A mutual aid scheduling control method based on a virtual power plant

Abstract

The present application relates to the technical field of supervisory control and data acquisition system, specifically to a mutual aid scheduling control method based on virtual power plant, comprising the following steps: collecting actual and target power calculation deviation, processing deviation level to generate label table, configuring timer to generate trigger label, acquiring state parameters to calculate resource boundary and adjustment direction, distributing adjustment power to generate instruction sequence, issuing instructions to collect execution deviation correction error, and updating boundary to generate correction configuration set. In the present application, the grid-connected point power deviation is collected to establish a level label, the closed-loop cycle balance response and load are dynamically matched, the high-delay data is filtered combined with time stamp verification, the strategy is ensured to be based on real-time effective information, communication congestion misjudgment is avoided, the adjustment boundary is calculated according to the resource state and the ordered instruction is generated, the safety of collaborative adjustment is guaranteed, the correction amount is calculated based on the execution feedback, the response error is compensated in a bidirectional closed loop, and the target power tracking accuracy and system robustness are improved.

Description

Technical Field

[0001] This invention relates to the field of monitoring, control and data acquisition systems, and in particular to a mutual dispatch control method based on a virtual power plant. Background Technology

[0002] The field of monitoring, control, and data acquisition systems primarily involves monitoring, control, and data acquisition technologies for power systems. Its core aspects include real-time monitoring of the operational status of substations, feeders, distributed power sources, energy storage devices, and controllable loads on the user side; collection of switch quantities, analog quantities, and alarm information; centralized management of telemetry, telesignaling, remote control, and remote adjustment information; and data aggregation and command issuance through communication links between the master station and field terminals. This technical field employs a control method consisting of a master station system, a communication network, and field acquisition terminals. It collects data such as voltage, current, power, frequency, state of charge, and equipment status through protocol communication, and generates control commands based on scheduling strategies to coordinate and control the power, output, and load at grid connection points. Among these, the traditional virtual power plant mutual assistance scheduling control method refers to a scheduling control method that, under the condition of aggregating multiple types of distributed resources in a virtual power plant, complementarily allocates adjustment quantities such as power generation output, energy storage charging and discharging, and load reduction or peak shifting based on the adjustability and operational constraints of each resource. This method addresses the mutual assistance power allocation and scheduling control among multiple resources within a virtual power plant. Traditional methods rely on clearly defined constraints such as the target power at the grid connection point, the available capacity of each resource, ramping constraints, upper and lower limits of state of charge, and adjustable load range. Combined with real-time collected power, voltage, current, and equipment status data, a time-segmented scheduling plan is formed. Power setpoints, charging and discharging commands, and load adjustment commands are then sent to distributed power inverters, energy storage converters, and load control terminals through a monitoring and control and data acquisition system to complete mutual assistance scheduling control.

[0003] Existing technologies rely on fixed time intervals for full data acquisition and command issuance during actual operation, lacking in-depth verification and filtering mechanisms for the timeliness of the data itself. When faced with signal transmission fluctuations or terminal reporting delays caused by complex network environments, the master station system often directly uses telemetry data with significant lag for scheduling calculations. This results in control strategies that are severely out of touch with real-time operating conditions, easily leading to contradictory adjustment directions or excessive adjustment amounts, thereby exacerbating power oscillations at grid connection points or even triggering safety limits. Traditional scheduling methods generally adopt an open-loop management mode after issuing power commands, focusing only on the delivery status of command data while ignoring the actual response quality of the controlled objects. They cannot perceive the execution deviations of distributed power inverters and energy storage converters caused by their own ramp rates, capacity saturation, or internal fault protection logic. When the field equipment fails to output power accurately according to the preset instructions, the system lacks real-time monitoring and dynamic compensation for the execution effect, which causes the scheduling error to accumulate and amplify continuously over a long period of time. This results in the failure of power tracking at the grid connection point over a long period of time, making it difficult to meet the strict assessment requirements of the power grid for high-precision regulation of virtual power plants. It also limits the system's ability to mitigate the randomness and volatility of distributed resources. Summary of the Invention

[0004] The purpose of this invention is to address the shortcomings of existing technologies by proposing a mutual dispatch control method based on a virtual power plant.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: a mutual assistance dispatch control method based on a virtual power plant, comprising the following steps: S1: Collect the actual power and target power of the grid connection point, obtain the frequency monitoring data and target power curve distribution information of the area to which the grid connection point belongs, calculate the active power deviation value, call the preset deviation level threshold set to process the active power deviation value, obtain the deviation level identifier, and obtain the area identifier based on the grid connection point number mapping, and establish a grid connection point deviation level label table by combining the deviation level identifier and the area identifier. S2: Configure the master station verification timer based on the grid connection point deviation level label table, calculate the telemetry data time difference, call the timeout judgment value to determine the telemetry data time difference, and generate a closed-loop verification trigger status flag. S3: Obtain the state of charge, maximum charging power, maximum discharging power, maximum adjustable capacity and inverter limiting state quantity according to the closed-loop verification trigger state flag, calculate the resource boundary and identify the adjustment direction, and generate a mutual assistance direction and resource boundary table. S4: Call the mutual assistance direction and resource boundary table to extract the resource adjustment boundary, allocate the energy storage adjustment power, load adjustment power and inverter power limit value, and integrate to generate a sequence of power setting instructions to be issued; S5: Issue instructions based on the issued power setting instruction sequence, collect the execution power and calculate the execution deviation value, call the deviation tolerance interval to compare the execution deviation value and calculate the scheduling error correction amount, collect energy storage mode and load switch status information, update the resource adjustment boundary and generate a scheduling callback correction configuration set.

[0006] As a further embodiment of the present invention, the grid connection point deviation level label table includes the grid connection point's region identifier, active power deviation value, and deviation level identification result; the closed-loop verification trigger status marker includes the closed-loop verification cycle parameter, telemetry data time difference comparison result, and scheduling logic start judgment value; the mutual assistance direction and resource boundary table includes the resource upward adjustment boundary, resource downward adjustment boundary, mutual assistance adjustment direction, and resource availability judgment status; the issued power setting command sequence includes the energy storage adjustment power amount, load adjustment power amount, inverter power limit value, and issuance sequence time information; the scheduling callback correction configuration set includes the scheduling error correction amount, energy storage mode status information, load switch status information, and current scheduling constraint items.

[0007] As a further aspect of the present invention, the set of deviation level thresholds includes at least three active power deviation interval thresholds, which are set in ascending order of absolute value, and each active power deviation interval threshold corresponds to a unique region identifier.

[0008] As a further aspect of the present invention, the step of obtaining S1 is as follows: S101: Collect the actual active power data of the grid connection point and the target power data of the grid connection point configured in the scheduling plan. Record the grid connection point number and the target power value and actual power value at the corresponding time point. Match the target power value and actual power value at the same time point. Obtain the active power deviation value by calculating the difference between the target power and the actual power. S102: Call the active power deviation value, combine it with the configured deviation level threshold set, perform interval matching identification, determine the threshold interval range that the deviation value of the grid connection point falls into at each time point, identify the corresponding level identifier, and introduce the power change of the grid connection point, the target power distribution status and the frequency fluctuation of the area, judge and classify the deviation value, and obtain the deviation level interval. S103: Call the deviation level range, and according to the regional identifier of each grid connection point, associate and integrate the grid connection point number, the region to which it belongs, and the identified level range, and arrange them in a structured manner according to the regional and time dimensions to establish a grid connection point deviation level label table.

