat map.

8. An energy analysis device as claimed in any one of claims 1 to 7, wherein the power consumption data for each of the plurality of power supply facilities is data indicating the power consumption per predetermined time unit for each of the plurality of distribution boards to which the plurality of energy demand facilities are connected.

9. An energy management system comprising: an energy analysis device according to any one of claims 1 to 8; and an energy management device that performs energy management of the plurality of energy demand facilities based on the analysis results obtained by the energy analysis device.

10. An energy management system as described in claim 9, further comprising a plurality of energy demand facilities, wherein the energy management device performs energy management of the plurality of energy demand facilities based on the analysis results obtained by the energy analysis device.

11. A program used in an energy analysis device that analyzes energy consumption in multiple energy demand facilities, the energy analysis program providing a computer with the following functions: a power consumption data acquisition unit that acquires power consumption data for each of multiple power supply facilities that supply power to the multiple energy demand facilities; a power pattern model generation unit that performs multivariate analysis of the power consumption data to generate a power pattern profile that classifies each of the multiple power supply facilities into multiple power fluctuation patterns and / or a power pattern operating state series that is the operating state of the power pattern profile over time; and a power load calculation unit that calculates the power load of the multiple energy demand facilities or the multiple power supply facilities in accordance with the multiple power fluctuation patterns, or the power load over time in accordance with the multiple power fluctuation patterns, based on the power pattern profile and / or the power pattern operating state series.

12. An energy analysis method for analyzing energy consumption in a plurality of energy demand facilities, comprising: acquiring power consumption data for each of a plurality of power supply facilities that supply power to the plurality of energy demand facilities; performing multivariate analysis on the power consumption data to generate a power supply pattern profile that classifies each of the plurality of power supply facilities into a plurality of power fluctuation patterns and / or a power supply pattern operating state series that is the operating state of the power supply pattern profile over time; and calculating, based on the power supply pattern profile and / or the power supply pattern operating state series, the power load of the plurality of energy demand facilities or the plurality of power supply facilities corresponding to the plurality of power fluctuation patterns, or the power load over time corresponding to the plurality of power fluctuation patterns.

Images