JP2022093652A - Device control system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To make it difficult for surplus power used in an electrical device to run short when the weather on the day is different from the forecast on the previous day.

SOLUTION: A device control system 100 includes power generation means 1, a control device 2, and an electrical device 3. The control device uses a trained model that is machine-learned using past weather information and calendar information as inputs and the power consumption as an output to output the forecasted power consumption of the next day by using the forecast weather information of the next day and the calendar information of the next day as inputs, calculates the predicted surplus electric energy of the next day on the basis of the output predicted surplus electric energy, determines whether the control date of the electrical device is to be the same day or the next day on the basis of the calculated predicted surplus electric energy, and controls the electrical device on the determined control day.

SELECTED DRAWING: Figure 1

COPYRIGHT: (C)2022,JPO&INPIT

Description

The present invention relates to a device control system.

With the end of the feed-in tariff system for electricity, it has become economically advantageous to reduce the cost of purchasing electricity by consuming it at home rather than selling surplus electricity. .. Therefore, conventionally, various techniques for utilizing surplus electric power in a building have been proposed.
For example, Patent Document 1 describes a surplus power prediction means for determining a surplus power prediction value in a specific period by using a power generation prediction value of a photovoltaic power generation device determined based on the predicted weather information, and a surplus power prediction means in the specific period. Photovoltaic power generation device based on the device power consumption prediction means that determines the device power consumption predicted value consumed by the hot water storage type hot water supply device and the cumulative value of the period in which the surplus power predicted value is equal to or greater than the device power consumption predicted value in a specific period. A control means for controlling the heating means so as to perform boiling operation by electric power from the power source, and a photovoltaic power generation device-linked hot water storage type hot water supply system having the control means are described.

Japanese Unexamined Patent Publication No. 2018-165596

By the way, the actual weather (weather, temperature, etc.) may not be as indicated by the weather information acquired in advance. However, the system described in Patent Document 1 is adapted to determine the surplus power prediction value based on the meteorological information acquired before the change even if such a change in the weather occurs. For this reason, when the sunshine hours are shorter than the weather information indicates, the amount of sunshine per unit time is less, or the peak of the sunshine is shifted, the surplus power is insufficient at least in some time zones. There is a possibility that it will end up. If the surplus electricity is insufficient, hot water cannot be used when needed, or the hot water storage type water heater must be operated by boiling during the daytime, which is more expensive than at night. become.

The present invention has been made in view of the above problems, and an object of the present invention is to make it difficult for a state in which surplus electric power used in an electric device is insufficient occurs when the weather on the day is different from the forecast on the previous day.

The invention according to claim 1 is, for example, as shown in FIGS. 1 to 4.
Power generation means 1 installed in the building and
A trained model M1 in which past weather information of the past day and calendar information of the day are input, and machine learning is performed using the power consumption, which is the amount of power consumed in the day in the building B, as an output.
Using the trained model M1, the forecasted weather information of the next day and the calendar information of the next day predicted on the current day are input, and the predicted power consumption amount, which is the amount of power expected to be consumed on the next day in the building B, is output. Output means 21 and
Based on the predicted power consumption of the next day output by the output means 21, the predicted surplus power amount, which is the amount of power expected to be surplus the next day due to the power generation by the power generation means 1 on the next day, is calculated. Calculation means 21 and
Based on the predicted surplus electric energy of the next day calculated by the calculation means 21, the determination means 21 for determining whether the control date of the electric device 3 is the same day or the next day,
The device control system 100 is characterized by comprising device control means 21 for controlling the electric device 3 or transmitting control information to the electric device on the control day determined by the determination means 21.

The invention according to claim 2, for example, as shown in FIG.
In the device control system 100 according to claim 1,
The electric device 3 is characterized in that it is at least one of an electric water heater and a storage battery.

The invention according to claim 3 is, for example, as shown in FIG.
In the device control system 100 according to claim 1 or 2.
A second calculation means 21 for calculating a predicted power generation amount, which is the amount of power expected to be generated by the power generation means 1 on the next day, based on the predicted sunshine duration or amount of solar radiation on the next day.
The calculation means 21 is characterized in that the predicted surplus power amount is calculated by subtracting the power consumption amount output by the output means 21 from the predicted power generation amount calculated by the second calculation means 21.

