KR20220031359A - Electric energy storage system for fire prevention and maximum profit and control method of the same - Google Patents

Abstract

According to one embodiment of the present invention, an electric power storage system includes: a solar photovoltaic inverter for generating and outputting a power from sunlight; a power storage unit for receiving the power from the solar photovoltaic inverter or an external power grid system to store energy; a DC/AC conversion unit for converting an AC power of the solar photovoltaic inverter or the external power grid system into a DC power to supply the DC power to the power storage unit, and transmitting the power stored in the power storage unit to a load as the AC power; and a control unit for receiving a past load power amount used by the load and a past solar power generation amount output from the solar photovoltaic inverter to predict a current solar power generation amount and a current load power amount, and forming a charging and discharging schedule of the power storage unit based on the predicted solar power generation amount, the predicted load power amount, and a set target load amount to control the solar photovoltaic inverter and the DC/AC conversion unit according to the charging and discharging schedule.

Description

Electricity storage system for fire prevention and maximum profit and operation method thereof

The present invention relates to a power storage system and a method for operating the same, and more particularly, to a power storage system for differently adjusting a charging method and a discharging method for each time period based on power consumption and solar power generation amount obtained from a power database, and a method for operating the same will be.

Electricity has the characteristic that the amount of electricity produced and the amount of electricity used must match in real time.

Therefore, unless a separate auxiliary system is used, it is impossible to efficiently operate an electric power system, such as storing electric energy during a time period of low electric power usage and using the stored electric energy during a time period of high electric power use.

In addition, recently, new and renewable energy such as solar power or wind power has been introduced. Since the amount of renewable energy is determined according to external factors such as weather, there is a problem that the amount of electricity must be adjusted according to the amount of electricity used. do.

As described above, a power storage device may be used as a method to solve problems related to the efficient operation of the power system and the time difference between power production/power usage.

Such a power storage device is a device that stores electrical energy and uses it when needed. The power storage device can be utilized as a system for reducing the usage fee of system power. That is, if the electric energy is stored in the light load section where the usage fee is low in the power storage device, the electric energy stored in the power storage device can be used as an alternative power source in the maximum load section where the price is high.

When an energy storage system (ESS) is connected to a solar power generation facility, the power generation revenue is estimated to be 2 to 25 times that of solar power alone. If the manufacturing industry, which uses a lot of electricity, etc. installs an energy storage device with a capacity of 1MWh for the purpose of lowering the peak power consumption, it can be expected to reduce electricity bills by 100 million won per year.

Power demand management by linking energy storage system (ESS) to solar power generation facilities is required to maximize operating profit.

Energy Storage System (ESS) is a technology used when necessary by storing the generated power energy in order to improve the characteristics of electrical energy that must be produced and consumed at the same time. The energy storage system can optimize power system operation through load leveling and secure power reserve by storing idle power in the light load section and using the stored power in the maximum load section.

The energy storage system can provide efficient operation of the power system and stable high-quality power by compensating for the disadvantages of new and renewable energy with high output fluctuations.

One technical problem to be solved by the present invention is to provide a method of operating an energy storage system for the most fire prevention and maximum profit.

A power storage system according to an embodiment of the present invention includes a solar inverter for generating and outputting power from sunlight; a power storage unit for storing energy by receiving power from the solar inverter or an external power grid system; a DC/AC converter for converting AC power of the solar inverter or external power grid system into DC power, supplying it to the power storage unit, and transferring the power stored in the power storage unit as AC to a load; and the past load power used in the load and the past solar power output from the solar inverter to predict the current solar power generation and load power, and based on the predicted solar power generation and load power and set target load to form a charging and discharging schedule of the power storage unit, and a control unit configured to control the solar inverter and the DC/AC converter according to the charging and discharging schedule.

In one embodiment of the present invention, the control unit, a data collection module for collecting the amount of solar power generation and the amount of load power; a database for storing data collected by the data collection module; a prediction unit for predicting the amount of solar power generation and load power by using the data stored in the database; a schedule generator for generating a charging schedule and a discharging schedule based on the target load; and a control module for controlling the solar inverter and the DC/AC converter according to the charging schedule and the discharging schedule.

