JP7551672B2 - Power management device and power management method - Google Patents

Description

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

In recent years, technologies that utilize the charging and discharging power (hereinafter referred to as regulated power) of energy storage devices (such as VPP; Virtual Power Plant) have been attracting attention for the purpose of stabilizing the supply and demand balance of the power grid.

Furthermore, technology is also known that develops a charging and discharging plan for an energy storage device based on the predicted value of the output power of a power generation device, the predicted value of the power consumption of a load device, and the predicted value of the power purchase price (for example, Patent Documents 1 and 2).

Patent No. 5672186 Patent No. 6386744

However, it is difficult to accurately predict the predicted value of the output power of a power generation device that uses natural energy, such as a solar cell device, and prediction errors are expected to occur. Similarly, it is difficult to accurately predict the predicted value of the power consumption of a load device, and prediction errors are expected to occur.

Under these assumptions, if a charging and discharging plan for the energy storage device is simply formulated based on the predicted value of the output power of the power generation device and the predicted value of the power consumption of the load equipment, if a prediction error occurs in the prediction value, it may not be possible to properly secure the regulating power of the energy storage device.

The present invention has been made to solve the above-mentioned problems, and aims to provide a power management device and a power management method that make it possible to formulate a charge/discharge plan that can appropriately ensure the regulating power of the storage device.

One aspect of the disclosure is a power management device that includes a control unit that formulates a charge/discharge plan to be applied to a power storage device, and the control unit identifies N (N is an integer less than M) scenario groups by clustering M (M is an integer equal to or greater than 2) scenarios that indicate a transition in a predetermined power related to the power demand of a facility having the power storage device, and formulates the charge/discharge plan by aggregating the identified N scenario groups.

One aspect of the disclosure is a power management method that includes the steps of: identifying N (N is an integer less than M) scenario groups by clustering M (M is an integer equal to or greater than 2) scenarios that indicate a change in a predetermined power level related to the power demand of a facility having a power storage device; and formulating a charge/discharge plan to be applied to the power storage device by aggregating the identified N scenario groups.

The present invention provides a power management device and a power management method that enable the formulation of a charge/discharge plan that can appropriately ensure the regulating power of a power storage device.

FIG. 1 is a diagram showing a power management system 100 according to an embodiment. FIG. 2 is a diagram showing a facility 300 according to the embodiment. FIG. 3 is a diagram showing a power management server 200 according to the embodiment. FIG. 4 is a diagram illustrating a local control device 360 according to an embodiment. FIG. 5 is a diagram for explaining a predicted value and a prediction error according to the embodiment. FIG. 6 is a diagram for explaining the specification of a scenario according to the embodiment. FIG. 7 is a diagram for explaining the specification of a scenario according to the embodiment. FIG. 8 is a diagram for explaining the specification of a scenario according to the embodiment. FIG. 9 is a diagram illustrating a power management method according to an embodiment. FIG. 10 is a diagram for explaining the identification of a scenario according to the first modification. FIG. 11 is a diagram for explaining the identification of a scenario according to the first modification. FIG. 12 is a diagram for explaining the identification of a scenario according to the first modification. FIG. 13 is a diagram showing a power management method according to the first modification. FIG. 14 is a diagram for explaining display control according to the second modification. FIG. 15 is a diagram for explaining display control according to the second modification.

The following describes the embodiments with reference to the drawings. In the following description of the drawings, the same or similar parts are denoted by the same or similar reference numerals. However, the drawings are schematic.

[Embodiment]
(Power Management System)
In the following, the power management system will be described.

As shown in FIG. 1, the power management system 100 includes a power management server 200, a facility 300, and a power company 400. In FIG. 1, facilities 300A to 300C are shown as examples of the facility 300.

Each facility 300 is connected to the power grid 110. In the following, the flow of power from the power grid 110 to the facility 300 is referred to as the power flow, and the flow of power from the facility 300 to the power grid 110 is referred to as the reverse power flow.

The power management server 200, the facility 300, and the power company 400 are connected to a network 120. The network 120 may provide lines between these entities. For example, the network 120 is the Internet. The network 120 may also include a dedicated line such as a VPN (Virtual Private Network).

The power management server 200 is a server managed by a business operator such as a power generation business operator, a power transmission and distribution business operator, a retail business operator, or a resource aggregator. The resource aggregator may be a power business operator that adjusts the power supply and demand balance of the power system 110 in a VPP (Virtual Power Plant). The adjustment of the power supply and demand balance may include a transaction (hereinafter, negawatt trading) in which a reduction in the forward flow power (also referred to as the demand power of the facility 300) supplied from the power system 110 to the facility 300 is exchanged for value. The adjustment of the power supply and demand balance may include a transaction in which an increase in the reverse flow power supplied from the facility 300 to the power system 110 is exchanged for value. Hereinafter, a request for adjustment of the power supply and demand balance may be referred to as a power saving request. The resource aggregator may be a power business operator that provides reverse flow power to a power generation business operator, a power transmission and distribution business operator, a retail business operator, or the like in a VPP.

