A wind power simulation method and device based on digital twinning and a medium

Abstract

The application discloses a wind power simulation method and device based on digital twinning, and a medium, relates to the technical field of wind power twinning grid-connected simulation, and comprises the following steps: collecting wind farm entity operation data, performing edge-time calibration on the wind farm entity operation data, and generating a wind power twinning input boundary; according to the wind turbine space position in the wind power twinning input boundary, the wind-affected partition is backstepped, the wind-affected partition is bound to the blade aerodynamic node, the aerodynamic load in the blade aerodynamic node is transmitted to the electromechanical coupling node and then connected to the grid-connected port, and a wind power coupling simulation body is generated; the slow variables of the flow field and the electromechanical fast variables are extracted from the wind power coupling simulation body, the slow variables of the flow field are extrapolated, the aggregated value of the electromechanical fast variables is written back, the error propagation direction is marked, and an error direction interface chain is generated. The application improves the time sequence consistency, error traceability, boundary credibility and operation deduction stability of wind power simulation.

Description

Technical Field

[0001] This invention relates to the field of wind power twin grid connection simulation technology, and in particular to a wind power simulation method, equipment and medium based on digital twins. Background Technology

[0002] With the large-scale development of wind farms, the increasing size of turbines, and the rising proportion of renewable energy connected to the grid, wind power simulation methods are gradually evolving from simple aerodynamic calculations and local electromechanical transient analysis to a combined approach that considers wind farm flow, turbine loads, transmission response, and grid connection status. Existing methods typically collect operational data such as wind speed, wind direction, power, speed, pitch angle, grid voltage, and grid frequency, and combine these with computational fluid dynamics, aerodynamic load models, electromechanical transient models, and power prediction models to simulate and analyze the wind farm operation. With the gradual application of digital twin methods in the wind power field, wind power simulation further emphasizes the dynamic mapping between physical operational data and virtual simulation models. By synchronously describing the turbine's spatial location, wake effects, blade loads, generator response, and grid connection port status, it provides data support for wind farm operation assessment, power prediction, control strategy verification, and fault risk analysis.

[0003] However, in the actual operation of wind farms, changes in incoming flow, wake shielding, spatial distribution of turbines, blade aerodynamic response, electromechanical coupling of the transmission chain, and fluctuations at the grid connection port often have different time scales and different transmission directions. Changes in the flow field usually manifest as a slow spatial diffusion and extrapolation process, while electromechanical response and grid connection disturbances may form rapid feedback in a short period of time. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides a wind power simulation method based on digital twins, which solves the problems of insufficient time-series coordination between slow variables in the wind farm and fast variables in electromechanical systems, and difficulty in tracking the error direction.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: In a first aspect, the present invention provides a wind power simulation method based on digital twins, comprising: collecting wind farm physical operation data, performing edge-time calibration on the wind farm physical operation data, and generating a wind power twin input boundary; deducing the wind-receiving zone based on the wind turbine spatial position in the wind power twin input boundary, binding the wind-receiving zone to the blade aerodynamic node, and transferring the aerodynamic load in the blade aerodynamic node to the electromechanical coupling node and then connecting it to the grid connection port, thereby generating a wind power coupled simulation body; extracting slow flow field variables and fast electromechanical variables from the wind power coupled simulation body, extrapolating the slow flow field variables and writing back the aggregated value of the fast electromechanical variables, marking the error propagation direction, and generating an error direction interface chain; determining the deviation propagation inflection point based on the error direction interface chain, pushing the extrapolated value of the slow variable forward along the inflection point, and pushing the aggregated value of the fast variable backward along the inflection point, thereby generating a real-time simulation operation trajectory; performing a peer comparison between the real-time simulation operation trajectory and the wind power twin input boundary, and backfeeding and correcting the positions where the deviation falls at the wind farm end and the electromechanical end, respectively, thereby generating a wind power simulation trajectory.

[0007] As a preferred embodiment of the wind power simulation method based on digital twins described in this invention, the specific steps for collecting wind farm entity operation data, performing edge-time calibration on the wind farm entity operation data, and generating wind power twin input boundaries are as follows: The wind farm physical operation data is classified and arranged to obtain the wind power boundary side position, and the wind power boundary side position is mapped to the measurement point position to generate the boundary measurement point sequence. Based on the boundary measurement point sequence, the acquisition time of different measurement point locations is uniformly aligned to the same simulation time axis, and the boundary positions of wind power operation parameters in the same simulation time axis are aligned to generate a boundary-time aligned sequence. By performing bidirectional boundary-time locking on the edge-time alignment sequence, the wind power operation parameters are simultaneously fixed to the corresponding wind power boundary side position, generating a wind power twin input boundary.

[0008] As a preferred embodiment of the wind power simulation method based on digital twins described in this invention, the specific steps for deriving the wind-receiving zone based on the spatial position of the wind turbine in the wind power twin input boundary are as follows: The spatial position of the wind turbine is extracted from the wind power twin input boundary. The spatial position of the wind turbine is obliquely expanded by the prevailing wind direction. The distance in the direction of the incoming flow is taken as the downwind projection and the distance deviating from the center line of the incoming flow is taken as the crosswind offset. The wind displacement sequence is obtained, and the wind intensity of each wind turbine is marked by the windward relationship to generate the wind inversion sequence. Based on the wind-induced sequence, wind turbines with adjacent wind strength, continuous downwind projection, and crosswind offset that do not cross the wake shielding boundary are grouped into the same area. The locations of wind turbines with abrupt changes in wind strength are used as partitioning breakpoints to generate wind-induced zones.

[0009] As a preferred embodiment of the wind power simulation method based on digital twins described in this invention, the specific steps of binding the wind-receiving zone to the blade aerodynamic node and transferring the aerodynamic load in the blade aerodynamic node to the electromechanical coupling node and then connecting it to the grid connection port to generate a wind power coupled simulation body are as follows: According to the wind-receiving zone, the windward adjustment parameters in each zone are connected to the blade aerodynamic nodes under the wind turbine number, and the load-bearing positions are arranged in the radial order from the blade root to the blade tip to generate an aerodynamic node binding chain. Based on the aerodynamic node binding chain, the aerodynamic load in the blade aerodynamic node is sequentially transferred along the electromechanical transfer path, and the transmission receiving position, torque receiving position and electrical receiving position are determined according to the electromechanical transfer path to generate the electromechanical coupling node. Electromechanical response parameters are extracted from the electromechanical coupling nodes and bound to the electromechanical coupling nodes to generate an electromechanical load transfer chain. Based on the electromechanical load transmission chain, the grid-connected electrical state corresponding to the electromechanical coupling node is connected to the grid-connected port, and the aerodynamic load and electromechanical response parameters are connected in series according to the electrical connection relationship of the grid-connected port to generate a wind power coupling simulation body.

[0010] As a preferred embodiment of the wind power simulation method based on digital twins described in this invention, the specific steps of extracting slow flow field variables and fast electromechanical variables from the wind power coupled simulation body, extrapolating the slow flow field variables, and writing back the aggregated values ​​of the fast electromechanical variables are as follows: Slow flow field variables and fast electromechanical variables are extracted from the wind power coupled simulation body and attached to the corresponding interface receiving positions. A slow-fast interface alignment table is obtained. According to the slow-fast interface alignment table, the slow flow field variables in the upstream interface receiving position are recursively pushed to the adjacent downstream interface receiving positions according to the windward sequence. The variables are written to the corresponding blade aerodynamic nodes to generate a slow flow field variable extrapolation chain. Based on the slow variable extrapolation chain of the flow field, the grid-connected electrical state is retrieved from the grid-connected port. The power fluctuations in the grid-connected electrical state are reversed and matched to the electromechanical transfer path. The electromechanical transfer path and the electromechanical response parameters are peak-valley matched and folded into a fast variable write-back cluster. The fast variable write-back cluster is pressed into the corresponding interface receiving position to generate a slow-fast residual comparison chain.

[0011] As a preferred embodiment of the wind power simulation method based on digital twins described in this invention, the specific steps for marking the error propagation direction and generating the error direction interface chain are as follows: Based on the slow and fast residual comparison chain, the flow field extrapolation deviation and electromechanical write-back deviation at each interface acceptance position are compared in opposite directions. The side with the dominant deviation is marked as the error initiation side, and the error propagation direction is marked along the interface acceptance sequence to generate error propagation direction markings. Based on the error propagation direction marking, positions with consistent error propagation directions and continuous interface acceptance order are connected in series to form the same error propagation segment, and positions where the error propagation direction is reversed are marked as direction transition points, thus generating an error direction interface chain.

