A method for coordinating and complementing energy storage power station and power grid based on swarm intelligence

Abstract

This invention discloses a collaborative and complementary regulation method between an energy storage power station and the power grid based on swarm intelligence, comprising the following steps: Step 1: Collecting energy storage operation data and power grid operation data and generating station-grid operation data; Step 2: Generating an energy storage collaborative regulation vector; Step 3: Generating a power grid regulation sensitivity field; Step 4: Constructing an improved mountain gazelle optimization algorithm; Step 5: Generating an energy storage role table; Step 6: Generating candidate collaborative regulation vectors; Step 7: Generating feasible collaborative regulation vectors through state of charge verification and power flow security verification; Step 8: Outputting collaborative and complementary regulation commands. This invention improves the matching degree between the energy storage regulation direction and the power grid operation requirements, and enhances the security and executability of the regulation commands.

Description

Technical Field

[0001] This invention relates to the field of power system dispatching technology, and in particular to a method for coordinated and complementary regulation of energy storage power stations and power grids based on swarm intelligence. Background Technology

[0002] With the continuous growth of new energy grid connection scale and the increasing demand for flexible adjustment of distribution networks, the coordinated and complementary adjustment technology for energy storage power stations and grid operation has received widespread attention. Existing energy storage dispatching methods mainly rely on fixed rules, economic targets, or conventional swarm intelligence algorithms to generate charging and discharging plans, but these methods generally suffer from the following problems in practical applications: The lack of correlation between energy storage operation data and grid operation data makes it difficult to form a unified scheduling basis based on data collection time, access nodes, and operating status, resulting in inaccurate matching between energy storage regulation capacity and grid regulation needs. Existing optimization methods often treat energy storage power as a single variable for optimization, making it difficult to form regulation directions oriented towards grid-sensitive states based on access node power deviation, line power flow margin, and node voltage deviation. For issues of state of charge exceeding limits and power flow exceeding limits, traditional algorithms usually use penalty functions or boundary truncation, which makes it difficult to roll back the out-of-limit regulation amount to energy storage power stations with state of charge margin and power flow margin, affecting the feasibility of energy storage coordinated regulation commands and grid operation safety.

[0003] Therefore, how to provide a method for coordinated and complementary regulation between energy storage power stations and the power grid based on swarm intelligence is a problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0004] One objective of this invention is to propose a collaborative and complementary regulation method between energy storage power stations and the power grid based on swarm intelligence. This invention employs an improved mountain gazelle optimization algorithm, embedding the grid regulation sensitivity field into the migration direction update rule and the state of charge margin into the over-boundary rollback rule, thereby achieving collaborative determination of energy storage regulation capacity, response time period, and access node. It has the advantages of accurate regulation direction, low over-boundary risk, and strong station-grid coordination capability.

[0005] According to an embodiment of the present invention, a method for coordinated and complementary regulation between an energy storage power station and a power grid based on swarm intelligence includes the following steps: Step 1: Collect energy storage operation data and grid operation data, associate them with access nodes according to the collection time, and generate station-grid operation data; Step 2: Determine the energy storage regulation capacity, response time period, and access nodes based on the station network operation data, and generate an energy storage coordinated regulation vector; Step 3: Generate a grid regulation sensitivity field based on the power deviation of the access node, the power flow margin of the line, and the voltage deviation of the node; Step 4: Write the grid regulation sensitivity field into the migration direction update rule of the mountain gazelle optimization algorithm, and write the state of charge margin into the out-of-bounds rollback rule to construct an improved mountain gazelle optimization algorithm. Step 5: The improved mountain gazelle optimization algorithm is used to process the energy storage coordinated regulation vector, and an energy storage role table is generated based on the grid regulation sensitivity field; Step 6: Update the charge and discharge regulation direction in the energy storage coordinated regulation vector according to the energy storage role table, and generate candidate coordinated regulation vectors; Step 7: Perform state of charge verification and power flow safety verification on the candidate coordinated regulation vectors, and roll back the out-of-bounds regulation amount to the energy storage power station with state of charge margin and power flow margin to generate a feasible coordinated regulation vector. Step 8: Determine the target regulation capacity, target response time period, and target access node based on the feasible coordinated regulation vector, and output the coordinated and complementary regulation command between the energy storage power station and the power grid.

[0006] Optionally, step one specifically includes: Energy storage operation data is collected according to the unified collection cycle of energy storage power station and power grid. The energy storage operation data includes power station identification, collection time, access node identification, state of charge, charging and discharging power and available capacity. Power grid operation data is collected according to a unified collection cycle. The power grid operation data includes the collection time, access node identifier, node power, line power flow, node voltage and operating frequency. Using the time of data collection as the time-related field, the energy storage operation data is matched with the execution time of the power grid operation data to generate time-related operation records; Using the access node identifier as the node association field, the node matching is performed on the time-related operation records to generate a station-network association record between the energy storage power station and the access node; Perform missing field validation, numerical boundary validation, and temporal continuity validation on the station network associated records, filter out abnormal operation records, and arrange them according to the collection time and access node identifier to generate station network operation data.

[0007] Optionally, step two specifically involves: Extract energy storage operation fields and power grid operation fields from the station network operation data, and form node operation records according to the collection time and access node identifier; The adjustable capacity of the energy storage power station is calculated based on the state of charge, charging and discharging power and available capacity in the node operation record, and the adjustment demand of the access node is determined based on the node power, line power flow and node voltage. Match the adjustable capacity of the energy storage power station with the adjustment demand of the access node to determine the energy storage adjustment capacity, and establish a correspondence between the energy storage adjustment capacity and the access node identifier; Based on the data collection time and changes in operating frequency, the response period for energy storage power stations to participate in regulation is determined, and a correspondence is established between the response period and the energy storage regulation capacity. Vector units are constructed according to energy storage regulation capacity, response time period and access node identifier, and the vector units are arranged according to the acquisition time to generate energy storage collaborative regulation vector.

