Adjustable load control method and device, computer equipment, medium and product

Abstract

The invention relates to an adjustable load control method and device, computer equipment, a medium and a product. The method comprises the steps of obtaining a peak regulation plan curve of each load virtual unit based on load control information of each load virtual unit; for each load virtual unit, determining a pre-control target value of the load virtual unit based on the peak regulation plan curve, the pre-control time parameter and the baseline fluctuation parameter of the load virtual unit; the pre-control time parameter is related to transmission delay information and load information, and the baseline fluctuation parameter is obtained by performing baseline prediction based on the pre-control time parameter; and generating a pre-control instruction based on the pre-control target value of the load virtual unit, and sending the pre-control instruction to a load aggregator corresponding to the load virtual unit so as to instruct the load aggregator to control the plurality of associated adjustable loads based on the pre-control instruction. According to the method, the control time delay of the adjustable load and the fluctuation change of the baseline are comprehensively considered, and the control efficiency and the resource utilization rate of the adjustable load can be improved.

Description

Technical Field

[0001] This application relates to the field of smart grid technology, and in particular to an adjustable load control method, apparatus, computer equipment, medium and product. Background Technology

[0002] Driven by the energy transition, adjustable loads such as electric vehicle charging stations and distributed energy storage have become important flexibility resources for the power grid. With the development of smart grids, how to efficiently utilize adjustable loads and improve the utilization rate of power grid resources is currently an important research direction.

[0003] Compared with traditional thermal and hydropower units, these resources are characterized by their large quantity, dispersed distribution, small individual capacity, and varying response characteristics, making them impossible to be directly and centrally controlled by the automatic generation control (AGC) system at the main station like conventional units. To address this, a control scheme has emerged that integrates resources through load aggregators, enabling the main station AGC to indirectly and efficiently control each adjustable load through the load aggregator.

[0004] However, traditional control methods for adjustable loads still suffer from a series of problems, such as low control efficiency and low resource utilization. Summary of the Invention

[0005] Therefore, it is necessary to provide an adjustable load control method, apparatus, computer equipment, computer-readable storage medium, and computer program product that can improve the control efficiency of adjustable loads and improve the resource utilization of adjustable loads, in order to address the above-mentioned technical problems.

[0006] In a first aspect, this application provides an adjustable load control method applied to a master station. The master station includes a load control area, within which at least one load virtual machine group is configured. Each load virtual machine group corresponds to a load aggregator, and multiple adjustable loads are associated with each load aggregator. The method includes:

[0007] Based on the load control information of each load virtual machine group, the peak shaving plan curve of each load virtual machine group is obtained; the load control information of the load virtual machine group is determined based on the load control information of the corresponding load aggregator.

[0008] For each load virtual machine group, the pre-control target value of the load virtual machine group is determined based on the peak shaving plan curve, pre-control time parameter, and baseline fluctuation parameter. The pre-control time parameter is related to transmission delay information and load information, and the baseline fluctuation parameter is obtained by baseline prediction based on the pre-control time parameter.

[0009] Pre-control instructions are generated based on the pre-control target value of the load virtual machine group, and then sent to the load aggregator corresponding to the load virtual machine group to instruct the load aggregator to control the multiple adjustable loads associated with the pre-control instructions.

[0010] In one embodiment, the pre-control target value of the load virtual machine group is determined based on the load virtual machine group's peak shaving plan curve, pre-control time parameters, and baseline fluctuation parameters, including:

[0011] Based on transmission delay information and / or load information, determine the pre-control time parameters corresponding to the load virtual machine group;

[0012] Based on the pre-controlled time parameters, historical baseline data, and the preset baseline prediction model, the baseline fluctuation parameters corresponding to the load virtual machine group are determined.

[0013] Based on the peak shaving plan curve, pre-control time parameters, and baseline fluctuation parameters of the load virtual machine group, the initial target value of the load virtual machine group is determined.

[0014] The initial target value is corrected to obtain the pre-control target value.

[0015] In one embodiment, the transmission delay information includes the first average load control delay of the master station, the second average load control delay of the load aggregator, the third average load control delay of the adjustable load, and transmission path data. The load information includes at least one of load resource type, adjustment capacity, and historical delay data. Based on the transmission delay information and / or load information, the pre-control time parameters corresponding to the load virtual machine group are determined, including:

[0016] The first delay time is determined based on the first average load control delay, the second average load control delay, and the third average load control delay;

[0017] The second delay time is determined based on transmission path data, load information, and a preset delay prediction model.

[0018] The smaller of the first delay time and the second delay time is determined as the pre-control time parameter corresponding to the load virtual machine group.

[0019] In one embodiment, the initial target value of the load virtual machine group is determined based on the peak shaving plan curve, pre-control time parameters, and baseline fluctuation parameters, including:

[0020] Based on the first peak shaving plan value of the current pre-control period and the second peak shaving plan value of the next pre-control period in the peak shaving plan curve of the load virtual machine group, the initial peak shaving plan value at the current moment is determined; the current pre-control period is determined based on the current period and the pre-control time parameter, and the next pre-control period is determined based on the next period and the pre-control time parameter;

[0021] Based on the initial peak shaving plan value and baseline fluctuation parameters, the initial target value of the load virtual machine group is determined.

[0022] In one embodiment, the initial target value is corrected to obtain the pre-controlled target value, including:

[0023] Obtain the previous pre-control target deviation and the current pre-control target deviation from the previous pre-control command; the current pre-control target deviation is the deviation obtained after simulating the initial control command generated based on the initial target value.