[0012] As a preferred embodiment of the wind power simulation method based on digital twins described in this invention, the specific steps for determining the deviation propagation inflection point based on the error direction interface chain, pushing the extrapolated value of the slow variable forward along the inflection point, and pushing the aggregated value of the fast variable backward along the inflection point to generate the real-time simulation trajectory are as follows: Based on the error direction interface chain, find the position where the error propagation direction is reversed along the interface acceptance sequence, and align the found position with the direction transition point to generate the deviation transmission inflection point. Based on the deviation propagation inflection point, the interface receiving position before the deviation propagation inflection point is determined as the slow variable forward push segment, and the interface receiving position after the deviation propagation inflection point is determined as the fast variable reverse push segment. The extrapolated values ​​of the slow variables in the flow field are pushed to the deviation transmission inflection point along the windward sequence, and the windward phase is matched with the interface receiving position at the deviation transmission inflection point to generate the slow variable inflection point push chain. Based on the slow variable inflection point push chain, the electromechanical fast variable aggregated value is pressed back along the electromechanical transfer path from the grid port to the deviation transfer inflection point, and the electrical end is reversed and connected to the interface receiving position at the deviation transfer inflection point to generate the fast variable inflection point push chain. Based on the forward chain of slow variable inflection points and the backward chain of fast variable inflection points, the deviation propagation inflection point is taken as the bidirectional continuation position. The extrapolated values ​​of slow variables in the flow field and the aggregated values ​​of fast electromechanical variables are sequentially sealed and continuously connected according to the simulation time slices to generate the real-time simulation running trajectory.

[0013] As a preferred embodiment of the wind power simulation method based on digital twins described in this invention, the specific steps for comparing the real-time simulation trajectory with the wind power twin input boundary, and correcting deviations at the wind farm end and electromechanical end respectively to generate the wind power simulation trajectory are as follows: The real-time simulation trajectory is aligned with the wind power twin input boundary in terms of time and position. The extrapolated values ​​of slow variables in the flow field and the aggregated values ​​of fast electromechanical variables are compared in the same slot to generate a double-end deviation comparison chain. Based on the double-end deviation comparison chain, the wind field deviation and electromechanical deviation are determined to be of their source, and the positions of the wind field end deviation, electromechanical end deviation, and double-end overlap are marked to generate deviation source markers. Based on the deviation source markers, the deviation positions at the wind farm end are fed back to the wind-receiving zone, and the deviation positions at the electromechanical end are fed back to the electromechanical transfer path. The boundary parameters and peak-valley engagement positions are corrected respectively to generate the wind power simulation trajectory.

[0014] In a second aspect, the present invention provides a computer device including a memory and a processor, wherein the memory stores a computer program, wherein the computer program, when executed by the processor, implements any step of the wind power simulation method based on digital twins as described in the first aspect of the present invention.

[0015] Thirdly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the wind power simulation method based on digital twins as described in the first aspect of the present invention.

[0016] The beneficial effects of this invention are as follows: By forming the wind power twin input boundary through digital twin edge-time calibration, and by back-deriving the aerodynamic, electromechanical connection and grid connection ports of the blades through wind-receiving zones, a wind power digital twin coupled simulation link is constructed; then, by aligning slow and fast variables, connecting the deviation transmission inflection point bidirectionally, and returning the deviation to its source at both ends, the slow diffusion of the wind field and the fast feedback of electromechanical systems are coordinated and aligned within the same simulation cycle, thereby improving the temporal consistency, error traceability, boundary credibility and operational simulation stability of wind power simulation. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart of the overall process for wind power simulation based on digital twins.

[0019] Figure 2 A flowchart for constructing a wind power coupled simulation system.

[0020] Figure 3 The flowchart for generating the error direction interface chain and deviation propagation inflection point.

[0021] Figure 4 This is a flowchart of the two-end deviation comparison and recharge correction.

[0022] Figure 5 This is a comparison chart of convergence of errors in the two-end deviation source attribution and backfilling.

[0023] Figure 6 A magnified comparison of peak and valley grid-connected power. Detailed Implementation

[0024] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0025] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0026] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0027] Reference Figures 1-6 As one embodiment of the present invention, this embodiment provides a wind power simulation method based on digital twins, comprising the following steps: S1. Collect wind farm physical operation data, perform edge-time calibration on the wind farm physical operation data, and generate wind power twin input boundaries.

[0028] S1.1 Classify and arrange the wind farm entity operation data, obtain the wind power boundary side position, and map the wind power boundary side position to the measurement point position to generate the boundary measurement point sequence.

[0029] It should be noted that when classifying and arranging the operational data of wind farm entities, the data is first sorted item by item according to the data source location, data type, and acquisition sequence. Wind speed, wind direction, rotor speed, pitch angle, power generation, and grid-connected voltage and current are respectively assigned to the corresponding operational categories. Each operational category is then arranged into location-based data groups according to the turbine number, spatial orientation, and acquisition time sequence. Based on the location-based data groups, the boundary assignment locations are identified item by item along the windward side, leeward side, turbine side, and grid-connected side of the wind farm. The boundary assignment locations that can characterize the wind farm input, turbine response, and grid-connected output are determined as wind power boundary positions. The wind power boundary positions are then mapped one by one to the actual deployed measurement points. The windward side wind power boundary positions are mapped to wind speed and wind direction measurement points, the turbine side wind power boundary positions are mapped to turbine operation measurement points, and the grid-connected side wind power boundary positions are mapped to electrical measurement points. The boundary measurement point sequence is generated by continuously arranging the wind power boundary positions according to the spatial order and acquisition sequence of the measurement points. S1.2 Based on the boundary measurement point sequence, the acquisition time of different measurement point locations is uniformly aligned to the same simulation time axis, and the boundary positions of wind power operation parameters in the same simulation time axis are aligned to generate a boundary time alignment sequence.

[0030] It should be noted that, based on the boundary measurement point sequence, the corresponding acquisition time and wind power operation parameters are extracted item by item according to the measurement point location, and the acquisition time is arranged into a measurement point time queue in chronological order. Using the time scale of the same simulation time axis as the fitting reference, the acquisition time of each measurement point location is mapped to the nearest time scale one by one. Acquisition times located between adjacent time scales are interpolated according to the adjacent acquisition times, so that the wind power operation parameters of different measurement point locations fall into the same simulation time axis. Continuing to follow the correspondence between the wind power boundary side position and the measurement point position in the boundary measurement point sequence, the wind power operation parameters within the same simulation time axis are placed into the corresponding boundary positions on the windward side, turbine side, leeward side, and grid-connected side, respectively. For wind power operation parameters with missing or misaligned boundary positions under the same time scale, adjacent measurement points are used to fill in the gaps and the boundary order is corrected. Finally, the same simulation time axis, measurement point position, boundary position, and wind power operation parameters are continuously bound together to generate an edge-time aligned sequence.

[0031] S1.3. Perform bidirectional boundary time locking on the edge-time alignment sequence to simultaneously fix the wind power operation parameters to the corresponding wind power boundary side position, generating wind power twin input boundaries.

[0032] It should be noted that when performing bidirectional boundary time locking on the edge-time aligned sequence, firstly, continuous time lock slots are divided along the same simulation time axis, and boundary side position lock slots are established according to the windward side, turbine side, leeward side, and grid-connected side. Based on the edge-time aligned sequence, the time scale, measurement point position, and wind power boundary side position corresponding to each wind power operation parameter are extracted one by one. The time scale is subtracted into the corresponding time lock slot, and the measurement point position is subtracted back into the corresponding boundary side position lock slot along the correspondence between the wind power boundary side position and the measurement point position, forming a bidirectional locking relationship between the time lock slot and the boundary side position lock slot. In the direction of the time lock slot, it is verified whether the wind power operation parameters on the windward side, turbine side, leeward side, and grid-connected side are complete at the same time scale, and any missing parameters are checked. The position is filled by the same type of wind power operation parameters of adjacent measuring points. For wind power operation parameters that fall into the wrong wind power boundary position within the same time scale, they are corrected to the corresponding boundary position locking slot according to the measuring point position. In the direction of the boundary position locking slot, the continuity of wind power operation parameters of adjacent time scales under the same wind power boundary position is verified. Wind power operation parameters with time scale offset are back-attached according to the front and back time locking slots so that the wind power operation parameters do not deviate from the corresponding time scale or the corresponding wind power boundary position. Finally, the same simulation time axis, time locking slot, wind power boundary position, measuring point position and wind power operation parameters after bidirectional coupling are fixedly bound to generate the wind power twin input boundary.

