MEDIUM

TECHNICAL FIELD [0001] The present disclosure relates to the field of electrical energy control technologies, and more particularly to a demand control method and system, and a storage medium.

BACKGROUND

[0002] For larger scale cities, due to the limited capacity of the distribution network in the urban area, and the increasing total demand for electricity consumption by users, the capacity of the distribution network available on the user side is becoming increasingly tight.

[0003] In the related art, the electricity fee for commercial electricity consumption is charged in two parts: the basic electricity fee and the meter reading electricity fee. The basic electricity fee is charged based on the maximum monthly demand, and the demand refers to the average power (in kilowatts) provided by the distribution network measured within a time period (such as 15 minutes). The meter reading electricity fee is charged based on the total electricity consumption (commonly known as kWh in kilowatt hours,) and the unit electricity price during the electricity consumption period. The unit electricity price is higher during peak hours and lower during off-peak hours. In order to reduce the basic electricity fee, the relevant technical personnel can control the maximum monthly demand based on experience, such as determining based on experience when the demand control needs to be performed, and at what power the energy storage device discharges during the demand control. [0004] In the above-mentioned related art, the relevant technical personnel make decisions on demand control merely based on manual experience, which results in low accuracy of demand control.

SUMMARY [0005] The embodiments of the present disclosure provide a demand control method and system, and a storage medium, which can solve the technical problem of low accuracy of demand control. The technical solution is as follows.

[0006] In one aspect, a demand control method is provided by an embodiment of the present disclosure. The demand control method is applicable to a demand control system including a controller, an energy storage device, a monitoring meter, and an energy storage cloud platform. The demand control method includes:

[0007] acquiring, by the energy storage cloud platform, a forecasted demand of an i ^th charging unit of a charging cycle through past load data by calling a load forecasting model for load forecasting; and sending, by the energy storage cloud platform, the forecasted demand of the i ^th charging unit to the controller, wherein the charging cycle includes n charging units, n being a positive integer, and i being a positive integer less than or equal to n; [0008] sending, by the monitoring meter, an actual demand of (i-1) charging units to the controller, wherein the (i-1) charging units are first (i-1) charging units of the charging cycle; and

[0009] determining, by the controller, a maximum forecasted demand of the i ^th charging unit based on the forecasted demand of the i ^th charging unit; determining, by the controller, a maximum actual demand of the (i-1) charging units based on the actual demand of the (i-1) charging units; determining, by the controller, that a demand control for the i ^th charging unit needs to be performed if the maximum forecasted demand of the i ^th charging unit is greater than the maximum actual demand of the (i-1) charging units; and controlling, by the controller, the energy storage device to discharge after determining that the demand control for the i ^th charging unit needs to be performed.

[0010] Optionally, the load forecasting model is obtained by:

[0011] acquiring at least one set of training samples, wherein each set of training samples includes forecasted load data within a past time period and past condition data corresponding to the past forecasted load data, the past condition data being configured to represent estimated environmental characteristics within the past time period;

[0012] training the load forecasting model by using the at least one set of the training samples; and

[0013] obtaining a trained load forecasting model by stopping, when a training-stop condition is satisfied, training the load forecasting model. [0014] Optionally, the training-stop condition includes:

[0015] a value of a loss function of the load forecasting model being less than a first threshold value, wherein the value of the loss function is obtained based on the forecasted load data within the past time period and an actual load data within the past time period; [0016] or, [0017] a number of training times of the load forecasting model being greater than a second threshold value.

[0018] Optionally, the demand control method further includes:

[0019] controlling, by the controller, a discharging power of the energy storage device as a target power;

[0020] wherein the target power is configured to control an actual demand of the i ^th charging unit to be less than or equal to a target demand of the i ^th charging unit, wherein the actual demand of the i ^th charging unit refers to a demand generated in real-time within the i ^th charging unit, and the target demand of the i ^th charging unit refers to a demand threshold of the i ^th charging unit.

[0021] Optionally, the demand control method further includes:

[0022] acquiring, by the controller, a maximum discharging power of the energy storage device; obtaining, by the controller, a first demand difference by calculating a difference between the maximum forecasted demand of the i ^th charging unit and the maximum discharging power; and determining, by the controller, a larger one of the first demand difference and the maximum actual demand of the (i-1) charging units as the target demand of the i ^th charging unit.

[0023] Optionally, the demand control method further includes:

[0024] setting, by the controller, the target demand of the i ^th charging unit equal to the actual demand of the i ^th charging unit if the target power reaches the maximum discharging power and the actual demand of the i ^th charging unit is greater than the target demand of the i ^th charging unit.

[0025] Optionally, the demand control method further includes:

[0026] determining, by the controller, whether the i ^th charging unit ends; determining, by the controller, a maximum actual demand of i charging units if the i ^th charging unit ends, wherein the i charging units are first i charging units of the charging cycle.

[0027] Optionally, the demand control method further includes:

[0028] determining, by the controller, a maximum actual demand of the i ^th charging unit; determining, by the controller, whether the maximum actual demand of the i ^th charging unit is greater than the maximum actual demand of the (i-1) charging units; setting, by the controller, the maximum actual demand of the i charging units equal to the maximum actual demand of the i ^th charging unit if the maximum actual demand of the i ^th charging unit is greater than the maximum actual demand of the (i-1) charging units.