[0009] As a further aspect of the present invention, the step of obtaining S2 is as follows: S201: Based on the grid connection point deviation level label table, extract the closed-loop verification cycle parameters corresponding to the grid connection point number, combine the region and time information, complete the archiving process of the cycle parameters, and write them into the main station verification timer to obtain the closed-loop verification cycle parameter set. S202: Call the closed-loop verification cycle parameter set, collect the telemetry sampling timestamp of the grid connection point and the current system clock time and calculate the time difference, compare the calculated time difference with the configured telemetry data timeout judgment value, combine the time difference change status of the grid connection point and the timing characteristics of the region to which it belongs, make a judgment and identification, and obtain the scheduling logic trigger judgment flag. S203: Based on the scheduling logic trigger judgment flag, filter the grid connection points with the flag value in the start state, combine the number, area identifier and current time information, record the logic state switching behavior, construct the corresponding scheduling response record, and generate a closed-loop verification trigger state flag.

[0010] As a further aspect of the present invention, the step of obtaining S3 is as follows: S301: Based on the closed-loop verification trigger state flag, collect the state of charge, maximum charging power and maximum discharging power reported by the energy storage equipment monitoring device, extract the maximum adjustable capacity reported by the controllable load device, record the inverter limit state quantity, summarize the resource parameters, and obtain the resource state parameter set. S302: Call the resource status parameter set, identify the available power boundary within the operating range based on the current limitation status of the resource equipment, perform upward and downward power range calculations in combination with the adjustment capabilities of the energy storage equipment and the load equipment, obtain the resource adjustment boundary value, determine the adjustment direction of all grid-connected points in the current cycle, and obtain the resource adjustment direction identifier. S303: Based on the resource adjustment direction identifier and the resource adjustment boundary value, aggregate the upward and downward adjustment directions and corresponding power range data in the current cycle according to the grid connection point number, structure and integrate the adjustment direction type and boundary value, and output them in association with the area label to generate a data structure mutual adjustment direction and resource boundary table.

[0011] As a further aspect of the present invention, the specific calculation formula for calculating the upward and downward power range based on the adjustment capabilities of the energy storage device and the load device is as follows: ; The calculation obtains the resource adjustment boundary value, determines the status of the adjustment direction of all grid-connected points in the current cycle, and obtains the resource adjustment direction identifier. in, Indicates the first Resource adjustment boundary value for each resource unit For the first The maximum up-adjustment capacity of each energy storage device For the first The maximum power limit for each energy storage device. For the first Maximum reduction capacity of each load device For the first Maximum power reduction limit for each load device For the first The charge and discharge sensitivity of an energy storage device For the first Additional performance parameters for each energy storage device.

[0012] As a further aspect of the present invention, the step of obtaining S4 is as follows: S401: Call the resource adjustment boundary data in the mutual assistance direction and resource boundary table, identify the time tag, resource type, adjustment capacity upper limit, and area number parameters associated with the resource boundary, filter energy storage resources, load resources and inverter resources under matching conditions as participating items, and filter according to the matching degree between the resource adjustment boundary and the adjustment demand of the region to obtain a list of resources with adjustment capabilities and generate an adjustable resource number sequence. S402: Based on the resources in the adjustable resource number sequence, obtain the power adjustment capability value, available time period, equipment status parameters, and execution frequency participation items within a specified time period, and perform resource adjustment power allocation calculation to obtain the allocation power setting value matrix. Combine the resource number and time tag to obtain the resource power allocation value sequence. S403: Based on the resource power allocation value sequence, combined with the corresponding resource type, control parameters, power limit boundary, and actual load response ratio data, the data is assembled into a data structure sequence with continuity and scheduling priority according to the time sorting principle, and written into the scheduling instruction framework to obtain the power scheduling setting instruction sequence.

[0013] As a further aspect of the present invention, the step of obtaining S5 is as follows: S501: Based on the issued power setting instruction sequence, complete the instruction issuance and record the instruction timestamp, monitor the timer operation status and collect the execution power information returned by the terminal when entering the verification period, call the instruction timestamp to perform time alignment processing on the execution power information, calculate the difference according to the correspondence between the execution power and the set power, and generate the execution deviation amount; S502: Based on the execution deviation amount, obtain the configured deviation tolerance range and perform the range comparison operation, collect the energy storage mode status information and load switch status information within the same time period, make a consistency judgment on the adjustment direction information and feedback status information, and make a correction calculation on the execution deviation amount based on the consistency judgment result to obtain the scheduling error correction amount. S503: Based on the aforementioned scheduling error correction amount, the current adjustment direction information is invoked and combined with the collected feedback status information to perform an adjustment boundary update operation. The corresponding scheduling constraint items are constructed according to the updated adjustment boundary, and the scheduling constraint items are aggregated and processed to establish a scheduling callback correction configuration set.

[0014] As a further aspect of the present invention, the resource boundary is obtained by constraining the state of charge, maximum charging power, maximum discharging power, maximum adjustable capacity, and inverter limiting state quantities, and the mutual assistance direction is determined by the positive or negative attribute of the active power deviation value.

[0015] Compared with the prior art, the advantages and positive effects of the present invention are as follows: In this invention, a level label is established by collecting the power deviation of the grid connection point, and the closed-loop cycle balance response and load are dynamically matched. High-latency data is filtered by combining timestamp verification to ensure that the strategy is based on real-time and effective information, avoid misjudgment due to communication congestion, calculate the adjustment boundary based on the resource status and generate ordered instructions to ensure the safety of coordinated adjustment, and generate correction amount based on the deviation calculation of execution feedback to construct a two-way closed-loop compensation response error, thereby improving the target power tracking accuracy and system robustness. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the steps of the present invention; Figure 2 This is a detailed schematic diagram of S1 of the present invention; Figure 3 This is a detailed schematic diagram of S2 of the present invention; Figure 4 This is a detailed schematic diagram of S3 of the present invention; Figure 5 This is a detailed schematic diagram of S4 of the present invention; Figure 6 This is a detailed schematic diagram of S5 of the present invention. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0018] In the description of this invention, it should be understood that the terms "length," "width," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientation or positional relationships, are based on the orientation or positional relationships shown in the accompanying drawings and are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Furthermore, in the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0019] Please see Figure 1 A mutual dispatch control method based on a virtual power plant includes the following steps: S1: Collect the actual power of the grid connection point received by the virtual power plant master station dispatch platform and the target power of the grid connection point configured in the dispatch plan, and obtain the frequency monitoring data and target power curve distribution information of the area to which the grid connection point belongs. Calculate the active power deviation value between the two, call the configured deviation level threshold set to identify the range of deviation value, and establish a grid connection point deviation level label table in combination with the grid connection point belonging area identifier. S2: Extract the corresponding closed-loop verification cycle parameters based on the grid connection point deviation level label table and write them into the main station verification timer. Collect the telemetry sampling timestamp and the current time and calculate the telemetry data time difference. Call the configured telemetry data timeout judgment value to compare and judge the time difference. Combine the comparison results to determine whether to start the scheduling logic and generate a closed-loop verification trigger status flag. S3: Based on the closed-loop verification trigger status flag, obtain the state of charge, maximum charging power, and maximum discharging power reported by the energy storage equipment monitoring device, extract the maximum adjustable capacity reported by the controllable load device, and determine the resource availability by combining the inverter limit status quantity, calculate the resource adjustment and reduction boundary, identify the adjustment direction corresponding to the current deviation trend, and generate a data structure mutual assistance direction and resource boundary table. S4: Call the mutual assistance direction and resource boundary table to extract the adjustment boundary of the resource, allocate the corresponding energy storage adjustment power, load adjustment power and inverter power limit value, integrate the three types of power values ​​and time information to generate a data sequence with a delivery order, write it into the scheduling instruction structure, and generate a power setting instruction sequence to be delivered. S5: Based on the sequence of power setting instructions, complete the instruction issuance and record the instruction time stamp. When the timer enters the verification period, collect the execution power data returned by the terminal, calculate the execution deviation value between the value and the set value, call the configured deviation tolerance range for comparison, and calculate the scheduling error correction amount. At the same time, collect the energy storage mode status information and load switch status information, compare the adjustment direction with the feedback status, update the adjustment boundary, construct the current scheduling constraint item, and generate the scheduling callback correction configuration set.