[0033] S2. Based on the wind turbine spatial position in the wind power twin input boundary, the wind receiving zone is deduced, the wind receiving zone is bound to the blade aerodynamic node, and the aerodynamic load in the blade aerodynamic node is transferred to the electromechanical coupling node and then connected to the grid connection port to generate a wind power coupling simulation body.

[0034] S2.1 Extract the spatial position of the wind turbine from the wind power twin input boundary, and diagonally expand the spatial position of the wind turbine by the main wind direction. Take the distance in the direction of the incoming flow as the downwind projection and the distance away from the center line of the incoming flow as the crosswind offset to obtain the wind-receiving displacement sequence. Combine the wind-facing sequence with the wind-facing sequence to mark the wind intensity of each wind turbine and generate the wind-receiving reverse sequence.

[0035] It should be explained that the spatial position of each wind turbine is extracted from the wind power twin input boundary. The prevailing wind direction is used as the diagonal expansion reference. The spatial position of the wind turbine is projected in the forward direction along the incoming flow direction and in the lateral direction perpendicular to the incoming flow direction. The distance in the incoming flow direction between the spatial position of the wind turbine and the windward reference position is taken as the downwind projection amount. The lateral deviation distance between the spatial position of the wind turbine and the center line of the incoming flow is taken as the crosswind offset amount. Then, the spatial positions of the wind turbines are arranged from the windward side to the leeward side according to the downwind projection amount, and the crosswind offset amount is incorporated into the arrangement order to form the windward displacement sequence.

[0036] Based on the windward displacement sequence, wind turbines with forward downwind projection and crosswind offset close to the center line of the incoming flow are marked as having strong winds, while wind turbines with backward downwind projection and crosswind offset falling within the range of the preceding wind turbine's obstruction are marked as having weak winds. Wind turbines with similar downwind projection but different crosswind offsets are corrected for wind intensity according to their degree of deviation from the center line of the incoming flow. Then, by combining the windward sequence, the wind intensity of each wind turbine is bound to the corresponding position in the windward displacement sequence to generate a windward reverse sequence.

[0037] It should also be noted that the wind-receiving reverse sequence is a record of wind-receiving order derived from the spatial position of the wind turbines in the wind power twin input boundary and the prevailing wind direction. It mainly determines the wind-facing order of each wind turbine by the downwind projection, judges the degree of deviation of each wind turbine from the center line of the incoming flow by the crosswind offset, and marks the strong wind, weak wind, or wind-receiving transition state by combining the wake shielding range. The wind-receiving reverse sequence is used to convert the spatial distribution of wind turbines into wind-receiving transmission relationship, providing a spatial wind-receiving basis for subsequent division of wind-receiving zones, binding of blade aerodynamic nodes, and recursive deduction of slow variables in the flow field.

[0038] S2.2 Based on the wind-induced sequence, wind turbines with adjacent wind strength, continuous downwind projection, and crosswind offset that do not cross the wake shielding boundary are grouped into the same area. The locations of wind turbines with abrupt changes in wind strength are used as partitioning breakpoints to generate wind-induced zones.

[0039] It should be noted that, based on the windward reverse sequence, the process is carried out sequentially from the windward side fan to the leeward side fan according to the windward order, and the wind intensity, downwind projection and crosswind offset between adjacent fans are continuously verified.

[0040] When the wind intensity of adjacent wind turbines remains at adjacent levels, the downwind projection is continuously connected along the direction of the incoming flow, and the crosswind offset is still within the same wake shielding boundary, the adjacent wind turbines are classified into the same area, and the wind turbine locations within the same area continue to be used as the starting range for the next round of continuous verification; when the wind intensity of adjacent wind turbines changes from strong to weak or from weak to strong and crosses adjacent levels, or the downwind projection becomes discontinuous, or the crosswind offset crosses the wake shielding boundary, the corresponding wind turbine location is locked as a zone breakpoint; the wind-receiving reverse sequence is divided according to the zone breakpoint, and each divided zone is bound to the wind turbine location, wind intensity, downwind projection, and crosswind offset within the zone to generate wind-receiving zones.

[0041] S2.3 According to the windward zone, the windward adjustment parameters in each zone are connected to the blade aerodynamic nodes under the wind turbine number, and the load-bearing positions are arranged in the radial order from the blade root to the blade tip to generate the aerodynamic node binding chain.

[0042] It should be noted that, based on the windward zones, the windward adjustment parameters within each zone are first extracted according to the wind intensity and the order of windward movement. These windward adjustment parameters are then paired with the corresponding turbine numbers to form windward parameter pairing items under each turbine number. Based on these windward parameter pairing items, the blade aerodynamic nodes corresponding to each turbine number are located, and the windward adjustment parameters are attached to the windward input positions of these blade aerodynamic nodes, ensuring that each turbine number's blade aerodynamic node receives the windward adjustment parameters within its corresponding zone. Continuing along the radial sequence from the blade root to the blade tip, the blade aerodynamic nodes are divided into continuous load-bearing positions, and the windward adjustment parameters are sequentially distributed radially to the load-bearing positions corresponding to the blade root, middle extension, and adjacent blade tip sections. Finally, the turbine number, blade aerodynamic nodes, windward adjustment parameters, and load-bearing positions are concatenated according to the zone order and radial order to generate an aerodynamic node binding chain.

[0043] It should also be noted that the aerodynamic node binding chain is used to connect the windward adjustment parameters in the wind-receiving zone to the blade aerodynamic nodes according to the wind turbine number, and to arrange the load-bearing positions in the radial order of the blade root section, the middle extension section, and the blade tip adjacent section, so that the wind intensity, the windward sequence, and the aerodynamic load position of the blade form a continuous correspondence; through the aerodynamic node binding chain, the wind field side wind difference can be accurately transmitted to the blade aerodynamic side, providing a clear load source basis for the subsequent aerodynamic load transmission along the electromechanical transfer path, the formation of electromechanical coupling nodes, and the grid connection port response tracking.

[0044] S2.4 Based on the aerodynamic node binding chain, the aerodynamic load in the blade aerodynamic node is transferred sequentially along the electromechanical transfer path, and the transmission receiving position, torque receiving position and electrical receiving position are determined according to the electromechanical transfer path to generate the electromechanical coupling node.

[0045] It should be noted that, based on the aerodynamic node binding chain, the aerodynamic loads in the blade aerodynamic nodes are first arranged according to the wind turbine number and load receiving position, and then arranged into aerodynamic load receiving rows in the radial order of the blade root section, the middle extension section and the blade tip adjacent section.

[0046] Based on the aerodynamic load receiving line, the receiving form of the aerodynamic load is traced step by step along the electromechanical transfer path when the aerodynamic load is transferred from the aerodynamic node of the blade to the rear. The first receiving position where the aerodynamic load changes from radial load on the blade to rotational transmission is determined as the transmission receiving position. At the transmission receiving position, the load direction is aligned and engaged with the transmission rotation direction to form the transmission load receiving relationship.

[0047] Based on the transmission load acceptance relationship, the positions where the rotational transmission quantity is converted into the torque transmission quantity are traced along the electromechanical transfer path. The positions that can accept the transmission rotation direction and form the torque transmission parameter are determined as torque acceptance positions, and torque load acceptance relationships are formed according to the order of the fan number, load acceptance position, and transmission acceptance position. Based on the torque load acceptance relationship, the positions where the torque transmission parameter is converted into the electrical response are traced along the electromechanical transfer path. The positions that can accept the torque transmission parameter and form a corresponding order with the voltage, frequency, power, and phase of the grid-connected port are determined as electrical acceptance positions, and electrical load acceptance relationships are formed. Finally, the transmission acceptance position, torque acceptance position, and electrical acceptance position are connected in series according to the order of the electromechanical transfer path, and the transmission load acceptance relationship, torque load acceptance relationship, and electrical load acceptance relationship are continuously bound to the fan number, blade aerodynamic node, and the source of aerodynamic load to generate electromechanical coupling nodes.