[0024] If the sum of the deviation of the previous pre-control target and the deviation of the current pre-control target is less than or equal to the preset deviation threshold, then the initial target value will be used as the pre-control target value.

[0025] If the sum of the previous pre-control target deviation and the current pre-control target deviation is less than the maximum adjustment deviation of the load aggregator corresponding to the load virtual machine group, then the sum of the initial target value, the previous pre-control target deviation, and the current pre-control target deviation will be used as the pre-control target value.

[0026] If the sum of the deviation of the previous pre-control target and the deviation of the current pre-control target is greater than or equal to the maximum adjustment deviation, then the sum of the initial target value and the maximum adjustment deviation shall be used as the pre-control target value.

[0027] In one embodiment, obtaining the current pre-control target deviation includes:

[0028] Generate initial control commands based on the initial target values;

[0029] The initial control command is input into the preset control response model to obtain the current predicted output.

[0030] Based on the deviation between the current predicted output and the initial target value, the deviation of the current pre-control target is determined.

[0031] Secondly, this application also provides an adjustable load control device applied to a master station. The master station includes a load control area, within which at least one load virtual machine group is configured. Each load virtual machine group corresponds to a load aggregator, and multiple adjustable loads are associated with each load aggregator. The device includes:

[0032] The acquisition module is used to acquire the peak shaving plan curve of each load virtual machine group based on the load control information of each load virtual machine group; the load control information of the load virtual machine group is determined based on the load control information of the corresponding load aggregator.

[0033] The determination module is used to determine the pre-control target value of each load virtual machine group based on the load virtual machine group's peak shaving plan curve, pre-control time parameter, and baseline fluctuation parameter. The pre-control time parameter is related to transmission delay information and load information, and the baseline fluctuation parameter is obtained by baseline prediction based on the pre-control time parameter.

[0034] The control module is used to generate pre-control instructions based on the pre-control target value of the load virtual machine group, and send the pre-control instructions to the load aggregator corresponding to the load virtual machine group, so as to instruct the load aggregator to control the multiple adjustable loads associated with the pre-control instructions.

[0035] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the adjustable load control method in the first aspect described above.

[0036] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the adjustable load control method in the first aspect described above.

[0037] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the adjustable load control method described in the first aspect above.

[0038] The aforementioned adjustable load control method, apparatus, computer equipment, storage medium, and computer program product are applied to a master station. The master station includes a load control area, within which at least one load virtual machine group is configured. Each load virtual machine group corresponds to a load aggregator, and multiple adjustable loads are associated with each load aggregator. When controlling the adjustable loads, the master station obtains the peak shaving plan curve of each load virtual machine group based on its load control information. The load control information of each load virtual machine group is determined based on the load control information of the corresponding load aggregator. For each load virtual machine group, the pre-control target value is determined based on its peak shaving plan curve, pre-control time parameter, and baseline fluctuation parameter. The pre-control time parameter is related to transmission delay information and load information, and the baseline fluctuation parameter is obtained by baseline prediction based on the pre-control time parameter. A pre-control command is generated based on the pre-control target value of the load virtual machine group, and the pre-control command is sent to the load aggregator corresponding to the load virtual machine group to instruct the load aggregator to control the multiple adjustable loads associated with it based on the pre-control command. In other words, the adjustable load control method proposed in this application, under the two-level control architecture of master station-aggregator, comprehensively considers the control delay and baseline fluctuation of adjustable load. For different load virtual machine groups, the corresponding pre-control time parameters and baseline fluctuation parameters are first determined. Then, based on the peak shaving plan curve, the pre-control target value of the load virtual machine group is determined according to the pre-control time parameters and baseline fluctuation parameters. This not only avoids control errors caused by control delay, but also further improves the control accuracy of adjustable load through baseline prediction correction. Thus, it comprehensively improves the control efficiency of the master station for a large number of adjustable loads and improves the resource utilization rate of adjustable loads. Attached Figure Description

[0039] To more clearly illustrate the technical solutions in the embodiments or related technologies of this application, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0040] Figure 1 This is an application environment diagram of an adjustable load control method in one embodiment;

[0041] Figure 2 This is a flowchart illustrating an adjustable load control method in one embodiment;

[0042] Figure 3 A flowchart illustrating an adjustable load control method in another embodiment;

[0043] Figure 4This is a structural block diagram of an adjustable load control device in one embodiment;

[0044] Figure 5 This is a structural block diagram of an adjustable load control device in another embodiment;

[0045] Figure 6 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0046] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0047] Driven by the energy transition, adjustable loads such as electric vehicle charging stations and distributed energy storage have become important flexibility resources for the power grid. However, compared with traditional thermal and hydropower units, these resources are characterized by their large number, dispersed distribution, small individual capacity, and diverse response characteristics, making them unsuitable for direct and centralized control by the master station's AGC (Automatic Generator Control) like conventional units. Existing AGC systems are mainly designed for conventional units, and their control models, command issuance cycles, and strategies are ill-suited to the aggregation and control requirements of adjustable loads.

[0048] Currently, although the industry has proposed the concept of resource integration through load aggregators, several key technical challenges remain to be addressed. These include: establishing an effective control model at a primary substation; designing control strategies that balance the response delay characteristics of adjustable loads with the real-time requirements of the power grid; coordinating the joint operation of adjustable loads and conventional generating units; and ensuring the safety and reliability of control commands. The lack of a mature and complete closed-loop control strategy severely restricts the ability of adjustable loads to participate in large-scale, high-quality real-time power grid regulation.

