KR102502724B1 - Apparatus and method for scheduling battery using demand forecasting based on AI - Google Patents

Abstract

Battery scheduling apparatus and method using artificial intelligence-based demand forecasting according to a preferred embodiment of the present invention, based on the predicted demand by time zone obtained using a long short-term memory (LSTM)-based demand forecasting model, By performing optimal charging/discharging scheduling of an electric vehicle (EV) or energy storage system (ESS), it determines the optimized energy scheduling for each time zone to reduce economic costs and improve the safety of system load. can make it

Description

Apparatus and method for scheduling battery using demand forecasting based on AI}

The present invention relates to a battery scheduling apparatus and method using artificial intelligence-based demand forecasting, and more particularly, to perform charging/discharging scheduling of an electric vehicle (EV) or an energy storage system (ESS), It relates to an apparatus and method.

Energy scheduling using an energy storage system (ESS) refers to adjusting supply and demand by additionally supplying insufficient demand and storing surplus energy in an energy storage system (ESS) when supply is excessive.

Power loads can be efficiently managed by scheduling multiple batteries included in an energy storage system (ESS) and an electric vehicle (EV), and an energy scheduling method can be flexibly changed according to demand and supply.

The characteristics of the grid and electric vehicle (EV) loads can be monitored for each time period, the characteristics of each load vary according to the time, and the charge based on usage can also vary according to the time-based rate system. An appropriate amount of energy must be charged/discharged according to the appropriate charge/discharge level for the time zone.

Scheduling should reduce economic costs as much as possible while considering supply and demand. In conventional scheduling methods, it is difficult to obtain economic effects because they do not consider energy costs that change by time. Alternatively, there is a problem of predicting the energy demand of an electric vehicle (EV) for each time period.

Scheduling in electric vehicles (EVs) requires minute-based scheduling and forecasting technology, not conventional hourly-based scheduling, and accordingly, appropriate scheduling must be presented according to the demand/supply required for each time zone based on accurate demand forecasting.

The object to be achieved by the present invention is based on the predicted demand by time zone obtained using a long short-term memory (LSTM)-based demand forecasting model, electric vehicle (EV) or energy storage system (Energy It is an object of the present invention to provide a battery scheduling device and method using artificial intelligence-based demand forecasting that performs optimal charging and discharging scheduling of a Storage System (ESS).

Other non-specified objects of the present invention may be additionally considered within the scope that can be easily inferred from the following detailed description and effects thereof.

Battery scheduling apparatus using artificial intelligence-based demand forecasting according to a preferred embodiment of the present invention for achieving the above object uses a long short-term memory (LSTM)-based demand forecasting model learned based on learning data a demand forecasting unit that obtains predicted demand for each time slot of a scheduling target; a determination unit for determining a time window (TW) used for charge scheduling and discharge scheduling of the scheduling target based on the predicted demand for each time period obtained through the demand estimation unit; And using a water-filling algorithm and a bin packing problem, based on the predicted demand for each time period obtained through the demand forecasting unit and the price and supply information for each time period set in advance, the determination is made. and a scheduling unit that obtains a charging schedule and a discharging schedule of the scheduling target in units of time according to the time window determined through the unit.

Here, the scheduling unit obtains the maximum charge capacity and maximum discharge capacity for each time period according to the time window using the water-filling algorithm, and uses the box filling problem to obtain the predicted demand for each time period, the rate for each time period, and The charging schedule and the discharging schedule of the scheduling target may be obtained in units of time according to the time window based on the supply amount information and the maximum charge capacity and maximum discharge capacity for each time period.

Here, the scheduling unit obtains a charging level based on the predicted demand for each time slot, and obtains the charging schedule of the scheduling target based on the charging level, the rate and supply information for each time slot, and the maximum charging capacity for each time slot. The charge level may be inversely converted to obtain a discharge level, and the discharge schedule of the scheduling target may be obtained based on the discharge level, the rate and supply information for each time period, and the maximum discharge capacity for each time period.

Here, the learning data may be driving data of an electric vehicle or advanced metering infrastructure (AMI) data of a power grid.

A battery scheduling method using artificial intelligence-based demand forecasting according to a preferred embodiment of the present invention for achieving the above object uses a long short-term memory (LSTM)-based demand forecasting model learned based on learning data obtaining a predicted demand for each time slot of a scheduling target; determining a time window (TW) used for charging scheduling and discharging scheduling of the scheduling target based on the predicted demand for each time period; And using a water-filling algorithm and a bin packing problem, based on the predicted demand for each time period and the price and supply information for each time period set in advance, the time unit according to the time window Acquiring a charging schedule and a discharging schedule of a scheduling target; includes.