According to the third aspect of the present invention, the expected surplus electric energy can be calculated relatively accurately by a simple calculation.

According to the present invention, when the weather on the current day is different from the forecast on the previous day, it is possible to prevent a state in which the surplus electric power used in the electric device is insufficient.

It is a block diagram which shows the equipment control system which concerns on embodiment of this invention. It is a block diagram which shows the relay device provided in the device control system of FIG. It is a block diagram which shows the control device provided in the device control system of FIG. It is a flowchart which shows the flow of the power management process executed by the control device of FIG.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.
However, although the embodiments described below are provided with various technically preferable limitations for carrying out the present invention, the technical scope of the present invention is not limited to the following embodiments and illustrated examples. do not have.

<1. Equipment control system>
First, a schematic configuration of a device control system (hereinafter, system 100) according to this embodiment will be described.
FIG. 1 is a block diagram showing a system 100.

The system 100 includes, for example, as shown in FIG. 1, a power generation means 1, a control device 2, and an electric device 3.
Further, the system 100 according to the present embodiment further includes a relay device 4.
These can communicate with each other via the communication network N.
Further, the system 100 according to the present embodiment can also communicate with an organization A (for example, a meteorological information company) that handles meteorological information via the communication network N.
It should be noted that the organization A (for example, the Japan Meteorological Agency) that handles the weather information to be the communication partner may be able to be added.

The power generation means 1 is for generating electricity and is installed in the building B.
The power generation means 1 according to the present embodiment is a solar panel installed on the roof of the building B.
Further, the power generation means 1 according to the present embodiment includes a power generation amount measuring unit 1a.
The power generation measuring unit 1a is adapted to repeatedly measure the hourly power generation amount every time a predetermined time (for example, 30 minutes or one hour) elapses.
The "hourly power generation amount" is the amount of power actually generated by the power generation means 1 in the past predetermined time (for example, one hour from 1 hour 30 minutes before the measurement to 30 minutes before).
The power generation means 1 may be something other than a solar panel (for example, a gas generator, a wind power generator, etc.).

The control device 2 is for calculating various predicted values and controlling the electric device 3 as needed, and is composed of a PC, a dedicated device, or the like.
Further, the control device 2 is connected to the database 2a or stores the database 2a.
The database 2a stores various information including various measured values acquired and various predicted values calculated.
The details of this control device 2 will be described later.

At least one electric device 3 is installed in the building B.
The electric device 3 includes an electric device 3 to be controlled by the control device 2 and another electric device 3 not to be controlled.
The electric device 3 to be controlled according to the present embodiment is an electric water heater.
The electric water heater is a heat pump type or a hybrid (combined with electricity and gas) type water heater.
The electric water heater according to the present embodiment is configured to automatically boil water based on the predicted surplus electric energy when the data of the predicted surplus electric energy described later is acquired.
Further, the electric device 3 according to the present embodiment includes a power consumption measuring unit 3a.
The power consumption measuring unit 3a measures the hourly power consumption every time a predetermined time (for example, 30 minutes or one hour) elapses.
The "equipment time power consumption amount" is the amount of power consumed by one electric device 3 in the past predetermined time (for example, one hour from 1 hour 30 minutes before to 30 minutes before the measurement).
The power consumption measuring unit 3a may be provided in at least another electric device 3.
Further, the power consumption measuring unit 3a may not be provided in each electric device 3, but may be provided in the distribution board or the relay device 4 of the building B.
Further, the electric device 3 can store the electricity generated by the power generation means 1 and the electricity supplied from the electric power company, and can supply electric power to the electric device 3 as needed. It may be an electric water heater and a storage battery.
The device time power consumption of the storage battery is the time power consumption stored in the past predetermined time.

The relay device 4 relays various information and various signals exchanged between the power generation means 1, the electric device 3, and the control device 2.
The details of the relay device 4 will be described later.

<2. Relay device>
Next, the relay device 4 included in the system 100 will be specifically described.
FIG. 2 is a block diagram showing a relay device 4.

[Relay device configuration]
As shown in FIG. 2, the relay device 4 includes a first control unit 41, a first communication unit 42, and a first storage unit 43.