[0008] Optionally, step three specifically includes: Extract access node identifier, acquisition time, node power, line power flow and node voltage from the station network operation data, and form node status records according to acquisition time and access node identifier; The access node power deviation is calculated based on the node power difference between adjacent acquisition times in the node status record, and a correspondence is established between the access node power deviation and the access node identifier. Calculate the line power flow margin based on the line power flow and line rated power flow in the node status record, and establish a correspondence between the line power flow margin and the access node identifier; Calculate the node voltage deviation based on the node voltage and node rated voltage in the node status record, and establish a correspondence between the node voltage deviation and the access node identifier; The dimensions of the access node power deviation, line power flow margin and node voltage deviation are unified and directionally calibrated, and then collected according to the access node identification to generate a power grid regulation sensitivity field.

[0009] Optionally, step four specifically includes: Extract the energy storage regulation capacity, response time period and access node identifier from the energy storage coordinated regulation vector, map the energy storage regulation capacity to the regulation candidate position in the mountain gazelle optimization algorithm, and write the response time period and access node identifier into the regulation candidate position to form a candidate position constraint record. Extract the access node power deviation, line power flow margin and node voltage deviation corresponding to the access node identifier from the power grid regulation sensitivity field, and perform matching between the access node identifier and the candidate location constraint record to generate candidate location sensitivity record; The migration direction weight is calculated based on the candidate position sensitivity record, and the migration direction weight is written into the migration direction update rule of the mountain gazelle optimization algorithm so that the adjustment candidate position is updated according to the adjustment direction determined by the power grid adjustment sensitivity field. Extract the state of charge and available capacity corresponding to the access node identifier from the station network operation data, calculate the state of charge margin based on the state of charge and available capacity, and write the state of charge margin into the adjustment candidate position to form the candidate position margin record. Based on the candidate position margin record, an out-of-bounds rollback rule is established. The out-of-bounds rollback rule is used to perform capacity rollback on the adjustment candidate positions that exceed the state of charge margin, and to limit the adjustment candidate positions after capacity rollback to the capacity range corresponding to the state of charge margin. The mountain gazelle optimization algorithm with written migration direction update rules and out-of-bounds rollback rules was identified as the improved mountain gazelle optimization algorithm.

[0010] Optionally, step five specifically includes: The energy storage coordinated regulation vector is input into the improved mountain gazelle optimization algorithm, and the energy storage regulation capacity, response time period and access node identifier are read according to the collection time to generate regulation candidate records; Match the candidate records of regulation with the grid regulation sensitivity field according to the access node identifier, and write the access node power deviation, line power flow margin and node voltage deviation into the candidate records of regulation to generate regulation sensitivity matching records. The main regulation score is generated based on the access node power deviation in the regulation-sensitive matching record, the boundary protection score is generated based on the line power flow margin, and the backup compensation score is generated based on the node voltage deviation. The main regulation score, boundary protection score, and backup compensation score are written into the same regulation-sensitive matching record to generate a role score record. Perform numerical sorting on the main adjustment score, boundary protection score and backup compensation score in the role rating record, write the role tag corresponding to the first ranked score into the role rating record, and generate a role tag record. Arrange the energy storage regulation capacity, response time period, access node identifier, and role tag in the role tag record according to the collection time to generate an energy storage role table.

[0011] Optionally, step six specifically includes: Read the data collection time, energy storage regulation capacity, response time period, access node identifier, and role tag from the energy storage role table, and generate a role direction update record; Match the role direction update record with the energy storage collaborative adjustment vector according to the acquisition time and access node identifier, write the role mark into the corresponding vector unit in the energy storage collaborative adjustment vector, and generate the role matching adjustment vector; The charging and discharging adjustment direction is determined based on the role marker in the role matching adjustment vector. When the role marker corresponds to the main adjustment score, the charging and discharging adjustment direction is determined according to the access node power deviation. When the role marker corresponds to the boundary protection score, the charging and discharging adjustment direction is determined according to the line power flow margin. When the role marker corresponds to the standby compensation score, the charging and discharging adjustment direction is determined according to the node voltage deviation. Write the charging / discharging adjustment direction into the corresponding vector unit in the character matching adjustment vector to generate the direction update adjustment vector; The vector units are organized according to the energy storage regulation capacity, response time period, access node identifier and charge / discharge regulation direction in the direction update regulation vector, and the vector units are arranged according to the acquisition time to generate candidate coordinated regulation vectors.

[0012] Optionally, step seven specifically includes: Extract the acquisition time, energy storage regulation capacity, response time period, access node identifier, and charging / discharging regulation direction from the candidate coordinated regulation vector to generate candidate verification records; The candidate verification records are matched with the station network operation data according to the collection time and access node identifier, the state of charge margin and line power flow margin are read, and margin matching verification records are generated. Based on the energy storage regulation capacity, charge / discharge regulation direction and state of charge margin in the margin matching verification record, perform state of charge verification, calculate the capacity difference between the energy storage regulation capacity and the state of charge margin, and generate the state of charge out-of-bounds amount. Based on the energy storage regulation capacity, charging and discharging regulation direction and line power flow margin in the margin matching verification record, perform power flow safety verification, calculate the power flow difference caused by the energy storage regulation capacity exceeding the line power flow margin, and generate the power flow exceeding limit. The over-limit quantity of state of charge and the over-limit quantity of power flow are merged into the over-limit regulation quantity, and the over-limit regulation quantity is written into the candidate verification record according to the collection time and access node identifier. Energy storage power stations with both state of charge margin and power flow margin are selected from the station network operation data. The out-of-bounds regulation is rolled back to the selected energy storage power stations, and the candidate coordinated regulation vector is updated to generate a feasible coordinated regulation vector.