The power management server 200 transmits a control message to the local control device 360 provided in the facility 300 to instruct the control of the distributed power source (e.g., the solar cell device 310, the power storage device 320, or the fuel cell device 330) provided in the facility 300. For example, the power management server 200 may transmit a forward flow control message (e.g., DR; Demand Response) requesting control of forward flow power, or may transmit a reverse flow control message requesting control of reverse flow power. Furthermore, the power management server 200 may transmit a power source control message to control the operating state of the distributed power source. The degree of control of the forward flow power or reverse flow power may be expressed as an absolute value (e.g., XX kW) or a relative value (e.g., XX%). Alternatively, the degree of control of the forward flow power or reverse flow power may be expressed in two or more levels. The degree of control of forward flow power or reverse flow power may be represented by the electricity price determined by the current electricity supply and demand balance (RTP: Real Time Pricing), or by the electricity price determined by the past electricity supply and demand balance (TOU: Time Of Use).

As shown in FIG. 2, the facility 300 has a solar cell device 310, a power storage device 320, a fuel cell device 330, a load device 340, a local control device 360, a power meter 380, and a power meter 390.

The solar cell device 310 is a distributed power source that generates power in response to light such as sunlight. The solar cell device 310 may be an example of a distributed power source that allows reverse power flow. The solar cell device 310 may be an example of a distributed power source to which a feed-in tariff (FIT) may be applied. The solar cell device 310 may be a distributed power source for which the application period of the feed-in tariff has expired. For example, the solar cell device 310 is composed of a PCS (Power Conditioning System) and solar panels.

Here, the power output from the solar cell device 310 can vary depending on the amount of light received, such as sunlight. Therefore, when the power generation efficiency of the solar cell device 310 is taken into consideration, the power output from the solar cell device 310 is variable power that can vary depending on the amount of light received by the solar panel.

The power storage device 320 is a distributed power source that charges and discharges power. The power storage device 320 may be an example of a distributed power source in which reverse power flow is not permitted. The power storage device 320 may be an example of a distributed power source to which a feed-in tariff cannot be applied. For example, the power storage device 320 is composed of a PCS and a power storage cell.

The fuel cell device 330 is a distributed power source that generates power using fuel. The fuel cell device 330 may be an example of a distributed power source that does not allow reverse power flow. The fuel cell device 330 may be an example of a distributed power source to which a feed-in tariff does not apply. For example, the fuel cell device 330 is composed of a PCS and a fuel cell.

For example, the fuel cell device 330 may be a solid oxide fuel cell (SOFC), a polymer electrolyte fuel cell (PEFC), a phosphoric acid fuel cell (PAFC), or a molten carbonate fuel cell (MCFC).

In an embodiment, the solar cell device 310, the power storage device 320, and the fuel cell device 330 may be regulated power sources used in the VPP. The regulated power sources are distributed power sources installed in the facility 300 that contribute to the VPP.

Load device 340 is a device that consumes power. For example, load device 340 is an air conditioner, a lighting device, an AV (audio visual) device, etc.

The local control device 360 is a device (EMS: Energy Management System) that manages the power of the facility 300. The local control device 360 may control the operating state of the solar cell device 310, may control the operating state of the power storage device 320 provided in the facility 300, or may control the operating state of the fuel cell device 330 provided in the facility 300. Details of the local control device 360 will be described later (see FIG. 4).

In the embodiment, communication between the power management server 200 and the local control device 360 is performed according to a first protocol. On the other hand, communication between the local control device 360 and the distributed power source (the solar cell device 310, the power storage device 320, or the fuel cell device 330) is performed according to a second protocol different from the first protocol. For example, the first protocol may be a protocol conforming to Open ADR (Automated Demand Response) or a unique dedicated protocol. For example, the second protocol may be a protocol conforming to ECHONET Lite, SEP (Smart Energy Profile) 2.0, KNX, or a unique dedicated protocol. Note that the first protocol and the second protocol may be different, and for example, even if both are unique dedicated protocols, they may be protocols created according to different rules. However, the first protocol and the second protocol may be protocols created according to the same rules.

The power meter 380 is an example of a core power meter that measures forward flow power from the power grid 110 to the facility 300 and reverse flow power from the facility 300 to the power grid 110. For example, the power meter 380 is a smart meter belonging to the power company 400.

Here, the power meter 380 transmits a message including an information element indicating the integrated value of the forward flow power or reverse flow power in the unit time to the local control device 360 for each unit time (e.g., 30 minutes). The power meter 380 may transmit the message autonomously, or may transmit the message in response to a request from the local control device 360. The power meter 380 may transmit a message including an information element indicating the forward flow power or reverse flow power in the unit time to the power management server 200 for each unit time.

The power meter 390 is an example of an individual power meter that measures the power of each distributed power source (the solar cell device 310, the power storage device 320, and the fuel cell device 330). The power meter 390 may be provided at the output end of the PCS of the distributed power source, and may be considered to be part of the distributed power source. In FIG. 2, power meter 391, power meter 392, and power meter 393 are set as the power meter 390. The power meter 391 measures the output power of the solar cell device 310. The power meter 392 measures the discharge power of the power storage device 320. The power meter 392 may measure the charging power of the power storage device 320. The power meter 393 measures the individual output power of the fuel cell device 330.