[0029] In another aspect, a demand control system is provided by an embodiment of the present disclosure. The demand control system includes a controller, an energy storage device, a monitoring meter, and an energy storage cloud platform; wherein [0030] the energy storage cloud platform is configured to acquire a forecasted demand of an i ^th charging unit of a charging cycle through past load data by calling a load forecasting model for load forecasting; and send the forecasted demand of the i ^th charging unit to the controller, wherein i is a positive integer;

[0031] the monitoring meter is configured to send an actual demand of (i-1) charging units to the controller; and

[0032] the controller is configured to determine a maximum forecasted demand of the i ^th charging unit based on the forecasted demand of the i ^th charging unit; and determine a maximum actual demand of the (i-1) charging units based on the actual demand of the (i-1) charging units; determine that a demand control for the i ^th charging unit needs to be performed if the maximum forecasted demand of the i ^th charging unit is greater than the maximum actual demand of the (i-1) charging units; and control the energy storage device to discharge after determining that the demand control for the i ^th charging unit needs to be performed.

[0033] In still another aspect, a computer-readable storage medium is provided by an embodiment of the present disclosure, wherein the computer-readable storage memory is configured to storing a computer program, wherein the computer program is loaded and executed by a processor to perform the method steps on a controller side of the demand control method described above.

[0034] The technical solution provided by the embodiments of the present disclosure includes at least the following beneficial effects.

[0035] The maximum forecasted demand of the i ^th charging unit is obtained by performing the load forecasting model for load forecasting, whether the demand control for the i ^th charging unit needs to be performed is determined in combination with the maximum actual demand of the (i-1) charging units, and the demand control is performed by controlling the energy storage device to discharge if the demand control for the i ^th charging unit needs to be performed. Compared with the related art, the relevant technical personnel make decisions on demand control according to manual experience. In the technical solution provided by the embodiments of the present disclosure, on the one hand, the demand control system acquires the forecasted demand according to past data and determines whether to perform the demand control based on the forecasted demand and how to control the energy storage device for the demand control, which achieves the automation of the demand control and improves the accuracy of the demand control; on the other hand, the load forecasting is performed by using the load forecasting model, which improves the accuracy of the forecasting and further improves the accuracy of the demand control.

[0036] It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and do not limit the present disclosure.

BRIEF DESCRIPTION OF THE DRAWINGS [0037] In order to illustrate the technical solution provided by the embodiments of the present disclosure more clearly, a brief introduction may be given hereinafter to the accompany drawings that may be used in the description of the embodiments. Apparently, the drawings in the description below are merely some embodiments of the present disclosure, and other drawings may be acquired by those of ordinary skilled in the art according to these drawings without any creative efforts.

[0038] FIG. 1 is a schematic diagram of a demand control system according to an embodiment of the present disclosure;

[0039] FIG. 2 is a flowchart of a demand control method according to an embodiment of the present disclosure;

[0040] FIG. 3 is a flowchart of a demand control method according to another embodiment of the present disclosure;

[0041] FIG. 4 is an exemplary graph illustrating a relationship among real-time demand, target demand, and discharging power of an energy storage system;

[0042] FIG. 5 exemplarily illustrates a flowchart of a demand control method; and [0043] FIG. 6 is a structural block diagram of a controller according to an embodiment of the present disclosure.

DETAILED DESCRIPTION

[0044] Description will now be made in detail to exemplary embodiments, examples of which are represented in the accompanying drawings. The following description refers to the accompanying drawings in which the same numbers in different drawings represent the same or similar elements unless otherwise represented. The implementations set forth in the following description of exemplary embodiments do not represent all implementations consistent with the present disclosure. Instead, they are merely examples of some methods consistent with aspects of the present disclosure as recited in the appended claims. [0045] First, a brief introduction is given to the terms involved in the embodiments of the present disclosure.

[0046] The demand refers to the average power (in kilowatts) provided by the distribution network measured during a measurement period. The electricity fee is charged based on the maximum demand, that is, within a certain settlement period, the electricity fee is charged according to a maximum value of the average power of the user's electricity consumption. Since the instantaneous maximum power is meaningless and difficult to detect, the power supply company often use the "slip" measurement method to measure users' electricity consumption demand. Currently, the measurement rule adopted by the power supply company is as follows: a smart meter enters a measurement period (15 minutes) every 1 minute. In this way, there are 1440 measurements per day, and the smart meter only accesses the maximum value of the current day. If the value measured in the next day is greater than the value recorded the previous day, the value recorded by the smart meter is automatically overwritten by the new value. [0047] The energy storage device is a device that stores energy, which can collect and output energy. The stored energy and outputted energy of different energy storage devices may be different. For example, an energy storage device using solar energy may first convert the solar energy into chemical energy, and then convert the chemical energy into electrical energy for outputting. In the embodiments of the present disclosure, the energy storage device may input electrical energy and convert the electrical energy into other forms of energy (such as chemical energy, potential energy, thermal energy, and the like) for storage, and according to the actual situation, the stored energy is converted into electrical energy when or where it is needed, and thus is released.

[0048] The load forecasting refers to the load data determined at a specific time in the future based on factors such as the system's operating properties, capacity-increment decisions, natural conditions and social influences, and the like, under the condition satisfying certain accuracy requirements. The load refers to the demand for electrical energy or the electricity consumption. The methods applying the load forecasting may include unit consumption method, trend extrapolation method, elastic coefficient method, regression analysis method, time series method, gray model method, optimal combination forecasting method, wavelet analysis forecasting method, and the like; and the load forecasting may be divided into ultra-short-term, short-term, mid-term and long-term load forecasting depending on different purposes.

[0049] The ultra-short -term load forecasting refers to load forecasting within 1 hour in the future. In the state of security monitoring, it generally requires a forecasted value of 5 to 10 seconds or 1 to 5 minutes. The preventive control and emergency handling generally require a forecasted value of 10 minutes to 1 hour.