[0013] Optionally, step eight specifically includes: Extract the acquisition time, energy storage regulation capacity, response time period, access node identifier and charging / discharging regulation direction from the feasible coordinated regulation vector to generate a feasible regulation record; The feasible adjustment record is matched with the grid adjustment sensitivity field according to the access node identifier. The access node power deviation, line power flow margin and node voltage deviation are written into the feasible adjustment record to generate the target screening record. Based on the access node power deviation in the target screening record, a capacity priority value is generated; based on the line power flow margin, a safety priority value is generated; based on the node voltage deviation, a voltage priority value is generated; and the capacity priority value, safety priority value, and voltage priority value are merged into the target adjustment priority value. The target screening records are numerically sorted according to the target adjustment priority value, and the target screening record with the first position in the sort is determined as the target vector unit. The target regulation capacity, target response time period, and target access node are determined from the target vector unit, and the target regulation capacity, target response time period, target access node, and charging / discharging regulation direction are written into the instruction record. The energy storage power station and the power grid are then output as a coordinated and complementary regulation instruction.

[0014] The beneficial effects of this invention are: This invention collects energy storage operation data and grid operation data, and correlates them with access nodes according to the collection time to form station-grid operation data that simultaneously reflects the status of both energy storage and the grid. Based on this data, an energy storage collaborative regulation vector and a grid regulation sensitivity field are generated, enabling the coordinated determination of energy storage regulation capacity, response time, and access nodes around access node power deviation, line power flow margin, and node voltage deviation. This avoids energy storage power stations participating in regulation solely according to fixed charging and discharging rules, improves the matching degree between energy storage regulation direction and the actual operating needs of the grid, and enhances the support capability of energy storage power stations for load fluctuations, node voltage deviations, and line power flow pressure.

[0015] This invention incorporates the grid regulation sensitivity field into the migration direction update rule of the mountain gazelle optimization algorithm and the state of charge margin into the out-of-bounds rollback rule, constructing an improved mountain gazelle optimization algorithm. This allows candidate coordinated regulation vectors to be jointly constrained by the grid sensitivity state and the energy storage safety margin during the optimization process. When a candidate coordinated regulation vector experiences a state of charge out-of-bounds or power flow out-of-bounds, the out-of-bounds regulation amount is rolled back to an energy storage station with both state of charge margin and power flow margin to generate a feasible coordinated regulation vector. This reduces the risks of overcharging and over-discharging of energy storage and line out-of-bounds risks, and improves the executability, stability, and grid operation safety of coordinated and complementary regulation commands. Attached Figure Description

[0016] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings: Figure 1 This is a flowchart of a collaborative and complementary regulation method between an energy storage power station and the power grid based on swarm intelligence, as proposed in this invention. Figure 2 This is a schematic diagram illustrating the construction of an improved mountain gazelle optimization algorithm for a collaborative and complementary regulation method between energy storage power stations and power grids based on swarm intelligence, as proposed in this invention. Detailed Implementation

[0017] The present invention will now be described in further detail with reference to the accompanying drawings. These drawings are simplified schematic diagrams, illustrating only the basic structure of the invention, and therefore only show the components relevant to the invention.

[0018] refer to Figures 1-2 A method for coordinated and complementary regulation between energy storage power stations and power grids based on swarm intelligence includes the following steps: Step 1: Collect energy storage operation data and grid operation data, associate them with access nodes according to the collection time, and generate station-grid operation data; Step 2: Determine the energy storage regulation capacity, response time period, and access nodes based on the station network operation data, and generate an energy storage coordinated regulation vector; Step 3: Generate a grid regulation sensitivity field based on the power deviation of the access node, the power flow margin of the line, and the voltage deviation of the node; Step 4: Write the grid regulation sensitivity field into the migration direction update rule of the mountain gazelle optimization algorithm, and write the state of charge margin into the out-of-bounds rollback rule to construct an improved mountain gazelle optimization algorithm. Step 5: The improved mountain gazelle optimization algorithm is used to process the energy storage coordinated regulation vector, and an energy storage role table is generated based on the grid regulation sensitivity field; Step 6: Update the charge and discharge regulation direction in the energy storage coordinated regulation vector according to the energy storage role table, and generate candidate coordinated regulation vectors; Step 7: Perform state of charge verification and power flow safety verification on the candidate coordinated regulation vectors, and roll back the out-of-bounds regulation amount to the energy storage power station with state of charge margin and power flow margin to generate a feasible coordinated regulation vector. Step 8: Determine the target regulation capacity, target response time period, and target access node based on the feasible coordinated regulation vector, and output the coordinated and complementary regulation command between the energy storage power station and the power grid.

[0019] In this embodiment, step one specifically includes: Energy storage operation data is collected according to the unified collection cycle of energy storage power station and power grid. During the collection process, energy storage power stations are distinguished by power station identifier, data generation time is marked by collection time, and the location of energy storage power station connected to the power grid is recorded by access node identifier. The state of charge, charging and discharging power and available capacity are recorded simultaneously. The data fields corresponding to the same power station identifier at the same collection time are organized into energy storage operation data. Power grid operation data is collected according to a unified collection cycle. During the collection process, the data generation time of the power grid side is marked by the collection time, and the access node is identified as the access location of the corresponding energy storage power station. The node power, line power flow, node voltage and operating frequency corresponding to the access node are recorded. The data fields corresponding to the same access node at the same collection time are organized into power grid operation data. Using the acquisition time as the time association field, the acquisition time in the energy storage operation data is matched with the acquisition time in the power grid operation data. Energy storage operation data and power grid operation data with consistent time are written into the same operation record unit to generate time-associative operation records. Using the access node identifier as the node association field, the energy storage power station access node identifier in the time-related operation record is matched with the grid-side access node identifier. The energy storage operation field and grid operation field under the same access node identifier are written into the same association record unit to generate a station-network association record between the energy storage power station and the access node. The system performs missing field validation, numerical boundary validation, and temporal continuity validation on the station network associated records. Missing field validation is used to identify running records with empty fields. Numerical boundary validation is used to identify running records that exceed the range of field values. Temporal continuity validation is used to identify running records whose data collection time is interrupted. Running records that fail the validation are filtered out as abnormal running records. The remaining running records are then arranged according to the data collection time and access node identifier to generate station network running data.