[0020] The grid connection point deviation level label table includes the grid connection point's region identifier, active power deviation value, and deviation level identification result; the closed-loop verification trigger status flag includes the closed-loop verification cycle parameter, telemetry data time difference comparison result, and dispatch logic start judgment value; the mutual assistance direction and resource boundary table includes the resource upward adjustment boundary, resource downward adjustment boundary, mutual assistance adjustment direction, and resource availability judgment status; the issued power setting command sequence includes the energy storage adjustment power amount, load adjustment power amount, inverter power limit value, and issuance sequence time information; the dispatch callback correction configuration set includes the dispatch error correction amount, energy storage mode status information, load switch status information, and current dispatch constraints.

[0021] Please see Figure 2 The steps to obtain S1 are as follows: S101: Collect the actual active power data of the grid connection point and the target power data of the grid connection point configured in the scheduling plan. Record the grid connection point number and the target power value and actual power value at the corresponding time point. Match the target power value and actual power value at the same time point. Obtain the active power deviation value by calculating the difference between the target power and the actual power. When collecting active power and target power data from grid-connected points, a unique index key is established on the main station side for the same grid-connected point number. The index key is generated by concatenating the grid-connected point number, metering channel number, and time stamp. The time stamp uses a UTC timestamp and is retained to a 1-second resolution. When reading the target power curve from the scheduling plan library, the data is first filtered by the grid-connected point number, and then linear interpolation is performed by the timestamp to ensure that the target power has a unique value at each sampling time. When reading the actual active power from the measurement side, data cleaning is performed first: invalid codes, out-of-bounds values, and sudden jump points are removed, and the original values ​​are retained for future reference. The out-of-bounds threshold is set at 1.20 times the rated grid-connected capacity of the grid-connected point. Sudden jump points are identified by the rule that the difference between two adjacent sampling points is greater than 0.15 times the rated capacity and the duration is less than two sampling periods. The cleaned data is deduplicated by grid-connected point number and timestamp. If there are multiple records for the same timestamp, only one record is retained in the order of "measurement quality indicator priority, lower communication delay priority, and earlier arrival time priority". After completion, target power and actual power are paired one-to-one under the same time tag. Time points where pairing fails are not included in subsequent level classification, and the missing measurement reason code is recorded. When calculating the deviation value after successful pairing, the sign convention of "actual power minus target power" is uniformly adopted, retaining the positive and negative signs. The unit of deviation value is uniformly kW, and the deviation value is truncated to avoid abnormal expansion. The truncation upper limit is 1.10 times the rated capacity of the grid connection point. For example: the rated capacity of grid connection point number 101 is 2000kW. At a certain moment, the target power is 1200kW, and the actual power after cleaning is 1340kW, then the deviation value is 140kW; at the next moment of the same grid connection point, the target power is 1200kW, and the actual power is 980kW, then the deviation value is -220kW; neither of these two deviation values ​​exceeds the truncation upper limit of 2200kW, and they are directly written into the deviation sequence. To ensure reproducibility, the sampling period is fixed at 5s, and the time alignment tolerance is set to 1s. If the difference between the target power timestamp and the actual power timestamp exceeds 1s, the pairing is re-matched according to the rule of "target power forward alignment, actual power remaining at the original time". If the rule is still not met, the pairing is judged as a failure and a missing test reason code is written.

[0022] S102: Call the active power deviation value, combine it with the configured deviation level threshold set, perform interval matching identification, determine the threshold interval range that the deviation value of the grid connection point falls into at each time point, identify the corresponding level identifier, and introduce the power change of the grid connection point, the target power distribution status and the frequency fluctuation of the area, judge the deviation value and classify the level, and obtain the deviation level interval. The construction of the deviation level threshold set adopts a combination of offline calibration and online fixation: In the offline stage, 30 consecutive days of historical data are selected, and the deviation rate is obtained by normalizing it according to the rated capacity of the grid connection point = deviation value / rated capacity. The quantiles of the absolute value of the deviation rate are counted and candidate thresholds are determined in combination with scheduling tolerance requirements. The threshold is then selected through field trial operation. In the online stage, the threshold set is fixed as 4-level intervals and written into the configuration file. The interval boundaries are represented by the absolute value of the deviation rate, with values ​​of [0, 0.02), [0.02, 0.05), [0.05, 0.10), and [0.10, +∞). This threshold setting is constrained by the strictness of frequency control, and the normal frequency deviation is maintained within a very small range. Therefore, a finer low deviation gradation is used for active power deviation to trigger correction actions in advance. When performing interval matching identification, the absolute value of the deviation value at each time point is first taken, then the deviation rate is calculated and the interval is matched to obtain level labels 1, 2, 3, and 4. Subsequently, power change, target power distribution status, and regional frequency fluctuation are introduced for further judgment and classification: 1) Power change is obtained by the first difference of the deviation value sequence to obtain the change rate, and the change rate threshold is set at 0.03 / 5s of the rated capacity. If the deviation rate is at level 2 but the change rate exceeds the threshold for 3 consecutive cycles, the level is upgraded by 1 level to avoid sudden upslope causing subsequent verification delays. 2) The target power distribution status is quantified by the "high load interval ratio" of the target power curve of the day. The high load interval is defined as the target power being greater than 0.80 of the rated capacity. If the ratio is greater than 0.40 and the deviation level is 3, the level is maintained at 3 without downgrading. If the ratio is less than 0.10 and the deviation level is 3, a downgrade to 2 is allowed to reduce false triggering during low load periods. 3) Regional frequency fluctuations are quantified by the frequency monitoring value Δf of the area where the grid connection point is located. Δf is the difference between the current frequency and the rated frequency of 50Hz, retaining the positive or negative sign. The high range of frequency fluctuation is defined as |Δf| ≥ 0.05Hz. If it is in the high range and the deviation level is 2 or 3, it will be uniformly upgraded by 1 level to enhance the response strength to frequency disturbances. Example: Grid connection point 101 has a rated capacity of 2000kW. The deviation value of a certain period is 140kW, so the deviation rate is 0.07, and the initial matching is level 3. If the current |Δf| of the area to which the grid connection point belongs is 0.06Hz, the level will be upgraded to 4. If the change rate in the same period does not exceed the threshold, the level will remain at 4 and written into the level range field.