According to the battery scheduling apparatus and method using artificial intelligence-based demand forecasting according to a preferred embodiment of the present invention, based on the predicted demand by time zone obtained using a long short-term memory (LSTM)-based demand forecasting model , By performing the optimal charging and discharging scheduling of an electric vehicle (EV) or energy storage system (ESS), it is possible to determine the optimal energy scheduling for each time zone to reduce economic costs and ensure system load safety. can improve

The effects of the present invention are not limited to the effects mentioned above, and other effects not mentioned will be clearly understood by those skilled in the art from the description below.

1 is a block diagram illustrating a battery scheduling device using artificial intelligence-based demand forecasting according to a preferred embodiment of the present invention.
2 is a diagram for explaining a long short-term memory (LSTM) according to a preferred embodiment of the present invention.
3 is a diagram for explaining a process of obtaining a charge level and a discharge level according to a preferred embodiment of the present invention.
4 is a diagram for explaining a process of obtaining maximum charge capacity and maximum discharge capacity according to a preferred embodiment of the present invention.
5 is a diagram for explaining a process of obtaining a charge schedule and a discharge schedule according to a preferred embodiment of the present invention, and shows a process of obtaining weights for each time period.
6 is a diagram for explaining a process of obtaining a charge schedule and a discharge schedule according to a preferred embodiment of the present invention, showing a process of optimizing a charge schedule and a discharge schedule using a bin packing problem.
7 is a diagram for explaining an example of a charging schedule and a discharging schedule of a scheduling target according to a preferred embodiment of the present invention. FIG. 7 (a) shows a case where a time window (TW) is 5 minutes, Figure 7 (b) shows the case where the time window (TW) is 45 minutes, Figure 7 (c) shows the case where the time window (TW) is 60 minutes, Figure 7 (d) shows the time window (TW) represents the case where is 180 minutes.
8 is a diagram for explaining optimization results and cost of a charge schedule and a discharge schedule according to a time window (TW) according to a preferred embodiment of the present invention.
9 is a flowchart illustrating a battery scheduling method using artificial intelligence-based demand forecasting according to a preferred embodiment of the present invention.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings. Advantages and features of the present invention, and methods for achieving them, will become clear with reference to the embodiments described below in detail in conjunction with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below and may be implemented in various different forms, only the present embodiments make the disclosure of the present invention complete, and common knowledge in the art to which the present invention belongs. It is provided to completely inform the person who has the scope of the invention, and the present invention is only defined by the scope of the claims. Like reference numbers designate like elements throughout the specification.

Unless otherwise defined, all terms (including technical and scientific terms) used in this specification may be used in a meaning commonly understood by those of ordinary skill in the art to which the present invention belongs. In addition, terms defined in commonly used dictionaries are not interpreted ideally or excessively unless explicitly specifically defined.

In this specification, terms such as "first" and "second" are used to distinguish one component from another component, and the scope of rights should not be limited by these terms. For example, a first element may be termed a second element, and similarly, a second element may be termed a first element.

In this specification, identification codes (eg, a, b, c, etc.) for each step are used for convenience of explanation, and identification codes do not describe the order of each step, and each step is clearly Unless a specific order is specified, it may occur in a different order from the specified order. That is, each step may occur in the same order as specified, may be performed substantially simultaneously, or may be performed in the reverse order.

In this specification, expressions such as "has", "may have", "includes" or "may include" indicate the existence of a corresponding feature (eg, numerical value, function, operation, or component such as a part). indicated, and does not preclude the presence of additional features.

In addition, the term '~unit' described in this specification means software or a hardware component such as a field-programmable gate array (FPGA) or ASIC, and '~unit' performs certain roles. However, '~ part' is not limited to software or hardware. '~bu' may be configured to be in an addressable storage medium and may be configured to reproduce one or more processors. Therefore, as an example, '~unit' refers to components such as software components, object-oriented software components, class components, and task components, processes, functions, properties, and procedures. , subroutines, segments of program code, drivers, firmware, microcode, circuitry, data structures and variables. Functions provided within components and '~units' may be combined into smaller numbers of components and '~units' or further separated into additional components and '~units'.