[0020] In this embodiment, step two specifically includes: The energy storage operation field and the power grid operation field are read from the station network operation data according to the acquisition time. The energy storage operation field includes the state of charge, charging and discharging power and available capacity, while the power grid operation field includes node power, line power flow, node voltage and operating frequency. Then, the energy storage operation field and the power grid operation field corresponding to the same acquisition time are written into the same recording unit according to the access node identifier to form a node operation record. The remaining charging margin and remaining discharging margin of the energy storage power station are determined based on the state of charge in the node operation record. The power regulation boundary of the energy storage power station is determined based on the charging and discharging power. The capacity regulation boundary of the energy storage power station is determined based on the available capacity. The remaining charging margin, remaining discharging margin, power regulation boundary and capacity regulation boundary are subjected to intersection constraints to calculate the adjustable capacity of the energy storage power station. The node power deviation is formed based on the node power, the power flow occupancy margin is formed based on the line power flow, and the voltage deviation is formed based on the node voltage. The node power deviation, power flow occupancy margin and voltage deviation are aggregated into the regulation demand of the access node. Match the adjustable capacity of the energy storage power station with the adjustment demand of the access node. If the adjustable capacity of the energy storage power station is not less than the adjustment demand of the access node, the adjustment demand of the access node is determined as the energy storage adjustment capacity. If the adjustable capacity of the energy storage power station is less than the adjustment demand of the access node, the adjustable capacity of the energy storage power station is determined as the energy storage adjustment capacity, and a corresponding relationship is established between the energy storage adjustment capacity and the access node identifier. The node operation records are arranged according to the acquisition time, and the frequency change duration interval is determined according to the operation frequency change between adjacent acquisition times. The frequency change duration interval is mapped to the response period of the energy storage power station participating in the regulation, and the response period is associated with the energy storage regulation capacity under the same access node identifier. Vector units are constructed based on energy storage regulation capacity, response time period, and access node identifier. Each vector unit contains a set of energy storage regulation capacity, response time period, and access node identifier. The vector units are then arranged according to the acquisition time to generate an energy storage coordinated regulation vector.

[0021] In this embodiment, step three specifically includes: Read the access node identifier, node power, line power flow and node voltage from the station network operation data according to the acquisition time. Write the node power, line power flow and node voltage corresponding to the same access node identifier at the same acquisition time into the same recording unit. Arrange the recording units according to the order of acquisition time and access node identifier to form a node status record. The power difference is calculated based on the node power at adjacent collection times under the same access node identifier in the node status record. The node power deviation is obtained by subtracting the node power at the previous collection time from the node power at the later collection time. The access node power deviation is written under the corresponding access node identifier to form a correspondence between the access node power deviation and the access node identifier. The line power flow margin is calculated based on the line power flow and the line rated power flow in the node status record. The line power flow margin is obtained by subtracting the line power flow from the line rated power flow. The ratio of the remaining power flow margin to the line rated power flow is determined as the line power flow margin. The line power flow margin is written under the corresponding access node identifier to form a correspondence between the line power flow margin and the access node identifier. The node voltage deviation is calculated based on the node voltage and node rated voltage in the node status record. The node voltage is subtracted from the node rated voltage to obtain the voltage difference. The ratio of the voltage difference to the node rated voltage is determined as the node voltage deviation. The node voltage deviation is written under the corresponding access node identifier to form a correspondence between the node voltage deviation and the access node identifier. The dimensions of the access node power deviation, line power flow margin, and node voltage deviation are unified. The access node power deviation is converted into a power deviation normalized value, the line power flow margin is converted into a power flow margin normalized value, and the node voltage deviation is converted into a voltage deviation normalized value. Then, the power deviation normalized value, power flow margin normalized value, and voltage deviation normalized value are directionally calibrated according to the adjustment direction, and collected according to the access node identifier to generate a power grid regulation sensitivity field.

[0022] In this embodiment, step four specifically includes: Read vector units from the energy storage coordinated regulation vector according to the acquisition time, extract the energy storage regulation capacity, response time period and access node identifier from the vector unit, use the energy storage regulation capacity as the numerical position participating in the position iteration in the mountain gazelle optimization algorithm, write the response time period as the time limit field into the numerical position, write the access node identifier as the node limit field into the numerical position, and bind the numerical position, response time period and access node identifier as regulation candidate positions to form candidate position constraint records; Read the access node power deviation, line power flow margin and node voltage deviation from the grid regulation sensitivity field according to the access node identifier. Match the access node identifier in the candidate position constraint record with the access node identifier in the grid regulation sensitivity field. Write the matching regulation candidate position, response time, access node power deviation, line power flow margin and node voltage deviation into the same recording unit to generate a candidate position sensitivity record. The migration direction weight is calculated based on the sensitivity record of the candidate position. During the calculation process, the positive and negative directions of the access node power deviation are used to determine the charging and discharging direction of the candidate position, the magnitude of the access node power deviation is used to determine the power regulation intensity, the line power flow margin is used to determine the available power flow amplitude of the candidate position, and the node voltage deviation is used to determine the voltage correction direction of the candidate position. The charging and discharging direction, power regulation intensity, available power flow amplitude, and voltage correction direction are combined into the migration direction weight, and the migration direction weight is written into the migration direction update rule of the mountain gazelle optimization algorithm so that the candidate position is updated according to the regulation direction determined by the power grid regulation sensitivity field. The state of charge (SOC) and available capacity are read from the station network operation data according to the access node identifier. The SOC margin is calculated based on the SOC and available capacity. During the calculation, the difference between the SOC and available capacity is multiplied by the available capacity to obtain the remaining charging margin. The SOC is multiplied by the available capacity to obtain the remaining discharging margin. The remaining charging margin and the remaining discharging margin are used together as the SOC margin. The SOC margin is written to the corresponding adjustment candidate position to form a candidate position margin record. Based on the candidate location margin record, an out-of-bounds rollback rule is established. The out-of-bounds rollback rule uses the state of charge margin as the capacity boundary for adjusting the candidate location. When the energy storage adjustment capacity corresponding to the candidate location exceeds the capacity boundary corresponding to the state of charge margin, the difference between the energy storage adjustment capacity and the capacity boundary is calculated as the out-of-bounds capacity value. Capacity rollback is then performed on the out-of-bounds capacity value according to the capacity boundary, thus limiting the adjusted candidate location after capacity rollback to the capacity range corresponding to the state of charge margin. The mountain gazelle optimization algorithm that incorporates migration direction update rules and out-of-bounds rollback rules is identified as the improved mountain gazelle optimization algorithm. The improved mountain gazelle optimization algorithm uses candidate position constraint records to limit the response time period and access node identifier of the adjustment candidate position, uses candidate position sensitivity records to determine the migration direction weight of the adjustment candidate position, and uses candidate position margin records to limit the capacity range of the adjustment candidate position.