[0023] S103: Call the deviation level range, and based on the area identifier of each grid connection point, associate and integrate the grid connection point number, the area to which it belongs, and the identified level range, and arrange them in a structured manner according to the area and time dimensions to establish a grid connection point deviation level label table. After completing the grade range, the seven items—grid connection point number, region identifier, time label, deviation value, deviation rate, grade, and upward / downward adjustment reason code—are linked and integrated. The data is then structured and arranged in ascending order of region identifier, time label, and grid connection point number. For multiple grid connection point records under the same time label in the same region, a region summary field is added, including the proportion of Grade 4 within the region, the mean deviation value, and the standard deviation of the deviation value. This summary field is used as the index condition for subsequent closed-loop cycle parameter extraction. To facilitate on-site reproduction, the table field types are strictly fixed: time labels are integer seconds, deviation values ​​are signed floating-point numbers with one decimal place, deviation rates are floating-point numbers with four decimal places, and grades are integers.

[0024] Table 1. Example table of target power and actual power at grid connection points; As shown in Table 1, after the deviation values ​​are matched one-to-one under the same time label, they are directly entered into the level classification and summarized in the regional dimension to form a data base that can be extracted and stably referenced in subsequent cycles.

[0025] Please see Figure 3 The steps to obtain S2 are as follows: S201: Based on the grid connection point deviation level label table, extract the closed-loop verification cycle parameters corresponding to the grid connection point number, combine the region and time information, complete the archiving process of the cycle parameters, and write them into the main station verification timer to obtain the closed-loop verification cycle parameter set; After reading the deviation level label table, when retrieving the closed-loop verification cycle parameters by grid connection point number, two types of cycles are first fixed in the parameter library for each grid connection point: the basic verification cycle Tb and the tightened verification cycle Ts; Tb is used for levels 1 and 2, and Ts is used for levels 3 and 4; the cycle unit is uniformly s and the value range is limited to [10, 300]. The cycle setting is determined through offline experiments: using 30 days of operation data as a sample, the scheduling process is replayed under the combinations of Tb=60s, 90s, 120s and Ts=10s, 15s, 20s, respectively, and the proportion of "events with deviation duration exceeding 60s" and the number of "invalid triggers" are statistically analyzed; when Tb=90s and Ts=15s, the event capture ratio reaches 96.4%, and the number of invalid triggers decreases by 28.7% compared to the combination of Tb=60s and Ts=10s. This set of parameters is then dynamically selected according to the grid connection point level and archived into the "grid connection point-time label-cycle" triplet. During extraction, the region and time information of the grid connection point are written into the archived record: the region field is used for subsequent staggered triggering by region, and the time field is used to align with the master station timer; when writing to the master station verification timer, the next trigger point is generated using the method of "next trigger time = current time tag + selected period", and the generated result is written back to the archived record to prevent duplicate triggering. Example: Grid connection point 101 has a level of 4 at time tag 1700000000, so Ts=15s is selected, and the next trigger point is 1700000015; Grid connection point 205 has a level of 1 at the same time tag, so Tb=90s is selected, and the next trigger point is 1700000090.

[0026] S202: Call the closed-loop verification cycle parameter set, collect the telemetry sampling timestamp of the grid connection point and the current system clock time and calculate the time difference, compare the calculated time difference with the configured telemetry data timeout judgment value, combine the time difference change status of the grid connection point and the timing characteristics of the region to make a judgment and obtain the scheduling logic trigger judgment flag; When collecting telemetry sampling timestamps and the current system clock time and calculating the time difference, the time source is uniformly the system clock after master station time synchronization, with an accuracy better than 1ms; the telemetry timestamp comes from the sampling time field in the acquisition frame; to avoid misjudgment caused by network jitter, the time difference sequence is first subjected to sliding median filtering, with a window length of 3 sampling periods; the time difference Δt obtained after filtering is used for comparison with the telemetry data timeout judgment value Ttimeout. Ttimeout is calibrated according to the characteristics of the communication link. When using the IEC60870-5-104 link, the link timeout related parameters are used as a reference and determined in combination with the on-site round-trip delay statistics, with a value range of [2, 10]s. During the trial operation, 5s is used as the default value, and it is increased to 8s in areas with poor link quality to suppress false alarms.

[0027] The comparison and judgment are not triggered by a single point timeout, but are performed by combining the time difference change state and regional time series characteristics: 1) The time difference change state is judged by the Δt increment of K consecutive periods, where K is fixed at 3; if Δt increases for 3 consecutive times and Δt≥Ttimeout, it is judged as "timeout trend is established".

[0028] 2) The regional time series characteristics use the 95th percentile of Δt at the grid-connected points within the region as the regional benchmark Δt95. If Δt ≥ Δt95 at a certain grid-connected point and the timeout trend is established, the trigger flag for that grid-connected point is set to start. If Δt ≥ Ttimeout but does not reach Δt95, the flag is set to warning and triggering is not initiated. Example: At a certain moment, Δt95 in region A is 6.2s and Ttimeout is 5s. After filtering, the Δt sequence of grid-connected point 101 is 4.6s, 5.4s, 6.1s, and 6.8s. The timeout trend is established and Δt = 6.8s ≥ Δt95, so the trigger flag is set to start. The Δt of grid-connected point 205 is 5.2s, 5.1s, 5.3s, and 5.4s, which does not form a continuous upward trend, so the trigger flag remains closed.

[0029] S203: Based on the scheduling logic trigger judgment flag, filter the grid connection points with the flag value in the start state, combine the number, area identifier and current time information, record the logical state switching behavior, construct the corresponding scheduling response record, and generate the closed-loop verification trigger state flag; When filtering startup states and constructing scheduling response records based on trigger judgment flags, seven items are written into the response record: grid connection point number, area identifier, current timestamp, trigger reason code, pre-trigger level, selection period, and Δt value. Simultaneously, logical state switching behavior is recorded, with the state field transitioning between "closed," "warning," and "started," and the transition conditions are consistent with the judgment above. Repeated startups of the same grid connection point within the same trigger window are suppressed, with the suppression window length being twice the selection period to avoid duplicate distributions caused by link fluctuations. When generating closed-loop verification trigger state flags, in addition to the flag value, two time boundaries are attached: "earliest time allowed to enter resource verification" and "latest time required to complete resource verification." The former is taken from the current timestamp, and the latter is taken from the current timestamp + 2 × selection period, used to constrain subsequent resource status acquisition and power allocation to be completed within a controllable time window.