Hereinafter, with reference to the accompanying drawings, preferred embodiments of a battery scheduling apparatus and method using artificial intelligence-based demand forecasting according to the present invention will be described in detail.

First, a battery scheduling device using artificial intelligence-based demand forecasting according to a preferred embodiment of the present invention will be described with reference to FIGS. 1 to 6 .

1 is a block diagram for explaining a battery scheduling apparatus using artificial intelligence-based demand forecasting according to a preferred embodiment of the present invention, and FIG. 2 is a long short-term memory (LSTM) according to a preferred embodiment of the present invention. ), FIG. 3 is a diagram for explaining a process of obtaining a charge level and a discharge level according to a preferred embodiment of the present invention, and FIG. 4 is a view for explaining the maximum charge capacity and maximum charge level according to a preferred embodiment of the present invention 5 is a diagram for explaining a process of obtaining a discharge capacity, and FIG. 5 is a diagram for explaining a process of obtaining a charge schedule and a discharge schedule according to a preferred embodiment of the present invention, and shows a process of obtaining weights for each time period. A diagram for explaining a process of obtaining a charge schedule and a discharge schedule according to a preferred embodiment of the present invention, showing a process of optimizing a charge schedule and a discharge schedule using a bin packing problem.

Referring to FIG. 1, a battery scheduling apparatus (hereinafter referred to as a 'battery scheduling apparatus') 100 using artificial intelligence-based demand forecasting according to a preferred embodiment of the present invention is based on long short-term memory (LSTM) Based on the predicted demand for each time period obtained using the demand prediction model, optimal charging and discharging scheduling of an electric vehicle (EV) or an energy storage system (ESS) is performed.

To this end, the battery scheduling device 100 may include a demand estimation unit 110 , a determination unit 130 and a scheduling unit 150 .

The demand forecasting unit 110 obtains the predicted demand for each time slot of the scheduling target by using a long short-term memory (LSTM) based demand prediction model learned based on the learning data.

Here, the scheduling target refers to an electric vehicle (EV) or an energy storage system (ESS).

And, the learning data may be driving data of an electric vehicle (EV) or advanced metering infrastructure (AMI) data of a power grid.

That is, the demand forecasting unit 110 may obtain predicted demand for each time slot for the future of a scheduling target through the long and short term memory (LSTM) as shown in FIG. 2 . Referring to FIG. 2, long short-term memory (LSTM) solves the long-term dependency problem by adding an input gate, a forget gate, and an output gate to the memory cells of the hidden layer, erasing unnecessary memories, and storing data values to be remembered. It refers to models based only on artificial intelligence and deep neural networks. x _t is an input value and represents demand data of an electric vehicle (EV) or a power grid (Grid).

The determination unit 130 determines a time window (TW) used for charging scheduling and discharging scheduling of a scheduling target based on the predicted demand for each time slot obtained through the demand forecasting unit 110 .

That is, the determination unit 130 may determine a time window (TW) suitable for the scheduling target based on the predicted demand for each time slot for the future of the scheduling target.

Here, the time window TW may be determined in consideration of the sampling rate of the AMI of the grid and the maximum duration of the output of the energy storage system (ESS) (maximum duration that can be sustained with the same output). For example, when the AMI sampling rate of the grid is 5 minutes, the minimum value of the time window (TW) is 5 minutes, and the maximum value of the time window (TW) is the maximum duration of the output of the energy storage system (ESS). there is. That is, the time window TW may be determined as a value between 5 minutes and the maximum duration of the output of the ESS.

The scheduling unit 150 uses a water-filling algorithm and a bin packing problem to determine the predicted demand for each time period obtained through the demand forecasting unit 110 and the pre-set price and supply amount for each time period. Based on the information, the charging schedule and the discharging schedule of the scheduling target are obtained in units of time according to the time window TW determined through the determination unit 130 .

For example, as shown in FIG. 3, a charge level (C-Level) and a discharge level (D-Level) are obtained based on predicted demand for each time period and price and supply information for each time period. At this time, the maximum charge capacity (W _C ) and maximum discharge capacity (W _D ) can be obtained. Since the power price is low at low load (Off-Peak), charging scheduling is mainly considered, and at high load (On-Peak), it can be scheduled in the direction of reducing electricity cost as much as possible through discharge.

That is, the scheduling unit 150 may obtain the maximum charge capacity (W _C ) and maximum discharge capacity (W _D ) for each time period according to the time window (TW) by using a water-filling algorithm.