The first control unit 41 is composed of a CPU (Central Processing Unit), a RAM (Random Access Memory), and the like.
Then, the CPU of the first control unit 41 reads various programs stored in the first storage unit 43, expands them in the RAM, executes various processes according to the expanded programs, and operates each unit of the relay device 4. It is designed to be centrally controlled.

The first communication unit 42 is composed of a communication module or the like.
Then, the first communication unit 42 transmits and receives various signals and various data to and from other devices (for example, power generation means 1, control device 2, electric device 3, etc.) connected via the communication network N. It has become.

The first storage unit 43 is composed of a non-volatile semi-dynamic memory, a hard disk, or the like.
Further, the first storage unit 43 stores various programs and the like executed by the first control unit 41.

[Operation of relay device]
The first control unit 41 of the relay device 4 configured in this way has the following functions.
The first control unit 41 has a function of acquiring the hourly power generation amount from the power generation means 1 via the first communication unit 42.
Further, the first control unit 41 has a function of acquiring the device time power consumption from at least another electric device 3 among the electric devices 3 via the first communication unit 42.
The first control unit 41 according to the present embodiment acquires these values every time a predetermined time (for example, 30 minutes) elapses (for example, 00 minutes and 30 minutes of each hour).
Further, when the building B is provided with a plurality of power generation means 1, the first control unit 41 acquires the time power generation amount of each power generation means 1 and calculates the total value thereof. The total value is taken as the hourly power generation amount.
Further, when the building B is provided with a plurality of other electric devices 3, the first control unit 41 acquires the device time power consumption of each of the other electric devices 3, and the first control unit 41 acquires the device time power consumption of each of the other electric devices 3. The total value is calculated as the time power consumption.

Further, the first control unit 41 transfers the acquired value or the calculated value (time power generation amount, time power consumption amount) to the control device 2 via the first communication unit 42 in response to the request from the control device 2. It has a function to send.
Further, the first control unit 41 has a function of transmitting a control signal received from the control device 2 to the electric device 3 via the first communication unit 42.

<3. Control device>
Next, the control device 2 included in the system 100 will be specifically described.
FIG. 3 is a block diagram showing the control device 2, and FIG. 4 is a flowchart showing the flow of the power management process executed by the control device 2.

[Control device configuration]
As shown in FIG. 3, the control device 2 includes a second control unit 21, a second communication unit 22, and a second storage unit 23.

The second control unit 21 is composed of a CPU (Central Processing Unit), a RAM (Random Access Memory), and the like.
Then, the CPU of the second control unit 21 reads out various programs stored in the second storage unit 23, expands them in the RAM, executes various processes according to the expanded programs, and operates each unit of the control device 2. It is designed to be centrally controlled.

The second communication unit 22 is composed of a communication module or the like.
The second communication unit 22 transmits and receives various signals and various data to and from another device (for example, an electric device 3 or the like) connected via the communication network N.

The second storage unit 23 is composed of a non-volatile semi-dynamic memory, a hard disk, or the like.
Further, the second storage unit 23 stores various programs (including power management processing described later) executed by the second control unit 21.
Further, the second storage unit 23 stores the learned model M1 used in the power management process described later.
The trained model M1 is machine-learned by inputting past weather information of the past day and calendar information of the day and using the power consumption, which is the amount of power consumed in 24 hours of the day, as an output in the building B. There is.
The specific method of machine learning will be described later.

This "past weather information" includes the maximum and minimum temperatures of the past day.
The past weather information according to the present embodiment also includes sunshine duration, amount of solar radiation, weather (for example, fine weather, cloudy weather, etc.), wind direction, wind speed, etc. in the past day.
Further, the "calendar information" includes today's date, today's day of the week, information on whether today is a weekday or a holiday, and the like.
In addition, the "power consumption" here is a summary of the power consumption for each hour of the day.
As described above, since the hourly power consumption is the amount of power consumed by the other electric device 3 in one hour (total value), the power consumption according to the present embodiment is determined by the other electric device 3. It is the total amount of electricity consumed in 24 hours of the day.

Further, the second storage unit 23 according to the present embodiment stores the second learned model M2.
The second trained model M2 is machine-learned by inputting the calendar information of the past day, the measured value of the amount of power generation, and the amount of solar radiation, and using the amount of power generation as an output.
The "power generation amount" here is a summary of the power generation amount for each hour of the day.

