A method and system for adaptive adjustment of suction fan partitioned flow

Abstract

This invention provides a method and system for adaptive adjustment of suction flow rate in different zones of a suction sail. It collects incoming wind speed data and sail angle of attack data to extract dynamic features and construct a flow field state database. Boundary layer flow patterns are identified from the database, and flow separation analysis is performed to generate a separation risk index. Spatial mapping is used to convert the separation risk index into a suction demand distribution map. An angle of attack-lift coupling response curve is generated based on the sail angle of attack data, and sensitivity analysis is performed to obtain flow rate adjustment coefficients, forming a regional flow rate configuration table. Flow rate-lift-drag ratio response surface analysis is performed on the suction demand distribution map to generate a lift-drag ratio gain value. Efficiency and power consumption are co-optimized to generate a flow rate priority sequence and construct an adaptive control strategy. Time-varying differentiated suction sequences are generated by regional decomposition, and flow gradient constraints are applied to generate a smooth transition suction sequence. Real-time flow rate adjustment commands are output to complete the adaptive control of the suction flow rate.

Description

Technical Field

[0001] This invention relates to the field of marine energy conservation technology, and in particular to a method and system for adaptive adjustment of suction flow rate in a suction sail zone. Background Technology

[0002] As a novel type of wind-assisted propulsion device for ships, suction sails utilize an array of suction holes on their surface to apply active suction, thereby delaying boundary layer flow separation and improving the lift-to-drag ratio, thus generating greater propulsion. Compared to traditional rigid wing sails, suction sails can maintain attached flow over a wider angle-of-attack range, significantly expanding the effective working area and showing broad application prospects in the field of ship energy conservation and emission reduction.

[0003] However, existing suction sail control systems have the following problems in flow regulation: the suction flow rate is usually adjusted to a fixed value or a simple proportional adjustment, failing to adaptively adjust according to the incoming wind speed, sail angle of attack, and boundary layer state, resulting in energy waste at low wind speeds and insufficient separation suppression at high wind speeds; a uniform suction strategy is adopted in each region of the sail, without considering the differences in flow separation risk and suction efficiency in the leading edge, middle chord, and trailing edge regions, making it difficult to achieve optimal allocation of flow resources; the flow control lacks synergistic optimization of aerodynamic efficiency and power consumption, and cannot reasonably balance the relationship between lift-to-drag ratio gain and energy consumption when the pump power is limited. Summary of the Invention

[0004] This invention provides a method and system for adaptive adjustment of suction flow rate in different zones of a suction sail. It collects incoming wind speed data and sail angle of attack data to extract dynamic features and construct a flow field state database. From this database, it identifies boundary layer flow patterns and performs flow separation analysis to generate a suction demand distribution map. Based on the sail angle of attack data, it generates angle of attack-lift coupling response curves and performs sensitivity analysis to form a regional flow configuration table. It then performs flow-lift-drag ratio response surface analysis on the suction demand distribution map to generate a lift-drag ratio gain value and performs performance-power consumption co-optimization to construct an adaptive control strategy. Finally, it decomposes the adaptive control strategy by region and applies flow gradient constraints to generate real-time flow adjustment commands.

[0005] The first aspect of this invention proposes a method for adaptive adjustment of the suction flow rate of a suction sail in different zones, comprising the following steps: Collect incoming wind speed data and sail angle of attack data in the operating environment of the suction sail, and perform dynamic feature extraction on the incoming wind speed data and sail angle of attack data to construct a flow field state database; Identify boundary layer flow patterns from the flow field state database, perform flow separation analysis on the boundary layer flow patterns to generate a separation risk index, and use chord-direction spatial mapping processing to convert the separation risk index into a pumping demand distribution map. Based on the dynamic change characteristics of the sail angle of attack data, an angle of attack-lift coupling response curve is generated. Sensitivity analysis is performed on the angle of attack-lift coupling response curve to obtain the flow regulation coefficient. Based on the flow regulation coefficient, the sail is divided into leading edge region, mid-chord region and trailing edge region to form a regional flow configuration table. A flow-lift-drag ratio response surface analysis is performed on the suction demand distribution map to generate the lift-drag ratio gain value for each region. A performance-power consumption co-optimization analysis is performed on the lift-drag ratio gain value to generate a flow priority sequence. An adaptive control strategy is constructed by fusing the flow priority sequence with the region flow configuration table. The adaptive control strategy is decomposed into regions to generate time-varying differentiated suction sequences. Adjacent region flow gradient constraints are applied to the time-varying differentiated suction sequences to generate smooth transition suction sequences. Real-time flow regulation commands are generated based on the smooth transition suction sequences.