For example, as shown in FIG. 4, the maximum charge capacity (W _C ) or the maximum discharge capacity (W _D ) can be obtained in units of time according to the time window (TW) based on a water-filling algorithm. can The water-filling algorithm determines how much capacity to deploy at each time zone is the most optimal energy level for the entire time period, and uses the bin packing problem described later to determine each time period. This will optimize the charging capacity and discharging capacity.

In addition, the scheduling unit 150 uses the bin packing problem to estimate demand for each time slot, price and supply information for each time slot, and maximum charge capacity (W _C ) and maximum discharge capacity (W _D ) for each time slot based on As a result, the charging schedule and the discharging schedule of the scheduling target can be obtained in units of time according to the time window TW.

For example, as shown in FIG. 5 , energy (α _i ) supplied in each time period may be determined in two domains. That is, α _i is determined according to the function W _k =(α _i , M, t _k ), the y-axis domain is 1/Φ(t _k ), and the x-axis domain is M-demand(t _k ). Here, Φ(t _k ) represents the rate for each time slot. M represents the maximum charge capacity (W _C ) or maximum discharge capacity (W _D ). In this way, the charging capacity optimized for each time slot may be determined and allocated by simultaneously considering the charge and the available charging capacity.

In more detail, the scheduling unit 150 may obtain a charging level (C-Level) based on predicted demand for each time period. In addition, the scheduling unit 150 may obtain a charging schedule of a scheduling target based on a charging level (C-Level), rate and supply information for each time slot, and maximum charging capacity (W _C ) for each time slot. Also, the scheduling unit 150 may inversely convert the charge level (C-Level) to obtain the discharge level (D-Level). Also, the scheduling unit 150 may obtain a discharge schedule of a scheduling target based on the discharge level (D-Level), rate and supply amount information per time period, and maximum discharge capacity (W _D ) per time period.

That is, the charge level (C-Level) for determining the economically suitable maximum charge amount is determined in consideration of the total capacity of the installed energy storage system (ESS) and the electric rate system as shown in FIG. 4 . When the energy storage system (ESS) is charged 100% in a low-load rate period (Off-Peak), the charging level (C-Level) may be determined to enable charging at the lowest cost. In order to determine the charging level (C-Level), the determination of the maximum charging capacity (Wc) in each time zone must be preceded.

는 해당 충전 용량 사용시 부과되는 전기 요금을 나타낸다.Here, S represents the number of scheduling per day according to the time window (TW). C _t represents the charge capacity allocated to each unit of time.

represents the electric charge charged when using the corresponding charging capacity.

)은 에너지 저장 시스템(ESS)의 설치 용량과 같거나 그 이하이여야 한다. 이때, 결정된 각 시간대별 최대 충전 용량(W_C)을 배치하는 경우(도 4의 하늘색 영역), 도 4에 도시된 바와 같이 충전 레벨(C-Level)이 결정될 수 있다.The water-filling algorithm maximizes the maximum charging capacity (W _C ) while simultaneously maximizing the sum of the charging capacities scheduled through the water-filling algorithm (

) should be equal to or less than the installed capacity of the energy storage system (ESS). In this case, when the determined maximum charging capacity (W _C ) for each time zone is arranged (light blue area in FIG. 4), the charging level (C-Level) may be determined as shown in FIG. 4 .

As above, in the case of the discharge level (D-Level), discharge scheduling is determined through a water-filling algorithm in a similar way.

는 해당 방전 용량 사용시 부과되는 전기 요금을 나타낸다.Here, D _t represents the discharge capacity allocated to each unit of time.

represents the electric charge charged when using the corresponding discharge capacity.

)은 충전 용량의 총합 (

)과 같거나 그 이하이여야 한다. 마찬가지로 각 시간대 별로 최대 방전 용량(W_D)을 과부하 구간(On-Peak)에 배치하는 경우(도 4의 노란색 영역), 도 4에 도시된 바와 같이 방전(D-Level)이 결정될 수 있다.In the case of discharge scheduling, the total amount of discharge capacity (

) is the sum of the charging capacities (

) must be equal to or less than. Similarly, when the maximum discharge capacity (W _D ) is placed in the overload section (On-Peak) for each time period (yellow area in FIG. 4), the discharge (D-Level) can be determined as shown in FIG.