[0030] Please see Figure 4 The steps to obtain S3 are as follows: S301: Based on the closed-loop verification trigger status flag, collect the state of charge, maximum charging power and maximum discharging power reported by the energy storage equipment monitoring device, extract the maximum adjustable capacity reported by the controllable load device, record the inverter limit status quantity, summarize the resource parameters, and obtain the resource status parameter set. When collecting resource status parameters based on the closed-loop verification trigger status mark, a one-time collection request is first initiated according to the grid connection point number list recorded in the mark. The collection channels include the energy storage equipment's state of charge, maximum charging power, maximum discharging power, the maximum adjustable capacity of controllable load equipment, and inverter limit status quantities. During collection, a collection time tag is attached to each type of data and aligned with the trigger time tag. If the difference between the returned time tag and the trigger time tag exceeds 2 seconds, the data is marked as expired and a re-collection is triggered. If the re-collection is still expired, the resource is marked as unavailable in the current cycle. The state of charge (SOC) is represented by the value reported by the battery management unit (BMU) and ranges from 0% to 100%. To improve lifespan and avoid extreme operating conditions, this embodiment limits the adjustable SOC window to [10, 90]%, and immediately performs window trimming after data acquisition: values ​​below 10% are included in the adjustment calculation at 10%, but the downward adjustment capability is set to 0; values ​​above 90% are included in the adjustment calculation at 90%, but the upward adjustment capability is set to 0. This window setting conforms to the engineering practice recommendation to avoid long-term extreme SOC values. The maximum charging power and maximum discharging power are jointly determined from the rated value of the energy storage converter and the current temperature limit: the rated power Prated and temperature T are collected, and the available rate factor α is mapped to [0.50, 1.00] according to the temperature limit curve. The maximum available power is then obtained by multiplying Prated by α. Example: A certain energy storage unit has a rated power of 500kW, and the current temperature is in the derating range with α=0.80. Therefore, the maximum charging power and maximum discharging power are both reported as 400kW for subsequent boundary calculations. The maximum adjustable capacity of the controllable load is obtained from the "upper limit of adjustable power" and "upper limit of adjustable power" reported by the load controller. If only one direction is reported, the other direction is treated as 0. A consistency check is first performed on the load data: if the sum of the upper and lower limits exceeds 1.10 times the rated power of the load, it is scaled proportionally by 1.10 times to prevent abnormal reporting. The inverter limit status variables are given by the grid-connected inverter control strategy. The status variable values ​​are limited to 0 or 1, where 0 indicates no power limitation and 1 indicates power limitation. After acquisition, the status variables undergo de-jitter processing: a status switch is confirmed only if two consecutive sampling periods are consistent, avoiding transient jitter from misleading boundary judgments.

[0031] S302: Call the resource status parameter set, identify the available power boundary within the operating range based on the current limitation status of the resource equipment, and calculate the upward and downward power range based on the adjustment capabilities of the energy storage equipment and the load equipment, using the following formula: ; The calculation obtains the resource adjustment boundary value, determines the status of the adjustment direction of all grid-connected points in the current cycle, and obtains the resource adjustment direction identifier. in, Indicates the first Resource adjustment boundary value for each resource unit (unit: power, kW). For the first Maximum up-adjustment capacity of an energy storage device (unit: power, kW). For the first Maximum power limit for each energy storage device (unit: power, kW). For the first Maximum down-regulation capacity of each load device (unit: power, kW). For the first Maximum power reduction limit for each load device (unit: power, kW). For the first The charge / discharge sensitivity of an energy storage device (unit: power / time, kW / s). For the first Additional performance parameters for each energy storage device (unit: power, kW); When calling the resource status parameter set to identify the available power boundary within the operating range, the up-adjustment and down-adjustment capabilities of each resource are first uniformly converted into two values: "available up-adjustment capacity" and "available down-adjustment capacity". The available up-adjustment capacity of energy storage is determined by the maximum discharge power and the upper limit constraint of SOC, and the available down-adjustment capacity is determined by the maximum charging power and the lower limit constraint of SOC. The available up-adjustment capacity of load is the upper limit of up-adjustment, and the available down-adjustment capacity is the upper limit of down-adjustment. If the inverter is in a limited-generation state, the up-adjustment capacity associated with the inverter is treated as 0 and "limited-generation suppression" is noted in the reason code. Subsequently, based on the upward and downward adjustment capabilities, the upward and downward adjustment power range calculations are performed, and the resource adjustment boundary value is calculated according to the original formula. The participating items involved are not directly set, but are obtained through collection and calculation: 1) C_i^+ takes the maximum upward adjustment capacity of the energy storage unit at the current SOC and temperature, which is equal to the maximum discharge power, and decays linearly when the SOC is close to the lower limit. The decay start point is set to SOC=15%, and decays to 0 when SOC=10%. Example: SOC=12%, maximum discharge power 400kW, then C_i^+ becomes 160kW after decay. 2) L_i^+ is the upward adjustment limit power of the energy storage unit after being limited by the inverter and the grid connection point power ramping constraint. The ramping constraint is converted according to the inverter's "power ramp rate" configuration. The ramp rate allowable range is configured as 0-100% / min. In this embodiment, 20% / min is used and converted to kW / s based on the rated power. Example: Rated 500kW, 20% / min is converted to 1.67kW / s. Within the verification window of 15s, the allowable change is about 25kW. Therefore, L_i+ is the smaller value between C_i+ and 25kW. 3) C_j- and L_j- come from the load's adjustable capacity and the downward adjustment limit power. The downward adjustment limit power is given by the minimum operating power of the equipment and process constraints. After collecting the minimum operating power, the "upper limit of downward adjustment = current power - minimum operating power" is calculated.

[0032] 4) F_i is calculated from the energy storage charge and discharge sensitivity. The changes in power and SOC at 6 consecutive sampling points are obtained by taking the median of the absolute value of "power change / time change", and the unit is kW / s. To ensure reproducibility, the sampling point interval is fixed at 5s and the total window length is 30s. 5) G_i is quantified from the additional performance parameters of energy storage. The three items, "health status score", "communication quality score" and "temperature margin score", are normalized to [0, 1] and weighted by (0.50, 0.30, 0.20) to obtain the comprehensive score η, which is then mapped to G_i by η×50kW. The weights are optimized through offline experiments: on 20 sets of field data, (0.40, 0.40, 0.20), (0.50, 0.30, 0.20) and (0.60, 0.20, 0.20) are used respectively to compare the correction success rate within 15s after triggering. The correction success rate of (0.50, 0.30, 0.20) reaches 92.0%, which is 6.5% higher than that of (0.40, 0.40, 0.20). Substituting the above parameter values ​​into the formula yields B_i, which is then used to adjust the direction determination: if the deviation value of the grid connection point is positive and the level is ≥3, the direction indicator is set to downward; if the deviation value is negative and the level is ≥3, the direction indicator is set to upward; if the level is ≤2, the direction indicator is set to inactive and B_i is used only for capability assessment. Example: The current deviation of grid connection point 101 is 140kW and its level is 4, with the direction marked as downward adjustment. Its corresponding energy storage unit, after SOC=12% and temperature derating, has C_i^+=160kW and L_i^+=25kW. On the load side, n=2 and the average value of |C_j-−L_j-| for both types of loads is 30kW, F_i=2.0kW / s and G_i=35kW. Substituting these values ​​into the formula, we get B_i as approximately 108kW. This result shows that, after considering the limiting and slope constraints, the effective adjustment boundary that this energy storage unit can provide in the current cycle is in the hundreds of kilowatts, which can cover part of the 140kW deviation of the grid connection point. It needs to be supplemented in conjunction with load resources in the subsequent allocation stage.