[0023] In this embodiment, step five specifically includes: The energy storage coordinated regulation vector is input into the improved mountain gazelle optimization algorithm. The energy storage regulation capacity, response time period and access node identifier are read item by item according to the collection time. A set of energy storage regulation capacity, response time period and access node identifier under the same collection time is organized into a regulation candidate record, so that the regulation candidate record carries the capacity information, time information and node information of the energy storage power station participating in the coordinated regulation. The access node identifier in the candidate record is matched with the access node identifier in the grid regulation sensitivity field. The matching access node power deviation, line power flow margin and node voltage deviation are written into the same candidate record, so that the candidate record contains both the regulation information of the energy storage coordinated regulation vector and the sensitivity information of the grid regulation sensitivity field, and a regulation sensitivity matching record is generated. The power deviation magnitude is calculated based on the access node power deviation in the regulation sensitive matching record, and the main regulation score is generated based on the power deviation magnitude. The power flow margin is calculated based on the power flow boundary proximity, and the boundary protection score is generated based on the power flow boundary proximity. The voltage deviation magnitude is calculated based on the node voltage deviation, and the backup compensation score is generated based on the voltage deviation magnitude. The main regulation score, boundary protection score, and backup compensation score are then written into the same regulation sensitive matching record to generate a role score record. Perform numerical sorting on the main adjustment score, boundary protection score, and backup compensation score in the role rating record. Take the score with the highest numerical value as the target score, read the role tag corresponding to the target score, write the role tag into the same role rating record, and generate a role tag record. Read the energy storage regulation capacity, response time period, access node identifier and role tag from the role tag record, write the energy storage regulation capacity, response time period, access node identifier and role tag into the same table entry, and then arrange the table entries according to the collection time to generate the energy storage role table.

[0024] In this embodiment, step six specifically includes: Read entries from the energy storage role table according to the collection time, extract the energy storage regulation capacity, response time period, access node identifier and role mark from the entries, organize the energy storage regulation capacity, response time period, access node identifier and role mark under the same collection time into the same record unit, and generate role direction update record; The acquisition time and access node identifier in the role direction update record are matched with the acquisition time and access node identifier in the energy storage coordinated adjustment vector. If the match is consistent, the role marker in the role direction update record is written into the corresponding vector unit in the energy storage coordinated adjustment vector to generate the role matching adjustment vector. The charging and discharging adjustment direction is determined based on the role marker in the role matching adjustment vector. When the role marker corresponds to the main adjustment score, the energy storage station is determined to perform charging or discharging adjustment according to the positive or negative direction of the access node power deviation. When the role marker corresponds to the boundary protection score, the charging and discharging adjustment direction of the energy storage station to reduce power flow occupancy is determined according to the line power flow margin. When the role marker corresponds to the standby compensation score, the charging and discharging adjustment direction of the energy storage station to support voltage recovery is determined according to the node voltage deviation. Write the charging and discharging adjustment direction into the vector unit in the role matching adjustment vector that is consistent with the acquisition time and access node identifier, so that the same vector unit simultaneously contains the energy storage adjustment capacity, response time period, access node identifier, role mark and charging and discharging adjustment direction, and generate the direction update adjustment vector; The vector units are organized according to the energy storage regulation capacity, response time period, access node identifier and charge / discharge regulation direction in the direction update regulation vector, so that each vector unit corresponds to one energy storage power station regulation content. Then, the vector units are arranged according to the acquisition time to generate candidate coordinated regulation vectors.

[0025] In this embodiment, step seven specifically includes: Extract the acquisition time, energy storage regulation capacity, response time period, access node identifier and charge / discharge regulation direction from the candidate coordinated regulation vector. Write the energy storage regulation capacity, response time period and charge / discharge regulation direction corresponding to the same acquisition time and the same access node identifier into the same recording unit to generate candidate verification records. The collection time and access node identifier in the candidate verification record are matched with the collection time and access node identifier in the station network operation data. The state of charge, available capacity and line power flow are extracted from the matched station network operation data. The state of charge margin is determined based on the state of charge and available capacity, and the line power flow margin is determined based on the line power flow. Margin matching verification record is generated. Based on the energy storage regulation capacity, charge / discharge regulation direction and state of charge margin in the margin matching verification record, the state of charge verification is performed. When the charge / discharge regulation direction is the charging direction, the energy storage regulation capacity is compared with the remaining charging margin. When the charge / discharge regulation direction is the discharging direction, the energy storage regulation capacity is compared with the remaining discharging margin. The capacity difference between the energy storage regulation capacity and the state of charge margin is determined as the state of charge overrun. Based on the energy storage regulation capacity, charging and discharging regulation direction and line power flow margin in the margin matching verification record, the power flow safety verification is performed. The energy storage regulation capacity is converted into line power flow change according to the charging and discharging regulation direction, and the line power flow change is compared with the line power flow margin. If the line power flow change exceeds the line power flow margin, the excess part is determined as the power flow exceeding the limit. The out-of-bounds charge state and power flow exceedance are merged into out-of-bounds regulation quantities, and the out-of-bounds regulation quantities are written into the candidate verification record according to the collection time and access node identifier, so that the candidate verification record simultaneously records the energy storage regulation capacity, charging and discharging regulation direction, out-of-bounds charge state and power flow exceedance. Energy storage power stations with both state of charge margin and power flow margin are selected from the station network operation data. The out-of-bounds regulation amount is rolled back to the selected energy storage power station according to the collection time and access node identifier. The candidate coordinated regulation vector is updated according to the rolled-back energy storage regulation capacity to generate a feasible coordinated regulation vector.