[0048] S2.4 Extract the electromechanical response parameters from the electromechanical coupling node and bind the electromechanical response parameters to the electromechanical coupling node to generate the electromechanical load transfer chain.

[0049] It should be noted that, based on the electromechanical coupling node, the transmission receiving position, torque receiving position and electrical receiving position are first located according to the fan number and electromechanical transfer path. Then, the speed change, torque change, electric power change and phase change caused by the transfer of aerodynamic load are extracted in each receiving position as electromechanical response parameters.

[0050] The electromechanical response parameters are used to establish response alignment rows according to the aerodynamic load source, the receiving position source, and the time scale. The response alignment rows are then deducted back to the corresponding electromechanical coupling nodes, so that speed changes correspond to transmission receiving positions, torque changes correspond to torque receiving positions, and electric power changes and phase changes correspond to electrical receiving positions. The response alignment rows are then verified for forward and backward transfer according to the electromechanical transfer path. The electromechanical response parameters formed at the previous receiving position are used as the transfer basis for the next receiving position. The path deduction is performed on the electromechanical response parameters whose receiving sequence is broken. The electromechanical response parameters that have completed the transfer verification are fixedly bound to the electromechanical coupling nodes to generate an electromechanical load transfer chain.

[0051] S2.5. Based on the electromechanical load transmission chain, connect the grid-connected electrical state corresponding to the electromechanical coupling node to the grid-connected port, and connect the aerodynamic load and electromechanical response parameters in series according to the electrical connection relationship of the grid-connected port to generate a wind power coupling simulation body.

[0052] It should be noted that, based on the electromechanical load transmission chain, the electromechanical coupling node is located according to the wind turbine number and electromechanical transfer path, and the grid-connected electrical state corresponding to the electromechanical response parameters is extracted from the electromechanical coupling node.

[0053] Connect the grid-connected electrical status to the grid-connected port in the order of voltage connection, frequency connection, power connection, and phase connection, and align the electrical connection relationships in the grid-connected port with the electromechanical connection paths item by item, so that the grid-connected port can accept the electrical changes in the electromechanical response parameters.

[0054] Based on the aligned electrical connection relationships, the aerodynamic loads in the blade aerodynamic nodes, the electromechanical response parameters in the electromechanical coupling nodes, and the grid-connected electrical states in the grid-connected ports are connected in series in the order of aerodynamic input, electromechanical conversion, and grid connection. During the series connection process, the correspondence between the wind turbine number, load connection position, and electromechanical transfer path is preserved. The electrical connection relationships of the aerodynamic loads, electromechanical response parameters, grid-connected electrical states, and grid-connected ports are merged into a continuous simulation chain to generate a wind power coupling simulation body.

[0055] It should also be noted that the electrical connection relationship is used to clarify the connection sequence of voltage, frequency, power, and phase at the grid connection port, so that the electromechanical response parameters are continuously connected to the grid connection electrical status, which facilitates the tracking of the source of grid connection fluctuations and subsequent deviation feedback.

[0056] S3. Extract the slow flow field variables and fast electromechanical variables from the wind power coupled simulation body, extrapolate the slow flow field variables and write back the aggregated values ​​of the fast electromechanical variables, mark the error propagation direction, and generate the error direction interface chain.

[0057] S3.1 Extract the slow flow field variables and fast electromechanical variables from the wind power coupled simulation body, and attach them to the corresponding interface receiving positions. Obtain the slow and fast interface alignment table. According to the slow and fast interface alignment table, based on the windward sequence, push the slow flow field variables in the upstream interface receiving position to the adjacent downstream interface receiving position, and write them into the corresponding blade aerodynamic node to generate the flow field slow variable extrapolation chain.

[0058] It should be noted that, from the wind power coupling simulation body, according to the order of wind-receiving zone, blade aerodynamic node, electromechanical coupling node and grid connection port, the slow variables of the flow field and the fast variables of the electromechanical system are extracted one by one, and the slow variables of the flow field and the fast variables of the electromechanical system are connected to the corresponding interface receiving positions according to the wind turbine number, the windward sequence and the electromechanical connection path, so as to obtain the slow and fast interface alignment table.

[0059] Based on the slow-fast interface alignment table, the slow variables of the flow field at the upstream interface receiving position are locked. Then, the adjacent downstream interface receiving positions are found according to the windward sequence. The slow variables of the flow field at the upstream interface receiving position are recursively extrapolated to the adjacent downstream interface receiving positions along the downwind projection direction. During the recursion process, the deviation degree corresponding to the crosswind offset and the shielding range corresponding to the wake shielding boundary are incorporated into the recursion correction so that the adjacent downstream interface receiving positions obtain the extrapolated values ​​of the slow variables of the flow field that match the windward zone. The extrapolated values ​​of the slow variables of the flow field are written into the corresponding blade aerodynamic nodes according to the fan number. The upstream interface receiving position, the adjacent downstream interface receiving position, the extrapolated values ​​of the slow variables of the flow field, and the blade aerodynamic nodes are connected in series to generate the extrapolation chain of the slow variables of the flow field.

[0060] It should also be noted that the slow variable extrapolation chain of the flow field is used to extrapolate slowly changing parameters such as wind speed, wind direction, downwind projection, crosswind offset, and wake shielding boundary at the upstream interface receiving position to the adjacent downstream interface receiving position according to the windward sequence, and to form a corresponding connection with the aerodynamic nodes of the blades. Through the slow variable extrapolation chain of the flow field, the spatial diffusion changes on the wind field side can be continuously transmitted to the downstream wind turbine receiving position, so that the subsequent electromechanical fast variable backwriting, slow-fast residual comparison, and error propagation direction marking have a clear basis for the wind field source, thereby improving the temporal continuity and boundary reliability of the simulation trajectory.

[0061] The expression for locking the slow flow field variables at the upstream interface receiving location is: ; in, For the target interface receiving position Locked-in slow variables in the upstream flow field; This is the set of candidate upstream interface receiving locations; Locking in the overall value upstream; The target interface is the receiving location; This is the candidate upstream interface receiving location; These are time scales within the same simulation time axis.

[0062] It should also be noted that the slow variables in the flow field originate from the wind speed, wind direction, wind-receiving zones, downwind projection, crosswind offset, and wake shielding boundary in the wind power twin input boundary. Their main function is to characterize the slow change process of the wind field in spatial diffusion and wind transmission. Their significance lies in providing boundary basis for the extrapolation of slow variables in the flow field, the recursive extrapolation of wind-receiving zones, and the wind field end deviation reinjection.

[0063] Electromechanical fast variables originate from the electromechanical response parameters and grid-connected electrical state formed after the aerodynamic load is transmitted along the electromechanical transfer path. Their main function is to characterize the rapid changes in blade loading, transmission conversion, power fluctuation, and grid-connected response. Their significance lies in providing response basis for electromechanical fast variable aggregation and rewriting, error propagation direction judgment, and electromechanical end deviation feedback.

[0064] S3.2 Based on the slow variable extrapolation chain of the flow field, retrieve the grid-connected electrical state from the grid-connected port, reverse the position of the power fluctuation in the grid-connected electrical state to the electromechanical transfer path, and fold the electromechanical transfer path and electromechanical response parameters into a fast variable write-back cluster after peak-valley interlocking. Press the fast variable write-back cluster into the corresponding interface receiving position to generate a slow-fast residual comparison chain.

[0065] It should be noted that, based on the slow variable extrapolation chain of the flow field, according to the transfer relationship between the interface receiving position and the grid connection port, the grid connection electrical status under the corresponding wind turbine number is retrieved from the grid connection port, and the power fluctuation start position, power fluctuation peak position and power fluctuation valley position are located in the grid connection electrical status. The power fluctuation start position, power fluctuation peak position and power fluctuation valley position are jointly determined as the power fluctuation arrival position.