[0050] The short-term load forecasting refers to daily load forecasting and weekly load forecasting, which are used to arrange daily dispatching plans and weekly dispatching plans respectively, including determining unit start and stop, hydro-thermal power coordination, tie-lines exchanged power, load economic distribution, reservoir dispatching, device maintenance and the like. The short-term forecasted value is closely related to factors such as weather factors, day types and short-term load changes. [0051] The mid-term load forecasting refers to monthly to yearly load forecasting, which is mainly used to determine the unit operation mode and device overhaul plan.

[0052] The long-term load forecasting refers to load forecasting within the next 3 to 5 years or even longer, which is mainly the long-term plan of the power grid transformation and expansion made by the power grid planning department according to the development of the national economy and the requirement for electrical load. For the mid-term and long-term load forecasting, special research should be made on the influence of national economic development and national policies.

[0053] Referring to FIG. 1, which illustrates a schematic diagram of a demand control system according to an embodiment of the present disclosure. As shown in FIG. 1, the demand control system 100 may include an energy storage device 110, a controller 120, an energy storage cloud platform 130, and a monitoring meter 140.

[0054] The above-mentioned energy storage device 110 is configured to store electrical energy and release the stored electrical energy when the demand control needs to be performed, and the charging power and discharging power of the energy storage device 110 may be controlled. The energy storage device 110 may send its operating state (such as charging state and discharging state) and operating parameters (such as operating power, operating time, stored electricity, released electricity, remaining electricity, or the like) to the controller 120.

[0055] The energy storage cloud platform 130 is configured to obtain a load forecasting result by performing load forecasting. The energy storage cloud platform 130 may send the load forecasting result to the controller 120, such that the controller 120 may control the operating state and operating parameters of the energy storage device 110. The load forecasting is a method of forecasting load data at a specific time in the future, and the load forecasting result is configured to represent the load data at the specific time in the future. [0056] The monitoring meter 140 may be configured to monitor electricity consumption information of an electricity consumption device, such as the electricity consumption time and real-time demand of the electricity consumption device, and the devices which are operating. The above-mentioned electricity consumption device may be one electricity consumption device, or may include a plurality of electricity consumption devices. For enterprises, the above-mentioned electricity consumption device may include a production device such as machine tool and blast furnace, and an auxiliary production device such as illuminating lamp and crown block. The demand involved in the embodiments of the present disclosure is the sum of the demand of the electricity consumption devices. The monitoring meter 140 may send the electricity consumption situation of the electricity consumption device to the controller 120.

[0057] The controller 120 is configured to control the operating state and operating parameters of the energy storage device 110 according to operating state and operating parameters of the energy storage device 110, the load forecasting result, and the electricity consumption information of the electricity consumption device. The controller 120 may control the operating state and operating parameters of the energy storage device 110 according to the electricity consumption information of the electricity consumption device and the load forecasting result. The electricity consumption situation of the electricity consumption device monitored by the monitoring meter 140 may also be transmitted to the energy storage cloud platform 130 via the controller 120, such that the energy storage cloud platform 130 may perform a load forecasting based on the electricity consumption situation of the electricity consumption device.

[0058] Referring to FIG. 2, which illustrates a flowchart of a demand control method according to an embodiment of the present disclosure. In this embodiment, it is mainly illustrated by taking the demand control method applicable to the demand control system introduced in the embodiment of FIG. 1 as an example. The demand control method may include the following steps.

[0059] In step 201, the energy storage cloud platform acquires a forecasted demand of an i ^th charging unit of a charging cycle through past load data by calling a load forecasting model for load forecasting.

[0060] The charging cycle includes n charging units, wherein n is a positive integer, and i is a positive integer less than or equal to n.

[0061] The basic electricity fee is paid once in each charging cycle, and the basic electricity fee in each charging cycle is paid according to a maximum demand generated in the charging cycle.

[0062] Optionally, the n charging units are successive time units.

[0063] In some possible embodiments, the charging cycle may be one month, two months, or three months, which is not limited in the embodiments of the present disclosure. [0064] In some possible embodiments, one charging unit may be 4 hours, 6 hours, 12 hours, one day, two days, or one week, which is not limited in the embodiments of the present disclosure.

[0065] In the above-mentioned demand control system, the energy storage cloud platform is used for load forecasting to obtain the forecasted demand of the i ^th charging unit. The load forecasting has been explained in the above introduction to the terms, which will not be repeated herein. The above-mentioned forecasted demand of the i ^th charging unit refers to the demand of the i ^th charging unit which is forecasted.

[0066] The energy storage cloud platform may acquire the above-mentioned forecasted demand of the i ^th charging unit through past load data by calling the load forecasting model for load forecasting. The above-mentioned past load data is load data within a past time period, and the past load data includes a past actual demand.

[0067] Optionally, the above-mentioned load forecasting model may be a long short term memory (LSTM) model, a recurrent neural network (RNN) model, a deep residual network (DRN) model, an autoregressive integrated moving average (ARIMA) model, and the like. Furthermore, the above-mentioned load forecasting model may also be other models, which is not limited in the embodiments of the present disclosure.

[0068] Optionally, before the load forecasting model is called for load forecasting, the past load data may also be preprocessed to eliminate abnormal data therein, thereby improving the accuracy of the forecasted result. In the embodiments of the present disclosure, a combination of short-term forecasting (daily forecasting) and mid-term forecasting (monthly forecasting) may be used to perform a load forecasting, which improves the accuracy of load forecasting.

[0069] In step 202, the energy storage cloud platform sends the forecasted demand of the i ^th charging unit to the controller. [0070] After acquiring the above-mentioned forecasted demand of the i ^th charging unit, the energy storage cloud platform may send the forecasted demand of the i ^th charging unit to the controller.