In one embodiment of the present invention, the schedule generating unit generates a solar charging schedule and a light load interval charging schedule based on the precedence of the target load excess start time and the maximum load period start time.

A power storage system according to an embodiment of the present invention includes a solar inverter for generating and outputting power from sunlight; a power storage unit for receiving power from the solar inverter or grid and storing energy; a DC/AC converter for converting AC power of the solar inverter or external power grid system into DC power, supplying it to the power storage unit, and transferring the power stored in the power storage unit as AC to a load; and the past load power used in the load and the past photovoltaic power output from the photovoltaic inverter to predict the current photovoltaic power generation and load power, and to the predicted photovoltaic power generation and load power and set target load and a controller configured to form a charging and discharging schedule of the power storage unit based on the charging and discharging schedule and controlling the solar inverter and the DC/AC converting unit according to the charging and discharging schedule. The operating method of the power storage system includes the steps of estimating the amount of solar power generation and the amount of load power; setting a target load and generating a charging schedule of the power storage unit based on the target load; generating a discharge schedule of the power storage unit based on the target load amount; and controlling the DC/AC converter according to the charging schedule and the discharging schedule.

In an embodiment of the present invention, the step of setting a target load amount and generating the charging schedule of the power storage unit based on the target load amount includes: calculating a target load excess time exceeding the target load amount from the predicted load power amount step; comparing the target load excess time and the maximum load section start time; predicting a first solar power generation amount from the end time of the light load section to the start time of the target load excess when the target load excess time is earlier than the maximum load section start time; calculating the first amount of charge of the power storage unit before the end of the light load period by subtracting the first solar power generation amount from the upper limit of the charge of the power storage unit; generating a solar charging schedule with the first solar power generation amount from the end time of the light load section to the start time of exceeding the target load; calculating a system charging start time of the power storage unit for system charging through an external system within the light load section in consideration of the charging speed of the power storage unit; and generating a charging schedule for the light load section from the start time of the system charging to the end time of the light load section.

In an embodiment of the present invention, when the target load exceeding time is slower than the maximum load section start time, predicting a second solar power generation amount from the light load section end time to the maximum load section start time; calculating a second amount of charge of the power storage unit before the end of the light load period by subtracting the second solar power generation amount from the upper limit of the charge of the power storage unit; and generating a solar charging schedule with the second solar power generation amount from the end time of the light load period to the start time of the maximum load period; include more

In one embodiment of the present invention, the generating of the discharge schedule of the power storage unit based on the target load amount includes: calculating a first excess amount of power exceeding the target load amount from the predicted load amount of electricity, and the target amount of electricity and setting a first priority discharge schedule by calculating an excess section exceeding .

In one embodiment of the present invention, generating a second priority discharge schedule to discharge the remaining amount of charge remaining after subtracting the excess amount of first power from the upper limit of the charge of the power storage unit in a maximum load section; and generating a third priority discharge schedule to discharge in an intermediate load section with the remaining charge after the second priority discharge schedule.

In one embodiment of the present invention, predicting the amount of solar power generation and load power may include: converting past solar power generation and load power data into unstructured data; extracting machine learning training data using the unstructured data; constructing a layer model of Long Short-Term Memory models (LSTM) of a recurrent neural network using the machine learning training data; repeatedly learning the layer model; and estimating the solar power generation amount and the load power amount based on the repetitive learning in a 24-hour time series.

The method of operating a power storage device according to an embodiment of the present invention may provide fire prevention and operation for maximum profit.