[0066] Using the arrival location of power fluctuations as the starting point for reverse bonding, the electromechanical bonding path is traced back along the opposite bonding direction of the grid connection port, electromechanical coupling node, and blade aerodynamic node. The arrival location of power fluctuations is bonded step by step to the corresponding transmission receiving position, torque receiving position, and electrical receiving position in the electromechanical bonding path. The electromechanical response parameters corresponding to each receiving position in the electromechanical bonding path are then unfolded sequentially in time. The peak segments in the electromechanical response parameters are aligned with the peak positions of power fluctuations, and the valley segments in the electromechanical response parameters are aligned with the valley positions of power fluctuations. The turning point between the peak and valley segments is taken as the peak-valley biting position. Based on the peak-valley biting position, the electromechanical response parameters are folded to the adjacent interface receiving position to form a fast variable write-back cluster containing the correspondence between the arrival location of power fluctuations, the electromechanical bonding path, and the electromechanical response parameters. The fast variable write-back cluster is pushed into the corresponding interface receiving position in the flow field slow variable extrapolation chain and placed in parallel with the flow field slow variable extrapolation value in the corresponding interface receiving position to generate a slow-fast residual comparison chain.

[0067] It should also be noted that the location of power fluctuation arrival refers to the time and location of the first power change to appear, the peak value to be concentrated, or the valley value to be prominent in the grid-connected electrical state, and the electrical connection location. Its function is to reverse the fluctuation phenomenon of the grid-connected port to the electromechanical connection path; its significance is to provide a positioning basis for peak-valley alignment, fast variable write-back cluster construction, and electromechanical end deviation feedback.

[0068] The fast variable write-back cluster is a set of fast variables formed by folding the power fluctuations in the grid-connected electrical state to the peak segment, valley segment, and peak-valley engagement position of the electromechanical response parameters after the power fluctuations arrive at the location and are reversed along the electromechanical connection path. Its function is to write back the fast fluctuations of the grid-connected port to the corresponding interface receiving position. Its significance is to provide a basis for slow and fast residual comparison, error propagation direction judgment, and electromechanical end deviation feedback.

[0069] The slow-fast residual comparison chain is used to place the extrapolated values ​​of slow variables and the write-back clusters of fast variables in the same interface for parallel comparison. By comparing the wind field extrapolation deviation and the electromechanical write-back deviation, it is determined whether the deviation mainly comes from the upstream flow field transmission or the grid-connected electromechanical response write-back. Its significance lies in incorporating the forward recursive process of slow variables and the reverse write-back process of fast variables into the same alignment relationship, so that the subsequent error propagation direction marking, direction transition point positioning, and wind field end or electromechanical end recharge correction have a clear basis, thereby improving the accuracy of deviation source tracking and the pertinence of operation trajectory correction in wind power simulation.

[0070] S3.3. Based on the slow-fast residual comparison chain, the flow field extrapolation deviation and electromechanical write-back deviation at each interface acceptance position are compared in opposite directions. The side with the dominant deviation is marked as the error starting side, and the error propagation direction is marked along the interface acceptance sequence to generate error propagation direction markings.

[0071] It should be noted that, based on the slow-fast residual comparison chain, each interface acceptance position is unfolded according to the interface acceptance sequence, and the simulation time axis, time scale, wind turbine number, and wind power boundary side position corresponding to each interface acceptance position are used as the basis for co-positioning.

[0072] Based on the co-location criteria, wind power operation parameters at the same time scale, turbine number, and boundary side are extracted from the wind power twin input boundary. These extracted wind power operation parameters are used as co-location wind power operation parameters. Co-location reference relationships are established between these co-location wind power operation parameters and the extrapolated values ​​of slow variables and the write-back clusters of fast variables in the flow field. Based on these reference relationships, the deviation of the extrapolated values ​​of slow variables from the co-location wind power operation parameters on the windward or leeward side is defined as the flow field extrapolation deviation. This deviation is normalized according to the reference amplitude of the corresponding co-location wind power operation parameters to obtain the flow field normalized deviation. Similarly, the deviation of the power fluctuation arrival position, peak-valley overlap position, and the fluctuation amplitude corresponding to the electromechanical response parameters in the write-back clusters from the co-location wind power operation parameters on the turbine side or grid-connected side is defined as the electromechanical write-back deviation. This deviation is normalized according to the reference amplitude of the corresponding co-location wind power operation parameters to obtain the electromechanical write-back deviation. Electromechanical normalization deviation: The flow field normalization deviation and electromechanical normalization deviation within the same interface acceptance position are compared in opposite directions. If the flow field normalization deviation is higher than the electromechanical normalization deviation, the side where the flow field extrapolation deviation is located is marked as the error initiation side. If the electromechanical normalization deviation is higher than the flow field normalization deviation, the side where the electromechanical write-back deviation is located is marked as the error initiation side. Using the error initiation side as the starting point of the direction, the continuity relationship of the normalization deviation in adjacent interface acceptance positions is verified one by one along the interface acceptance sequence. The direction in which the normalization deviation continues from the upstream interface acceptance position to the downstream interface acceptance position is marked as forward propagation, and the direction in which the normalization deviation reverses from the grid connection port to the blade aerodynamic node is marked as reverse propagation. The corresponding wind power operation parameters, flow field normalization deviation, electromechanical normalization deviation, error initiation side, deviation continuity relationship, and propagation direction for each interface acceptance position are bound item by item to generate an error propagation direction mark.

[0073] S3.4 Based on the error propagation direction marking, the positions with the same error propagation direction and continuous interface acceptance sequence are connected in series to form the same error propagation segment, and the position where the error propagation direction is reversed is marked as the direction transition point, thus generating the error direction interface chain.

[0074] It should be noted that, based on the error propagation direction marking, each interface acceptance position is arranged sequentially according to the interface acceptance order, and the error start side, deviation continuation relationship and error propagation direction corresponding to each interface acceptance position are written into the same position.

[0075] Starting from the first interface acceptance position and proceeding to the next, if the error propagation direction of both the preceding and following interface acceptance positions is either forward or backward, and the interface acceptance sequence is continuous and unbroken, then the preceding and following interface acceptance positions are connected in series to form the same error propagation segment, and the newly merged position is used as the starting point for the next round of verification. If the error propagation direction of both the preceding and following interface acceptance positions changes from forward to backward, or from backward to forward, then the interface acceptance position where the flip occurs is marked as a direction transition point, and the preceding error propagation segment is truncated and the following error propagation segment is started at the direction transition point. Finally, each error propagation segment is connected according to the interface acceptance sequence, and the direction transition point is embedded between adjacent error propagation segments to form a continuous link containing interface acceptance positions, error propagation directions, error propagation segments, and direction transition points, generating an error direction interface chain.

[0076] S4. Based on the error direction interface chain, determine the deviation propagation inflection point, push the extrapolated value of the slow variable forward along the inflection point, and push the aggregated value of the fast variable backward along the inflection point to generate the real-time simulation running trajectory.

[0077] like Figure 5 As shown in the figure, this graph illustrates the fit between three curves: the measured grid-connected power, the standard simulated power, and the simulated power of this scheme. The simulated power of this scheme corresponds to the real-time simulation trajectory after the extrapolated slow variable of the flow field and the aggregated fast variable of the electromechanical system are bidirectionally connected via the deviation propagation inflection point. By magnifying the peak and valley positions, as well as the locations of the greatest differences between the standard and simulated power in the figure, it is clearly evident that this scheme more closely approximates the measured grid-connected power in terms of peak and valley variations and overall trends, thus demonstrating improved timing consistency and operational simulation stability.

[0078] The measured grid-connected power is a benchmark curve, representing the actual power change of the wind farm collected by the grid connection port within a continuous simulation time slice. It is used to reflect the actual output state under the combined effects of main wind speed fluctuations, wind-receiving zone differences, blade aerodynamic response, electromechanical handover, and grid connection port feedback. In the figure, the peak value, valley value, and trend turning point of the measured grid-connected power are used to verify whether the simulated power of this scheme can accurately match the actual operating trajectory.

[0079] The ordinary simulation power is a reference curve, representing the simulation power obtained without using wind power twin input boundaries, slow and fast variable reciprocation, and bidirectional connection of deviation transmission inflection points. As can be seen from the figure, the ordinary simulation power lags or deviates at some peaks, valleys, and trend turning points, indicating that conventional simulation methods are difficult to simultaneously capture the dynamic changes of slow wind field diffusion and fast electromechanical feedback.