[0026] In this embodiment, step eight specifically includes: Extract the energy storage regulation capacity, response time period, access node identifier and charge / discharge regulation direction from the feasible coordinated regulation vector according to the acquisition time. Write the energy storage regulation capacity, response time period and charge / discharge regulation direction corresponding to the same acquisition time and the same access node identifier into the same recording unit to generate a feasible regulation record. Perform a matching process between the access node identifier in the feasible adjustment record and the access node identifier in the power grid adjustment sensitivity field. Write the matching access node power deviation, line power flow margin and node voltage deviation into the same feasible adjustment record, so that the feasible adjustment record contains both feasible adjustment content and power grid sensitive information, and generate a target screening record. The power deviation amplitude is calculated based on the access node power deviation in the target screening record, and the power deviation amplitude is converted into a capacity priority value. The power flow safety margin is calculated based on the line power flow margin, and the power flow safety margin is converted into a safety priority value. The voltage deviation amplitude is calculated based on the node voltage deviation, and the voltage deviation amplitude is converted into a voltage priority value. Finally, the capacity priority value, safety priority value and voltage priority value are merged into the target adjustment priority value. The target screening records are numerically sorted according to the target adjustment priority value. The target screening record corresponding to the target adjustment priority value with the first value is determined as the target vector unit, so that the target vector unit corresponds to the access node with the highest adjustment priority in the power grid adjustment sensitivity field. Extract the energy storage regulation capacity from the target vector unit and determine it as the target regulation capacity. Extract the response time period and determine it as the target response time period. Extract the access node identifier and determine it as the target access node. Then write the target regulation capacity, target response time period, target access node and charging / discharging regulation direction into the instruction record and output the energy storage power station and grid coordinated and complementary regulation instruction.

[0027] Example 1: To verify the feasibility of this invention in practice, it was applied to a coordinated regulation scenario of a distribution network energy storage power station cluster in a certain region. In this scenario, the distribution network is connected to distributed photovoltaic and industrial / commercial loads. During midday, there is a risk of reverse power flow due to a concentrated increase in photovoltaic output, and in the evening, there is a problem of node power deviation and local line power flow approaching limits due to rapid load ramp-up. The original regulation method mainly involves charging and discharging energy storage according to fixed time periods. There is a lack of coordinated allocation between energy storage power stations based on the status of the connected nodes. Some energy storage power stations discharge too quickly during high-load periods, and cannot continue to participate in regulation during low-load periods because their state of charge is close to the boundary. This results in uneven utilization of line power flow margin and untimely correction of node voltage deviations.

[0028] During implementation, edge dispatch terminals access energy storage operation data and grid operation data according to a unified collection cycle. Energy storage operation data includes power station identification, collection time, access node identification, state of charge, charging / discharging power, and available capacity. Grid operation data includes node power, line power flow, node voltage, and operating frequency. The system correlates these two types of data according to collection time and access node to form station-grid operation data. Based on this data, it calculates the adjustable capacity of the energy storage power station, the adjustment demand of the access node, and the response time period, generating an energy storage coordinated adjustment vector. The system further calculates the access node power deviation based on the node power difference between adjacent collection times, calculates the line power flow margin based on line power flow and rated line power flow, and calculates the node voltage deviation based on node voltage and rated node voltage. After dimensional unification and direction calibration, a grid regulation sensitivity field is formed.

[0029] In the coordinated regulation calculation, the system maps the energy storage regulation capacity to candidate regulation positions in the mountain gazelle optimization algorithm, and writes the response time period and access node identifier into the candidate regulation positions to form candidate position constraint records. The access node power deviation, line power flow margin, and node voltage deviation in the grid regulation sensitivity field are written into the migration direction update rules to guide the adjustment candidate positions to update towards access nodes with larger power deviations, tighter power flow margins, and more significant voltage deviations. The state of charge and available capacity are calculated to form the state of charge margin and written into the over-boundary rollback rules. When the candidate coordinated regulation vector causes the energy storage regulation capacity to exceed the state of charge margin, or causes the line power flow to exceed the line power flow margin, the system calculates the over-boundary regulation amount and rolls it back to the energy storage station that still has state of charge margin and power flow margin, generating a feasible coordinated regulation vector, and then outputs the coordinated and complementary regulation command between the energy storage station and the grid.