1 is a conceptual diagram illustrating a power storage device according to an embodiment of the present invention.
FIG. 2 is a flowchart illustrating a method of operating the power storage device of FIG. 1 .
3 shows a section according to the usage fee for each time period provided by Korea Electric Power Corporation.
4 is a flowchart for estimating the amount of solar power generation and the amount of load power according to an embodiment of the present invention.
5 is a flowchart of generating a charging schedule of a power storage device according to an embodiment of the present invention.
6 is a flowchart of generating a discharge schedule of a power storage device according to an embodiment of the present invention.
7 is a timing chart illustrating a charging schedule and a discharging schedule according to an embodiment of the present invention.
8 is a timing chart illustrating a charging schedule and a discharging schedule according to another embodiment of the present invention.
9 is a timing chart illustrating a charging schedule and a discharging schedule according to another embodiment of the present invention.

The power storage device according to an embodiment of the present invention uses a charging/discharging algorithm in consideration of fire prevention and optimum profit during operation. By setting the target load, the base fee can be reduced through peak reduction. According to KEPCO's discount rate support, it is possible to generate profits through maximum discount by charging in the light load section and discharging in the maximum load section.

According to the fire accident investigation of power storage devices, the occurrence of fires was frequently caused by battery voltage imbalance during full charge standby. Accordingly, it provides a charge/discharge control algorithm for maximum profit while minimizing the waiting state for full charge during charge/discharge control.

The present invention predicts the amount of solar power generation and load for 24 hours, and charges in advance during light load time. In consideration of the amount of solar power generation, it is charged to the minimum during light load time. According to the 24-hour load energy prediction, the charging time of the external system is calculated inversely in the light load section in consideration of the discharge time.

The present invention provides a charging/discharging technology for maximum profit while minimizing the waiting time for full battery and performing stable charging and discharging through speed control in the upper and lower limit sections of charging and discharging.

Hereinafter, the present invention will be described in more detail with reference to the accompanying drawings. Hereinafter, the present invention will be described in more detail with reference to preferred embodiments. However, these Examples are intended to explain the present invention in more detail, and it will be apparent to those of ordinary skill in the art that the present invention is not limited or limited by experimental conditions, material types, and the like. The present invention is not limited to the embodiments described herein and may be embodied in other forms. Rather, the embodiments introduced herein are provided so that the disclosed subject matter may be thorough and complete, and the spirit of the present invention may be sufficiently conveyed to those skilled in the art. In the drawings, components are exaggerated for clarity. Parts indicated with like reference numerals throughout the specification indicate like elements.

1 is a conceptual diagram illustrating a power storage device according to an embodiment of the present invention.

FIG. 2 is a flowchart illustrating a method of operating the power storage device of FIG. 1 .

3 shows a section according to the usage fee for each time period provided by Korea Electric Power Corporation.

4 is a flowchart for estimating the amount of solar power generation and the amount of load power according to an embodiment of the present invention.

5 is a flowchart of generating a charging schedule of a power storage device according to an embodiment of the present invention.

6 is a flowchart of generating a discharge schedule of a power storage device according to an embodiment of the present invention.

7 is a timing chart illustrating a charging schedule and a discharging schedule according to an embodiment of the present invention.

8 is a timing chart illustrating a charging schedule and a discharging schedule according to another embodiment of the present invention.

9 is a timing chart illustrating a charging schedule and a discharging schedule according to another embodiment of the present invention.

1 to 7 , the power storage system 100 includes a solar inverter 120 for generating and outputting power from sunlight; a power storage unit 130 for receiving power from the solar inverter 120 or an external power grid system 150 and storing energy; AC power of the solar inverter 120 or the external power grid system 150 is converted into DC power and supplied to the power storage unit 130 , and the power stored in the power storage unit 130 is supplied to the load 160 . DC/AC conversion unit 140 for transferring to AC; and the past load power used by the load 160 and the past solar power generation output from the solar inverter 120 to predict the current photovoltaic power generation amount and load power amount, and the predicted solar power generation amount and load Forming a charging and discharging schedule of the power storage unit 130 based on the amount of power and the set target load to control the solar inverter 120 and the DC/AC converter 140 according to the charging and discharging schedule a control unit 110 .