[0006] A second aspect of the present invention provides a suction sail zoned suction flow adaptive adjustment system, comprising: The data acquisition module is used to collect incoming wind speed data and sail angle of attack data in the operating environment of the suction sail, and to perform dynamic feature extraction on the incoming wind speed data and sail angle of attack data to construct a flow field state database. The separation analysis module is used to identify boundary layer flow patterns from the flow field state library, perform flow separation analysis on the boundary layer flow patterns to generate a separation risk index, and use chord-direction spatial mapping processing to convert the separation risk index into a pumping demand distribution map. The region division module is used to generate an angle-of-attack-lift coupling response curve based on the dynamic change characteristics of the sail angle-of-attack data, perform sensitivity analysis on the angle-of-attack-lift coupling response curve to obtain the flow regulation coefficient, and divide the sail into leading edge region, mid-chord region and trailing edge region according to the flow regulation coefficient to form a regional flow configuration table. The optimization decision module is used to perform flow-lift-drag ratio response surface analysis on the suction demand distribution map to generate lift-drag ratio gain values ​​for each region, perform efficiency and power consumption co-optimization analysis on the lift-drag ratio gain values ​​to generate a flow priority sequence, and construct an adaptive control strategy based on the flow priority sequence and the regional flow configuration table. The flow control module is used to decompose the adaptive control strategy into time-varying differentiated suction sequences by region, apply adjacent region flow gradient constraints to the time-varying differentiated suction sequences to generate smooth transition suction sequences, and generate real-time flow adjustment commands based on the smooth transition suction sequences.

[0007] The beneficial effects of this invention are reflected in the following points: 1. By extracting wind speed temporal characteristics and turbulence intensity parameters from incoming wind speed data and performing spectral decomposition to identify the dominant frequency of wind speed fluctuations, the dominant frequency of wind speed fluctuations is fused with sail angle of attack data to generate a wind condition-angle of attack coupling matrix to construct a flow field state library. Based on the flow field state library, boundary layer flow patterns are identified, and a separation risk index is generated by combining the separation tendency coefficient and boundary layer shape factor, thus realizing refined perception of sail flow state and quantitative assessment of separation risk. 2. By performing flow-lift-drag ratio response surface analysis on the suction demand distribution map to generate lift-drag ratio gain values ​​for each region, the lift gain component and drag reduction component are decomposed from the lift-drag ratio gain value to construct an efficiency evaluation space. Combined with power consumption cost indicators, collaborative optimization is performed to generate a flow priority sequence, solving the optimization decision problem of flow allocation in each region under limited pump power, and prioritizing the allocation of limited suction resources to the region with the highest efficiency. 3. The adaptive control strategy is decomposed into regions and the dynamic adjustment range is determined through feedforward-feedback decoupling analysis. Flow gradient constraints are applied at the boundary between adjacent regions to generate a smooth transition suction sequence, which eliminates flow jumps between regions and avoids artificially introduced flow field disturbances, thus ensuring the time-varying differentiated control and spatial continuity of the sail suction flow.

[0008] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0009] The accompanying drawings illustrate specific examples of the technical solutions described in this invention and, together with the detailed embodiments, form part of the specification, serving to explain the technical solutions, principles, and effects of this invention.

[0010] Figure 1 This is a schematic flowchart of a method for adaptive adjustment of suction flow rate in a suction sail partition according to the present invention.

[0011] Figure 2 This is a structural block diagram of a suction sail partitioned suction flow adaptive adjustment system according to the present invention. Detailed Implementation

[0012] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, circuits, and methods are omitted so as not to obscure the description of this application with unnecessary detail.

[0013] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.

[0014] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0015] The technical solutions of the embodiments of this application will be described below.