[Operation of control device]
The second control unit 21 of the control device 2 configured in this way has a function of executing the power management process as shown in FIG.
The second control unit 21 according to the present embodiment repeatedly executes this power management process every day when a predetermined time (for example, an arbitrary timing from 20:00 to 00:00 every night) is reached.

In this power management process, the second control unit 21 first executes the acquisition process (step S1).
In this acquisition process, the second control unit 21 acquires the forecasted weather information for the next day predicted on the current day.
In the acquisition process according to the present embodiment, the second control unit 21 receives the forecast weather information from the organization A that handles the weather information via the second communication unit 22.
This "forecast weather information" includes the maximum temperature and the minimum temperature of the predicted day or the next day.
The forecasted weather information according to the present embodiment also includes the predicted sunshine duration, the amount of solar radiation, the weather (for example, fine weather, cloudy weather, etc.), the wind direction, the wind speed, and the like on the day or the next day.

Further, in this acquisition process, the second control unit 21 acquires the calendar information of the next day.
Further, in the acquisition process according to the present embodiment, the second control unit 21 acquires various measured values from the relay device 4 provided in the building B.
In the acquisition process according to the present embodiment, the second control unit 21 acquires the power generation amount and the power consumption amount.

After acquiring various information, the second control unit 21 executes the output process (step S2).
In this output process, the second control unit 21 outputs the predicted power consumption.
This "estimated power consumption" is a summary of each predicted time power consumption of the next day.
The "estimated time power consumption" is the amount of power that is expected to be consumed by an electric device 3 other than the electric device 3 at a predetermined time (for example, one hour from n o'clock) on the next day.
In the output process according to the present embodiment, the second control unit 21 uses a machine learning method for calculating the predicted power consumption.
Specifically, the trained model M1 stored in the second storage unit 23 is used, and the predicted power consumption amount is output by inputting the forecasted weather information of the next day and the calendar information of the next day acquired in the acquisition process.

Machine learning for obtaining the trained model M1 is roughly divided into regression and classification, but regression is used for prediction of time-series data such as power generation amount and power consumption amount as in this case.
Machine learning algorithms often used in regression include, for example, linear regression methods such as ridge regression and lasso regression, support vector machines, random forests, gradient boosting decision trees, neural networks, ARIMA models, state space models, and the like.
Hereinafter, the case where the gradient boosting decision tree is used as a machine learning algorithm will be described, but the algorithm is not particularly limited to this algorithm.

The specific calculation flow is as follows.
First, using the past weather information, calendar information, power generation amount, and actual power consumption data for a predetermined period (up to one year in the past), the past weather information and calendar for each day in the predetermined period (up to one year in the past). Machine learning is performed using the above algorithm with information, the amount of power generation up to the current day, and the amount of power consumption as explanatory variables, and the amount of power generation and power consumption of the next day as objective variables. As a result, the trained model M1 stored in the storage unit 23 is generated.
Then, when the trained model M1 is loaded with the forecasted weather information, calendar information, power generation amount up to the predicted time, and power consumption amount for 24 hours after the predicted time as application data, the second control unit 21 consumes the predicted amount. Outputs the amount of power.
The second control unit 21 serves as an output means by executing such an output process.

After acquiring various measured values, the second control unit 21 executes the calculation process (step S3).
In this calculation process, the second control unit 21 calculates the predicted surplus power amount based on the predicted power consumption amount of the next day output in the output process.
This "predicted surplus electric energy" is a summary of the surplus electric energy for each predicted time on the next day.
The "estimated time surplus electric energy" is an electric energy that is expected to become surplus the next day when the power generation means 1 generates power at a predetermined time (for example, one hour from n o'clock) on the next day.

In the calculation process according to the present embodiment, the second control unit 21 first calculates the predicted power generation amount based on the sunshine duration or the sunshine amount of the next day acquired in the acquisition process.
This "predicted power generation amount" is a summary of the power generation amount for each predicted time of the next day.
The “estimated time power generation amount” is the amount of power that is expected to be generated by the power generation means 1 at a predetermined time (for example, one hour from n o'clock) on the next day.
As described above, the power generation means 1 according to the present embodiment is a solar panel. Therefore, in the calculation process according to the present embodiment, the second control unit 21 calculates the predicted power generation amount.