[0030] The trial operation used a dual-track comparison of historical load curves and real-time collected curves from consecutive operating days, maintaining consistent scheduling cycles, total rated power of the energy storage power station, available capacity, number of connected nodes, and rated power flow parameters of the lines. Conventional rule-based scheduling executes fixed charging and discharging according to peak and valley periods, while conventional group intelligent scheduling only optimizes based on power deviation and operating costs. The method of this invention introduces a grid regulation sensitivity field, an energy storage role table, and out-of-bounds rollback processing on the same data basis. The comparison results are shown in Table 1: Table 1 Comparison of Station-Network Coordination Adjustment Effects

[0031] As shown in Table 1, under the same energy storage scale and grid operation data, the method of this invention can significantly reduce the maximum line load factor, the number of node voltage overruns, and the duration of line power flow overruns. Conventional rule-based scheduling, lacking sensitivity judgment of access nodes, is prone to continuing to execute fixed charging and discharging commands when the local line power flow is already close to the limit, causing concentrated power flow pressure. Although conventional group intelligent scheduling improves the overall regulation effect, it lacks rollback processing for state of charge margin and power flow margin, still generating unexecutable commands. This invention determines the regulation direction through the grid regulation sensitivity field and transfers the regulation amount exceeding the state of charge margin and line power flow margin to the energy storage station with margin through overrun rollback, reducing the number of SOC overruns to 0 and increasing the executability rate of regulation commands to 98.6%. This demonstrates that this invention can improve the grid-side power balance capability and node voltage support capability while ensuring the safe operation of energy storage.

[0032] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A method for coordinated and complementary regulation between energy storage power stations and power grids based on swarm intelligence, characterized in that, Includes the following steps: Step 1: Collect energy storage operation data and grid operation data, associate them with access nodes according to the collection time, and generate station-grid operation data; Step 2: Determine the energy storage regulation capacity, response time period, and access nodes based on the station network operation data, and generate an energy storage coordinated regulation vector; Step 3: Generate a grid regulation sensitivity field based on the power deviation of the access node, the power flow margin of the line, and the voltage deviation of the node; Step 4: Write the grid regulation sensitivity field into the migration direction update rule of the mountain gazelle optimization algorithm, and write the state of charge margin into the out-of-bounds rollback rule to construct an improved mountain gazelle optimization algorithm. Step 5: The improved mountain gazelle optimization algorithm is used to process the energy storage coordinated regulation vector, and an energy storage role table is generated based on the grid regulation sensitivity field; Step 6: Update the charge and discharge regulation direction in the energy storage coordinated regulation vector according to the energy storage role table, and generate candidate coordinated regulation vectors; Step 7: Perform state of charge verification and power flow safety verification on the candidate coordinated regulation vectors, and roll back the out-of-bounds regulation amount to the energy storage power station with state of charge margin and power flow margin to generate a feasible coordinated regulation vector. Step 8: Determine the target regulation capacity, target response time period, and target access node based on the feasible coordinated regulation vector, and output the coordinated and complementary regulation command between the energy storage power station and the power grid.

2. The method for coordinated and complementary regulation of energy storage power stations and power grids based on swarm intelligence according to claim 1, characterized in that, Step one specifically involves: Energy storage operation data is collected according to the unified collection cycle of energy storage power station and power grid. The energy storage operation data includes power station identification, collection time, access node identification, state of charge, charging and discharging power and available capacity. Power grid operation data is collected according to a unified collection cycle. The power grid operation data includes the collection time, access node identifier, node power, line power flow, node voltage and operating frequency. Using the time of data collection as the time-related field, the energy storage operation data is matched with the execution time of the power grid operation data to generate time-related operation records; Using the access node identifier as the node association field, the node matching is performed on the time-related operation records to generate a station-network association record between the energy storage power station and the access node; Perform missing field validation, numerical boundary validation, and temporal continuity validation on the station network associated records, filter out abnormal operation records, and arrange them according to the collection time and access node identifier to generate station network operation data.

3. The method for coordinated and complementary regulation of energy storage power stations and power grids based on swarm intelligence according to claim 1, characterized in that, Step two specifically involves: Extract energy storage operation fields and power grid operation fields from the station network operation data, and form node operation records according to the collection time and access node identifier; The adjustable capacity of the energy storage power station is calculated based on the state of charge, charging and discharging power and available capacity in the node operation record, and the adjustment demand of the access node is determined based on the node power, line power flow and node voltage. Match the adjustable capacity of the energy storage power station with the adjustment demand of the access node to determine the energy storage adjustment capacity, and establish a correspondence between the energy storage adjustment capacity and the access node identifier; Based on the data collection time and changes in operating frequency, the response period for energy storage power stations to participate in regulation is determined, and a correspondence is established between the response period and the energy storage regulation capacity. Vector units are constructed according to energy storage regulation capacity, response time period and access node identifier, and the vector units are arranged according to the acquisition time to generate energy storage collaborative regulation vector.

4. The method for coordinated and complementary regulation of energy storage power stations and power grids based on swarm intelligence according to claim 1, characterized in that, Step three specifically involves: Extract access node identifier, acquisition time, node power, line power flow and node voltage from the station network operation data, and form node status records according to acquisition time and access node identifier; The access node power deviation is calculated based on the node power difference between adjacent acquisition times in the node status record, and a correspondence is established between the access node power deviation and the access node identifier. Calculate the line power flow margin based on the line power flow and line rated power flow in the node status record, and establish a correspondence between the line power flow margin and the access node identifier; Calculate the node voltage deviation based on the node voltage and node rated voltage in the node status record, and establish a correspondence between the node voltage deviation and the access node identifier; The dimensions of the access node power deviation, line power flow margin and node voltage deviation are unified and directionally calibrated, and then collected according to the access node identification to generate a power grid regulation sensitivity field.