[0016] like Figure 1 As shown, this embodiment of the invention provides a method for adaptive adjustment of suction flow rate in a suction sail partition, including the following steps S110-S150: Step S110: Collect incoming wind speed data and sail angle of attack data in the operating environment of the suction sail, and perform dynamic feature extraction on the incoming wind speed data and sail angle of attack data to construct a flow field state database.

[0017] Specifically, the system collects incoming wind speed data and sail angle of attack data from the suction sail's operating environment. An array of wind speed sensors is deployed at the top of the suction sail mast and the leading edge of the sail to collect wind speed information from the incoming flow area. The wind speed sensors employ ultrasonic anemometers or propeller-driven anemometers, with the sampling frequency set according to the ship's speed and wind conditions. The wind speed sensors measure the speed and direction angle of the incoming flow, organizing the results into a time-series data set containing fields such as instantaneous wind speed, wind direction angle, acquisition time, and sensor location. An angle encoder or tilt sensor is deployed at the sail rotation mechanism to measure the sail's geometric angle of attack relative to the incoming flow direction in real time. As the ship navigates at sea, the sail needs to continuously adjust its attitude according to the wind direction to obtain optimal propulsion. The angle encoder converts the sail's rotation angle into a digital signal output. The angle measurement results are fused with the ship's heading information to calculate the effective angle of attack of the sail, forming sail angle of attack data. This data records the sail's angle of attack value, rate of change of angle of attack, and acceleration of angle of attack at each moment. Incoming wind speed data and sail angle of attack data are collected synchronously and timestamped. When a ship encounters gusts, the wind speed fluctuates drastically in a short period of time. The synchronous recording of the two types of data ensures accurate matching in the time dimension during subsequent feature extraction.

[0018] In some embodiments, the step of dynamically extracting features from the incoming wind speed data and the sail angle of attack data to construct a flow field state library includes: extracting wind speed temporal features and turbulence intensity parameters from the incoming wind speed data; performing spectral decomposition on the turbulence intensity parameters to identify the dominant frequency of wind speed fluctuations; fusing the dominant frequency of wind speed fluctuations with the sail angle of attack data to generate a wind condition-angle of attack coupling matrix; and integrating the wind speed temporal features with the wind condition-angle of attack coupling matrix to form a flow field state library.

[0019] Wind speed temporal features and turbulence intensity parameters are extracted from the incoming wind speed data. A sliding window analysis is performed on the time series of the incoming wind speed data. Within each time window, the statistics of the wind speed data are calculated. The wind speed temporal features include the average wind speed, wind speed variance, wind speed skewness, and wind speed kurtosis within the window, reflecting the amplitude characteristics and distribution pattern of the wind speed signal. The deviation between the instantaneous wind speed and the average wind speed is calculated to obtain the fluctuating wind speed. The root mean square value of the fluctuating wind speed is divided by the average wind speed to obtain the turbulence intensity parameter. The turbulence intensity parameter characterizes the relative intensity of the turbulent component in the incoming flow; a larger value indicates a more unstable incoming flow. The incoming wind speed data of the suction sail in a certain section shows an average wind speed of 12 m / s and a wind speed variance of 2.3 (m / s)² within a 10-minute window. The extracted wind speed temporal features indicate that the wind speed fluctuation is moderate during this period. Simultaneously, the calculated turbulence intensity parameter is 0.18, indicating that the turbulent component of the incoming flow is at a moderate level during this period. When ships navigate in open waters, turbulence intensity parameters are typically low and change gradually. However, when swells occur or wind direction changes, turbulence intensity parameters increase significantly. In such cases, suction control needs to increase the response frequency to cope with fluctuations in the incoming flow. Wind speed temporal characteristics and turbulence intensity parameters are recorded in a time series to jointly describe the dynamic changes in the incoming flow.