[0033] S303: Based on the resource adjustment direction identifier and resource adjustment boundary value, aggregate the upward and downward adjustment directions and corresponding power range data in the current cycle according to the grid connection point number, structure and integrate the adjustment direction type and boundary value, and output them in association with the area label to generate a data structure mutual adjustment direction and resource boundary table. When aggregating the upward and downward adjustment directions and power range data by grid connection point number, the direction identifier, Bi, available upward adjustment capacity, available downward adjustment capacity, and limiting reason code of each resource are written into the grid connection point aggregation record. When there are multiple resources under the same grid connection point, the upward and downward adjustment capabilities are grouped and summarized separately to obtain the "total upward adjustment boundary" and "total downward adjustment boundary" at the grid connection point level. Then, the data is associated with the area label to form a mutual assistance direction and resource boundary table, and the "resource to grid connection point mapping relationship" field is retained in the table to ensure that subsequent power allocation can be traced back to the specific equipment on a resource-by-resource basis.

[0034] Please see Figure 5 The steps to obtain S4 are as follows: S401: Call the resource adjustment boundary data in the mutual assistance direction and resource boundary table, identify the time tag, resource type, adjustment capacity upper limit, and area number parameters associated with the resource boundary, filter energy storage resources, load resources and inverter resources under matching conditions as participating items, and filter according to the matching degree between the resource adjustment boundary and the adjustment demand of the region to obtain a list of resources with adjustment capabilities and generate an adjustable resource number sequence. After reading the mutual assistance direction and resource boundary table, when identifying resources participating in regulation, the time tag associated with the resource boundary, resource type, upper limit of regulation capacity, and area number are used as filtering conditions and executed in a strict order: first, resources in the same area as the current triggering grid connection point are filtered by area number; then, resources within the window of "earliest time allowed to enter resource verification - latest time must complete resource verification" are filtered by time tag; then, resources associated with energy storage, load, and inverter are filtered by resource type respectively; subsequently, a "matching score" is calculated for each resource, and the top K scores are truncated from high to low to form an adjustable resource number sequence. The value of K ranges from [1, 50] and is determined by the area size. In this embodiment, K is taken as the total number of resources when the number of resources in the area does not exceed 30, and K=30 when the number of resources exceeds 30. The matching score adopts a score system of 0-100 and consists of 3 items: capacity coverage, execution stability, and communication reliability. Capacity coverage is mapped to the ratio of B_i to the absolute value of the grid connection point deviation. A score of 1.00 is awarded as 60 points, while scores below 1.00 are linearly scaled. Execution stability is mapped to the success rate of the resource completing commands within the last 24 hours. A success rate ≥98% is awarded as 25 points, and scores below 98% are deducted by 5 points for every 1% decrease. Communication reliability is mapped to the number of timeouts within the last hour. Zero timeouts are awarded as 15 points, and ≥3 timeouts are awarded as 0 points, resulting in the resource being removed from the list. Example: A certain energy storage resource B_i has approximately 108kW and an absolute value of the grid connection point deviation of 140kW. Its capacity coverage is approximately 46 points; its 24-hour success rate is 99%, execution stability is 25 points; and its 1-hour timeout count is 0, resulting in a communication reliability score of 15 points. The total score is 86 points, and the resource is included in the sequence.

[0035] S402: Based on the resources in the adjustable resource number sequence, obtain the power regulation capability value, available time period, equipment status parameters, and execution frequency parameters for a specified time period, and perform resource regulation power allocation calculation using the formula: ; The allocated power setting value matrix is ​​obtained by calculation, and combined with the resource number and time tag to obtain the resource power allocation value sequence; in, Representing resources In time slice The power setting value, , Representing resources The Middle Sub-resources in time slices The maximum and minimum adjustable power values, Representing resources The operational status rating of the first category of resources. Representing resources The Middle Sub-resources in time slices Available time within, Representing resources The Middle The cumulative execution frequency of the resource class Represents all the first in the system Average execution frequency of class resources Representing resources The included number The total number of sub-resources; ; When obtaining the power regulation capability value, available time period, equipment status parameters, and execution frequency participation items within a specified time period, the time period is based on the trigger window and divided into equal-length slices according to the scheduling granularity, with a fixed slice length of 5s. For each resource, the maximum and minimum adjustable power are calculated in each time slice: the maximum and minimum of energy storage are jointly constrained by the maximum discharge power, maximum charging power, SOC window, and ramp rate; the maximum and minimum of load are jointly constrained by the process-allowed upper and lower adjustment boundaries and the minimum duration, with the minimum duration given by the field configuration and limited to [10, 300]s; inverter resources only participate when there is no power throttling and reactive / active power linkage permission is available. If the power throttling status variable is 1, this type of resource will not be included in the allocation calculation in this cycle. The available time period T_avail is quantified according to the equipment status parameters: if the temperature of energy storage is in the derating range and the continuous derating time exceeds 600s, then T_avail is reduced by 0.70 within the trigger window; if the load is in the frequent switching protection period, then T_avail is set to 0 and removed. The operational status rating S_rate is generated from three items: "health status rating, maintenance alarm status, and recent anomaly count". The rating level ranges from 1 to 5 and is mapped to [0.20, 1.00], where level 5 is mapped to 1.00 and level 1 is mapped to 0.20. The rating rules are fixed: no alarms and 0 anomalies is level 5; general alarms or 1-2 anomalies are level 4; important alarms or 3-5 anomalies are level 3; severe alarms or 6-10 anomalies are level 2; and locked alarms or more than 10 anomalies are level 1. The execution frequency f_exec is obtained by accumulating the number of commands issued to this resource in the past 7 days. The system average execution frequency f_sys is calculated based on the average value of similar resources in the past 7 days, and a lower limit ε is set for the absolute difference of the denominator after calculation to avoid excessive amplification. ε is set to 0.50. The value of ε was determined through a parameter sweep experiment: 30 days of data were replayed when ε was set to 0.10, 0.50, and 1.00, respectively, and the peak power multiplier and resource balance (measured by the standard deviation of execution frequency for similar resources) were statistically analyzed. When ε=0.50, the peak power multiplier did not exceed 1.25 and the balance improvement was most significant, with the execution frequency standard deviation decreasing by 31.2% compared to ε=0.10. Parameter values ​​were all obtained from the previous data collection and quantization step, without direct setting: R_max and R_min were derived from power boundary calculations; S_rate was derived from the scoring mapping; T_avail was derived from state reduction; f_exec and f_sys were derived from historical counts; and strictly satisfied |f_exec−f_sys|≥ε.Example: A resource has a total of n=2 sub-resources within time slice t. Sub-resource 1 has an adjustable range of 80kW, a rating mapping of 0.90, and an available time of 5s; sub-resource 2 has an adjustable range of 40kW, a rating mapping of 0.60, and an available time of 5s. The execution frequency difference of similar resources is 1.20 and satisfies ≥0.50. Substituting these values ​​into the formula, the setpoint of this resource in this time slice is approximately 400kW. This result shows that this resource can be allocated an adjustment command close to 100kW in the current time slice, and due to the large difference in execution frequency, there will be no extreme amplification. The advantage of the formula is that by introducing the execution frequency difference term and the operation rating term, resources with high health and fewer recent executions receive higher allocation weights. In the experimental comparison described later, when this allocation method is used, the standard deviation of the execution frequency of similar resources over 7 days decreases from 12.4 times to 7.9 times, while the closed-loop correction success rate increases from 88.3% to 94.6%.