By setting the target load, the base fee can be reduced through peak reduction. Specifically, the electricity rate peak interlocking system is a rate system that induces customers to restrain their maximum demand in order to reduce electricity rates by imposing a basic rate based on the maximum power throughout the year. Therefore, it is necessary to set a target load and not use more than the target load in the external power grid as a whole. When the target load is exceeded, the power stored in the power storage unit 130 may be discharged to use the external power grid with a priority below the target load.

Korea Electric Power divides the external power grid system into a light load section (A), a medium load section (B), and a maximum load section (C) by time zone. KEPCO may set different rates depending on the section. The light load section (A) may be the cheapest, and the maximum load section (C) may be the most expensive. Accordingly, the power storage device may be charged in the light load period (A) and discharged in the maximum load period (C). However, in this case, the power storage unit does not efficiently use the amount of solar power generation. In addition, the power storage unit 130 is maintained in a fully charged state for a long time to increase the possibility of fire.

Referring to FIG. 7 , the power storage unit 130 charges by using the amount of solar power until just before the start time of the target load amount. Subsequently, when the amount of load power exceeding the target load amount is generated, the power storage unit may immediately perform discharging and may not maintain the full charge state for a long time.

Referring to FIG. 8 , the power storage unit is charged using the amount of solar power until just before the start of the time of the maximum load section. Subsequently, when the maximum load period C starts, the power storage unit may immediately perform discharging and may not maintain the full charge state for a long time.

The control unit 110 includes a data collection module 112 for collecting the amount of solar power generation and the amount of load power; a database 114 for storing the data collected by the data collection module 112; a prediction unit 116 for predicting the amount of solar power generation and load power by using the data stored in the database 114; and a schedule generator 118 for generating a charging schedule and a discharging schedule based on the target load. and a control module 119 for controlling the solar inverter and the DC/AC converter according to the charging schedule and the discharging schedule.

The data collection module 112 may collect the amount of load power and the measurement time measured through the load meter 162 and the amount and time of photovoltaic generation measured through the power meter installed inside the solar inverter in real time or at regular intervals.

The database 114 may store the load power amount and measurement time, and the solar power generation amount and time. The database 114 may convert and store the load power amount and measurement time and the solar power generation amount and time into unstructured data. The unstructured data may be used for machine learning. The unstructured data may be converted into machine learning training data for machine learning learning. Some of the machine learning training data may be used as input data for machine learning, and the remaining data may be used for machine learning verification.

The prediction unit 116 may use an artificial neural network. Specifically, the prediction unit uses a recurrent neural network, and more specifically, a Long Short-Term Memory models (LSTM) layer model may be used. The prediction unit may predict the current amount of solar power generation and the amount of load power. The prediction of the amount of solar power generation and load power may utilize season and weather information.

The schedule generation unit 118 may generate a solar charging schedule and a light load interval charging schedule based on a precedence of the target load excess start time and the maximum load period C start time.

The target load amount may depend on a maximum value among the predicted load power amounts, and a maximum discharge output amount of the power storage system. For example, the target load amount may be set by subtracting the maximum discharge output amount of the power storage system from the maximum value among the predicted load power amounts.

The schedule generator 118 may generate a charging schedule and a discharging schedule. The charging schedule may include a solar charging schedule and a light load period charging schedule. The discharge schedule includes a first priority discharge schedule set in the excess section (D) exceeding the target load, a second priority discharge schedule set in the maximum load section (C), and a third priority discharge schedule set in the middle load section may include

The control module 119 may control the DC/AC converter 140 according to the charging schedule and the charging schedule. Also, the control module 119 may operate the solar inverter 120 when the amount of solar radiation is equal to or greater than a threshold value.

The operation method of the power storage system includes the steps of predicting the amount of solar power generation and the amount of load power (S10); setting a target load and generating a charging schedule of the power storage unit 130 based on the target load (S20); generating a discharge schedule of the power storage unit 130 based on the target load (S30); and controlling the DC/AC converter 140 according to the charging schedule and the discharging schedule (S40).