[0020] The dominant frequency of wind speed fluctuations is identified by spectral decomposition of turbulence intensity parameters. A Fast Fourier Transform (FFT) is performed on the fluctuating wind speed signal corresponding to the turbulence intensity parameters to convert the time-domain signal into a frequency-domain representation. The energy distribution of each frequency component in the spectrum is analyzed to identify frequency ranges where energy is concentrated. Peak positions are searched in the energy spectrum; the frequencies corresponding to these peaks are the dominant frequency components of the fluctuating wind speed. The highest-energy peak frequencies are extracted to form a set of dominant wind speed fluctuation frequencies. After spectral decomposition, the turbulence intensity parameters collected by the suction sail under gust conditions show two energy peaks at 0.05Hz and 0.3Hz. These two peak frequencies are extracted to form a set of dominant wind speed fluctuation frequencies. 0.05Hz corresponds to the periodic disturbance of large-scale turbulent eddies, and 0.3Hz corresponds to the rapid fluctuation of small-scale turbulent eddies. The dominant wind speed fluctuation frequencies reflect the characteristic frequencies of periodic disturbances in the incoming flow; low-frequency components correspond to large-scale turbulent eddies, and high-frequency components correspond to small-scale turbulent eddies. The dominant frequency of wind speed fluctuations in offshore wind fields is typically concentrated in the range of 0.01 Hz to 1 Hz, with corresponding periods ranging from several seconds to several minutes. Pump control adjusts the control bandwidth according to the distribution of the dominant wind speed fluctuation frequency, enabling the pump flow rate to be adjusted in accordance with the main disturbance frequency. Identifying the dominant wind speed fluctuation frequency helps determine the energy distribution characteristics of turbulence; a single-peak distribution indicates the presence of a dominant disturbance source, while a multi-peak distribution indicates the simultaneous action of turbulent eddies at multiple scales.

[0021] A wind condition-angle of attack coupling matrix is ​​generated by fusing wind speed fluctuation frequency and sail angle of attack data. The correlation between the frequency components of the wind speed fluctuation frequency and the changes in sail angle of attack data is analyzed. The amplitude and phase of the angle of attack response at each frequency are calculated. The sail angle of attack data undergoes the same spectral decomposition as the wind speed fluctuation frequency, extracting the spectral components of the angle of attack signal at each frequency. A two-dimensional matrix structure is constructed, with row indices corresponding to the frequency components of the wind speed fluctuation frequency, column indices corresponding to the characteristic parameters of the angle of attack response, and matrix elements representing the coupling strength. The coupling strength is represented by the coherence function or transfer function gain; a larger value indicates a more significant impact of wind speed fluctuation on the angle of attack at that frequency. The wind speed fluctuation frequency of the suction sail in a certain segment includes two frequencies: 0.05Hz and 0.3Hz. The sail angle of attack response amplitude at 0.05Hz is 2.1° with a phase lag of 15°, and the response amplitude at 0.3Hz is 0.4° with a phase lag of 72°. These values ​​are filled into the corresponding positions in the matrix to form the wind condition-angle of attack coupling matrix. The wind-angle of attack coupling matrix reveals how wind speed disturbances of different frequencies are transmitted to changes in the angle of attack of the sail. High low-frequency coupling strength in the matrix indicates that the sail can adjust its attitude according to slowly changing wind conditions, while low high-frequency coupling strength indicates that the sail has a good ability to suppress fast disturbances. The balance between the two determines the overall aerodynamic performance of the sail.

[0022] A flow field state database is formed by integrating wind speed temporal features and the wind-angle of attack coupling matrix. Wind speed temporal features serve as a static description of the flow field state, characterizing the basic attributes of the incoming flow under the current operating condition, including wind speed level, wave intensity, and extreme value characteristics. The wind-angle of attack coupling matrix serves as a dynamic description of the flow field state, characterizing the response characteristics of the incoming flow disturbance transmitted to the sail surface, including frequency response, phase delay, and coupling bandwidth. A standardized data structure for the flow field state is designed, including an operating condition identifier, a wind speed temporal feature field group, and a wind-angle of attack coupling matrix field group. The collected data is classified into operating conditions based on average wind speed and average angle of attack, with each operating condition corresponding to one state record. During a voyage, the suction sail categorized its operating conditions into low-wind-speed and medium-wind-speed levels based on the average wind speed. The low-wind-speed level showed an average wind speed of 8 m / s and a variance of 1.2 m / s². The wind-angle-of-attack coupling matrix indicated that the coupling strength at each frequency was below 0.5. This characteristic data was packaged into a single state record and written into the flow field state database. All state records were imported into the database and indexed to support fast queries. The flow field state database supports multiple query methods, including searching by wind speed range, by angle-of-attack range, and by coupling characteristics. When the vessel encounters similar wind conditions again, historical data can be directly retrieved to accelerate the matching of control strategies and reduce the adaptive time for suction flow adjustment.