Here, the power meter 390 may transmit a message including an information element indicating the power of the distributed power source to the local control device 360 at intervals shorter than the unit time (e.g., one minute). The individual output power of the regulated power source may be represented by an instantaneous value or an integrated value. The power meter 390 may transmit the message autonomously or in response to a request from the local control device 360.

Returning to FIG. 1, the electric power company 400 is an entity that provides infrastructure such as the electric power system 110, and is, for example, a power generation company or a power transmission and distribution company. The electric power company 400 may outsource various operations to an entity that manages the power management server 200.

(Power management server)
The power management server will be described below. As shown in Fig. 3, the power management server 200 includes a management unit 210, a communication unit 220, and a control unit 230. The power management server 200 may be an example of a VTN (Virtual Top Node). In the embodiment, the power management server 200 is an example of a power management device.

The management unit 210 is configured with storage media such as a hard disk drive (HDD), a solid state drive (SSD), and a non-volatile memory, and manages information related to the facility 300. For example, the information related to the facility 300 includes the type of distributed power source (solar cell device 310, power storage device 320, or fuel cell device 330) installed in the facility 300, and the specifications of the distributed power source (solar cell device 310, power storage device 320, or fuel cell device 330) installed in the facility 300. The specifications may include the rated power generation of the solar cell device 310, the rated charging power of the power storage device 320, the rated charging power of the power storage device 320, and the rated output power of the fuel cell device 330. The specifications may include the rated capacity of the power storage device 320, the maximum charging and discharging power, etc.

The communication unit 220 is configured by a communication module, and communicates with the local control device 360 via the network 120. The communication module may be a wireless communication module that complies with standards such as IEEE802.11a/b/g/n, ZigBee, Wi-SUN, LTE, 5G, and 6G, or may be a wired communication module that complies with standards such as IEEE802.3.

As described above, the communication unit 220 communicates according to the first protocol. For example, the communication unit 220 transmits a first message to the local control device 360 according to the first protocol. The communication unit 220 receives a first message response from the local control device 360 according to the first protocol.

The control unit 230 may include at least one processor. The at least one processor may be configured by a single integrated circuit (IC), or may be configured by multiple circuits (such as integrated circuits and/or discrete circuits) communicatively connected.

The control unit 230 controls each component provided in the power management server 200. For example, the control unit 230 instructs the local control device 360 provided in the facility 300 to control the distributed power sources (the solar cell device 310, the power storage device 320, or the fuel cell device 330) provided in the facility 300 by transmitting a control message. The control message may be a power flow control message, a reverse power flow control message, or a power source control message, as described above.

In the embodiment, the control unit 230 constitutes a control unit that formulates a charge/discharge plan to be applied to the power storage device 320. The control unit 230 identifies N (N is an integer less than M) scenario groups by clustering M (M is an integer equal to or greater than 2) scenarios as scenarios that indicate a transition in a predetermined power related to the power demand of a facility having the power storage device 320. The control unit 230 formulates a charge/discharge plan by aggregating the identified N scenario groups. The method of aggregating the scenario groups will be described later.

Although not limited thereto, a scenario may be represented by a progression of a given power. That is, a scenario may be defined by a given power and time. Clustering is a procedure for classifying two or more scenarios that are similar to each other into one group.

The specified power may include the power consumption of the load device 340, or may include the output power of the power generation device (the solar cell device 310 and the fuel cell device 330). The specified power may be the demand power (i.e., tidal power or demand power), or may be the power consumption of the load device 340 minus the output power of the power generation device.

The control unit 230 controls the charging and discharging of the power storage device 320 based on the formulated charging and discharging plan. The control unit 230 may indirectly control the power storage device 320 by transmitting the formulated charging and discharging plan to the local control device 360. Alternatively, the control unit 230 may directly control the power storage device 320 by transmitting a control command based on the formulated charging and discharging plan to the power storage device 320.

Here, if the specified power deviates from the scenario, the charging and discharging of the power storage device 320 may be controlled to adjust the deviation between the scenario and the specified power. Such charging and discharging may be controlled by the power management server 200 or by the local control device 360.

In the embodiment, the charging power of the power storage device 320 is an example of adjustment power used to adjust the supply and demand balance of the power system 110. Similarly, the discharging power of the power storage device 320 is an example of adjustment power used to adjust the supply and demand balance of the power system 110. In other words, the charging and discharging plan may be considered to be a schedule for securing adjustment power used to adjust the supply and demand balance of the power system 110.

(Local Control Device)
The local control device 360 will be described below. As shown in Fig. 4, the local control device 360 includes a first communication unit 361, a second communication unit 362, and a control unit 363. The local control device 360 may be an example of a VEN (Virtual End Node).

The first communication unit 361 is configured by a communication module, and communicates with the power management server 200 via the network 120. The communication module may be a wireless communication module that complies with standards such as IEEE802.11a/b/g/n, ZigBee, Wi-SUN, LTE, 5G, and 6G, or may be a wired communication module that complies with standards such as IEEE802.3.