[0036] S403: Based on the resource power allocation value sequence, combined with the corresponding resource type, control parameters, power limit boundary, and actual load response ratio data, it is assembled into a data structure sequence with continuity and scheduling priority according to the time sorting principle, and written into the scheduling instruction framework to obtain the power scheduling setting instruction sequence. When assembling the scheduling instruction framework based on the resource power allocation value sequence, all resource setting values ​​under the same grid connection point are arranged in ascending order by time tag, and the setting values ​​are bound to resource type, control parameters, power limit boundary, and actual load response ratio: energy storage control parameters include charging and discharging direction and power ramp upper limit, and load control parameters include minimum switching hold time and allowable gradual change step size; the actual load response ratio is obtained through historical comparison, taking the median of "actual change / setting change" of the last 30 instructions. If the median is lower than 0.60, the load allocation value in this cycle is reduced by 0.60 and the gap is backfilled to energy storage resources. After binding is completed, it is written into the power scheduling setting instruction sequence, and the sequence is checked for consistency: the setting values ​​of each resource in the same time slice of the same grid connection point must be consistent in direction. If there is a directional conflict, it is retained according to the rule of "priority to grid connection points with higher level and priority to resources with higher matching degree score". The conflicting resource setting value is set to 0 and the conflict reason code is recorded. Table 2. Deviation Levels and Verification Cycle Configuration Table; Referring to Table 2, a fixed mapping is established between the deviation level and the verification cycle and timeout judgment value, so that subsequent triggering and allocation can operate continuously under the same set of reproducible parameter system.

[0037] Please see Figure 6 The steps to obtain S5 are as follows: S501: Based on the power setting instruction sequence, complete the instruction issuance and record the instruction timestamp, monitor the timer operation status and collect the execution power information returned by the terminal when entering the verification period, call the instruction timestamp to perform time alignment processing on the execution power information, calculate the difference according to the correspondence between the execution power and the set power, and generate the execution deviation amount; When issuing the power scheduling setting command sequence, the commands are grouped according to the grid connection point number, and sent sequentially in ascending order of time slice label within each group. Before issuance, each command generates a unique command identifier, which is formed by concatenating the grid connection point number, resource number, time slice label, and issuance sequence number, used for accurate backtracking of subsequent execution results. Immediately after the command is sent through the scheduling communication channel, the issuance time stamp is recorded, using the master station system clock and accurate to the millisecond level. Upon entering the verification period, the execution power monitoring timer is started according to the command time stamp, triggering the acquisition action according to the time slice cycle. When collecting the execution power information returned by the terminal, the sampling timestamp is read and aligned with the command timestamp. If the sampling timestamp falls within the corresponding time slice interval, it is directly assigned to that time slice; if the deviation is no more than 2 seconds, it is assigned to the nearest time slice; if the deviation exceeds 2 seconds, it is considered invalid. After time alignment, the execution power information is matched one-to-one with the set power value, and the execution deviation is calculated by subtracting the set power value from the execution power value, retaining the positive or negative sign to represent the actual execution direction.

[0038] S502: Based on the execution deviation, obtain the configured deviation tolerance range and perform an interval comparison operation. Collect the energy storage mode status information and load switch status information within the same time period. Make a consistency judgment on the adjustment direction information and feedback status information. Based on the consistency judgment result, correct the execution deviation and obtain the scheduling error correction amount. When entering the deviation tolerance judgment process based on the execution deviation, the deviation tolerance parameters corresponding to the resource type and adjustment direction are read from the configuration table. For energy storage resources, the deviation tolerance parameter is set to 5% of the absolute value of the set power with a minimum lower limit of 2kW; for load resources, it is set to 8% of the absolute value of the set power with a minimum lower limit of 3kW. The absolute value of each execution deviation is calculated and compared with the corresponding upper limit of the tolerance range. Deviations falling within the tolerance range are considered within tolerance; deviations exceeding the tolerance range are considered outside tolerance, and the excess amplitude is recorded. Simultaneously, energy storage mode status information and load switch status information are collected within the same time slice and time alignment is achieved. Consistency judgment is performed on the feedback status information in conjunction with the adjustment direction information. If the adjustment direction does not match the resource operating status, consistency is deemed unsuccessful. Correction calculations are performed based on the consistency judgment result and the tolerance judgment result. If consistency is successful and within tolerance, the correction amount is 0. If consistency is successful but exceeds tolerance, only the excess amplitude is converted to a correction amount using a coefficient of 0.50. If consistency fails, the absolute value of the set power of the resource is directly used as the correction amount.

[0039] S503: Based on the scheduling error correction amount, call the current adjustment direction information and combine it with the collected feedback status information to perform the adjustment boundary update operation. Construct the corresponding scheduling constraint items according to the updated adjustment boundary, collect and process the scheduling constraint items, and establish a scheduling callback correction configuration set. When updating resource adjustment boundaries based on scheduling error correction, the resource adjustment boundary values ​​stored in the previous cycle are called and updated in conjunction with the current adjustment direction information. When resource consistency is passed and the correction is 0, the adjustment boundary value is increased by 5% but not exceeding the available power limit. When resource consistency is passed but the correction is not zero, the boundary value is shrunk according to the ratio of the correction to the absolute value of the set power, and the shrunk boundary value is not less than 50% of the original boundary value. When resource consistency fails, the adjustment boundary value is directly set to 0 and the cooling flag is recorded. After the boundary update is completed, scheduling constraint items containing adjustment boundary values, power change rate, minimum hold time and cooling status are generated for each resource. The scheduling constraint items under the same grid connection point are aggregated to form a scheduling callback correction configuration set, which is used for resource screening and power allocation constraints in the next round of closed-loop verification.

[0040] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A mutual dispatch control method based on a virtual power plant, characterized in that, Includes the following steps: S1: Collect the actual power and target power of the grid connection point, obtain the frequency monitoring data and target power curve distribution information of the area to which the grid connection point belongs, calculate the active power deviation value, call the preset deviation level threshold set to process the active power deviation value, obtain the deviation level identifier, and obtain the area identifier based on the grid connection point number mapping, and establish a grid connection point deviation level label table by combining the deviation level identifier and the area identifier. S2: Configure the master station verification timer based on the grid connection point deviation level label table, calculate the telemetry data time difference, call the timeout judgment value to determine the telemetry data time difference, and generate a closed-loop verification trigger status flag. S3: Obtain the state of charge, maximum charging power, maximum discharging power, maximum adjustable capacity and inverter limiting state quantity according to the closed-loop verification trigger state flag, calculate the resource boundary and identify the adjustment direction, and generate a mutual assistance direction and resource boundary table. S4: Call the mutual assistance direction and resource boundary table to extract the resource adjustment boundary, allocate the energy storage adjustment power, load adjustment power and inverter power limit value, and integrate to generate a sequence of power setting instructions to be issued; S5: Issue instructions based on the issued power setting instruction sequence, collect the execution power and calculate the execution deviation value, call the deviation tolerance interval to compare the execution deviation value and calculate the scheduling error correction amount, collect energy storage mode and load switch status information, update the resource adjustment boundary and generate a scheduling callback correction configuration set.