Predicting the amount of solar power generation and load power (S10), the steps of converting past solar power generation and load power data into unstructured data (S101); extracting machine learning learning data using the unstructured data (S110); constructing a layer model of Long Short-Term Memory models (LSTM) of a recurrent neural network using the machine learning training data (S120); repeatedly learning the layer model (S130); and estimating the solar power generation amount and the load power amount based on the repetitive learning in a 24-hour time series (S140).

Setting a target load amount and generating the charging schedule of the power storage unit based on the target load amount (S20) may include calculating a target load excess time exceeding the target load amount from a predicted load power amount (S201); Comparing the target load exceeding time and the maximum load section start time (S210); If the target load excess time is earlier than the maximum load period start time, predicting a first solar power generation amount from the light load period end time to the target load excess start time (S220a); calculating the first amount of charge of the power storage unit before the end of the light load period by subtracting the first solar power generation amount from the upper limit of the charge of the power storage unit (S230a); generating a solar charging schedule with the first solar power generation amount from the end time of the light load section to the start time of exceeding the target load (S240a); calculating a system charging start time of the power storage unit for system charging through an external system within the light load section in consideration of the charging speed of the power storage unit (S250); and generating a charging schedule for the light load section from the start time of the system charging to the end time of the light load section (S260).

When the target load excess time is slower than the maximum load section start time (S210), the second solar power generation amount from the light load section end time to the maximum load section start time can be predicted (S220b).

then, calculating the second amount of charge of the power storage unit before the end of the light load period by subtracting the second solar power generation amount from the upper limit of the charge of the power storage unit (S230b);

Subsequently, a solar charging schedule may be generated with the second solar power generation amount from the end time of the light load period to the start time of the maximum load period (S240b).

Next, in consideration of the charging speed of the power storage unit, a system charging start time of the power storage unit for system charging through an external system within the light load section may be calculated (S250);

Subsequently, a light load section charging schedule may be generated from the system charging start time to the light load section end time (S260).

In the step (S30) of generating a discharge schedule of the power storage unit based on the target load amount, a first excess amount of power exceeding the target load amount is calculated from the predicted amount of load power, and an excess section exceeding the target amount of power ( and D) calculating and setting a first priority discharge schedule (S301).

The step of generating the discharge schedule of the power storage unit based on the target load amount (S30) includes subtracting the first excess amount of power from the upper limit of the charge of the power storage unit and discharging the remaining charge in the maximum load section. generating a discharge schedule (S310); and generating (S320) a third priority discharge schedule to discharge in an intermediate load section with the remaining charge after the second priority discharge schedule.

Again, referring to FIG. 7 , when the target load excess time is earlier than the start time of the maximum load section, the charging and discharging schedules are displayed. In the middle load section (B) from 9:00 to 10:00, the predicted load wattage reaches the target load. A target load overrun start time in which the predicted load wattage coincides with the target load is calculated. The amount of solar power generation is predicted from the end of the light load section to the start time of exceeding the target load, and the power storage unit sets a solar charge schedule according to the amount of photovoltaic power generation in this section.

Then, the start of grid charging is calculated in consideration of the charging speed in order to charge the remaining amount of charge of the power storage unit to the external power grid system in the light load section. In this embodiment, the start of grid charging is 23:00.

After the target load excess time, the discharge schedule is set. The first priority discharge schedule is set in the excess section D in which the predicted load power exceeds the target load. Preferably, assuming that the total charging strategy of the power storage unit is 100%, the excess amount of power in the excess section may be at the level of 20%. Accordingly, the remaining 80% of the charging amount is allocated to the maximum load section (C) as the second priority. However, when the excess section (D) and the maximum load section (C) overlap, the overlapping section is excluded from the second priority schedule. In the second priority schedule, 70% is used, leaving 10%.

Then, the remaining 10% is allocated to the intermediate load section (B). However, when the excess section (D) and the intermediate load section (B) overlap, the overlapping section is excluded from the third priority schedule.

The amount of solar power generation is set for charging the power storage unit between the end time of the light load section and the start time of exceeding the target load, and power is directly supplied to the load in the remaining section.