As described above, the first communication unit 361 communicates according to the first protocol. For example, the first communication unit 361 receives a first message from the power management server 200 according to the first protocol. The first communication unit 361 transmits a first message response to the power management server 200 according to the first protocol.

The second communication unit 362 is configured by a communication module and communicates with the distributed power source (the solar cell device 310, the power storage device 320, or the fuel cell device 330). The communication module may be a wireless communication module that complies with standards such as IEEE802.11a/b/g/n, ZigBee, Wi-SUN, LTE, 5G, or 6G, or may be a wired communication module that complies with standards such as IEEE802.3 or a proprietary protocol.

As described above, the second communication unit 362 communicates according to the second protocol. For example, the second communication unit 362 transmits a second message to the distributed power source according to the second protocol. The second communication unit 362 receives a second message response from the distributed power source according to the second protocol.

For example, the second communication unit 362 may receive a message from the power meter 380 that includes an information element that identifies the forward flow power or the reverse flow power. The second communication unit 362 may receive a message from the power meter 390 that includes an information element that identifies the power of each distributed power source.

The control unit 363 may include at least one processor. The at least one processor may be configured by a single integrated circuit (IC), or may be configured by multiple circuits (such as integrated circuits and/or discrete circuits) communicatively connected.

The control unit 363 controls each component provided in the local control device 360. Specifically, the control unit 363 instructs the devices to set the operating state of the distributed power source by sending a second message and receiving a second message response in order to control the power of the facility 300. The control unit 363 may instruct the distributed power source to report information about the distributed power source by sending a second message and receiving a second message response in order to manage the power of the facility 300.

The control unit 363 constitutes a control unit that executes charging control to charge the power storage device using surplus power generated by the output power of the power generation device. The control unit 363 may also execute discharging control of the power storage device.

For example, the power generation device may be a solar cell device 310 that allows reverse power flow. The power storage device may be a power storage device 320 that is charged using surplus power. The surplus power may be the difference between the power generated by the solar cell device 310 and the power consumption of the facility 300. The power consumption of the facility 300 may be the power consumption of the load device 340. When the fuel cell device 330 is outputting power, the power consumption of the facility 300 may be the power consumption of the load device 340 minus the output power of the fuel cell device 330. The power consumption of the facility 300 may be the total power consumption of the devices (e.g., the solar cell device 310, the power storage device 320, the fuel cell device 330, the load device 340, the local control device 360, the power meter 380, and the power meter 390) installed in the facility 300. In the embodiment, the power consumption of the facility 300 may not include the charging power of the power storage device 320.

In the following, the surplus power may be used as reverse flow power from the facility 300 to the power grid 110, or may be used as charging power for the power storage device 320.

The power demand of the facility 300 may be synonymous with the forward flow power from the power grid 110 to the facility 300. When the power storage device 320 is being charged, the power demand of the facility 300 may be the sum of the power consumed by the facility 300 and the charging power of the power storage device 320. When the power storage device 320 is being discharged, the power demand of the facility 300 may be the power consumed by the facility 300 minus the discharged power of the power storage device 320.

(Range of predicted value)
The following describes the range of fluctuation of the predicted value of the specified power related to the power demand. The following illustrates the prediction error of the predicted value as the range of fluctuation of the predicted value. FIG. 5 illustrates the predicted value and the prediction error for a specified period. The specified period may be a period for which a charge/discharge plan should be formulated. Although not particularly limited, here, the specified power is exemplified as the power demand, and the specified period is exemplified as one day. The predicted value and the prediction error may be acquired at a specified granularity (e.g., 30 minutes).

The predicted value and prediction error may be determined by learning actual values of power demand. For example, a prediction model may be generated by learning actual values of power demand and learning parameters that affect power demand, and input parameters corresponding to the learning parameters may be input to the prediction model, thereby outputting a predicted value and prediction error of power demand from the prediction model. The learning parameters may include at least a parameter that specifies the time. The learning parameters may include parameters that specify the day of the week, month, season, weather, temperature, humidity, etc.

Although FIG. 5 illustrates an example in which the specified power is the demand power, the specified power may be either the generated power or the demand power. In such a case, the learning parameters may include parameters that affect the generated power (demand power). For example, the learning parameters may include parameters that specify the weather, temperature, humidity, amount of solar radiation, etc. that affect the generated power of the solar cell device 310.

(How scenarios are compiled)
A method for aggregating scenarios will be described below. In the following, the operation of the control unit 230 of the power management server 200 will be mainly described. In the following, the explanation will be continued assuming that a scenario is defined by a predetermined power (vertical axis) and time (horizontal axis).

First, as shown in FIG. 6, the control unit 230 sets M scenarios based on the predicted value of the specified power (demanded power in FIG. 6). The control unit 230 may set M scenarios within the fluctuation range of the predicted value of the specified power.

For example, the control unit 230 may set M scenarios by changing the predicted value within the fluctuation range of the predicted value using random numbers for each predetermined granularity (e.g., 30 minutes) as described above. Setting M scenarios may be referred to as sampling of the predicted value within the fluctuation range of the predicted value. The control unit 230 may set M scenarios by sampling using machine learning or a mathematical model, without using random numbers. The control unit 230 may reflect white noise in the M scenarios.