[0071] Optionally, the data may be transmitted between the energy storage cloud platform and the controller through a network connection or through an electrical connection, which is not limited in the embodiments of the present disclosure.

[0072] Optionally, the energy storage cloud platform may send the forecasted demand of the i ^th charging unit to the controller before the i ^th charging unit or when the i ^th charging unit just starts. [0073] In step 203, the monitoring meter sends an actual demand of (i-1) charging units to the controller.

[0074] The (i-1) charging units are first (i-1) charging units of the charging cycle.

[0075] The monitoring meter may acquire the electricity consumption information of the electricity consumption device by monitoring the electricity consumption device. The electricity consumption information includes the actual demand of the (i-1) charging units. Then, the monitoring meter may send the actual demand of the (i-1) charging units to the controller. The above-mentioned actual demand of the (i-1) charging units refers to the demand actually generated by the (i-1) charging units.

[0076] Optionally, the data may be transmitted between the monitoring meter and the controller through a network connection or through an electrical connection, which is not limited in the embodiments of the present disclosure.

[0077] Optionally, the data may be transmitted between the energy storage cloud platform and the monitoring meter through a network connection or through an electrical connection, which is not limited in the embodiments of the present disclosure. [0078] In step 204, the controller determines a maximum forecasted demand of the i ^th charging unit based on the forecasted demand of the i ^th charging unit.

[0079] The maximum forecasted demand of the i ^th charging unit refers to the maximum demand of the i ^th charging unit which is forecasted.

[0080] Optionally, the forecasted demand of the i ^th charging unit for each time period may be obtained by load forecasting, and then the maximum value of the forecasted demand of the i ^th charging unit for each time period is determined as the maximum forecasted demand of the i ^th charging unit.

[0081] Optionally, a forecasted demand curve of the i ^th charging unit may be obtained by load forecasting, and the forecasted demand curve of the i ^th charging unit represents the forecasted change of the forecasted demand of the i ^th charging unit. The demand corresponding to the highest point of the forecasted demand curve of the i ^th charging unit is determined as the maximum forecasted demand of the i ^th charging unit.

[0082] In step 205, the controller determines a maximum actual demand of the (i-1) charging units based on the actual demand of the (i-1) charging units. [0083] The maximum actual demand of the (i-1) charging units refers to a maximum of the demand actually generated by the (i-1) charging units.

[0084] Optionally, the controller may acquire actual demands of the (i-1) charging units for each time period, and then a maximum of the actual demands of the (i-1) charging units for each time period is determined as the maximum actual demand of the (i-1) charging units. [0085] Optionally, the controller may acquire an actual demand curve of the (i-1) charging units, and the actual demand curve of the (i-1) charging units represents a change of the actual demand of the (i-1) charging units. The demand corresponding to the highest point of the actual demand curve of the (i-1) charging units is determined as the maximum actual demand of the (i-1) charging units.

[0086] Optionally, i is equal to 1, and the maximum actual demand of the (i-1) charging units may be set to 0.

[0087] In step 206, the controller determines that a demand control for the i ^th charging unit needs to be performed if the maximum forecasted demand of the i ^th charging unit is greater than the maximum actual demand of the (i-1) charging unit.

[0088] After acquiring the above-mentioned maximum forecasted demand of the i ^th charging unit and the maximum actual demand of the (i-1) charging units, the maximum forecasted demand of the i ^th charging unit may be compared with the maximum actual demand of the (i-1) charging units; if the maximum forecasted demand of the i ^th charging unit is greater than the maximum actual demand of the (i-1) charging units, the controller determines that the demand control for the i ^th charging unit needs to be performed.

[0089] The maximum forecasted demand of the i ^th charging unit is configured to represent a maximum demand that the i ^th charging unit will generate. Based on the maximum forecasted demand of the i ^th charging unit, it may be determined whether the demand control for the i ^th charging unit needs to be performed.

[0090] When the maximum forecasted demand of the i ^th charging unit is greater than the maximum actual demand of the (i-1) charging units, if the demand control is not performed, the maximum demand of the i ^th charging unit that is generated actually is likely to exceed the maximum actual demand of the (i-1) charging units that is originally recorded, and then a higher basic electricity fee may need to be paid for the charging cycle; that is, in this case, the demand control needs to be performed to prevent the maximum actual demand of the i ^th charging unit from exceeding the maximum actual demand of the (i-1) charging units as much as possible, so as to reduce the expenditure of the basic electricity fee.

[0091] In addition, if the maximum forecasted demand of the i ^th charging unit is less than or equal to the maximum actual demand of the (i-1) charging units, it is determined that the demand control for the i ^th charging unit does not need to be performed.

[0092] In step 207, the controller controls the energy storage device to discharge after determining that the demand control for the i ^th charging unit needs to be performed. [0093] The demand control is performed by controlling the energy storage device, and the demand control may be performed by controlling the discharging power of the energy storage device.

[0094] Optionally, when determining that the demand control for the i ^th charging unit needs to be performed, the energy storage device outputs power to supplement the power needed to be consumed by the electricity consumption device, such that the real-time demand which is generated does not exceed the maximum actual demand of the (i-1) charging units as much as possible.

[0095] If it is predicted that the demand control for the i ^th charging unit needs to be performed, part of the electricity may be reserved in the energy storage device to use for the demand control in the future during the time period in which the demand control does not need to be performed but it is within the peak electricity price period.

[0096] Optionally, the reserved electricity may be 15%, 20%, or 25% of the total capacity of the energy storage device. The reserved electricity is specifically set by the relevant technical personnel according to actual needs, which is not limited in the embodiments of the present disclosure.