[0080] S4.1. Based on the error direction interface chain, find the position where the error propagation direction is reversed along the interface acceptance sequence, and align the found position with the direction transition point to generate the deviation transmission inflection point.

[0081] It should be noted that, according to the error direction interface chain, the interface acceptance positions are first expanded one by one according to the interface acceptance order, and the error propagation direction, error propagation segment and direction transition point corresponding to each interface acceptance position are organized into a direction position record.

[0082] Based on the direction sequence record, the error propagation direction is verified item by item from the previous interface acceptance position to the next interface acceptance position to see if it changes from forward propagation to reverse propagation, or from reverse propagation to forward propagation. The interface acceptance position where the flip occurs is marked as the search position. According to the previous and next error propagation segments where the search position is located, the propagation termination edge before the search position and the propagation start edge after the search position are extracted respectively. The propagation termination edge, the search position, and the propagation start edge are then continuously aligned according to the interface acceptance order to form a flipped clamping record. Based on the flipped clamping record, the search position is... The position is aligned with the marked direction transition point in the error direction interface chain. If the search position and the direction transition point are at the same interface receiving position, the search position is directly confirmed as the folding continuation position. If the search position and the direction transition point are adjacent and offset, the search position is moved to the interface receiving position corresponding to the direction transition point according to the extension direction before and after the error propagation direction flip, forming a folding alignment record. Finally, the interface receiving position, error propagation direction flip relationship, previous error propagation segment and next error propagation segment in the folding alignment record are fixedly bound to generate the deviation transmission inflection point.

[0083] S4.2 Based on the deviation propagation inflection point, the interface receiving position before the deviation propagation inflection point is determined as the slow variable forward push segment, and the interface receiving position after the deviation propagation inflection point is determined as the fast variable reverse push segment.

[0084] It should be noted that, based on the deviation propagation inflection point, the interface acceptance position where the deviation propagation inflection point is located is first taken as the dividing position according to the interface acceptance order. Then, the interface acceptance position is traced back one by one along the interface chain in the error direction to the front of the deviation propagation inflection point. The interface acceptance positions that still accept the extrapolated values ​​of the slow variables of the flow field and whose error propagation direction points to the deviation propagation inflection point are continuously collected to form the front interface acceptance range.

[0085] Based on the front interface acceptance range, the windward sequence, the extrapolated values ​​of slow variables in the flow field, and the corresponding interface acceptance positions are aligned item by item. Interface acceptance positions where the error propagation direction has crossed the deviation transmission inflection point are eliminated to determine the slow variable forward push segment. The interface acceptance positions are then extended step by step along the interface chain in the error direction towards the deviation transmission inflection point. Interface acceptance positions that still accept the electromechanical fast variable aggregate value and whose error propagation direction points from the grid connection port in the opposite direction to the deviation transmission inflection point are continuously collected to form the rear interface acceptance range. Based on the rear interface acceptance range, the electromechanical transfer path, the electromechanical fast variable aggregate value, and the corresponding interface acceptance positions are aligned item by item. Interface acceptance positions where the error propagation direction does not continue to the deviation transmission inflection point are eliminated to determine the fast variable reverse push segment.

[0086] S4.3. Push the extrapolated values ​​of the slow variables in the flow field to the deviation transmission inflection point along the windward sequence, and match them with the windward phase of the interface receiving position at the deviation transmission inflection point to generate the slow variable inflection point push chain.

[0087] It should be noted that, based on the slow variable forwarding segment, the interface receiving positions on the front side of the deviation transmission inflection point are arranged sequentially from the windward side to the deviation transmission inflection point according to the windward sequence. The extrapolated values ​​of the slow variables in the flow field at each interface receiving position are then correlated in the same direction with the corresponding downwind projection, crosswind offset, and wake shielding boundary to form the slow variable forwarding sequence.

[0088] Based on the forward sequence of slow variables, the extrapolated values ​​of the flow field slow variables at the windward interface receiving positions are pushed forward sequentially along the adjacent interface receiving positions. During the pushing process, the forward distance is corrected according to the downwind projection, the deviation direction is corrected according to the crosswind offset, and the shading attenuation is corrected according to the wake shading boundary, so that the extrapolated values ​​of the flow field slow variables maintain the transmission order consistent with the windward sequence when they arrive at each interface receiving position. When the extrapolated values ​​of the flow field slow variables are pushed to the deviation transmission inflection point, the arrival time and windward direction of the extrapolated values ​​of the flow field slow variables are determined. The direction of the deviation continuation is aligned with the interface receiving position at the deviation transmission inflection point by windward phase. The extrapolated values ​​of slow variables in the flow field that arrive ahead of the time sequence are shifted backward and aligned forward. The extrapolated values ​​of slow variables in the flow field that arrive behind the time sequence are shifted forward and aligned. The extrapolated values ​​of slow variables in the flow field that deviate from the windward direction are corrected according to the windward orientation of the interface receiving position. Finally, the extrapolated values ​​of slow variables in the flow field after windward phase alignment, the windward sequence relationship, the forward push segment of slow variables and the deviation transmission inflection point are continuously bound together to generate the forward push chain of slow variable inflection point.

[0089] S4.4. Based on the slow variable inflection point forward push chain, the electromechanical fast variable aggregated value is pressed in reverse from the grid-connected port to the deviation transmission inflection point along the electromechanical transfer path, and the electrical end is reversed and connected with the interface receiving position at the deviation transmission inflection point to generate the fast variable inflection point reverse push chain.

[0090] It should be noted that, based on the slow variable inflection point forward push chain, the interface receiving position at the deviation transmission inflection point is first taken as the reverse pressing end point, and the fast variable reverse push sequence is arranged step by step back from the grid connection port to the electrical receiving position, torque receiving position and transmission receiving position according to the electromechanical transfer path. According to the fast variable reverse push sequence, the electromechanical fast variable aggregate value corresponding to the grid connection port is reversely distributed according to the receiving relationship of power fluctuation arrival position, peak and valley engagement position and electromechanical response parameter, so that the electromechanical fast variable aggregate value is pressed from the grid connection port to the electrical receiving position, torque receiving position and transmission receiving position in sequence, and the reverse pressing sequence is corrected according to the electromechanical transfer path in each receiving position.

[0091] When the electromechanical fast variable aggregation value is pressed to the deviation transmission inflection point, the electrical connection sequence, peak-valley engagement position, and reverse transmission direction corresponding to the electromechanical fast variable aggregation value are electrically reversed and connected to the interface connection position at the deviation transmission inflection point. For electromechanical fast variable aggregation values ​​with advanced electrical connection sequence, they are back-fitted; for electromechanical fast variable aggregation values ​​with lagging electrical connection sequence, they are front-filled; and for electromechanical fast variable aggregation values ​​with reverse transmission direction deviating from the electromechanical delivery path, the path is corrected. Finally, the electromechanical fast variable aggregation value, electromechanical delivery path, deviation transmission inflection point, and slow variable inflection point push chain after the electrical reverse connection are completed are continuously bound together to generate a fast variable inflection point reverse push chain.

[0092] S4.5. Based on the forward chain of slow variable inflection points and the reverse chain of fast variable inflection points, the deviation transmission inflection point is taken as the bidirectional continuation position. The extrapolated values ​​of slow variables in the flow field and the aggregated values ​​of electromechanical fast variables are sequentially sealed and continuously connected according to the simulation time slices to generate the real-time simulation running trajectory.

[0093] It should be noted that, based on the slow variable inflection chain and the fast variable inflection chain, the interface receiving position at the deviation transmission inflection point is first determined as the bidirectional connection position. Then, according to the simulation time slice, the extrapolated values ​​of the slow variable in the slow variable inflection chain and the aggregated values ​​of the electromechanical fast variable in the fast variable inflection chain are arranged at the same time to form the bidirectional connection record of the inflection point.