2. The mutual dispatch control method based on a virtual power plant according to claim 1, characterized in that: The grid connection point deviation level label table includes the grid connection point's region identifier, active power deviation value, and deviation level identification result; the closed-loop verification trigger status marker includes the closed-loop verification cycle parameter, telemetry data time difference comparison result, and scheduling logic start judgment value; the mutual assistance direction and resource boundary table includes the resource upward adjustment boundary, resource downward adjustment boundary, mutual assistance adjustment direction, and resource availability judgment status; the issued power setting command sequence includes the energy storage adjustment power amount, load adjustment power amount, inverter power limit value, and issuance sequence time information; The scheduling callback correction configuration set includes scheduling error correction amount, energy storage mode status information, load switch status information, and current scheduling constraints.

3. The mutual dispatch control method based on a virtual power plant according to claim 1, characterized in that: The set of deviation level thresholds includes at least three active power deviation interval thresholds, which are set in ascending order of absolute value, and each active power deviation interval threshold corresponds to a unique region identifier.

4. The mutual dispatch control method based on a virtual power plant according to claim 1, characterized in that, The steps to obtain S1 are as follows: S101: Collect the actual active power data of the grid connection point and the target power data of the grid connection point configured in the scheduling plan. Record the grid connection point number and the target power value and actual power value at the corresponding time point. Match the target power value and actual power value at the same time point. Obtain the active power deviation value by calculating the difference between the target power and the actual power. S102: Call the active power deviation value, combine it with the configured deviation level threshold set, perform interval matching identification, determine the threshold interval range that the deviation value of the grid connection point falls into at each time point, identify the corresponding level identifier, and introduce the power change of the grid connection point, the target power distribution status and the frequency fluctuation of the area, judge and classify the deviation value, and obtain the deviation level interval. S103: Call the deviation level range, and according to the regional identifier of each grid connection point, associate and integrate the grid connection point number, the region to which it belongs, and the identified level range, and arrange them in a structured manner according to the regional and time dimensions to establish a grid connection point deviation level label table.

5. The mutual dispatch control method based on a virtual power plant according to claim 1, characterized in that, The steps to obtain S2 are as follows: S201: Based on the grid connection point deviation level label table, extract the closed-loop verification cycle parameters corresponding to the grid connection point number, combine the region and time information, complete the archiving process of the cycle parameters, and write them into the main station verification timer to obtain the closed-loop verification cycle parameter set. S202: Call the closed-loop verification cycle parameter set, collect the telemetry sampling timestamp of the grid connection point and the current system clock time and calculate the time difference, compare the calculated time difference with the configured telemetry data timeout judgment value, combine the time difference change status of the grid connection point and the timing characteristics of the region to which it belongs, make a judgment and identification, and obtain the scheduling logic trigger judgment flag. S203: Based on the scheduling logic trigger judgment flag, filter the grid connection points with the flag value in the start state, combine the number, area identifier and current time information, record the logic state switching behavior, construct the corresponding scheduling response record, and generate a closed-loop verification trigger state flag.

6. The mutual dispatch control method based on a virtual power plant according to claim 1, characterized in that, The steps to obtain S3 are as follows: S301: Based on the closed-loop verification trigger state flag, collect the state of charge, maximum charging power and maximum discharging power reported by the energy storage equipment monitoring device, extract the maximum adjustable capacity reported by the controllable load device, record the inverter limit state quantity, summarize the resource parameters, and obtain the resource state parameter set. S302: Call the resource status parameter set, identify the available power boundary within the operating range based on the current limitation status of the resource equipment, perform upward and downward power range calculations in combination with the adjustment capabilities of the energy storage equipment and the load equipment, obtain the resource adjustment boundary value, determine the adjustment direction of all grid-connected points in the current cycle, and obtain the resource adjustment direction identifier. S303: Based on the resource adjustment direction identifier and the resource adjustment boundary value, aggregate the upward and downward adjustment directions and corresponding power range data in the current cycle according to the grid connection point number, structure and integrate the adjustment direction type and boundary value, and output them in association with the area label to generate a data structure mutual adjustment direction and resource boundary table.

7. The mutual dispatch control method based on a virtual power plant according to claim 6, characterized in that, The specific calculation formula for calculating the upward and downward power range by combining the adjustment capabilities of energy storage equipment and load equipment is as follows: ； The calculation obtains the resource adjustment boundary value, determines the status of the adjustment direction of all grid-connected points in the current cycle, and obtains the resource adjustment direction identifier. in, Indicates the first Resource adjustment boundary value for each resource unit For the first The maximum up-adjustment capacity of each energy storage device For the first The maximum power limit for each energy storage device. For the first Maximum reduction capacity of each load device For the first Maximum power reduction limit for each load device For the first The charging and discharging sensitivity of an energy storage device For the first Additional performance parameters for each energy storage device.

8. The mutual dispatch control method based on a virtual power plant according to claim 1, characterized in that, The steps to obtain S4 are as follows: S401: Call the resource adjustment boundary data in the mutual assistance direction and resource boundary table, identify the time tag, resource type, adjustment capacity upper limit, and area number parameters associated with the resource boundary, filter energy storage resources, load resources and inverter resources under matching conditions as participating items, and filter according to the matching degree between the resource adjustment boundary and the adjustment demand of the region to obtain a list of resources with adjustment capabilities and generate an adjustable resource number sequence. S402: Based on the resources in the adjustable resource number sequence, obtain the power adjustment capability value, available time period, equipment status parameters, and execution frequency participation items within a specified time period, and perform resource adjustment power allocation calculation to obtain the allocation power setting value matrix. Combine the resource number and time tag to obtain the resource power allocation value sequence. S403: Based on the resource power allocation value sequence, combined with the corresponding resource type, control parameters, power limit boundary, and actual load response ratio data, the data is assembled into a data structure sequence with continuity and scheduling priority according to the time sorting principle, and written into the scheduling instruction framework to obtain the power scheduling setting instruction sequence.

9. The mutual dispatch control method based on a virtual power plant according to claim 1, characterized in that, The steps to obtain S5 are as follows: S501: Based on the issued power setting instruction sequence, complete the instruction issuance and record the instruction timestamp, monitor the timer operation status and collect the execution power information returned by the terminal when entering the verification period, call the instruction timestamp to perform time alignment processing on the execution power information, calculate the difference according to the correspondence between the execution power and the set power, and generate the execution deviation amount; S502: Based on the execution deviation amount, obtain the configured deviation tolerance range and perform the range comparison operation, collect the energy storage mode status information and load switch status information within the same time period, make a consistency judgment on the adjustment direction information and feedback status information, and make a correction calculation on the execution deviation amount based on the consistency judgment result to obtain the scheduling error correction amount. S503: Based on the aforementioned scheduling error correction amount, the current adjustment direction information is invoked and combined with the collected feedback status information to perform an adjustment boundary update operation. The corresponding scheduling constraint items are constructed according to the updated adjustment boundary, and the scheduling constraint items are aggregated and processed to establish a scheduling callback correction configuration set.

10. The mutual dispatch control method based on a virtual power plant according to claim 1, characterized in that: The resource boundary is obtained by constraining the state of charge, maximum charging power, maximum discharging power, maximum adjustable capacity, and inverter limiting state quantities. The mutual assistance direction is determined by the positive or negative attribute of the active power deviation value.

Images