[0049] Existing technologies typically employ the traditional AGC direct control mode for conventional generator sets. In this mode, the master station AGC directly issues precise active power setpoints to individual generator sets, which then track based on their own adjustment capabilities. Applying existing technologies directly to adjustable loads presents several key technical challenges: the master station cannot directly manage and control massive, heterogeneous adjustable loads at the aggregator level; an efficient two-tiered coordinated control architecture between the master station and aggregators is lacking, resulting in ineffective control command issuance and execution. Furthermore, the instruction cycle and control logic of conventional AGC do not account for the significant control delays caused by the further decomposition of adjustable loads at the aggregator level, nor for the potential cumulative deviations after aggregating numerous terminals, leading to poor control performance and weak tracking ability of the planned curve. In summary, existing control methods for adjustable loads still suffer from a series of problems, including low control efficiency and low resource utilization.

[0050] Based on this, the embodiments of this application provide an adjustable load control method. Based on a two-level AGC control architecture of master station-aggregator, by introducing "time advance" and feedback correction, the method effectively compensates for system control delay, improves tracking accuracy, and thus improves the control efficiency of adjustable load and the resource utilization rate of adjustable load.

[0051] The adjustable load control method provided in this application embodiment can be applied to, for example... Figure 1 The application environment is shown. The master station-aggregator two-level AGC control architecture includes a master control area and a load control area. The master control area is used for conventional regional control of the interconnected power grid, while the adjustable load control area is used for indirect control of various adjustable loads within the provincial dispatch center. Multiple adjustable load virtual machine groups are established in the adjustable load control area. In the master station AGC system, each adjustable load virtual machine group is an equivalent control model established for the resources aggregated by one or more load aggregators. It possesses attributes similar to conventional generating units, such as installed capacity, actual output, and adjustment range, allowing the master station to control a large adjustable load cluster like a traditional generator. Each load aggregator can correspond to one or more load virtual machine groups, and multiple adjustable loads, such as distributed energy storage, electric vehicles, and electric heating equipment, are associated with each load aggregator. The load aggregator accumulates and aggregates load control information such as actual output, installed capacity, controllable signals, and adjustment range of each adjustable load to form the overall load control information of the adjustable load aggregator, including actual output, installed capacity, controllable signals, and adjustment range. This information is then sent to the provincial dispatch AGC as the load control parameters for the corresponding adjustable load virtual machine group of the provincial dispatch AGC.

[0052] The aforementioned master station-aggregator two-tier AGC control architecture is a hierarchical control system for adjustable loads. The master station (provincial dispatch center) is responsible for grid-level optimization calculations and issues overall control objectives to aggregators; aggregators are responsible for further decomposing instructions and directly controlling the massive adjustable loads they aggregate. This architecture serves as a core bridge connecting the macro-dispatch of the power grid with the micro-resources on the user side.

[0053] The above-mentioned two-level control architecture of master station and aggregator, and the establishment of "adjustable load control area" and "virtual machine group" model on the master station side, that is, the construction of adjustable load virtual machine group, can realize unified modeling and intensive control of massive distributed and heterogeneous adjustable loads, and perform indirect and efficient management of massive distributed resources. This avoids the problem of excessive communication and computing pressure and system scalability caused by the master station system directly facing terminal resources.

[0054] In one exemplary embodiment, such as Figure 2 As shown, an adjustable load control method is provided, which can be applied to... Figure 1 Taking the main station AGC as an example, the explanation includes the following steps 201 to 203. Among them:

[0055] Step 201: Based on the load control information of each load virtual machine group, obtain the peak shaving plan curve of each load virtual machine group.

[0056] In this embodiment, the load control information of a load virtual machine group is determined based on the load control information of the corresponding load aggregator. When there is a one-to-one correspondence between load virtual machine groups and load aggregators, the master station AGC can use the load control information of the load aggregator as the load control information for the corresponding load virtual machine group. When one load aggregator corresponds to multiple load virtual machine groups, the master station AGC can determine the load control information for each load virtual machine group based on the load control information of the load aggregator. For example, the load control information of the load aggregator can be proportionally divided to serve as the load control information for each load virtual machine group. Of course, other allocation methods can also be used, such as allocating load control information to each load virtual machine group based on its configuration information. This application does not specifically limit this method.

[0057] For example, when the master station AGC obtains the load control information of each load virtual machine group, it can send the load control information of each load virtual machine group to the market trading system. Based on the control information of each conventional unit and the load control information of each load virtual machine group, the market trading system organizes bidding, clearing, and settlement of various resources according to the current market electricity demand (such as peak-shaving demand), and generates a market clearing result. This market clearing result can include the peak-shaving plan curve of each load virtual machine group. For example, the peak-shaving plan curve of the load virtual machine group can be a curve representing multiple time nodes formed at fixed time intervals, such as 288 points, and the curve value corresponding to each time node can be the load adjustment amount.

[0058] Step 202: For each load virtual machine group, determine the pre-control target value of the load virtual machine group based on the load virtual machine group's peak shaving plan curve, pre-control time parameter, and baseline fluctuation parameter.

[0059] The pre-control time parameter is related to transmission delay information and load information, while the baseline fluctuation parameter is obtained by baseline prediction based on the pre-control time parameter. It should be noted that the pre-control time parameter and / or baseline fluctuation parameter can be different for different load virtual machine groups. For each load virtual machine group, the corresponding pre-control time parameter can be determined in advance based on transmission delay information and / or load information. This pre-control time parameter is used to characterize the control delay of the master station AGC for adjustable loads. It should be noted that the pre-control time parameter corresponding to the load virtual machine group can be a fixed value or a dynamically changing value. For example, the pre-control time parameter can be recalculated before each instruction is issued, or it can be recalculated at a preset time interval (e.g., 1 day), and the same pre-control time parameter can be used within the preset time interval. The master station AGC can set different pre-control time parameters for adjustable load virtual machine groups of different types, control paths, and adjustment capabilities, and can also set different pre-control time parameters for the same load virtual machine group at different times.