Secondly, as shown in FIG. 7, the control unit 230 may identify N scenario groups by clustering M scenarios. Known clustering methods (e.g., Ward's method, K-means method, etc.) may be used. FIG. 7 illustrates an example in which M scenarios are clustered into four groups (scenario groups).

Although not particularly limited, scenario group #1 may be a "morning peak type", scenario group #2 may be an "evening peak type", scenario group #3 may be a "morning/evening peak type", and scenario group #4 may be a "non-peak type". However, at the clustering stage, it is not necessary to assign meanings such as "morning peak type", "evening peak type", "morning/evening peak type", and "non-peak type".

For example, the control unit 230 may exclude scenario groups in which the number of scenarios included in the scenario group is less than a threshold. The threshold may be defined as a relative value with respect to M (e.g., when M=10,000, the threshold=10).

The control unit 230 may exclude scenario groups in which the peak of the specified power occurs a specified number of times (e.g., three times) or more in a specified period. The specified number may be defined by the number of charge/discharge cycles that are acceptable for the power storage device 320 in the specified period.

The number N may be arbitrarily set by the user. Alternatively, the number N may be determined by deep learning, such as AI (Artificial Intelligence). Alternatively, the number N may be determined by the number of types of lifestyle patterns, if the number of types of lifestyle patterns of the user of facility 300 can be determined. Although not particularly limited, in the case where the user of facility 300 is a business person living alone, the lifestyle patterns may include patterns for when going to the office, patterns for when working from home, and patterns for when on holidays (or paid leave). For example, the number N may be the number of types of lifestyle patterns. In such a case, it may be assumed that the family structure of the user of facility 300 is known.

Here, the control unit 230 determines the trend of the specified power for each of the N scenario groups based on at least one of the average value, weighted average value, median value, mode value, maximum value, and minimum value of the scenarios classified into each of the N scenario groups. The trend of the specified power for each of the N scenario groups may be considered to be a representative value for the N scenario groups.

Third, as shown in FIG. 8, the control unit 230 calculates the probability of each of the N scenario groups based on the number of scenarios classified into each of the N scenario groups. The control unit 230 aggregates (representative values of) the identified N scenario groups based on the calculated probabilities.

For example, assume that M=10,000 and the number of scenarios included in scenario group #1 is 2,000, the number of scenarios included in scenario group #2 is 3,000, the number of scenarios included in scenario group #3 is 4,000, and the number of scenarios included in scenario group #4 is 1,000. In such a case, the probability of scenario group #1 is 0.2 (2,000/10,000), the probability of scenario group #2 is 0.3 (3,000/10,000), the probability of scenario group #3 is 0.4 (4,000/10,000), and the probability of scenario group #4 is 0.1 (1,000/10,000).

For example, the control unit 230 may aggregate the representative values of the N scenario groups so as to minimize the electricity bill corresponding to the representative values of the N scenario groups. The objective function to be minimized may be expressed as shown below.

The control unit 230 charges and discharges the energy storage device 320 based on the aggregated scenario group (representative value). Specifically, the control unit 230 charges and discharges the energy storage device 320 so as to approach the scenario group (representative value). In other words, the charge and discharge plan for the energy storage device 320 can be considered as a plan to approach the scenario group (representative value).

(Power Management Method)
In the following, the power management method is described.

As shown in FIG. 9, in step S11, the power management server 200 acquires information necessary for setting a scenario. The information necessary for setting a scenario may include a predicted value of a specified power. The information necessary for setting a scenario may include a fluctuation range of the predicted value of a specified power.

In step S12, the power management server 200 sets M scenarios based on the predicted value of the specified power. The power management server 200 may set M scenarios within the fluctuation range of the predicted value of the specified power (see FIG. 6).

In step S13, the power management server 200 identifies N scenario groups by clustering the M scenarios. The power management server 200 identifies a representative value for each of the N scenario groups based on at least one of the average, weighted average, median, mode, maximum, and minimum values of the scenarios classified into each of the N scenario groups (see FIG. 7).

In step S14, the power management server 200 calculates the probability of each of the N scenario groups based on the number of scenarios classified into each of the N scenario groups (see FIG. 8).

In step S15, the power management server 200 identifies the scenario to be used in formulating the charge/discharge plan for the energy storage device 320 by aggregating (representative values of) the identified N scenario groups based on the calculated probability (see FIG. 8).

In step S16, the power management server 200 formulates a charging and discharging plan for the power storage device 320 based on the scenario identified in step S14.

Although FIG. 9 describes steps S15 and S16 as separate steps, the aggregation of (representative values of) the N scenario groups and the formulation of a charge/discharge plan for the power storage device 320 may be performed in a single step. For example, the power management server 200 may formulate a charge/discharge plan for the power storage device 320 so as to minimize the electricity bill corresponding to the representative values of the N scenario groups. Aggregating representative values of the N scenario groups may be considered to be synonymous with formulating a charge/discharge plan for the power storage device 320.

In step S17, the power management server 200 controls the charging and discharging of the power storage device 320 based on the formulated charging and discharging plan. For example, the power management server 200 may indirectly control the power storage device 320 based on the formulated charging and discharging plan, or may directly control the power storage device 320.