[0094] Based on the bidirectional connection record of the inflection point, the windward arrival time sequence corresponding to the extrapolated value of the slow flow field variable and the electrical reverse timing sequence corresponding to the aggregated value of the electromechanical fast variable are aligned within the bidirectional connection position. The extrapolated value of the slow flow field variable arriving earlier than the bidirectional connection position is moved backward to seal the edge, and the aggregated value of the electromechanical fast variable arriving later than the bidirectional connection position is moved forward to supplement the connection. The extrapolated value of the slow flow field variable and the aggregated value of the electromechanical fast variable are then time-sequentially sealed according to the same simulation time slice to form a sealed operation segment. Based on the sealed operation segment, the forward push chain of the slow variable inflection point continues to connect to the windward side interface receiving position, and the reverse push chain of the fast variable inflection point continues to connect to the grid connection port direction interface receiving position. Adjacent sealed operation segments are continuously connected in series according to the chronological relationship of the simulation time slices, so that the forward push path of the extrapolated value of the slow flow field variable, the reverse push path of the aggregated value of the electromechanical fast variable, and the bidirectional connection position at the deviation transmission inflection point remain continuously corresponding, generating a real-time simulation operation trajectory.

[0095] S5. Compare the real-time simulation trajectory with the wind power twin input boundary, and correct the deviations at the wind farm end and the electromechanical end respectively to generate the wind power simulation trajectory.

[0096] like Figure 6 As shown in the figure, this diagram illustrates the convergence differences of three curves: the error without backfeed correction, the error with ordinary error correction, and the error with backfeed correction for the deviation source in this scheme. The error with backfeed correction for the deviation source in this scheme corresponds to the residual error after backfeed correction of the double-ended deviation comparison chain, the deviation source marker, and the deviation positions at the wind farm end and the electromechanical end. The figure marks the deviation positions at the wind farm end, the electromechanical end, and the double-ended overlap position, and locally magnifies the error convergence process after the double-ended overlap position. This demonstrates that this scheme can trace the deviation source and quickly reduce the residual error, reflecting improvements in error traceability, boundary reliability, and operational simulation stability.

[0097] The no-backfeed correction error is a baseline curve, representing the residual error formed when no double-ended deviation comparison chain is used, no deviation source marking is performed, and no backfeed correction is performed for the wind farm end deviation position and the electromechanical end deviation position. As can be seen from the figure, after the wind farm end deviation position, the electromechanical end deviation position, and the double-ended overlap position appear, the peak value of the no-backfeed correction error is relatively high and decreases slowly, indicating that without deviation source attribution processing, subsequent simulation errors are prone to continuous accumulation.

[0098] The ordinary error correction error is the conventional correction comparison curve, which represents the residual error formed after only reducing or uniformly correcting the simulation deviation. The ordinary error correction error is lower than the error without recharge correction, but there are still obvious residual fluctuations near the overlapping position of the two ends. This indicates that although the ordinary correction method can reduce the error amplitude, it is difficult to distinguish whether the deviation comes from the wind-receiving zone or the electromechanical transfer path. Therefore, the error convergence speed and later stability are still weaker than this scheme.

[0099] S5.1 Align the real-time simulation trajectory with the wind power twin input boundary in terms of time and position, and compare the extrapolated values ​​of slow flow variables and the aggregated values ​​of fast electromechanical variables in the same slot to generate a double-end deviation comparison chain.

[0100] It should be noted that, based on the real-time simulation operation trajectory, the simulation time slice, trajectory writing slot, interface receiving position, extrapolated value of slow flow field variable and aggregated value of fast electromechanical variable are extracted segment by segment according to the simulation time axis. Then, the wind power boundary side position, measurement point position and wind power operation parameters under the same time scale are extracted from the wind power twin input boundary to form a time-position alignment record.

[0101] Based on the time-position alignment records, the simulation time slice is time-locked with the time scale, the interface receiving position is side-positioned with the wind power boundary position, and the trajectory written slot is position-locked with the measurement point position, forming a time-position boundary alignment row under the same time, the same side position, and the same slot. Based on the time-position boundary alignment row, the extrapolated values ​​of the slow flow field variables are compared with the wind power operation parameters corresponding to the windward and leeward sides in the wind power twin input boundary in the same slot to extract the wind field deviation. The aggregated values ​​of the electromechanical fast variables are compared with the wind power operation parameters corresponding to the unit side and grid-connected side in the wind power twin input boundary in the same slot to extract the electromechanical deviation. Finally, the wind field deviation, electromechanical deviation, simulation time slice, wind power boundary side position, trajectory written slot, and interface receiving position are continuously concatenated according to the time-position boundary alignment row to generate a double-ended deviation comparison chain.

[0102] It should also be noted that the dual-end deviation comparison chain is used to compare the extrapolated values ​​of slow flow field variables and the aggregated values ​​of fast electromechanical variables in the real-time simulation trajectory with the corresponding wind power operation parameters in the wind power twin input boundary, and to continuously connect the wind field deviation, electromechanical deviation, simulation time slice and interface connection position. Through bidirectional parallel tracking of the forward diffusion deviation at the wind field end and the reverse write-back deviation at the electromechanical end, the deviation is no longer limited to a single simulation error judgment, but can be located along the wind-receiving zone and the electromechanical transfer path, providing a clear basis for subsequent wind field end recharge, electromechanical end recharge and dual-end overlap position correction.

[0103] S5.2. Based on the double-end deviation comparison chain, determine the source of wind field deviation and electromechanical deviation, mark the position of wind field end deviation, electromechanical end deviation and double-end overlap position, and generate deviation source marker.

[0104] It should be noted that, based on the double-end deviation comparison chain, the wind field deviation and electromechanical deviation are located according to the simulation time slice, trajectory writing slot, and interface receiving position. The wind field deviation is deducted back to the windward zone along the windward sequence. If the deviation continuously expands on the windward or leeward side, it is marked as the wind field end deviation position. The electromechanical deviation is deducted back to the electromechanical transfer path along the error propagation direction. If the deviation changes abruptly first on the unit side or grid connection side, it is marked as the electromechanical end deviation position. If the wind field deviation and electromechanical deviation intersect simultaneously in the same trajectory writing slot, it is marked as the double-end overlap position. Finally, the above positions are bound according to the extension order of the double-end deviation comparison chain to generate the deviation source mark.

[0105] S5.3 Based on the deviation source marker, the deviation position at the wind farm end is fed back to the wind-receiving zone, and the deviation position at the electromechanical end is fed back to the electromechanical transfer path. The boundary parameters and peak-valley engagement positions are corrected respectively to generate the wind power simulation trajectory.

[0106] It should be noted that, based on the deviation source markers, the deviation positions at the wind farm end, electromechanical end, and double-end overlap positions are first extracted according to the simulation time slice and trajectory written into the slot. The wind farm end deviation position is then fed back to the windward zone along the windward sequence, and the boundary parameters of the downwind projection, crosswind offset, and wake shielding boundary in the deviation zone are adjusted to form the wind farm end feed-back section. The electromechanical end deviation position is then fed back to the electromechanical transfer path along the error propagation direction, and the peak-valley correspondence between the power fluctuation arrival position and the electromechanical response parameters is corrected by back-calculation, and the peak-valley engagement position is re-determined to form the electromechanical end feed-back section. Then, using the double-end overlap position as the common correction intersection point, the wind farm end feed-back section and the electromechanical end feed-back section are connected in series, and the corrected boundary parameters and peak-valley engagement positions are backfilled into the real-time simulation operation trajectory to generate the wind power simulation trajectory.

[0107] This embodiment also provides a computer device applicable to the wind power simulation method based on digital twins, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the wind power simulation method based on digital twins as proposed in the above embodiment.

[0108] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0109] This embodiment also provides a storage medium storing a computer program that, when executed by a processor, implements the wind power simulation method based on digital twins as proposed in the above embodiments. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0110] In summary, this invention constructs a wind power digital twin coupled simulation link by: forming the wind power twin input boundary through digital twin edge-time calibration, and using the wind-receiving zone to back-calculate the aerodynamic, electromechanical, and grid connection ports of the blades; and then, through slow and fast variable reconciliation, bidirectional connection of deviation transmission inflection points, and double-end deviation source return and reinjection, the slow diffusion of the wind field and the fast feedback of electromechanical systems are coordinated and aligned within the same simulation cycle, thereby improving the temporal consistency, error traceability, boundary credibility, and operational simulation stability of wind power simulation.