In the calculation process according to the present embodiment, the second control unit 21 calculates the predicted power generation amount by using the machine learning method as in the case of the predicted power consumption amount.
The specific calculation flow is as follows.
First, machine learning with the above algorithm using calendar information, actual measurement data of power generation amount, actual measurement data or estimated data of solar radiation amount as explanatory variables, and power generation amount of the next day as objective variables for each day for a predetermined period (up to one year in the past). Let me. As a result, the second trained model M2 stored in the storage unit 23 is generated.
The estimated data of the amount of solar radiation can be estimated from the measured amount of power generation and the amount of solar radiation predicted in the past day.
Then, when the calendar information of the predicted day and the amount of solar radiation for 24 hours after the predicted time are read into the second trained model M2 as application data, the second control unit 21 outputs the predicted power generation amount.
It should be noted that the sunshine duration may be used instead of the amount of solar radiation to generate the second trained model M2 and the predicted power consumption.

After calculating the predicted power generation amount, the second control unit 21 calculates the predicted surplus power amount by subtracting the predicted power consumption amount output in the output process from the calculated predicted power generation amount.
By doing so, the expected amount of surplus power can be calculated relatively accurately by a simple calculation.

In the calculation process according to the present embodiment, the second control unit 21 calculates the expected surplus power amount, and then specifies the time (1 hour) at which the predicted time surplus power amount becomes maximum based on the expected surplus power amount. , The estimated time surplus power amount for 3 hours including the time and 1 hour before and after this specified time is taken out and used as control information.
Table 1 below shows an example of the predicted surplus electric energy (transition of the predicted time surplus electric energy) calculated in the calculation process.
In this case, the maximum predicted time surplus electric energy is in the 14:00 range. Therefore, in such a case, the estimated time surplus electric energy for 3 hours (from 13:00 to 15:00) including 1 hour before and after that is used as control information.

なお、取り出す予測時間余剰電力量は３時間分に限られず、前後２時間以上の予測時間余剰電力量を取り出してもよい。
例えば、予測時間余剰電力量が最大となる時間を中央にする必要はなく、予測時間余剰電力量が最大となる時間の始まりの時刻（例えば００分）を特定し、特定した時刻を起点とする３時間としてもよい。

The amount of estimated time surplus power to be taken out is not limited to 3 hours, and the predicted time surplus power amount of 2 hours or more before and after may be taken out.
For example, it is not necessary to center the time when the predicted time surplus power is maximum, but specify the start time (for example, 00 minutes) of the time when the predicted time surplus power is maximum, and use the specified time as the starting point. It may be 3 hours.

The second control unit 21 is adapted to generate the above-mentioned control information twice a day.
The first time is the time of the above calculation process, and it is performed until the time zone when the unit price of electric power is low at night arrives. This is because the electric device 3 to be controlled boil or store hot water during the time when the unit price of electric power is low. This is to determine in advance whether to perform the operation during the time when the surplus power is generated.
The second generation of control information will be described later.
The second control unit 21 forms a calculation means by executing such a calculation process.

After calculating the predicted surplus electric energy, the second control unit 21 executes the determination process (step S4).
In this determination process, the second control unit 21 determines whether the control date of the electric device 3 is the same day or the next day based on the predicted surplus electric power amount of the next day calculated in the calculation process.
In the determination process according to the present embodiment, the second control unit 21 determines whether the calculated predicted surplus power amount (control information) of the next day is equal to or higher than a predetermined threshold value (the surplus power amount of the next day is boiled or stored). (Is it a sufficient amount?) To determine whether or not.
The second control unit 21 makes a determination means by executing this determination process.

In this determination process, when it is determined that the control date of the electric device 3 is the current day (step S4: Yes), the second control unit 21 executes the device control process (step S5).
In this device control process, the second control unit 21 controls the electric device 3 or transmits control information to the electric device 3.
Specifically, the data (control information) of the predicted surplus electric power is transmitted to the electric device 3 via the second communication unit 22 and the relay device 4.
Then, the electric device 3 automatically starts boiling water based on the predicted surplus electric energy.