The charge level (C-Level), maximum charge capacity (W _C ), discharge level (D-Level), and maximum discharge capacity (W _D ) are based on the grid, electricity rate, and energy storage system (ESS). When, the recommended maximum conditions and limits are indicated.

In order to arrange the optimized capacity for each time zone, it is necessary to optimize through the bin packing problem. α _i , which is determined as the charge/discharge capacity, changes the economic cost and the load of the power grid depending on the time of charge/discharge. As shown in FIG. 5, minimizing the y-axis domain (1/Φ (t _k )) that can consider economic cost and the x-axis domain (M-demand (t _k )) that can consider the load of the power grid at the same time Deploy the capacity at the right time. At this time, the disposed charge/discharge capacity is disposed in a time period that does not exceed the maximum charge/discharge level condition determined by the water-filling algorithm. The better the optimized layout, the higher the optimization score (Balance Index).

Next, performance of a battery scheduling operation using artificial intelligence-based demand forecasting according to a preferred embodiment of the present invention will be described with reference to FIGS. 7 and 8 .

7 is a diagram for explaining an example of a charging schedule and a discharging schedule of a scheduling target according to a preferred embodiment of the present invention. FIG. 7 (a) shows a case where a time window (TW) is 5 minutes, Figure 7 (b) shows the case where the time window (TW) is 45 minutes, Figure 7 (c) shows the case where the time window (TW) is 60 minutes, Figure 7 (d) shows the time window (TW) represents the case where is 180 minutes.

Referring to FIG. 7 , since the capacity optimized for each time period is supplied, economic cost can be reduced and load stability can be increased. In addition, flexible scheduling is possible according to the time window (TW), so it can be applied not only to a large grid (Grid) but also to an electric vehicle (EV), which is a small energy unit.

8 is a diagram for explaining optimization results and cost of a charge schedule and a discharge schedule according to a time window (TW) according to a preferred embodiment of the present invention.

8, as a result of comparing the optimization score (Balance Index) and the daily electricity cost (Daily Cost) according to each time window (TW), 5 minutes / 10 minutes / 15 minutes / 20 minutes / 30 minutes / 60 minutes It can be seen that the economic cost is the same in the unit schedule, but the optimization score shows the most effective result in the 60-minute unit scheduling.

Next, a battery scheduling method using artificial intelligence-based demand prediction according to a preferred embodiment of the present invention will be described with reference to FIG. 9 .

9 is a flowchart illustrating a battery scheduling method using artificial intelligence-based demand forecasting according to a preferred embodiment of the present invention.

Referring to FIG. 9 , the battery scheduling apparatus 100 obtains predicted demand for each time period of a scheduling target by using a long and short-term memory (LSTM)-based demand prediction model (S110).

Then, the battery scheduling apparatus 100 determines the time window TW used for charging scheduling and discharging scheduling of the scheduling target based on the predicted demand for each time period (S130).

Then, the battery scheduling device 100 uses a water-filling algorithm and a bin packing problem, based on the predicted demand for each time period and the price and supply information for each time period set in advance. A charging schedule and a discharging schedule of a scheduling target are obtained in units of time according to the window (S150).

That is, the battery scheduling device 100 obtains the maximum charge capacity (W _C ) and maximum discharge capacity (W _D ) for each time period according to the time window (TW) using a water-filling algorithm, and Using the Bin Packing Problem, the time according to the time window (TW) based on the predicted demand for each hour, rate and supply information for each hour, and the maximum charge capacity (W _C ) and maximum discharge capacity (W _D ) for each hour It is possible to obtain a charging schedule and a discharging schedule of a scheduling target in units.

In more detail, the battery scheduling device 100 obtains the charge level (C-Level) based on the predicted demand for each time period, and the charge level (C-Level), information on rate and supply per time period, and maximum charge capacity per time period. Acquire the charging schedule of the scheduling target based on (W _C ), inversely transform the charge level (C-Level) to obtain the discharge level (D-Level), and obtain the discharge level (D-Level), rate and supply by time A discharge schedule of a scheduling target may be obtained based on the information and the maximum discharge capacity (W _D ) for each time period.

Even though all components constituting the embodiments of the present invention described above are described as being combined or operated as one, the present invention is not necessarily limited to these embodiments. That is, within the scope of the object of the present invention, all of the components may be selectively combined with one or more to operate. In addition, although all of the components may be implemented as a single independent piece of hardware, some or all of the components are selectively combined to perform some or all of the combined functions in one or a plurality of pieces of hardware. It may be implemented as a computer program having. In addition, such a computer program can implement an embodiment of the present invention by being stored in a computer readable recording medium such as a USB memory, a CD disk, a flash memory, etc., and read and executed by a computer. A recording medium of a computer program may include a magnetic recording medium, an optical recording medium, and the like.