[0097] In summary, in the embodiments of the present disclosure, the maximum forecasted demand of the i ^th charging unit is obtained by performing the load forecasting model for load forecasting, and whether the demand control for the i ^th charging unit needs to performed is determined in combination with the maximum actual demand of the (i-1) charging units, and the demand control is performed by controlling the energy storage device to discharge if the demand control for the i ^th charging unit needs to be performed. Compared with the related art, the relevant technical personnel make decisions on demand control based on manual experience. In the technical solution provided by the embodiments of the present disclosure, on the one hand, the demand control system acquires the forecasted demand according to past data and determines whether to perform the demand control based on the forecasted demand and how to control the energy storage device for the demand control, which achieves the automation of the demand control and improves the accuracy of the demand control; on the other hand, the load forecasting is performed by using the load forecasting model, which improves the accuracy of the forecasting and further improves the accuracy of the demand control.

[0098] Referring to FIG. 3, which illustrates a flowchart of a demand control method according to an embodiment of the present disclosure. In this embodiment, it is mainly illustrated by taking the method applicable to the demand control system introduced above as an example. The method may include the following steps.

[0099] In step 301, the energy storage cloud platform acquires a forecasted demand of the i ^th charging unit through past load data by calling a load forecasting model for load forecasting.

[00100] The content of this step is the same as or similar to that of step 201 in the embodiment of FIG. 2, which will not be repeated herein.

[00101] Optionally, the load forecasting model may be obtained by the following ways. [00102] 1. At least one set of training samples are acquired.

[00103] Each set of training samples includes forecasted load data within a past time period and past condition data corresponding to the past forecasted load data. The above-mentioned past time period represents a time period in the past, each time period may be spaced at a fixed time. The above-mentioned past condition data is configured to represent estimated environmental characteristics within the past time period. The past condition data may include data such as the external temperature (such as minimum temperature, maximum temperature, real-time temperature, or the like), humidity (such as relative humidity, absolute humidity, real-time humidity, or the like), weather type (such as sunny, rainy, snowy, overcast, or the like) and the like corresponding to the past time period.

[00104] 2. The load forecasting model is trained by using the at least one set of training samples.

[00105] After acquiring the above-mentioned at least one set of training samples, the above-mentioned at least one set of training samples may be used to train the load forecasting model to adjust various parameters of the load forecasting model.

[00106] 3. A trained load forecasting model is obtained by stopping, when a training-stop condition is satisfied, training the load forecasting model.

[00107] Optionally, the above-mentioned training-stop condition may include: a value of a loss function of the load forecasting model being less than a first threshold value; or, a number of training times of the load forecasting model being greater than a second threshold value.

[00108] The above-mentioned value of the loss function is configured to represent a difference degree between the forecasted load data and an actual load data, and the value of the loss function is obtained based on the forecasted load data within the past time period and the actual load data within the past time period.

[00109] The threshold value may also be called critical value, which refers to the lowest or highest value that an effect can generated. In the present disclosure, the first threshold value refers to the minimum value of the loss function of the load forecasting model when the training-stop condition is satisfied; the second threshold value refers to the maximum value of the number of training times of the load forecasting model when the training-stop condition is satisfied.

[00110] The above-mentioned first threshold value and second threshold value may be set according to actual experience, which is not limited in the embodiments of the present disclosure.

[00111] In addition, when the value of the loss function is greater than or equal to the first threshold value, the above-mentioned various parameters of the load forecasting model are adjusted, and the training of the load forecasting model is continued until the value of the loss function is less than the first threshold value.

[00112] In the embodiments of the present disclosure, the load forecasting is performed by the load forecasting model. The load forecasting model is trained by using the forecasted load data within the past time period and the past condition data corresponding to the past forecasted load data, thereby improving the accuracy of the load forecasting result and also improving the efficiency of the load forecasting.

[00113] In step 302, the energy storage cloud platform sends the forecasted demand of the i ^th charging unit to the controller.

[00114] The content of this step is the same as or similar to that of step 202 in the embodiment of FIG. 2, which will not be repeated herein. [00115] In step 303, the monitoring meter sends an actual demand of (i-1) charging units to the controller.

[00116] The content of this step is the same as or similar to that of step 203 in the embodiment of FIG. 2, which will not be repeated here.

[00117] In step 304, the controller determines a maximum forecasted demand of the i ^th charging unit based on the forecasted demand of the i ^th charging unit.

[00118] The content of this step is the same as or similar to that of step 204 in the embodiment of FIG. 2, and which not be repeated here.

[00119] In step 305, the controller determines a maximum actual demand of the (i-1) charging units based on the actual demand of the (i-1) charging units. [00120] The content of this step is the same as or similar to that of step 205 in the embodiment of FIG. 2, which will not be repeated here.

[00121] In step 306, the controller determines that a demand control for the i ^th charging unit needs to be performed if the maximum forecasted demand of the i ^th charging unit is greater than the maximum actual demand of the (i-1) charging units.

[00122] The content of this step is the same as or similar to that of step 206 in the embodiment of FIG. 2, which will not be repeated here.

[00123] In step 307, the controller controls the energy storage device to discharge after determining that the demand control for the i ^th charging unit needs to be performed. [00124] A target demand of the i ^th charging unit may be determined first, and then the energy storage device is controlled to discharge based on the target demand of the i ^th charging unit.

[00125] Optionally, determining the target demand of the i ^th charging unit may include the following sub-steps: [00126] 1. acquiring, by the controller, a maximum discharging power of the energy storage device;

[00127] 2. obtaining, by the controller, a first demand difference by calculating a difference between the maximum forecasted demand of the i ^th charging unit and the maximum discharging power; and [00128] 3. determining, by the controller, a larger one of the first demand difference and the maximum actual demand of the (i-1) charging units as the target demand of the i ^th charging unit.