[0060] Similarly, for the baseline fluctuation parameter, the baseline can be predicted based on the pre-control time parameter and historical baseline data before the instruction is issued, and the baseline fluctuation parameter can be further determined. The baseline fluctuation parameter is used to characterize the fluctuation difference between the baseline after the pre-control time parameter and the current baseline. The historical baseline data can be the baseline data of the load virtual machine group in the historical time period before the current time, or it can include the baseline data of other load virtual machine groups in the historical time period before the current time. By referring to the historical baseline change pattern of other load virtual machine groups, the baseline prediction model can be corrected to improve the prediction accuracy of the baseline prediction model.

[0061] For example, for each load virtual machine group, after obtaining the peak shaving plan curve, pre-control time parameters, and baseline fluctuation parameters corresponding to that load virtual machine group, the pre-control target value of that load virtual machine group at the current moment can be determined based on the peak shaving plan curve, pre-control time parameters, and baseline fluctuation parameters of the load virtual machine group. In the master station-aggregator two-level AGC control architecture, the master station issues control commands to the adjustable load through the aggregator. Compared to conventional thermal power units, the adjustable load has more intermediate links when participating in the master station AGC closed-loop control, and each link has a certain control delay. After the load aggregator receives the control command issued by the master station, it needs to redistribute and issue control commands again, which also takes a certain amount of time. Therefore, the delay between the master station AGC issuing control commands and the load control execution device performing actual control is relatively long. Based on this, this embodiment of the application pre-obtains the corresponding pre-control time parameters for each adjustable virtual machine group as a pre-control time lead, so that the control commands issued at the current moment can match the actual control requirements of the adjustable load at the execution moment, improving the control accuracy of the adjustable load.

[0062] In addition, in this embodiment of the application, in order to further improve the tracking capability of the planning curve, the pre-control target of the load virtual machine group is also modified in combination with the baseline fluctuation of the load virtual machine group. That is, in this example, the control target value in the peak shaving planning curve is modified in combination with the pre-control time parameter and the baseline fluctuation parameter to obtain the pre-control target value of the load virtual machine group at the current time.

[0063] Step 203: Generate a pre-control instruction based on the pre-control target value of the load virtual machine group, and send the pre-control instruction to the load aggregator corresponding to the load virtual machine group to instruct the load aggregator to control the multiple adjustable loads associated with it based on the pre-control instruction.

[0064] For example, after determining the pre-control target value corresponding to each load virtual machine group, a pre-control instruction for that load virtual machine group can be generated based on the pre-control target value. Then, the master station AGC sends the pre-control instruction to the load aggregator corresponding to that load virtual machine group, so that the load aggregator can allocate pre-control targets to each associated adjustable load based on the pre-control target value in the pre-control instruction, and generate sub-pre-control instructions for each adjustable load, which are then sent to each adjustable load, ultimately realizing the control of the adjustable load.

[0065] For example, for each load aggregator, it can assign pre-control targets to controllable adjustable loads based on the load status and control information of each associated adjustable load. The control periods corresponding to different adjustable loads may be different. During uncontrollable periods, the adjustable load control object automatically switches to local control mode, and the master station AGC does not issue any control commands to the adjustable load control object. That is to say, during uncontrollable periods, the adjustable load does not participate in the coordinated control of the master station AGC; while during controllable periods, the master station AGC can implement control on the adjustable load control object according to different control requirements.

[0066] Furthermore, it should be noted that after the market trading system organizes bidding, clearing, and settlement of various resources based on current market electricity demand (such as peak-shaving demand), the market clearing results generated may also include the peak-shaving plan curves of each conventional unit. In other words, for conventional thermal power units, if a conventional thermal power unit participates in and wins the bid in the peak-shaving service market, the master station AGC will require the unit to strictly track the peak-shaving clearing results, i.e., the day-ahead and intraday power generation plans, based on the unit's operating status, and will put the unit into plan tracking mode. In this mode, the unit's control target is strictly the unit's day-ahead and intraday plans. When the deviation between the unit's actual output and planned output exceeds the unit's control dead zone, the unit's current planned output is immediately issued as the unit's target output, achieving strict tracking of the unit's plan. For conventional units, since the master station AGC can directly control the conventional units, the unit's adjusted output can basically track changes in the peak-shaving ancillary service market environment clearing results, the unit's response delay is relatively short, and the tracking effect between the unit's actual output and the clearing results is relatively good.

[0067] The above technical solution is suitable for the spot market model. In this model, adjustable load control objects such as electric vehicles and distributed energy storage participate in the whole-network peak shaving market, and the optimized control target for the period is derived from the peak shaving market optimization. Based on the peak shaving market clearing results, AGC obtains the control target for the next 5 minutes. After certain safety checks (such as adjustment range checks, SOC checks, and grid frequency safety checks), it automatically generates minute-level control target instructions, i.e., the aforementioned pre-control instructions. These pre-control instructions can be forwarded to the load aggregator through telemetry or messaging, and the load aggregator then decomposes the overall target and implements control on each sub-control object, i.e., the associated adjustable load.