The above description is merely an example of the technical idea of the present invention, and those skilled in the art can make various modifications, changes, and substitutions without departing from the essential characteristics of the present invention. will be. Therefore, the embodiments disclosed in the present invention and the accompanying drawings are not intended to limit the technical idea of the present invention, but to explain, and the scope of the technical idea of the present invention is not limited by these embodiments and the accompanying drawings. . The protection scope of the present invention should be construed according to the claims below, and all technical ideas within the equivalent range should be construed as being included in the scope of the present invention.

100: battery scheduling device,
110: demand forecasting unit,
130: decision unit,
150: scheduling unit

Claims (5)

a demand forecasting unit that obtains predicted demand for each time slot of a scheduling target by using a long short-term memory (LSTM)-based demand forecasting model learned based on the learning data;
a determination unit for determining a time window (TW) used for charge scheduling and discharge scheduling of the scheduling target based on the predicted demand for each time period obtained through the demand estimation unit; and
Using a water-filling algorithm and a bin packing problem, based on the predicted demand for each time period obtained through the demand forecasting unit and the price and supply information for each time period set in advance, the determining unit a scheduling unit that obtains a charging schedule and a discharging schedule of the scheduling target in units of time according to the time window determined through;
Including,
The scheduling unit obtains a maximum charge capacity and a maximum discharge capacity for each time period according to the time window using the water-filling algorithm, and uses the box filling problem to obtain the predicted demand for each time period and the price and supply information for each time period. and obtaining the charging schedule and the discharging schedule of the scheduling target in units of time according to the time window based on the maximum charge capacity and maximum discharge capacity for each time period;
The scheduling unit obtains a charging level based on the predicted demand for each time slot, and obtains the charging schedule of the scheduling target based on the charging level, the rate and supply information for each time slot, and the maximum charging capacity for each time slot; Obtaining a discharge level by inversely transforming the charge level, obtaining the discharge schedule of the scheduling target based on the discharge level, the rate and supply information for each time period, and the maximum discharge capacity for each time period;
The time window is determined by considering the sampling rate of the power grid's advanced metering infrastructure (AMI) and the maximum output duration time of the energy storage system (ESS).
Battery scheduling device using artificial intelligence-based demand forecasting. delete delete In paragraph 1,
The learning data,
Electric vehicle driving data or power grid AMI (Advanced Metering Infrastructure) data,
Battery scheduling device using artificial intelligence-based demand forecasting. obtaining predicted demand for each time slot of a scheduling target by using a long short-term memory (LSTM)-based demand prediction model learned based on the learning data;
determining a time window (TW) used for charging scheduling and discharging scheduling of the scheduling target based on the predicted demand for each time period; and
Using a water-filling algorithm and a bin packing problem, the scheduling is performed in units of time according to the time window, based on the predicted demand for each time period and the price and supply information for each time period set in advance. obtaining a charging schedule and a discharging schedule of the target;
Including,
The obtaining of the charging schedule and the discharging schedule of the scheduling target may include obtaining a maximum charge capacity and a maximum discharge capacity for each time period according to the time window using the water-filling algorithm, and using the box filling problem to obtain the maximum charge capacity and maximum discharge capacity for each time period. Acquiring the charging schedule and the discharging schedule of the scheduling target in units of time according to the time window based on the predicted demand for each unit, the rate and supply information for each time slot, and the maximum charge capacity and maximum discharge capacity for each time slot ,
The obtaining of the charging schedule and the discharging schedule of the scheduling target may include obtaining a charge level based on the predicted demand for each time slot, and based on the charge level, the charge and supply information for each time slot, and the maximum charge capacity for each time slot. The charging schedule of the scheduling target is obtained, the charging level is inversely converted to obtain a discharging level, and the discharging of the scheduling target is based on the discharging level, the rate and supply information for each time slot, and the maximum discharge capacity for each time slot. It consists of obtaining a schedule,
The time window is determined by considering the sampling rate of the power grid's advanced metering infrastructure (AMI) and the maximum output duration time of the energy storage system (ESS).
A battery scheduling method using artificial intelligence-based demand forecasting.

Images