Again, referring to FIG. 8 , when the target load exceeding time is slower than the start time of the maximum load section, the charging and discharging schedules are displayed. In the middle load section (B) from 12:00 to 13:00, the predicted load wattage reaches the target load. A target load overrun start time in which the predicted load wattage coincides with the target load is calculated. The amount of solar power generation is predicted from the end of the light load section to the start time of the maximum load section, and the power storage unit sets a solar charge schedule according to the amount of photovoltaic power generation in this section.

Then, the start of grid charging is calculated in consideration of the charging speed in order to charge the remaining amount of charge of the power storage unit to the external power grid system in the light load section. In this embodiment, the start of grid charging is 1 o'clock.

After the target load overrun start time, the discharge schedule is set. The first priority discharge schedule is set in the excess section D in which the predicted load power exceeds the target load. Preferably, assuming that the total charging strategy of the power storage unit is 100%, the excess amount of power in the excess section may be at the level of 20%. Accordingly, the remaining 80% of the charging amount is allocated to the maximum load section (C) as the second priority. However, when the excess section (D) and the maximum load section (C) overlap, the overlapping section is excluded from the second priority schedule. In the second priority schedule, 70% is used, leaving 10%.

Then, the remaining 10% is allocated to the intermediate load section (B). However, when the excess section (D) and the intermediate load section (B) overlap, the overlapping section is excluded from the third priority schedule.

The amount of solar power generation is set for charging the power storage unit between the end time of the light load section and the start time of the maximum load section, and power is directly supplied to the load in the remaining section.

Again, referring to FIG. 9 , when the target load exceeding time is slower than the start time of the maximum load section, the charging and discharging schedules are displayed. In the maximum load section (C) between 10:00 and 12:00, the predicted load wattage reaches the target load. A target load overrun start time in which the predicted load wattage coincides with the target load is calculated. The amount of solar power generation is predicted from the end of the light load section to the start time of the maximum load section, and the power storage unit sets a solar charge schedule according to the amount of photovoltaic power generation in this section.

Then, the start of grid charging is calculated in consideration of the charging speed in order to charge the remaining amount of charge of the power storage unit to the external power grid system in the light load section. In this embodiment, the start of grid charging is 1 o'clock.

After the target load overrun start time, the discharge schedule is set. The first priority discharge schedule is set in the excess section D in which the predicted load power exceeds the target load. Preferably, assuming that the total amount of charge of the power storage unit is 100%, the excess amount of power in the excess section D may be at the level of 20%. Accordingly, the remaining charge amount of 80% is allocated to the maximum load section (C) as the second priority. However, when the excess section (D) and the maximum load section (C) overlap, the overlapping section is excluded from the second priority schedule. In the second priority schedule, 70% is used, leaving 10%.

Then, the remaining 10% is allocated to the intermediate load section (B). However, when the excess section (D) and the intermediate load section (B) overlap, the overlapping section is excluded from the third priority schedule.

The amount of solar power generation is set for charging the power storage unit between the end time of the light load section and the start time of the maximum load section, and power is directly supplied to the load in the remaining section.

In the above, the present invention has been illustrated and described with respect to specific preferred embodiments, but the present invention is not limited to these embodiments, and those of ordinary skill in the art to which the present invention pertains in the claims. It includes all of the various types of embodiments that can be implemented without departing from the technical spirit.

110: control unit
120: solar inverter
130: power storage unit
140: DC / AC conversion unit
160: load