[0023] Step S120: Identify the boundary layer flow mode from the flow field state library, perform flow separation analysis on the boundary layer flow mode to generate a separation risk index, and use spatial mapping along the chord direction to convert the separation risk index into a pumping demand distribution map.

[0024] Specifically, boundary layer flow patterns are identified from the flow field state database. Based on the current incoming wind speed and sail angle of attack, matching flow field state records are retrieved from the database to obtain the corresponding wind speed time-series characteristics and wind-angle of attack coupling matrix. The development characteristics of the sail boundary layer are analyzed based on the flow field state parameters retrieved from the database, combined with sail geometry and surface roughness information. The boundary layer is a thin layer of fluid closely adhering to the sail surface, where the flow velocity gradually transitions from zero at the wall to the external incoming velocity. The flow regime of the boundary layer is determined based on the Reynolds number range recorded in the database: low Reynolds numbers correspond to laminar boundary layers, high Reynolds numbers correspond to turbulent boundary layers, and transition phenomena exist in the intermediate region. The flow field state database record for a certain segment of the journey shows a current Reynolds number of 3.2 × 10⁻⁶. 6Based on this, it was determined that the boundary layer transitions into turbulent flow at 5% of the chord length from the leading edge, and the boundary layer thickness gradually increases along the chord direction from 0.8 mm at the leading edge to 12 mm at the trailing edge. The development law of the boundary layer along the sail chord direction was analyzed, including the boundary layer thickness increase, velocity pattern evolution, and wall friction changes. The flow regime type, thickness distribution, and velocity pattern parameters of the boundary layer were integrated to form a boundary layer flow model. The boundary layer flow model records the boundary layer parameters at each location using the chord position as an index. When ships are sailing at sea, wind speeds usually cause the sail to operate in a higher Reynolds number range. Typical boundary layer flow models exhibit laminar or transitional flow near the leading edge and turbulent flow in the middle and rear sections.

[0025] In some embodiments, the step of performing flow separation analysis on the boundary layer flow mode to generate a separation risk index includes: extracting wall pressure distribution data from the boundary layer flow mode to generate adverse pressure gradient distribution characteristics; evaluating the separation tendency of the boundary layer flow mode based on the adverse pressure gradient distribution characteristics to generate a separation tendency coefficient; performing boundary layer shape factor analysis on the separation tendency coefficient to generate a separation sensitivity distribution field; and comparing the separation sensitivity distribution field with a preset critical separation threshold to determine the separation risk index.

[0026] Wall pressure distribution data is extracted from the boundary layer flow model to generate adverse pressure gradient distribution characteristics. The boundary layer flow model records pressure data at various chordal positions on the sail surface. The pressure data is discretely distributed along the chordal position. The rate of change of pressure along the chordal direction, i.e., the pressure gradient, is obtained by differential calculation of the values ​​at adjacent pressure measurement points in the boundary layer flow model. A positive pressure gradient indicates an increase in pressure along the flow direction (adverse pressure gradient), while a negative pressure gradient indicates a decrease in pressure along the flow direction (compatibility pressure gradient). The adverse pressure gradient is the main factor leading to boundary layer flow separation. The boundary layer flow model of the suction sail at an inflow wind speed of 12 m / s shows that the pressure coefficient on the suction surface of the sail gradually increases from -2.8 at the leading edge to -0.3 at the trailing edge. Differential calculation of adjacent measurement points shows that the pressure gradient in the region from 30% to 80% of the chord length is positive. The pressure gradient value in this region is extracted to form the adverse pressure gradient distribution characteristics. The positive portion of the pressure gradient is recorded along the chord direction, along with the numerical value and spatial range of the adverse pressure gradient. The distribution characteristics of the adverse pressure gradient at each location are dimensionless, with dynamic pressure as the reference quantity to eliminate the influence of wind speed. The pressure distribution on the suction surface of the sail exhibits a typical pattern: a suction peak at the leading edge, a pressure rise in the middle section, and a pressure at the trailing edge approaching the static pressure of the incoming flow. The adverse pressure gradient distribution characteristics begin to appear in the region downstream of the suction peak and gradually intensify towards the trailing edge. The dimensionless adverse pressure gradient values ​​at each chord direction are stored in array form.