(Action and Effects)
In the embodiment, the power management server 200 identifies N scenario groups by clustering M scenarios as scenarios showing a transition of a predetermined power related to the power demand of a facility having the power storage device 320. The power management server 200 identifies a scenario to be used in formulating a charge/discharge plan for the power storage device 320 by aggregating the identified N scenario groups. With this configuration, it is possible to bias the distribution of predicted values for each predetermined granularity (e.g., 30 minutes) by clustering and aggregation, and it is possible to appropriately identify a scenario to be used in formulating a charge/discharge plan for the power storage device 320. In other words, even if the prediction of the predetermined power is incorrect, a robust charge/discharge plan can be formulated.

In an embodiment, the power management server 200 calculates the probability of each of the N scenario groups based on the number of scenarios classified into each of the N scenario groups. With this configuration, it is possible to appropriately aggregate (representative values of) the N scenario groups.

[Change Example 1]
Modification 1 of the embodiment will be described below. Differences from the embodiment will be mainly described below.

In the above-described embodiment, M scenarios are set based on a predicted value of the predetermined power, whereas in the first modification, M scenarios are set based on an actual value of the predetermined power.
(Scenario Identification)

The following describes how to identify the scenario used to formulate the charge/discharge plan. The following mainly describes the operation of the control unit 230 of the power management server 200.

First, as shown in FIG. 10, the control unit 230 sets M scenarios based on the actual value of a specified power. The actual value of the specified power may be a reference period (e.g., the past month, the past three months, the past year, the past three years, etc.).

For example, if the specified period for which a charge/discharge plan is to be formulated is one day, then if the reference period is one month, the scenarios with the number of days equivalent to one month are set as M scenarios, and if the reference period is one year, the scenarios with the number of days equivalent to one year are set as M scenarios.

Furthermore, the reference period may be selected to be a period similar to the target period based on parameters such as season, weather, temperature, humidity, etc. The reference period may be a continuous period or a non-continuous period.

Secondly, as shown in FIG. 11, the control unit 230 may identify N scenario groups by clustering M scenarios. Known clustering methods (e.g., Ward's method, K-means method, etc.) may be used. FIG. 11 illustrates an example in which M scenarios are clustered into four groups (scenario groups).

Although not particularly limited, scenario group #1 may be a "morning peak type", scenario group #2 may be an "evening peak type", scenario group #3 may be a "morning/evening peak type", and scenario group #4 may be a "non-peak type". However, at the clustering stage, it is not necessary to assign meanings such as "morning peak type", "evening peak type", "morning/evening peak type", and "non-peak type".

For example, the control unit 230 may exclude scenario groups in which the number of scenarios included in the scenario group is less than a threshold. The threshold may be defined as a relative value with respect to M (e.g., when M=10,000, the threshold=10).

The control unit 230 may exclude scenario groups in which the peak of the specified power occurs a specified number of times (e.g., three times) or more in a specified period. The specified number may be defined by the number of charge/discharge cycles that are acceptable for the power storage device 320 in the specified period.

The number N may be arbitrarily set by the user. Alternatively, the number N may be determined by deep learning, such as artificial intelligence (AI).

Here, the control unit 230 determines the trend of the specified power for each of the N scenario groups based on at least one of the average value, weighted average value, median value, mode value, maximum value, and minimum value of the scenarios classified into each of the N scenario groups. The trend of the specified power for each of the N scenario groups may be considered to be a representative value for the N scenario groups.

Thirdly, as shown in FIG. 12, the control unit 230 calculates the similarity between the scenarios classified into each of the N scenario groups and the predicted value of the specified power. The control unit 230 aggregates (representative values of) the identified N scenario groups based on the calculated similarity. Known methods (such as Euclidean distance and cosine similarity) may be used to calculate the similarity.

Furthermore, the similarity of each of the N scenario groups may be converted into a probability of each of the N scenario groups. For example, the similarity may be converted into a probability by a softmax function (f(x)=exp(x _i )/Σexp(x _i )). However, if each similarity is in the range of 0 to 1 and the sum of the similarities is 1, the process of converting the similarity into a probability may be omitted.

For example, the control unit 230 may aggregate the representative values of the N scenario groups so as to minimize the electricity bill corresponding to the representative values of the N scenario groups. The objective function to be minimized may be expressed as shown below.

(Power Management Method)
In the following, the power management method is described.

As shown in FIG. 13, in step S21, the power management server 200 acquires information necessary for setting a scenario. The information necessary for setting a scenario may include an actual value of a specified power. The actual value of a specified power may be considered to be the M scenarios themselves.

In step S22, the power management server 200 identifies N scenario groups by clustering the M scenarios. The power management server 200 identifies a representative value for each of the N scenario groups based on at least one of the average, weighted average, median, mode, maximum, and minimum values of the scenarios classified into each of the N scenario groups (see FIG. 10).

In step S23, the power management server 200 calculates the similarity between the scenarios classified into each of the N scenario groups and the predicted value of the specified power (see FIG. 11). The power management server 200 may convert the similarity into a probability.