Claims (9)

a solar inverter for generating and outputting power from sunlight;
a power storage unit for storing energy by receiving power from the solar inverter or an external power grid system;
a DC/AC conversion unit for converting AC power of the solar inverter or external power grid system into DC power, supplying it to the power storage unit, and transferring the power stored in the power storage unit as AC to a load; and
Receives the past load power used in the load and the past solar power output from the solar inverter, predicts the current photovoltaic power generation and load power, and based on the predicted solar power generation and load power and set target load to form a charging and discharging schedule of the power storage unit, and a control unit for controlling the solar inverter and the DC/AC conversion unit according to the charging and discharging schedule. According to claim 1,
The control unit is:
a data collection module for collecting the solar power generation amount and the load power amount;
a database for storing data collected by the data collection module;
a prediction unit for predicting the amount of solar power generation and load power by using the data stored in the database;
a schedule generator for generating a charging schedule and a discharging schedule based on the target load; and
and a control module for controlling the solar inverter and the DC/AC converter according to the charging schedule and the discharging schedule. According to claim 1,
The schedule generating unit generates a solar charging schedule and a light load interval charging schedule based on the preceding and following the target load excess start time and the maximum load period start time. a solar inverter for generating and outputting power from sunlight; a power storage unit for receiving power from the solar inverter or grid and storing energy; a DC/AC conversion unit for converting AC power of the solar inverter or external power grid system into DC power, supplying it to the power storage unit, and transferring the power stored in the power storage unit as AC to a load; and the past load power used in the load and the past photovoltaic power output from the photovoltaic inverter to predict the current photovoltaic power generation and load power, and to the predicted photovoltaic power generation and load power and set target load In the operating method of a power storage system comprising a control unit that forms a charge and discharge schedule of the power storage unit based on the charge and discharge schedule and controls the solar inverter and the DC/AC conversion unit according to the charging and discharging schedule,
Predicting the amount of solar power generation and load power;
setting a target load and generating a charging schedule of the power storage unit based on the target load;
generating a discharge schedule of the power storage unit based on the target load amount; and
and controlling the DC/AC converter according to the charging schedule and the discharging schedule. 5. The method of claim 4 .
Setting a target load amount and generating a charging schedule of the power storage unit based on the target load amount includes:
calculating a target load excess time exceeding the target load amount from the predicted load power amount;
comparing the target load excess time and the maximum load section start time;
predicting a first solar power generation amount from the end time of the light load section to the start time of the target load excess when the target load excess time is earlier than the maximum load section start time;
calculating the first amount of charge of the power storage unit before the end of the light load period by subtracting the first solar power generation amount from the upper limit of the charge of the power storage unit;
generating a solar charging schedule with the first solar power generation amount from the end time of the light load section to the start time of exceeding the target load;
calculating a system charging start time of the power storage unit for system charging through an external system within the light load section in consideration of the charging speed of the power storage unit; and
and generating a charging schedule for the light load section from the start time of the grid charging to the end time of the light load section. 6. The method of claim 5.
predicting a second solar power generation amount from the end time of the light load section to the start time of the maximum load section when the target load exceeding time is slower than the start time of the maximum load section;
calculating a second amount of charge of the power storage unit before the end of the light load period by subtracting the second solar power generation amount from the upper limit of the charge of the power storage unit; and
generating a solar charging schedule with the second solar power generation amount from the end time of the light load period to the start time of the maximum load period; The method of operating a power storage system, characterized in that it further comprises. 5. The method of claim 4 .
The generating of the discharge schedule of the power storage unit based on the target load may include:
Calculating a first power excess exceeding the target load amount from the predicted load power amount, calculating an excess section exceeding the target power amount, and setting a first priority discharge schedule. how it operates. 8. The method of claim 7 .
generating a second priority discharge schedule to discharge the remaining amount of charge remaining after subtracting the excess amount of first power from the upper limit of charge of the power storage unit; and
The method of operating a power storage system, further comprising the step of generating a third priority discharge schedule to discharge in an intermediate load section with the remaining charge after the second priority discharge schedule. 5. The method of claim 4 .
Predicting solar power generation and load power is:
converting past solar power generation and load power data into unstructured data;
extracting machine learning training data using the unstructured data;
constructing a layer model of Long Short-Term Memory models (LSTM) of a recurrent neural network using the machine learning training data;
repeatedly learning the layer model; and
The method of operating a power storage system comprising the step of estimating the amount of solar power generation and the amount of load power based on the repetitive learning in a 24-hour time series.

Images