[0027] Separation tendency assessment of boundary layer flow models is performed based on the adverse pressure gradient distribution characteristics to generate a separation tendency coefficient. The adverse pressure gradient distribution characteristics are correlated with boundary layer parameters in the boundary layer flow models. The momentum thickness recorded in the boundary layer flow models reflects the momentum deficit carried by the fluid within the boundary layer; a larger momentum thickness indicates more low-energy fluid within the boundary layer and a weaker ability to resist the adverse pressure gradient. The product of the adverse pressure gradient at each location in the adverse pressure gradient distribution characteristics and the corresponding momentum thickness in the boundary layer flow models is calculated; this product characterizes the local separation tendency intensity. When a suction sail encounters continuous crosswinds at sea, it needs to maintain a large angle of attack to obtain sufficient lateral thrust. At this time, the rear half of the sail's suction surface experiences a significant adverse pressure gradient. The boundary layer flow model shows that the boundary layer in this region is significantly thickened and the wall shear stress decreases, indicating the accumulation of low-energy fluid and flow deceleration within the boundary layer. Based on the correlation between the adverse pressure gradient distribution characteristics and boundary layer parameters, the separation tendency coefficient in this region is significantly increased, indicating that the trailing edge region of the sail has entered a separation warning state and suction intervention needs to be initiated. The separation tendency coefficient is calculated using the formula β = (θ / τ_w) × (dp / dx), where β is the separation tendency coefficient, dimensionless; θ is the boundary layer momentum thickness, in meters (m), representing the equivalent thickness of momentum deficit within the boundary layer; τ_w is the wall shear stress, in Pa (Pa), representing the tangential frictional force of the fluid against the wall; and dp / dx is the pressure gradient value in the adverse pressure gradient distribution characteristic, in Pa / m. A positive separation tendency coefficient indicates the presence of a separation tendency; the larger the value, the stronger the separation tendency. When the separation tendency coefficient reaches a critical value, boundary layer separation is imminent. The separation tendency coefficients at each chordal position are summarized to form a separation tendency distribution curve along the chord. As the sail angle of attack increases, the suction peak strengthens, but the adverse pressure gradient region also expands accordingly. The separation tendency coefficient increases significantly in the trailing edge region.

[0028] A boundary layer shape factor analysis was performed on the separation tendency coefficient to generate a separation sensitivity distribution field. The boundary layer shape factor is defined as the ratio of displacement thickness to momentum thickness, reflecting the fullness of the boundary layer velocity profile. A larger shape factor indicates that the boundary layer is less full and closer to separation, while a smaller shape factor indicates that the boundary layer is more stable and full. The boundary layer shape factor corresponding to each position on the separation tendency coefficient distribution curve was calculated. The shape factor of the laminar boundary layer is approximately 2.6, and the shape factor of the turbulent boundary layer is approximately 1.3 to 1.5. When the shape factor exceeds the critical value, the boundary layer is at the edge of separation. When a suction sail is sailing at sea, it needs a large angle of attack to obtain sufficient thrust. The trailing edge region of the sail is subjected to a strong adverse pressure gradient, which leads to an increase in the separation tendency coefficient. At the same time, the boundary layer velocity profile in this region gradually becomes less full, increasing the shape factor. The sensitivity value obtained by combining the separation tendency coefficient and the shape factor according to weights is significantly higher in the trailing edge region than in the leading edge region, indicating that the trailing edge region has the most significant response to suction control and is the target location for priority deployment of suction resources. The separation tendency coefficient and shape factor are weighted and combined, with the weighting coefficient determined based on their respective contributions to separation. The combined value characterizes the sensitivity of that location to suction control; a higher value indicates a more significant separation suppression effect when suction is applied at that location. The sensitivity values ​​for each chordal location are extended to the spanwise direction to form a two-dimensional distribution, namely the separation sensitivity distribution field. The separation sensitivity distribution field describes the separation sensitivity at each spatial location based on the physical coordinates of the sail surface. The grid resolution of the separation sensitivity distribution field is 2% of the chord length along the chordal direction and 5% of the spanwise length along the spanwise direction.