In step S24, the power management server 200 identifies the scenarios to be used in formulating the charge/discharge plan for the energy storage device 320 by aggregating (representative values of) the identified N scenario groups based on the calculated similarity (or probability) (see FIG. 12).

In step S25, the power management server 200 formulates a charging and discharging plan for the power storage device 320 based on the scenario identified in step S24.

In step S26, the power management server 200 controls the charging and discharging of the power storage device 320 based on the formulated charging and discharging plan. For example, the power management server 200 may indirectly control the power storage device 320 based on the formulated charging and discharging plan, or may directly control the power storage device 320.

(Action and Effects)
In the first modification, the power management server 200 identifies N scenario groups by clustering M scenarios as scenarios showing a transition of a predetermined power related to the power demand of a facility having the power storage device 320. The power management server 200 identifies a scenario to be used in formulating a charge/discharge plan for the power storage device 320 by aggregating the identified N scenario groups. With this configuration, it is possible to bias the distribution of predicted values for each predetermined granularity (e.g., 30 minutes) by clustering and aggregation, and it is possible to appropriately identify a scenario to be used in formulating a charge/discharge plan for the power storage device 320. In other words, even if the prediction of the predetermined power is incorrect, a robust charge/discharge plan can be formulated.

In modification example 1, the power management server 200 calculates the similarity between the scenarios classified into each of the N scenario groups and the predicted value of a specified power. With this configuration, it is possible to appropriately aggregate the N scenario groups (representative values).

[Change Example 2]
Modification 2 of the embodiment will be described below. Differences from the embodiment will be mainly described below.

Specifically, in Modification Example 2, display control for users of the facility 300 will be described in a case where the power storage device 320 is controlled based on the above-mentioned charge/discharge plan. Modification Example 2 illustrates a case where the display control is executed by the local control device 360.

First, an example of display control in charging control will be described. The local control device 360 executes display control to display the image shown in FIG. 14.

As shown in FIG. 14, the local control device 360 may execute display control to notify the user of information indicating at least one of the fact that the total amount of charging power is secured and the effect of charging control. For example, the character string "Charging completion time is scheduled for XX to XX" is an example of information indicating that the total amount of charging power is secured. The character string "If a power saving request is made, the electricity bill will be reduced by XX to XX yen" is an example of information indicating the effect of charging control.

The local control device 360 may execute display control to notify the user of information indicating that the power storage device 320 is being charged using surplus power and that reverse flow of the surplus power is occurring. For example, "charging power = xx kW" is an example of information indicating that the power storage device 320 is being charged using surplus power. "Power sold = xx kW" is an example of information indicating that reverse flow of the surplus power is occurring.

The local control device 360 may execute display control to notify the user of information indicating that costs have been optimized for the facility 300 while charging control is being performed. For example, character strings such as "Charging completion time is scheduled for XX to XX" and "If a power saving request is made, electricity bills will be reduced by XX to XX yen" are examples of information indicating that costs have been optimized for the facility 300. Information indicating that costs have been optimized for the facility 300 is not limited to this, and may also be a character string such as "Electricity bills have been optimized."

Secondly, an example of display control in discharge control will be described. The local control device 360 executes display control to display the image shown in FIG. 15.

As shown in FIG. 15, the local control device 360 may execute display control to notify the user of information indicating at least one of the fact that the total amount of discharge power is secured and the effect of the discharge control. For example, the character string "Discharge completion time is scheduled for XX to XX" is an example of information indicating that the total amount of discharge power is secured. The character string "If a power saving request is made, the electricity bill will be reduced by XX to XX yen" is an example of information indicating the effect of the discharge control.

The local control device 360 may execute display control to notify the user of information indicating that the power storage device 320 is discharging. For example, "Discharging power = xx kW" is an example of information indicating that the power storage device 320 is discharging. The local control device 360 may execute display control to notify the user of information indicating that power is being supplied from the power grid 110. For example, "Purchased power = xx kW" is an example of information indicating that power is being supplied from the power grid 110.

The local control device 360 may execute display control to notify the user of information indicating that costs have been optimized for the facility 300 while discharge control is being performed. For example, the character strings "Discharge completion time is scheduled for XX to XX" and "If a power saving request is made, electricity bills will be reduced by XX to XX yen" are examples of information indicating that costs have been optimized for the facility 300. The information indicating that costs have been optimized for the facility 300 is not limited to this, and may also be a character string such as "Electricity bills have been optimized."

[Other embodiments]
Although the present invention has been described by the above disclosure, the description and drawings forming a part of this disclosure should not be understood as limiting the present invention. From this disclosure, various alternative embodiments, examples and operating techniques will become apparent to those skilled in the art.

In the above disclosure, a case has been exemplified in which the power management device that formulates the charge/discharge plan is the power management server 200. However, the above disclosure is not limited to this. The power management device that formulates the charge/discharge plan may be the local control device 360.

In the above disclosure, the prediction error of the predicted value is exemplified as the range of variation of the predicted value. However, the above disclosure is not limited to this. A predictive distribution of the predicted value may be used as the range of variation of the predicted value. The predictive distribution of the predicted value may be determined by a method such as quantile regression or Bayesian linear regression. In such a case, the prediction error may be read as the predictive distribution in the above embodiment.