5. The method for coordinated and complementary regulation of energy storage power stations and power grids based on swarm intelligence according to claim 1, characterized in that, Step four specifically involves: Extract the energy storage regulation capacity, response time period and access node identifier from the energy storage coordinated regulation vector, map the energy storage regulation capacity to the regulation candidate position in the mountain gazelle optimization algorithm, and write the response time period and access node identifier into the regulation candidate position to form a candidate position constraint record. Extract the access node power deviation, line power flow margin and node voltage deviation corresponding to the access node identifier from the power grid regulation sensitivity field, and perform matching between the access node identifier and the candidate location constraint record to generate candidate location sensitivity record; The migration direction weight is calculated based on the candidate position sensitivity record, and the migration direction weight is written into the migration direction update rule of the mountain gazelle optimization algorithm so that the adjustment candidate position is updated according to the adjustment direction determined by the power grid adjustment sensitivity field. Extract the state of charge and available capacity corresponding to the access node identifier from the station network operation data, calculate the state of charge margin based on the state of charge and available capacity, and write the state of charge margin into the adjustment candidate position to form the candidate position margin record. Based on the candidate position margin record, an out-of-bounds rollback rule is established. The out-of-bounds rollback rule is used to perform capacity rollback on the adjustment candidate positions that exceed the state of charge margin, and to limit the adjustment candidate positions after capacity rollback to the capacity range corresponding to the state of charge margin. The mountain gazelle optimization algorithm with written migration direction update rules and out-of-bounds rollback rules was identified as the improved mountain gazelle optimization algorithm.

6. The method for coordinated and complementary regulation of energy storage power stations and power grids based on swarm intelligence according to claim 1, characterized in that, Step five specifically involves: The energy storage coordinated regulation vector is input into the improved mountain gazelle optimization algorithm, and the energy storage regulation capacity, response time period and access node identifier are read according to the collection time to generate regulation candidate records; Match the candidate records of regulation with the grid regulation sensitivity field according to the access node identifier, and write the access node power deviation, line power flow margin and node voltage deviation into the candidate records of regulation to generate regulation sensitivity matching records. The main regulation score is generated based on the access node power deviation in the regulation-sensitive matching record, the boundary protection score is generated based on the line power flow margin, and the backup compensation score is generated based on the node voltage deviation. The main regulation score, boundary protection score, and backup compensation score are written into the same regulation-sensitive matching record to generate a role score record. Perform numerical sorting on the main adjustment score, boundary protection score and backup compensation score in the role rating record, write the role tag corresponding to the first ranked score into the role rating record, and generate a role tag record. Arrange the energy storage regulation capacity, response time period, access node identifier, and role tag in the role tag record according to the collection time to generate an energy storage role table.

7. The method for coordinated and complementary regulation of energy storage power stations and power grids based on swarm intelligence according to claim 1, characterized in that, Step six specifically involves: Read the data collection time, energy storage regulation capacity, response time period, access node identifier, and role tag from the energy storage role table, and generate a role direction update record; Match the role direction update record with the energy storage collaborative adjustment vector according to the acquisition time and access node identifier, write the role mark into the corresponding vector unit in the energy storage collaborative adjustment vector, and generate the role matching adjustment vector; The charging and discharging adjustment direction is determined based on the role marker in the role matching adjustment vector. When the role marker corresponds to the main adjustment score, the charging and discharging adjustment direction is determined according to the access node power deviation. When the role marker corresponds to the boundary protection score, the charging and discharging adjustment direction is determined according to the line power flow margin. When the role marker corresponds to the standby compensation score, the charging and discharging adjustment direction is determined according to the node voltage deviation. Write the charging / discharging adjustment direction into the corresponding vector unit in the character matching adjustment vector to generate the direction update adjustment vector; The vector units are organized according to the energy storage regulation capacity, response time period, access node identifier and charge / discharge regulation direction in the direction update regulation vector, and the vector units are arranged according to the acquisition time to generate candidate coordinated regulation vectors.

8. The method for coordinated and complementary regulation of energy storage power stations and power grids based on swarm intelligence according to claim 1, characterized in that, Step seven specifically involves: Extract the acquisition time, energy storage regulation capacity, response time period, access node identifier, and charging / discharging regulation direction from the candidate coordinated regulation vector to generate candidate verification records; The candidate verification records are matched with the station network operation data according to the collection time and access node identifier, the state of charge margin and line power flow margin are read, and margin matching verification records are generated. Based on the energy storage regulation capacity, charge / discharge regulation direction and state of charge margin in the margin matching verification record, perform state of charge verification, calculate the capacity difference between the energy storage regulation capacity and the state of charge margin, and generate the state of charge out-of-bounds amount. Based on the energy storage regulation capacity, charging and discharging regulation direction and line power flow margin in the margin matching verification record, perform power flow safety verification, calculate the power flow difference caused by the energy storage regulation capacity exceeding the line power flow margin, and generate the power flow exceeding limit. The over-limit quantity of state of charge and the over-limit quantity of power flow are merged into the over-limit regulation quantity, and the over-limit regulation quantity is written into the candidate verification record according to the collection time and access node identifier. Energy storage power stations with both state of charge margin and power flow margin are selected from the station network operation data. The out-of-bounds regulation is rolled back to the selected energy storage power stations, and the candidate coordinated regulation vector is updated to generate a feasible coordinated regulation vector.

9. The method for coordinated and complementary regulation of energy storage power stations and power grids based on swarm intelligence according to claim 1, characterized in that, Step eight specifically involves: Extract the acquisition time, energy storage regulation capacity, response time period, access node identifier and charging / discharging regulation direction from the feasible coordinated regulation vector to generate a feasible regulation record; The feasible adjustment record is matched with the grid adjustment sensitivity field according to the access node identifier. The access node power deviation, line power flow margin and node voltage deviation are written into the feasible adjustment record to generate the target screening record. Based on the access node power deviation in the target screening record, a capacity priority value is generated; based on the line power flow margin, a safety priority value is generated; based on the node voltage deviation, a voltage priority value is generated; and the capacity priority value, safety priority value, and voltage priority value are merged into the target adjustment priority value. The target screening records are numerically sorted according to the target adjustment priority value, and the target screening record with the first position in the sort is determined as the target vector unit. The target regulation capacity, target response time period, and target access node are determined from the target vector unit, and the target regulation capacity, target response time period, target access node, and charging / discharging regulation direction are written into the instruction record. The energy storage power station and the power grid are then output as a coordinated and complementary regulation instruction.

Images