[0029] The separation risk index is determined by comparing the separation sensitivity distribution field with a preset critical separation threshold. A critical separation threshold is set as the benchmark for judging the occurrence of separation; this threshold is calibrated based on the aerodynamic characteristics of the sail and historical data, and the critical threshold varies for different airfoils. The sensitivity values ​​at each grid point in the separation sensitivity distribution field are iterated, and the values ​​at each location are compared with the critical separation threshold. Sensitivity values ​​below the threshold indicate a lower separation risk, while sensitivity values ​​close to or exceeding the threshold indicate a higher separation risk. When a suction sail encounters a sudden change in wind direction while sailing at sea, it needs to adjust its angle of attack to adapt to the new incoming flow direction. During this angle of attack adjustment, the pressure distribution on the suction surface of the sail changes drastically. The separation sensitivity distribution field shows that the sensitivity value in the trailing edge region of the sail rises rapidly and approaches the critical separation threshold. Comparing the sensitivity value in this region with the critical threshold yields a separation risk index approaching the critical state. Based on this, it is determined that flow separation is imminent on the sail surface, triggering an emergency increase in suction flow to maintain attached flow. The separation risk index is calculated using the formula R_sep=min(1,D / D_cr), where R_sep is the separation risk index, dimensionless, ranging from 0 to 1; D is the sensitivity value in the separation sensitivity distribution field, dimensionless; and D_cr is the critical separation threshold, dimensionless. This formula normalizes the sensitivity to the interval between 0 and 1 and saturates at the threshold. The normalized risk scores are recorded according to spatial location to form a separation risk index distribution. A separation risk index of 0 indicates no separation risk, while 1 indicates a critical separation state or that separation has already occurred. When the sail angle of attack approaches the stall angle of attack, the separation risk index approaches 1 in the trailing edge region, prompting the control layer to initiate maximum flow pumping to maintain lift.

[0030] The separation risk index is converted into a suction demand distribution map using chord-direction spatial mapping. The chord-direction distribution curve of the separation risk index is mapped to the physical coordinate system of the sail surface, establishing a correspondence between the separation risk index and the sail position. The suction demand intensity at each position is determined based on the magnitude of the separation risk index. Positions with high risk indices require stronger suction to delay separation; removing low-energy fluid from the boundary layer through suction enhances the boundary layer's resistance to adverse pressure gradients. Positions with low risk indices have weaker or no suction demand; excessive suction increases energy consumption with limited benefits. When the suction sail operates at a medium angle of attack, the separation risk index shows lower flow stability risk values ​​in the leading and middle sections of the sail, while the risk value increases significantly in the trailing edge due to the adverse pressure gradient. The suction demand distribution map generated after substituting the separation risk index at each position into the mapping function exhibits a distribution characteristic of low at the front and high at the back. Based on this, the control system concentrates the suction airflow to the high-risk area at the trailing edge to suppress the separation trend, while maintaining only the minimum suction volume in the low-risk area at the leading edge to save energy, thus achieving optimal allocation of suction resources. A mapping function is established between the separation risk index and the pumping demand. This function employs a piecewise linear or S-curve form to ensure zero pumping in low-risk areas, maximum pumping in high-risk areas, and a smooth transition in intermediate areas. The mapped pumping demand values ​​are then organized into a grid based on the sail's spatial location, forming a two-dimensional distribution field, i.e., a pumping demand distribution map. The map uses the chordal and spanwise directions of the sail as coordinate axes, with each grid point recording the pumping flow demand value at that location. The grid resolution of the pumping demand distribution map is consistent with the separation sensitivity distribution field for subsequent control calls.

[0031] Step S130: Generate angle-of-attack-lift coupling response curve based on the dynamic change characteristics of sail angle of attack data; perform sensitivity analysis on angle-of-attack-lift coupling response curve to obtain flow regulation coefficient; and divide the sail into leading edge region, mid-chord region and trailing edge region according to flow regulation coefficient to form regional flow configuration table.