[00129] The maximum discharging power of the energy storage device is a maximum value of the electrical power that the energy storage device can output; wherein the first demand difference is configured to represent a forecasted minimum value of the maximum demand of the i ^th charging unit when the energy storage device discharges; the target demand of the i ^th charging unit refers to a demand threshold of the i ^th charging unit; the target demand is a ideally set demand upper limit, and the purpose of the demand control is to control that the real-time demand of the i ^th charging unit does not exceed the set target demand. [00130] The above-mentioned target demand of the i ^th charging unit may be calculated by the following formula:

[00131] MDtarget=max(MDforecast - Pstorage, MD),

[00132] wherein MDtarget represents the target demand of the i ^th charging unit, MDforecast represents the maximum forecasted demand of the i ^th charging unit, Pstorage represents the maximum discharging power of the energy storage device, and MD represents the maximum actual demand of the (i-1) charging units.

[00133] A determined energy storage device has a determined maximum discharging power. For the demand control for the i ^th charging unit, the target demand thereof needs to be greater than or equal to the maximum actual demand of the (i-1) charging units. However, if the maximum forecasted demand of the i ^th charging unit is so large that when the energy storage device is expected to output electrical energy at the maximum power, the demand (i.e. the first demand difference) that needs to be acquired from the distribution network is still greater than the maximum actual demand of the (i-1) charging units. In this case, it is not appropriate to set the target demand as the maximum actual demand of the (i-1) charging units, because the maximum actual demand of the charging cycle is greater than or equal to the first demand difference. Therefore, when the first demand difference is greater than the maximum actual demand of the (i-1) charging units, the first demand difference is determined as the target demand.

[00134] Optionally, controlling the energy storage device to discharge includes controlling, by the controller, a discharging power of the energy storage device as a target power.

[00135] The target power is configured to control the actual demand of the i ^th charging unit to be less than or equal to the target demand of the i ^th charging unit; the actual demand of the i ^th charging unit refers to the demand generated in real-time within the i ^th charging unit. [00136] Optionally, the minimum value of the above-mentioned target power is the minimum discharging power of the energy storage device.

[00137] In some possible embodiments, further determining the minimum discharging power of the energy storage device may include the following sub -steps:

[00138] 1. acquiring a real-time electricity consumption power;

[00139] 2. obtaining, when the real-time electricity consumption power is greater than the target demand, a second demand difference is by making a difference between the real-time electricity consumption power and the target demand of the i ^th charging unit;

[00140] 3. determining, when the second demand difference is less than or equal to the maximum discharging power of the energy storage device, the second demand difference as a minimum discharging power of the energy storage device; and

[00141] 4. repeating the above steps to adjust the minimum discharging power of the energy storage device in real-time.

[00142] The real-time electricity consumption power may be instantaneous power, so the time to acquire the real-time electricity consumption power is much shorter than the demand cycle. Proper demand control in time can prevent the real-time demand of the i ^th charging unit from exceeding the target demand of the i ^th charging unit as much as possible.

[00143] For the above sub-step 3, when the second demand difference is greater than the maximum discharging power of the energy storage device, the energy storage device may discharge at its maximum discharging power.

[00144] In some specific embodiments, the discharging power of the energy storage device may be greater than the minimum discharging power of the energy storage device when the remaining electricity of the energy storage device is larger and/or it is within the peak electricity price period; the discharging power of the energy storage device may be equal to the minimum discharging power of the energy storage device when the remaining electricity of the energy storage device is less and/or it is within the valley electricity price period. [00145] In step 308, the controller sets the target demand of the i ^th charging unit equal to the actual demand of the i ^th charging unit, if the target power reaches the maximum discharging power and the actual demand of the i ^th charging unit is greater than the target demand of the i ^th charging unit.

[00146] When the real-time demand of the i ^th charging unit is greater than the target demand of the i ^th charging unit, maintaining the original target demand of the i ^th charging unit cannot reduce the total electricity fee as much as possible, and thus the target demand of the i ^th charging unit needs to be updated, the maximum demand of the charging cycle that has been generated is set as a current target demand, and the demand control is continued to be performed based on the current target demand.

[00147] In step 309, the controller determines whether the i ^th charging unit ends; and the controller determines a maximum actual demand of i charging units if the i ^th charging unit ends.

[00148] The i charging units are first i charging units of the charging cycle.

[00149] Optionally, the way to determine whether the i ^th charging unit ends may be counting by a timer or counting by a clock. Exemplarily, taking one charging unit as one day as an example, when the timer counts for 24 hours, it means that the i ^th charging unit ends; or, when reaching 24 o'clock of the i ^th charging unit, it means that the i ^th charging unit ends. [00150] If the i ^th charging unit does not end, the above steps are continued to perform the demand control; if the i ^th charging unit ends, the energy storage device may stop operating. [00151] Determining the maximum actual demand of the i charging units may include the following sub-steps:

[00152] 1. determining, by the controller, a maximum actual demand of the i ^th charging unit; [00153] 2. determining, by the controller, whether the maximum actual demand of the i ^th charging unit is greater than the maximum actual demand of the (i-1) charging units; and [00154] 3. setting, by the controller, the maximum actual demand of the i charging units equal to the maximum actual demand of the i ^th charging unit if the maximum actual demand of the i ^th charging unit is greater than the maximum actual demand of the (i-1) charging units .

[00155] Determining the maximum actual demand of the i charging units may determine the current maximum demand of the charging cycle, such that the current maximum demand of the charging cycle serves as a reference for performing the demand control for the (i+l) ^th charging unit.