[0111] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A wind power simulation method based on digital twins, characterized in that, include: Collect wind farm physical operation data, perform edge-time calibration on the wind farm physical operation data, and generate wind power twin input boundaries; Based on the wind turbine spatial location in the wind power twin input boundary, the wind receiving zone is deduced, the wind receiving zone is bound to the blade aerodynamic node, and the aerodynamic load in the blade aerodynamic node is transferred to the electromechanical coupling node and then connected to the grid connection port to generate a wind power coupling simulation body. Slow flow field variables and fast electromechanical variables are extracted from the wind power coupled simulation body. The slow flow field variables are extrapolated and the aggregated values ​​of the fast electromechanical variables are written back. The error propagation direction is marked and the error direction interface chain is generated. Based on the error direction interface chain, the deviation propagation inflection point is determined, the extrapolated value of the slow variable is pushed forward along the inflection point, and the aggregated value of the fast variable is pushed back along the inflection point to generate the real-time simulation running trajectory. The real-time simulation trajectory and the wind power twin input boundary are compared in the same position. The deviations that fall at the wind farm end and the electromechanical end are corrected by backfeeding, and the wind power simulation trajectory is generated.

2. The wind power simulation method based on digital twins as described in claim 1, characterized in that, The specific steps for collecting wind farm operational data, performing edge-time calibration on the wind farm operational data, and generating wind power twin input boundaries are as follows: The wind farm physical operation data is classified and arranged to obtain the wind power boundary side position, and the wind power boundary side position is mapped to the measurement point position to generate the boundary measurement point sequence. Based on the boundary measurement point sequence, the acquisition time of different measurement point locations is uniformly aligned to the same simulation time axis, and the boundary positions of wind power operation parameters in the same simulation time axis are aligned to generate a boundary-time aligned sequence. By performing bidirectional boundary-time locking on the edge-time alignment sequence, the wind power operation parameters are simultaneously fixed to the corresponding wind power boundary side position, generating a wind power twin input boundary.

3. The wind power simulation method based on digital twins as described in claim 2, characterized in that, The specific steps for deducing the wind-receiving zone based on the spatial location of the wind turbines in the wind power twin input boundary are as follows: The spatial position of the wind turbine is extracted from the wind power twin input boundary. The spatial position of the wind turbine is obliquely expanded by the prevailing wind direction. The distance in the direction of the incoming flow is taken as the downwind projection and the distance deviating from the center line of the incoming flow is taken as the crosswind offset. The wind displacement sequence is obtained, and the wind intensity of each wind turbine is marked by the windward relationship to generate the wind inversion sequence. Based on the wind-induced sequence, wind turbines with adjacent wind strength, continuous downwind projection, and crosswind offset that do not cross the wake shielding boundary are grouped into the same area. The locations of wind turbines with abrupt changes in wind strength are used as partitioning breakpoints to generate wind-induced zones.

4. The wind power simulation method based on digital twins as described in claim 3, characterized in that, The specific steps for binding the wind-receiving zone to the blade aerodynamic node and transferring the aerodynamic load in the blade aerodynamic node to the electromechanical coupling node and then connecting it to the grid connection port to generate a wind power coupling simulation are as follows: According to the wind-receiving zone, the windward adjustment parameters in each zone are connected to the blade aerodynamic nodes under the wind turbine number, and the load-bearing positions are arranged in the radial order from the blade root to the blade tip to generate an aerodynamic node binding chain. Based on the aerodynamic node binding chain, the aerodynamic load in the blade aerodynamic node is sequentially transferred along the electromechanical transfer path, and the transmission receiving position, torque receiving position and electrical receiving position are determined according to the electromechanical transfer path to generate the electromechanical coupling node. Electromechanical response parameters are extracted from the electromechanical coupling nodes and bound to the electromechanical coupling nodes to generate an electromechanical load transfer chain. Based on the electromechanical load transmission chain, the grid-connected electrical state corresponding to the electromechanical coupling node is connected to the grid-connected port, and the aerodynamic load and electromechanical response parameters are connected in series according to the electrical connection relationship of the grid-connected port to generate a wind power coupling simulation body.

5. The wind power simulation method based on digital twins as described in claim 1, characterized in that, The steps for extracting slow flow field variables and fast electromechanical variables from the wind power coupled simulation, extrapolating the slow flow field variables, and writing back the aggregated values ​​of the fast electromechanical variables are as follows: Slow flow field variables and fast electromechanical variables are extracted from the wind power coupled simulation body and attached to the corresponding interface receiving positions. A slow-fast interface alignment table is obtained. According to the slow-fast interface alignment table, the slow flow field variables in the upstream interface receiving position are recursively pushed to the adjacent downstream interface receiving positions according to the windward sequence. The variables are written to the corresponding blade aerodynamic nodes to generate a slow flow field variable extrapolation chain. Based on the slow variable extrapolation chain of the flow field, the grid-connected electrical state is retrieved from the grid-connected port. The power fluctuations in the grid-connected electrical state are reversed and matched to the electromechanical transfer path. The electromechanical transfer path and the electromechanical response parameters are peak-valley matched and folded into a fast variable write-back cluster. The fast variable write-back cluster is pressed into the corresponding interface receiving position to generate a slow-fast residual comparison chain.

6. The wind power simulation method based on digital twins as described in claim 5, characterized in that, The error propagation direction is marked, and an error direction interface chain is generated. The specific steps are as follows: Based on the slow and fast residual comparison chain, the flow field extrapolation deviation and electromechanical write-back deviation at each interface acceptance position are compared in opposite directions. The side with the dominant deviation is marked as the error initiation side, and the error propagation direction is marked along the interface acceptance sequence to generate error propagation direction markings. Based on the error propagation direction marking, positions with consistent error propagation directions and continuous interface acceptance order are connected in series to form the same error propagation segment, and positions where the error propagation direction is reversed are marked as direction transition points, thus generating an error direction interface chain.

7. The wind power simulation method based on digital twins as described in claim 1, characterized in that, The process of determining the deviation propagation inflection point based on the error direction interface chain, pushing the extrapolated value of the slow variable forward along the inflection point, and pushing the aggregated value of the fast variable backward along the inflection point to generate the real-time simulation trajectory, is as follows: Based on the error direction interface chain, find the position where the error propagation direction is reversed along the interface acceptance sequence, and align the found position with the direction transition point to generate the deviation transmission inflection point. Based on the deviation propagation inflection point, the interface receiving position before the deviation propagation inflection point is determined as the slow variable forward push segment, and the interface receiving position after the deviation propagation inflection point is determined as the fast variable reverse push segment. The extrapolated values ​​of the slow variables in the flow field are pushed to the deviation transmission inflection point along the windward sequence, and the windward phase is matched with the interface receiving position at the deviation transmission inflection point to generate the slow variable inflection point push chain. Based on the slow variable inflection point push chain, the electromechanical fast variable aggregated value is pressed back along the electromechanical transfer path from the grid port to the deviation transfer inflection point, and the electrical end is reversed and connected to the interface receiving position at the deviation transfer inflection point to generate the fast variable inflection point push chain. Based on the forward chain of slow variable inflection points and the backward chain of fast variable inflection points, the deviation propagation inflection point is taken as the bidirectional continuation position. The extrapolated values ​​of slow variables in the flow field and the aggregated values ​​of fast electromechanical variables are sequentially sealed and continuously connected according to the simulation time slices to generate the real-time simulation running trajectory.

8. The wind power simulation method based on digital twins as described in claim 7, characterized in that, The process of comparing the real-time simulation trajectory with the wind power twin input boundary, and correcting deviations at the wind farm end and electromechanical end respectively to generate the wind power simulation trajectory, is as follows: The real-time simulation trajectory is aligned with the wind power twin input boundary in terms of time and position. The extrapolated values ​​of slow variables in the flow field and the aggregated values ​​of fast electromechanical variables are compared in the same slot to generate a double-end deviation comparison chain. Based on the double-end deviation comparison chain, the wind field deviation and electromechanical deviation are determined to be of their source, and the positions of the wind field end deviation, electromechanical end deviation, and double-end overlap are marked to generate deviation source markers. Based on the deviation source markers, the deviation positions at the wind farm end are fed back to the wind-receiving zone, and the deviation positions at the electromechanical end are fed back to the electromechanical transfer path. The boundary parameters and peak-valley engagement positions are corrected respectively to generate the wind power simulation trajectory.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the wind power simulation method based on digital twins as described in any one of claims 1 to 8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the wind power simulation method based on digital twins as described in any one of claims 1 to 8.

Images