[0068] In the aforementioned adjustable load control method, the master station AGC obtains the peak shaving plan curve of each load virtual machine group based on the load control information of each load virtual machine group; and for each load virtual machine group, it determines the pre-control target value of the load virtual machine group based on the peak shaving plan curve, pre-control time parameter, and baseline fluctuation parameter; then, it generates a pre-control instruction based on the pre-control target value of the load virtual machine group and sends the pre-control instruction to the load aggregator corresponding to the load virtual machine group to instruct the load aggregator to control the associated multiple adjustable loads based on the pre-control instruction; wherein, the load control information of the load virtual machine group is determined based on the load control information of the corresponding load aggregator; the pre-control time parameter is related to the transmission delay information and load information, and the baseline fluctuation parameter is obtained by baseline prediction based on the pre-control time parameter. In other words, the adjustable load control method proposed in this application, under the two-level control architecture of master station-aggregator, comprehensively considers the control delay and baseline fluctuation of adjustable load. For different load virtual machine groups, the corresponding pre-control time parameters and baseline fluctuation parameters are first determined. Then, based on the peak shaving plan curve, the pre-control target value of the load virtual machine group is determined according to the pre-control time parameters and baseline fluctuation parameters. This not only avoids control errors caused by control delay, but also further improves the control accuracy of adjustable load through baseline prediction correction. Thus, it comprehensively improves the control efficiency of the master station for a large number of adjustable loads and improves the resource utilization rate of adjustable loads.

[0069] In one exemplary embodiment, such as Figure 3 As shown, step 202 above may include steps 301 to 304. Wherein:

[0070] Step 301: Determine the pre-control time parameters corresponding to the load virtual machine group based on transmission delay information and / or load information.

[0071] The transmission delay information may include the first average load control delay of the master station, the second average load control delay of the load aggregator, the third average load control delay of the adjustable load, and transmission path data. The load information may include at least one of the following: load resource type, adjustment capacity, and historical delay data. The first average load control delay of the master station may be the average of the master station load control delays of a preset number (e.g., 5) of the same type of historical days at the same time. The second average load control delay of the load aggregator may be the average of the load control delays of the load aggregator of a preset number (e.g., 5) of the same type of historical days at the same time. The third average load control delay of the adjustable load may be the average of the self-adjustment delays of the load control of a preset number (e.g., 5) of the same type of historical days at the same time.

[0072] For example, when determining the pre-control time parameter corresponding to a load virtual machine group, the pre-control time parameter can be determined based on the transmission delay information and / or load information corresponding to the load virtual machine group. That is, the pre-control time parameter can be determined based on the transmission delay information, the load information, or both. In other words, the pre-control time parameter can be determined based on only one reference item, or it can be determined by combining multiple reference items. Determining the pre-control time parameter based on multiple reference items can further improve the prediction accuracy of the pre-control time parameter.

[0073] For example, when determining the pre-control time parameter based on the load information corresponding to the load virtual machine group, the master station AGC can determine a first delay time based on the first average load control delay, the second average load control delay, and the third average load control delay, and use this first delay time as the pre-control time parameter. This can be expressed as: ,in, This is the first delay time. For the first average load control delay, For the second average load control delay, This is the third average load control delay.

[0074] For example, when determining the pre-control time parameter based on the load information corresponding to the load virtual machine group, the master station AGC can determine a second delay time based on transmission path data, load information, and a preset delay prediction model, and use this second delay time as the pre-control time parameter. For instance, the delay prediction model can be trained using load information data such as historical delay data of individual resource control delay, transmission path data, load resource type, and current adjustable capacity, to estimate the possible delay time in the next time period.

[0075] For example, when determining the pre-control time parameter based on the transmission delay information and load information corresponding to the load virtual machine group, the master station AGC can determine the pre-control time parameter based on the aforementioned first delay time and the aforementioned second delay time. For example, the average delay time of the first delay time and the second delay time can be used as the pre-control time parameter, or the smaller of the first delay time and the second delay time can be determined as the pre-control time parameter corresponding to the load virtual machine group. ,in, This is the second delay time.

[0076] Step 302: Based on the pre-controlled time parameters, historical baseline data, and preset baseline prediction model, determine the baseline fluctuation parameters corresponding to the load virtual machine group.

[0077] The historical baseline data may include multiple historical baseline data within the historical event segment preceding the current time for the corresponding load virtual machine group, or multiple historical baseline data within the historical event segment preceding the current time for other load virtual machine groups, etc.

[0078] For example, when performing baseline prediction, historical baseline data and a preset baseline prediction model are usually combined. However, in this application, a time lead is set, i.e., a pre-controlled time parameter. Therefore, when performing baseline prediction, the pre-controlled time parameter and historical baseline data are input together into the preset baseline prediction model to perform baseline prediction, thereby outputting the predicted baseline value corresponding to the pre-controlled time parameter after the current time, such as predicting the baseline value 2 minutes later.

[0079] Next, the master station AGC can obtain the baseline fluctuation parameters, such as the baseline fluctuation amount, of the load virtual machine group based on the difference between the actual operating value (such as actual output) of the load virtual machine group at the current moment and the predicted baseline value corresponding to the pre-control time parameter.

[0080] It should be noted that the aforementioned preset baseline prediction model can be any type of neural network model, obtained through iterative training based on sample baseline data and sample pre-control time parameters. This application will not elaborate further on this.

[0081] Step 303: Determine the initial target value of the load virtual machine group based on the peak shaving plan curve, pre-control time parameters, and baseline fluctuation parameters of the load virtual machine group.