[00156] For the above-mentioned embodiments, in some optional embodiments, as shown in FIG. 4, which is an exemplary graph illustrating a relationship among real-time demand, target demand, and discharging power of an energy storage device. In the time period 410, the energy storage device does not need to perform a demand control, and may discharge at any power without causing the actual demand to exceed the current target demand; in the time period 420, the energy storage device is controlled to discharge, such that the implemented power is equal to the current target demand; at the end of the time period 420, even though the energy storage device outputs electricity at the maximum discharging power, the actual demand cannot be kept below the current target demand line. Therefore, in the time period 430, the target demand line is adjusted immediately, such that the adjusted target demand is equal to the maximum demand that has been generated so far, and the discharging power of the energy storage device is then controlled based on the adjusted target demand to continue the demand control; and in the time period 440, the energy storage device may discharge at any power, without calculating the minimum discharging power of the energy storage device.

[00157] In summary, in the embodiments of the present disclosure, when the actual demand of the i ^th charging unit is greater than the target demand of the i ^th charging unit, the target demand is adjusted in time to set the target demand of the i ^th charging unit equal to the actual demand of the i ^th charging unit. According to the current actual situation as much as possible, the actual demand is lower than the maximum actual demand that has been generated in the past as much as possible, which optimizes the demand control method and increases the flexibility of the demand control.

[00158] The load forecasting is performed by the load forecasting model. The load forecasting model is trained by using the forecasted load data within the past time period and the past condition data corresponding to the past forecasted load data, thereby improving the accuracy of the load forecasting result and also improving the efficiency of the load forecasting.

[00159] Referring to FIG. 5, which exemplarily illustrates a flowchart of a demand control method. As shown in FIG. 5, in this embodiment, it is mainly illustrated by taking the demand control method applicable to the demand control system introduced above as an example. The demand control method may include the following steps.

[00160] In step 501, whether an i ^th charging unit is a 1 ^st charging unit of a charging cycle is determined; a maximum actual demand of (i-1) charging units is set equal to zero if the i ^th charging unit is the 1 ^st charging unit of the charging cycle.

[00161] i indicates a specific charging unit of the charging cycle; the maximum actual demand of the (i-1) charging units may represent as Maximum Demand (MD).

[00162] In step 502, the maximum actual demand of the (i-1) charging units is compared with a maximum forecasted demand of the i ^th charging unit, and a demand control for the i ^th charging unit needs to be performed if the maximum actual demand of the (i-1) charging units is less than the maximum forecasted demand of the i ^th charging unit.

[00163] The maximum forecasted demand of the i ^th charging unit may represent as Maximum Demand forecastrecast (MDforecast).

[00164] Optionally, if the maximum actual demand of the (i-1) charging units is greater than or equal to the maximum forecasted demand of the i ^th charging unit, the demand control for the i ^th charging unit does not need to be performed.

[00165] In step 503, a target demand of the i ^th charging unit is determined.

[00166] If the demand control for the i ^th charging unit needs to be performed, a first demand difference is obtained by calculating a difference between the maximum forecasted demand of the i ^th charging unit and a maximum discharging power; and a larger one of the first demand difference and the maximum actual demand of the (i-1) charging units is determined as the target demand of the i ^th charging unit.

[00167] The target demand of the i ^th charging unit may represent as Maximum Demand target (MDtarget). [00168] In step 504, the demand control is performed by controlling an energy storage device.

[00169] In step 505, if the maximum actual demand of the i ^th charging unit is still greater than the target demand of the i ^th charging unit when the energy storage device discharges at a maximum power, the target demand of the i ^th charging unit is set equal to the maximum actual demand of the i ^th charging unit.

[00170] An actual demand of the i ^th charging unit may represent as Maximum Demand new (MDnew), and the maximum discharging power of the energy storage device may represent as Pstorage.

[00171] In step 506, if the actual demand of the i ^th charging unit is less than the target demand of the i ^th charging unit, whether the i ^th charging unit ends is determined.

[00172] In step 507, if the i ^th charging unit ends, a maximum actual demand of i charging units is updated.

[00173] Optionally, whether a maximum value of the actual demand of the i ^th charging unit is greater than the maximum actual demand of the (i-1) charging units is determined. If the maximum value of the actual demand of the i ^th charging unit is greater than the maximum actual demand of the (i-1) charging units, the maximum actual demand of the i charging units is set equal to the maximum value of the actual demand of the i ^th charging unit; and if the maximum value of the actual demand of the i ^th charging unit is less than or equal to the maximum actual demand of the (i-1) charging units, the maximum actual demand of the i charging units remains unchanged.

[00174] After updating the maximum actual demand of the i charging units, the steps end. [00175] Another demand control system is further provided by an embodiments of the present disclosure. As shown in FIG. 1, the demand control system includes a controller, an energy storage device, a monitoring meter, and an energy storage cloud platform.

[00176] The energy storage cloud platform is configured to acquire a forecasted demand of an i ^th charging unit of a charging cycle through past load data by calling a load forecasting model for load forecasting; and send the forecasted demand of the i ^th charging unit to the controller, wherein the charging cycle comprises n charging units, n being a positive integer, and i being a positive integer less than or equal to n.

[00177] The monitoring meter is configured to send an actual demand of (i-1) charging units to the controller, wherein the (i-1) charging units are first (i-1) charging units of the charging cycle.