In the above disclosure, the solar cell device 310 is exemplified as a power generation device that allows reverse power flow. However, the above disclosure is not limited to this. A power generation device that allows reverse power flow may include a distributed power source that outputs electric power using renewable energy. Such a distributed power source may include one or more distributed power sources selected from a wind power generation device, a hydroelectric power generation device, a geothermal power generation device, and a biomass power generation device. Note that reverse power flow of the discharge power of the power storage device 320 may be permitted, and reverse power flow of the output power of the fuel cell device 330 may be permitted.

Although not specifically mentioned in the above disclosure, the power storage device 320 may include a stationary power storage device, or may include a power storage device mounted on an electric vehicle.

In the above disclosure, display control is exemplified as a notification control for notifying a user of information. However, the above disclosure is not limited to this. Notification control may include audio output control for notifying a user of information by audio.

In the above disclosure, a case where the local control device 360 is provided in the facility 300 has been exemplified. However, the above disclosure is not limited to this. The local control device 360 may be provided by a cloud service implemented by a server or the like provided on the network 120.

Although not specifically mentioned in the above disclosure, power may be an instantaneous value (kW) or an integrated value per unit time (kWh).

Although not specifically mentioned in the above disclosure, the above disclosure may contribute to the Sustainable Development Goals (SDGs): Goal 7 (Affordable and clean energy), Goal 9 (Industry, innovation and infrastructure), Goal 11 (Sustainable cities and communities), and Goal 13 (Climate action).

100...power management system, 110...power system, 120...network, 200...power management server, 210...management unit, 220...communication unit, 230...control unit, 300...facility, 310...solar cell device, 320...power storage device, 330...fuel cell device, 340...load device, 360...local control device, 361...first communication unit, 362...second communication unit, 363...control unit, 380...power meter, 390...power meter, power meter, 392...power meter, 393...power meter, 400...power company

Claims (9)

A control unit that formulates a charge/discharge plan to be applied to the power storage device,
The control unit is
identifying N (N is an integer less than M) scenario groups by clustering M (M is an integer equal to or greater than 2) scenarios as scenarios indicating a transition of a predetermined power related to the power demand of the facility having the power storage device;
Formulating the charge/discharge plan by aggregating the identified N scenario groups ;
The M scenarios are scenarios that indicate a transition of a predetermined power with respect to an overall power demand of the facility,
The N scenario groups are groups of scenarios that indicate a transition of a predetermined power with respect to an overall power demand of the facility,
The clustering is a procedure for classifying two or more scenarios that are similar to each other into one group. A control unit that formulates a charge/discharge plan to be applied to the power storage device,
The control unit is
identifying N (N is an integer less than M) scenario groups by clustering M (M is an integer equal to or greater than 2) scenarios as scenarios indicating a transition of a predetermined power related to the power demand of the facility having the power storage device;
Formulating the charge/discharge plan by aggregating the identified N scenario groups;
The control unit is
Calculating a probability for each of the N scenario groups based on the number of scenarios classified into each of the N scenario groups;
The power management apparatus aggregates the identified N scenario groups based on the calculated probabilities. 3. The power management device according to claim 1, wherein the control unit sets the M scenarios based on a predicted value of the predetermined power. The power management device of claim 1 or claim 2, wherein the control unit identifies a trend in the specified power for each of the N scenario groups based on at least one of an average value, a weighted average value, a median value, a mode value, a maximum value, and a minimum value of the scenarios classified into each of the N scenario groups. 3. The power management device according to claim 1, wherein the control unit sets the M scenarios based on an actual value of the predetermined power. The control unit is
Calculating a similarity between a scenario classified into each of the N scenario groups and the predicted value of the predetermined power;
The power management apparatus according to claim 5 , wherein the identified N scenario groups are aggregated based on the calculated similarity. The power management device according to claim 5 or 6, wherein the control unit determines the transition of the specified power for each of the N scenario groups based on at least one of the average, weighted average, median, mode, maximum, and minimum values of the scenarios classified into each of the N scenario groups. Identifying N (N is an integer less than M) scenario groups by clustering M (M is an integer equal to or greater than 2) scenarios as scenarios indicating a transition of a predetermined power related to a power demand of a facility having a power storage device;
and formulating a charge/discharge plan to be applied to the power storage device by aggregating the identified N scenario groups;
The M scenarios are scenarios that indicate a transition of a predetermined power with respect to an overall power demand of the facility,
The N scenario groups are groups of scenarios that indicate a transition of a predetermined power with respect to an overall power demand of the facility,
The power management method , wherein the clustering is a procedure for classifying two or more scenarios that are similar to each other into one group . A step A of identifying N (N is an integer less than M) scenario groups by clustering M (M is an integer equal to or greater than 2) scenarios as scenarios indicating a transition of a predetermined power related to a power demand of a facility having a power storage device;
and step B of formulating a charge/discharge plan to be applied to the power storage device by aggregating the identified N scenario groups,
The step B includes:
calculating a probability of each of the N scenario groups based on the number of scenarios classified into each of the N scenario groups;
aggregating the identified N scenario groups based on the calculated probabilities.