[0082] For example, for the peak shaving plan curve of the load virtual machine group formed based on the peak shaving market clearing results, after the master station AGC obtains the peak shaving plan curve, it can interpolate the peak shaving plan curve at fixed time intervals (e.g., once every minute) to obtain the load control target value at the current time point (i.e., the current moment). It should be noted that when the peak shaving plan curve is the adjustment amount for each time period, the adjustment amount corresponding to the time point is obtained by interpolating the peak shaving plan curve. On this basis, the baseline value at the current moment needs to be superimposed to obtain the load control target value at the current moment.

[0083] Furthermore, by considering the pre-control time parameters and baseline fluctuation parameters of the adjustable load virtual machine group, the initial target value of the load virtual machine group can be formed.

[0084] The above interpolation calculation of the peak shaving plan curve for the load virtual machine group can be expressed as follows:

[0085] (1)

[0086] in, Let t be the target adjustment amount for the load virtual machine group at time t; This refers to the peak-shaving plan value for the T+1 time period; This represents the planned peak-shaving value for time period T. The current output of the i-th load virtual machine group; The maximum number of commands per instance for the load virtual machine group; This represents the upper limit of the peak-shaving range for the i-th load virtual machine group; This is the lower limit of the peak-shaving range for the i-th load virtual machine group.

[0087] Based on the superposition of the above formula (1) and the baseline, the load control target value of the load virtual machine group at the current moment can be obtained. Considering the pre-control time parameter and the baseline fluctuation parameter, the above formula (1) can be adjusted as follows:

[0088] (2)

[0089] in, This refers to the pre-control adjustment amount of the load virtual machine group at the pre-control time corresponding to the pre-control time parameter. This refers to the peak-shaving plan value for the pre-control time parameter T+1 period; The peak-shaving plan value for the pre-controlled time period T; The time deviation is represented by T, which is the control period, such as 5 minutes. This is the baseline fluctuation parameter.

[0090] Next, based on the pre-control baseline predicted by the preset baseline prediction model, the pre-control adjustment amount obtained by the above formula (2) is superimposed to obtain the initial target value of the load virtual machine group.

[0091] Step 304: Correct the initial target value to obtain the pre-controlled target value.

[0092] For example, when correcting the initial target value, the control deviation corresponding to the previous pre-control command can be referenced, i.e., the previous pre-control target deviation. The previous pre-control target deviation can be determined based on the deviation between the actual output of the load virtual machine group collected after the previous pre-control command was issued and the previous pre-control target value.

[0093] For example, the deviation of the previous pre-control target can be obtained first. If the deviation of the previous pre-control target is less than or equal to a preset deviation threshold, the initial target value is used as the pre-control target value. If the deviation of the previous pre-control target is greater than the preset deviation threshold and less than the maximum adjustment deviation of the load aggregator corresponding to the load virtual machine group, the sum of the initial target value and the previous pre-control target deviation is used as the pre-control target value. If the deviation of the previous pre-control target is greater than or equal to the maximum adjustment deviation, the sum of the initial target value and the maximum adjustment deviation is used as the pre-control target value.

[0094] In other implementations, before issuing commands, the master station AGC can perform online simulation of the initial control commands and output the current predicted output. Based on the current predicted output and the initial target value, the current pre-control target deviation can be determined. Then, the master station AGC can correct the initial target value based on the previous pre-control target deviation and the current pre-control target deviation to obtain the pre-control target value.

[0095] For example, if the sum of the previous pre-control target deviation and the current pre-control target deviation is less than or equal to a preset deviation threshold, the initial target value can be used as the pre-control target value; if the sum of the previous pre-control target deviation and the current pre-control target deviation is less than the maximum adjustment deviation of the load aggregator corresponding to the load virtual machine group, the sum of the initial target value, the previous pre-control target deviation, and the current pre-control target deviation can be used as the pre-control target value; if the sum of the previous pre-control target deviation and the current pre-control target deviation is greater than or equal to the maximum adjustment deviation, the sum of the initial target value and the maximum adjustment deviation can be used as the pre-control target value.

[0096] For example, the method for obtaining the current pre-control target deviation may include: generating an initial control command based on the initial target value, inputting the initial control command into a preset control response model to obtain the current predicted output; and determining the current pre-control target deviation based on the deviation between the current predicted output and the initial target value. For instance, the deviation between the current predicted output and the initial target value can be used as the current pre-control target deviation, or a preset deviation amount can be added to the deviation between the current predicted output and the initial target value to obtain the current pre-control target deviation, wherein the preset deviation amount is used to compensate for the prediction deviation of the preset control response model. The master station AGC, through simulation of the local model, dynamically corrects the initial target value based on the simulated deviation value, which can fully consider the deviation during the execution process, predict the state after the delay in advance, and further improve the accuracy of pre-control.

[0097] In this embodiment, the master station AGC first determines the pre-control time parameters and baseline fluctuation parameters corresponding to each load virtual machine group. Then, for each load virtual machine group, based on the peak shaving plan curve, pre-control time parameters, and baseline fluctuation parameters, it determines the initial target value of the load virtual machine group and corrects the initial target value to obtain the pre-control target value. This method employs a pre-control peak shaving mode based on "time lead" and simulated feedback correction. By acquiring planning instructions in advance, simulating monitoring and tracking deviations, and continuously correcting control targets online, it effectively compensates for and precisely controls the inherent control delay of the adjustable load system. This ensures that the output of the adjustable load cluster can track the market-clearing peak shaving plan curve on time and accurately, avoiding problems such as poor tracking effect and substandard adjustment performance caused by delays and accumulated deviations. This improves the master station's control efficiency for a large number of adjustable loads and enhances resource utilization for adjustable loads.