[00178] The controller is configured to determine a maximum forecasted demand of the i ^th charging unit based on the forecasted demand of the i ^th charging unit; determine a maximum actual demand of the (i-1) charging units based on the actual demand of the (i-1) charging units; determine that a demand control for the i ^th charging unit needs to be performed if the maximum forecasted demand of the i ^th charging unit is greater than the maximum actual demand of the (i-1) charging units; and control the energy storage device to discharge after determining that the demand control for the i ^th charging unit needs to be performed.

[00179] In an exemplary embodiment, the load forecasting model is obtained by:

[00180] acquiring at least one set of training samples, wherein each set of training samples includes forecasted load data within a past time period and past condition data corresponding to the past forecasted load data, and the past condition data being configured to represent estimated environmental characteristics within the past time period;

[00181] training the load forecasting model by using the at least one set of training samples; and

[00182] obtaining a trained load forecasting model by stopping, when a training-stop condition is satisfied, training the load forecasting model.

[00183] In an exemplary embodiment, the training-stop condition includes: a value of a loss function of the load forecasting model being less than a first threshold value, wherein the value of the loss function is obtained based on the forecasted load data within the past time period and an actual load data within the past time period; or, a number of training times of the load forecasting model being greater than a second threshold value.

[00184] In an exemplary embodiment, the controller is configured to control a discharging power of the energy storage device as a target power.

[00185] The target power is configured to control an actual demand of the i ^th charging unit to be less than or equal to a target demand of the i ^th charging unit, wherein the actual demand of the i ^th charging unit refers to a demand generated in real-time within the i ^th charging unit, and the target demand of the i ^th charging unit refers to a demand threshold of the i ^th charging unit.

[00186] In an exemplary embodiment, the controller is further configured to set the target demand of the i ^th charging unit equal to the actual demand of the i ^th charging unit, if the target power reaches the maximum discharging power and the actual demand of the i ^th charging unit is greater than the target demand of the i ^th charging unit.

[00187] In an exemplary embodiment, the controller is further configured to determine whether the i ^th charging unit ends; and determine a maximum actual demand of i charging units if the i ^th charging unit ends, wherein the i charging units are first i charging units of the charging cycle.

[00188] In an exemplary embodiment, the controller is configured to determine a maximum actual demand of the i ^th charging unit; determine whether the maximum actual demand of the i ^th charging unit is greater than the maximum actual demand of the (i-1) charging units; and set the maximum actual demand of the i charging units equal to the maximum actual demand of the i ^th charging unit if the maximum actual demand of the i ^th charging unit is greater than the maximum actual demand of the (i-1) charging units.

[00189] FIG. 6 is a structural block diagram of a controller according to an embodiment of the present disclosure. With reference to FIG. 6, specifically:

[00190] a computer device 600 includes a central processing unit (CPU) 601, a system memory 604 including a random access memory (RAM) 602 and a read-only memory (ROM) 603, and a system bus 605 that connects the system memory 604 and the central processing unit 601. The computer device 600 further includes a basic input/output (I/O) system 606 that facilitates transmission of information between various components within a computer, and a mass storage device 607 for storing an operating system 613, an application 614, and other program modules 615.

[00191] The basic input/output system 606 includes a display 608 for displaying information and an input device 609 such as a mouse or keyboard for user input of information. The display 608 and the input device 609 are both connected to the central processing unit 601 via an input/output controller 610 connected to the system bus 605. The basic input/output system 606 may also include the input/output controller 610 for receiving and processing input from a plurality of other devices, such as a keyboard, mouse, or electronic stylus. Similarly, the input/output controller 610 also provides output to a display screen, printer, or other types of output device.

[00192] The mass storage device 607 is connected to the central processing unit 601 by a mass storage controller (not shown) connected to the system bus 605. The mass storage device 607 and its associated computer-readable medium provide non-volatile storage for the computer device 600. That is, the mass storage device 607 may include a computer-readable medium (not shown) such as a hard disk or a compact disc read-only memory (CD-ROM) drive.

[00193] Without loss of generality, the computer-readable medium may include a computer storage medium and a communication medium. The computer storage medium includes volatile and nonvolatile, removable and non-removable mediums implemented by any method or technology for storage of information such as computer-readable instructions, data structures, program modules or other data. The computer storage medium includes an RAM, an ROM, an erasable programmable read only memory (EPROM), an electrically erasable programmable read only memory (EEPROM), a flash memory or other solid state storage technologies, a CD-ROM, a digital video disc (DVD) or other optical storage, a tape cassette, a magnetic tape, a magnetic disk storage or other magnetic storage devices. Of course, those skilled in the art may know that the computer storage medium is not limited to the above. The system memory 604 and the mass storage device 607 described above may be collectively referred to as memories.

[00194] According to various embodiments of the present disclosure, the computer device 600 may also be operated by being connected via a network such as the Internet to a remote network computer. That is, the computer device 600 may be connected to a network 612 by a network interface unit 611 connected to the system bus 605, or that is, the computer device 600 may be connected to other types of networks or remote computer systems (not shown) by using the network interface unit 611. [00195] In an exemplary embodiment, a computer-readable storage medium is further provided, wherein the computer-readable storage memory stores a computer program, wherein the computer program is executed by a processor to perform the method steps on a controller side of the demand control method described above.

[00196] It should be understood that the numbers of the steps described herein only exemplarily illustrate a possible order of execution of the steps. In some other embodiments, the above steps may also be executed out of the order of the numbers, and for example, two steps with different numbers are executed simultaneously, or two steps with different numbers are executed in the reverse order of the drawings, which is not limited in the embodiments of the present disclosure. [00197] The foregoing descriptions are merely exemplary embodiments of the present disclosure, and are not intended to limit the present disclosure. Within the spirit and principles of the present disclosure, any modifications, equivalent substitutions, improvements, and the like, should be included within the protection scope of the present disclosure.