Further, in the power management process according to the present embodiment, the second control unit 21 controls the electric device 3 and then executes the storage process (step S6).
In this storage process, the second control unit 21 has acquired various measured values (power generation amount, sunshine time, power consumption amount), output / calculated various predicted values (predicted power generation amount, predicted power consumption amount, predicted surplus power amount). ) And the date and time when the predicted value was calculated is stored in the database.

On the other hand, when it is determined in the above determination process that the control date of the electric device 3 is not the same day, that is, the control date is determined to be the next day (step S4: No), the second control unit 21 determines the predetermined time of the next day (for example, in the morning). Until 6 o'clock), wait for the execution of the second acquisition process described later.
In the power management process according to the present embodiment, the determination of whether or not the predetermined time has been reached (step S7) is repeated until the predetermined time of the next day (for example, 6 o'clock in the morning) is reached.
In the above determination process, after determining that the control date of the electric device 3 is the next day, a timer for issuing a predetermined trigger is activated at a predetermined time on the next day, and the power management process is stopped until the trigger is detected. May be.

When it is determined in the process of step S6 that the predetermined time of the next day has come (step S76: Yes), the second control unit 21 executes the second acquisition process (step S8).
In this second acquisition process, the second control unit 21 acquires the forecast weather information for the day newly forecast for the day.
Further, in the second acquisition process according to the present embodiment, the second control unit 21 also acquires the sunshine hours.

After acquiring the forecast weather information for the day, the second control unit 21 executes the correction process (step S9).
In this correction process, the second control unit 21 uses the trained model M1 and outputs the corrected predicted power consumption of the day by inputting the forecast weather information of the day newly predicted on that day.
At this time, if the forecast weather information acquired on that day is different from the one acquired on the previous day, the corrected predicted power consumption output will be different from the value output on the previous day.
As a result, the second control unit 21 corrects the predicted power consumption output on the previous day in the output process after the next day.
As described above, the control device 2 according to the present embodiment obtains the corrected predicted power consumption that more reflects the actual weather by the forecast weather information of the day newly predicted on the next day and the learned model M1. It can be output (correct the predicted power consumption more accurately).
If the acquired forecast weather information and sunshine duration are the same as those acquired on the previous day or the difference is small in the above acquisition process, the predicted power consumption output in this correction process is the same as that output on the previous day. Is expected to be. In that case, the second control unit 21 skips this correction process, the second calculation process and the determination process described later, and performs the second device control process (step S12) described later based on the predicted surplus electric energy calculated on the previous day. It may be designed to run.

After executing the correction process, the second control unit 21 executes the second calculation process (step S10).
In this second calculation process, the second control unit 21 calculates the corrected predicted surplus electric energy based on the corrected predicted power consumption.
The corrected surplus electric energy is calculated by subtracting the corrected predicted power consumption from the predicted power generation amount.
In the correction process according to the present embodiment, the second control unit 21 calculates the predicted power generation amount based on the sunshine duration or the sunshine amount newly acquired on the next day.

Further, in the second calculation process, the second control unit 21 generates the control information for the second time.
The second time is performed until the daytime heat storage or storage is performed.
Usually, it is performed in the morning (for example, around 9 o'clock) before the amount of solar radiation increases. This is when, as a result of the judgment of the previous day, it is decided to store or store heat in a certain time zone in the daytime, and it is not appropriate to store or store heat in that time zone because the weather forecast is different from the previous day. This is to shift the operating time.
Table 2 below shows an example of the predicted surplus electric energy (transition of the predicted time surplus electric energy) calculated in the second calculation process.
In this case, the weather forecast changed at 9 o'clock the next day, the afternoon weather became worse, and the amount of surplus electricity became zero. In addition, the amount of surplus electricity is relatively large in the morning. Therefore, in such a case, the estimated time surplus electric energy for 3 hours (from 10:00 to 12:00) including 1 hour before and after 11:00 is used as control information.

このように、制御情報を２回生成することにより、すなわち、前日の予測だけでなく、制御直前の再予測を組み合わせることにより、より適切な余剰電力制御が可能となる。
第二制御部２１は、この第二算出処理及び上述した算出処理（ステップＳ３）を実行することにより算出手段をなす。

In this way, by generating the control information twice, that is, by combining not only the prediction of the previous day but also the re-prediction immediately before the control, more appropriate surplus power control becomes possible.
The second control unit 21 forms a calculation means by executing the second calculation process and the above-mentioned calculation process (step S3).