[0098] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0099] Based on the same inventive concept, this application also provides an adjustable load control device for implementing the adjustable load control method described above. The solution provided by this device is similar to the solution described in the above method; therefore, the specific limitations in one or more adjustable load control device embodiments provided below can be found in the limitations of the adjustable load control method described above, and will not be repeated here.

[0100] In one exemplary embodiment, such as Figure 4 As shown, an adjustable load control device is provided, applied to a master station. The master station includes a load control area, within which at least one load virtual machine group is configured. Each load virtual machine group corresponds to a load aggregator, and multiple adjustable loads are associated with each load aggregator. The device includes: an acquisition module 401, a determination module 402, and a control module 403, wherein:

[0101] The acquisition module 401 is used to acquire the peak shaving plan curve of each load virtual machine group based on the load control information of each load virtual machine group; the load control information of the load virtual machine group is determined based on the load control information of the corresponding load aggregator.

[0102] The determination module 402 is used to determine the pre-control target value of each load virtual machine group based on the peak shaving plan curve, pre-control time parameter and baseline fluctuation parameter of the load virtual machine group; the pre-control time parameter is related to the transmission delay information and load information, and the baseline fluctuation parameter is obtained by baseline prediction based on the pre-control time parameter.

[0103] The control module 403 is used to generate a pre-control instruction based on the pre-control target value of the load virtual machine group, and send the pre-control instruction to the load aggregator corresponding to the load virtual machine group, so as to instruct the load aggregator to control the multiple adjustable loads associated with the pre-control instruction.

[0104] In one embodiment, the determining module 402 includes:

[0105] The first determining submodule 4021 is used to determine the pre-control time parameters corresponding to the load virtual machine group based on transmission delay information and / or load information;

[0106] The second determining submodule 4022 is used to determine the baseline fluctuation parameters corresponding to the load virtual machine group based on the pre-controlled time parameters, historical baseline data and the preset baseline prediction model;

[0107] The third determination submodule 4023 is used to determine the initial target value of the load virtual machine group based on the peak shaving plan curve, pre-control time parameter and baseline fluctuation parameter of the load virtual machine group;

[0108] The correction submodule 4024 is used to correct the initial target value to obtain the pre-controlled target value.

[0109] In one embodiment, the transmission delay information includes the first average load control delay of the master station, the second average load control delay of the load aggregator, the third average load control delay of the adjustable load, and transmission path data. The load information includes at least one of load resource type, adjustment capacity, and historical delay data. The first determining submodule 4021 includes:

[0110] The first determining unit is used to determine the first delay time based on the first average load control delay, the second average load control delay, and the third average load control delay;

[0111] The second determining unit is used to determine the second delay time based on transmission path data, load information and a preset delay prediction model;

[0112] The third determining unit is used to determine the smaller of the first delay time and the second delay time as the pre-control time parameter corresponding to the load virtual machine group.

[0113] In one embodiment, the third determining submodule 4023 includes:

[0114] The fourth determining unit is used to determine the initial peak shaving plan value at the current moment based on the first peak shaving plan value of the current pre-control period and the second peak shaving plan value of the next pre-control period in the peak shaving plan curve of the load virtual machine group; the current pre-control period is determined based on the current period and the pre-control time parameter, and the next pre-control period is determined based on the next period and the pre-control time parameter;

[0115] The fifth determining unit is used to determine the initial target value of the load virtual machine group based on the initial peak shaving plan value and the baseline fluctuation parameter.

[0116] In one embodiment, the correction submodule 4024 includes:

[0117] The acquisition unit is used to acquire the previous pre-control target deviation and the current pre-control target deviation of the previous pre-control command; the current pre-control target deviation is the deviation obtained after simulating the initial control command generated based on the initial target value.

[0118] The sixth determining unit is used to take the initial target value as the pre-control target value when the sum of the previous pre-control target deviation and the current pre-control target deviation is less than or equal to a preset deviation threshold; when the sum of the previous pre-control target deviation and the current pre-control target deviation is less than the maximum adjustment deviation of the load aggregator corresponding to the load virtual machine group, take the sum of the initial target value, the previous pre-control target deviation, and the current pre-control target deviation as the pre-control target value; and when the sum of the previous pre-control target deviation and the current pre-control target deviation is greater than or equal to the maximum adjustment deviation, take the sum of the initial target value and the maximum adjustment deviation as the pre-control target value.

[0119] In one embodiment, the acquisition unit is specifically used to generate an initial control command based on an initial target value; input the initial control command into a preset control response model to obtain the current predicted output; and determine the current pre-control target deviation based on the deviation between the current predicted output and the initial target value.

[0120] Each module in the aforementioned adjustable load control device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the operations corresponding to each module.

[0121] In one exemplary embodiment, a computer device is provided, which may be a master station device, in which a master station AGC system may be deployed, and its internal structure diagram may be as follows. Figure 6 As shown, the computer device includes a processor, memory, input / output interfaces, a communication interface, a display unit, and an input device. The processor, memory, and input / output interfaces are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interfaces. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The input / output interfaces are used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, Near Field Communication (NFC), or other technologies. When the computer program is executed by the processor, it implements an adjustable load control method. The display unit is used to form a visually visible image and can be a display screen, a projection device, or a virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.

[0122] Those skilled in the art will understand that Figure 6 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0123] In one exemplary embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the adjustable load control method in any of the above embodiments.

[0124] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps of the adjustable load control method in any of the above embodiments.

[0125] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps of the adjustable load control method in any of the above embodiments.

[0126] It should be noted that the data involved in this application (including but not limited to data used for analysis, data stored, data displayed, etc.) are all information and data that have been fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0127] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0128] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0129] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. An adjustable load control method applied to a master station, the master station including a load control area, wherein at least one load virtual machine group is configured within the load control area, the load virtual machine group corresponds to a load aggregator, and multiple adjustable loads are associated under the load aggregator, characterized in that, The method includes: Based on the load control information of each load virtual machine group, the peak shaving plan curve of each load virtual machine group is obtained; the load control information of the load virtual machine group is determined based on the load control information of the corresponding load aggregator. For each of the aforementioned load virtual machine groups, the pre-control target value of the load virtual machine group is determined based on the peak shaving plan curve, pre-control time parameter, and baseline fluctuation parameter of the load virtual machine group; the pre-control time parameter is related to transmission delay information and load information, and the baseline fluctuation parameter is obtained by baseline prediction based on the pre-control time parameter; Based on the pre-control target value of the load virtual machine group, a pre-control instruction is generated and sent to the load aggregator corresponding to the load virtual machine group, so as to instruct the load aggregator to control the associated plurality of adjustable loads based on the pre-control instruction.

2. The method according to claim 1, characterized in that, The determination of the pre-control target value of the load virtual machine group based on the peak shaving plan curve, pre-control time parameters, and baseline fluctuation parameters includes: Based on the transmission delay information and / or the load information, determine the pre-control time parameters corresponding to the load virtual machine group; Based on the pre-controlled time parameters, historical baseline data, and preset baseline prediction model, the baseline fluctuation parameters corresponding to the load virtual machine group are determined; Based on the peak shaving plan curve of the load virtual machine group, the pre-control time parameter, and the baseline fluctuation parameter, the initial target value of the load virtual machine group is determined; The initial target value is corrected to obtain the pre-controlled target value.

3. The method according to claim 2, characterized in that, The transmission delay information includes the first average load control delay of the master station, the second average load control delay of the load aggregator, the third average load control delay of the adjustable load, and transmission path data. The load information includes at least one of the following: load resource type, adjustment capacity, and historical delay data. The step of determining the pre-control time parameter corresponding to the load virtual machine group based on the transmission delay information and / or the load information includes: The first delay time is determined based on the first average load control delay, the second average load control delay, and the third average load control delay; Based on the transmission path data, the load information, and the preset delay prediction model, a second delay time is determined. The smaller of the first delay time and the second delay time is determined as the pre-control time parameter corresponding to the load virtual machine group.

4. The method according to claim 2, characterized in that, The determination of the initial target value of the load virtual machine group based on the peak shaving plan curve of the load virtual machine group, the pre-control time parameter, and the baseline fluctuation parameter includes: Based on the first peak shaving plan value of the current pre-control period and the second peak shaving plan value of the next pre-control period in the peak shaving plan curve of the load virtual machine group, the initial peak shaving plan value at the current moment is determined; the current pre-control period is determined based on the current period and the pre-control time parameter, and the next pre-control period is determined based on the next period and the pre-control time parameter; Based on the initial peak shaving plan value and the baseline fluctuation parameter, the initial target value of the load virtual machine group is determined.

5. The method according to claim 2, characterized in that, The step of correcting the initial target value to obtain the pre-controlled target value includes: Obtain the previous pre-control target deviation and the current pre-control target deviation of the previous pre-control command; the current pre-control target deviation is the deviation obtained after simulating the initial control command generated based on the initial target value; If the sum of the previous pre-control target deviation and the current pre-control target deviation is less than or equal to a preset deviation threshold, then the initial target value is used as the pre-control target value; If the sum of the previous pre-control target deviation and the current pre-control target deviation is less than the maximum adjustment deviation of the load aggregator corresponding to the load virtual machine group, then the sum of the initial target value, the previous pre-control target deviation, and the current pre-control target deviation shall be used as the pre-control target value. If the sum of the previous pre-control target deviation and the current pre-control target deviation is greater than or equal to the maximum adjustment deviation, then the sum of the initial target value and the maximum adjustment deviation shall be used as the pre-control target value.

6. The method according to claim 5, characterized in that, The process of obtaining the current pre-control target deviation includes: The initial control command is generated based on the initial target value; The initial control command is input into a preset control response model to obtain the current predicted output. The deviation of the current pre-control target is determined based on the deviation between the current predicted output and the initial target value.

7. An adjustable load control device applied to a master station, the master station including a load control area, wherein at least one load virtual machine group is arranged in the load control area, the load virtual machine group corresponds to a load aggregator, and multiple adjustable loads are associated under the load aggregator, characterized in that, The device includes: The acquisition module is used to acquire the peak shaving plan curve of each load virtual machine group based on the load control information of each load virtual machine group; the load control information of the load virtual machine group is determined based on the load control information of the corresponding load aggregator. The determination module is used to determine the pre-control target value of each load virtual machine group based on the peak shaving plan curve, pre-control time parameter, and baseline fluctuation parameter of the load virtual machine group; the pre-control time parameter is related to transmission delay information and load information, and the baseline fluctuation parameter is obtained by baseline prediction based on the pre-control time parameter; The control module is configured to generate a pre-control instruction based on the pre-control target value of the load virtual machine group, and send the pre-control instruction to the load aggregator corresponding to the load virtual machine group, so as to instruct the load aggregator to control the associated plurality of adjustable loads based on the pre-control instruction.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images