After calculating the corrected predicted surplus electric energy, the second control unit 21 executes the determination process (step S11).
In this determination process, the second control unit 21 determines the control time of the electric device 3 based on the corrected predicted surplus electric energy calculated in the second calculation process.
In the determination process according to the present embodiment, the second control unit 21 controls so that the boiling operation of hot water is started from a time close to night (sunset) within a range in which a required amount of hot water can be boiled. It is designed to determine the time.
Since hot water is often used at night, by doing this, the electric device 3 (electric water heater) starts boiling hot water during the daytime, and as a result, extra hot water is boiled in anticipation that the hot water will cool down. It can be prevented from being stored.
The second control unit 21 forms a determination means by executing such a determination process.

After determining the control time, the second control unit 21 executes the second device control process (step S12).
In this second device control process, the second control unit 21 controls the electric device 3 or transmits control information to the electric device 3 at the control time determined in the determination process.
In the second device control process according to the present embodiment, the second control unit 21 corrects the predicted value of the surplus electric energy in the time zone other than the control time determined in the determination process to 0 [kWh]. The data is transmitted to the electric device 3.
When an electric water heater acquires data on the predicted amount of surplus power in a day, it often starts the boiling operation immediately from the time when the surplus power is generated even a little. However, by doing so, the electric device 3 starts the boiling operation from the control time determined in the determination process.
The second control unit 21 acts as a device control means by executing such a second device control process and the device control process (step S5) described above, and sets the electric device 3 on the control date determined in the determination process. It will be controlled or the control information will be transmitted to the electric device 3.

Further, in the power management process according to the present embodiment, the second control unit 21 executes the second storage process after controlling the electric device 3 (step S13).
In this second storage process, the second control unit 21 has acquired various measured values (power generation amount, sunshine time, power consumption amount), output / calculated various predicted values (predicted power generation amount, corrected predicted power consumption amount). , The predicted surplus electric energy after correction) and the date and time when the predicted value was calculated are stored in the database.

<4. Effect>
When the system 100 described above determines that the control date of the electric device 3 is the next day, the system 100 corrects the predicted power consumption after the next day.
Therefore, according to the system 100, when the weather on the day is different from the forecast on the previous day, it is possible to prevent a state in which the surplus power used in the electric equipment is insufficient. More specifically, it is possible to make it difficult for an electric water heater to be in a state where hot water cannot be boiled or a state in which electric power cannot be stored in a storage battery.

100 Equipment control system 1 Power generation means 1a Power generation amount measuring unit 2 Control device 2a Database 21 Second control unit (output means, calculation means, judgment means, equipment control means, correction means, determination means, second calculation means)
22 Second communication unit 23 Second storage unit 3 Electrical equipment 3a Power consumption measurement unit 4 Relay device 41 First control unit 42 First communication unit 43 First storage unit A Organization that handles weather information B Building M Learned model N Communication network

Claims (3)

The power generation means installed in the building and
A trained model in which past weather information of the past day and calendar information of the day are input, and machine learning is performed using the power consumption, which is the amount of power consumed in the building, as an output.
Using the trained model, output the predicted power consumption, which is the amount of power expected to be consumed the next day in the building, by inputting the forecast weather information of the next day and the calendar information of the next day predicted on the current day. Means and
Calculation to calculate the predicted surplus power amount, which is the amount of power expected to be surplus the next day due to the power generation by the power generation means the next day, based on the predicted power consumption amount of the next day output by the output means. Means and
Based on the predicted surplus electric energy of the next day calculated by the calculation means, a determination means for determining whether to set the control date of the electric device to the current day or the next day, and
A device control system comprising: a device control means for controlling the electric device or transmitting control information to the electric device on the control day determined by the determination means. In the device control system according to claim 1,
The device control system, wherein the electrical device is at least one of an electric water heater and a storage battery. In the device control system according to claim 1 or 2.
A second calculation means for calculating the predicted power generation amount, which is the amount of power expected to be generated by the power generation means on the next day, based on the predicted sunshine duration or amount of sunshine on the next day.
The calculation means is a device control system, characterized in that the predicted surplus power amount is calculated by subtracting the power consumption amount output by the output means from the predicted power generation amount calculated by the second calculation means.

Images