[0032] Specifically, an angle-of-attack (AOA)-lift coupled response curve is generated based on the dynamic variation characteristics of sail OAA data. The OAA data records the OAA value and its variation patterns at various times. Analysis of the temporal variation of the OAA data reveals the distribution frequency and dwell time of the OAA in different numerical ranges. When a ship is sailing at sea, the OAA typically fluctuates within a certain range, with the center of fluctuation depending on the angle between the wind direction and the heading. A correspondence between OAA and lift is established by combining a database of sail aerodynamic characteristics or real-time measured lift data. Within the OAA range covered by the sail OAA data, the lift coefficient corresponding to each OAA value is extracted. The lift coefficient reflects the magnitude of lift generated per unit area of ​​sail surface. Data from the suction sail showed that the angle of attack fluctuated between 4° and 14° during a certain flight segment. The lift coefficients for each 1° interval within this range, obtained from the aerodynamic database, were 0.45, 0.56, 0.67, 0.78, 0.88, 0.96, 1.02, 1.06, 1.08, 1.05, and 0.94. Connecting these angle-of-attack and lift data points formed the angle-of-attack-lift coupling response curve. Plotting the relationship between the angle of attack value on the x-axis and the lift coefficient on the y-axis, the angle-of-attack-lift coupling response curve exhibited a typical pattern of initial increase followed by a decrease. As the angle of attack increased from zero, the lift coefficient increased approximately linearly, but decreased sharply after reaching the stall angle of attack. The slope of the angle-of-attack-lift coupling response curve reflects the sensitivity of lift to changes in angle of attack. Suction control, by delaying flow separation, can alter the curve shape and improve the stall angle of attack.

[0033] In some embodiments, the step of performing sensitivity analysis on the angle-of-attack-lift coupling response curve to obtain the flow rate regulation coefficient includes: extracting the lift gradient distribution on the intake side and the pressure side from the angle-of-attack-lift coupling response curve; determining the angle-of-attack-flow rate joint sensitivity based on the lift gradient distribution to form a sensitivity distribution sequence; identifying inflection points in the sensitivity distribution sequence to determine the efficient control angle-of-attack range; and determining the corresponding flow rate regulation coefficient based on the efficient control angle-of-attack range.

[0034] The lift gradient distribution on the intake and pressure sides is extracted from the angle-of-attack-lift coupling response curve. The lift generated by the sail originates from the pressure difference between the intake and pressure sides. The contribution ratio of each side to the total lift varies with the angle of attack, with the low-pressure area on the intake side typically contributing the majority of the lift. The angle-of-attack-lift coupling response curve is decomposed to extract the lift components on the intake and pressure sides as a function of the angle of attack. The derivative of the lift component on the intake side with respect to the angle of attack is calculated to obtain the lift gradient on the intake side, and the derivative of the lift component on the pressure side with respect to the angle of attack is calculated to obtain the lift gradient on the pressure side. The total lift coefficient of the suction sail at an angle of attack of 8° is 0.88, with the intake side contributing 0.62 and the pressure side contributing 0.26. The derivative of the angle-of-attack-lift coupling response curve at this point yields a lift gradient of 0.095 / ° on the intake side and 0.025 / ° on the pressure side. The gradient values ​​on both sides are paired and recorded to form the lift gradient distribution data at this angle of attack. The lift gradients on both sides were paired according to the angle of attack to form a lift gradient distribution dataset. The lift gradient distribution reveals the distribution law of lift gain on both sides of the sail. Within the small angle of attack range, the lift gradient distribution shows that the gradients on both sides are positive and relatively stable. Within the large angle of attack range, the gradient on the intake side decreases significantly while the gradient on the pressure side remains relatively stable. This is because flow separation on the intake side leads to a weakening of suction. The changing trend of the lift gradient distribution indicates the development process of flow separation, and a sharp decrease in the gradient on the intake side usually indicates the occurrence of separation.

[0035] The angle-of-attack-flow combined sensitivity is determined based on the lift gradient distribution, forming a sensitivity distribution sequence. The lift gradient values ​​at each angle of attack in the lift gradient distribution reflect the potential benefits of applying suction control at that angle of attack. In angle-of-attack intervals with high lift gradients but already declining, suction control can significantly restore the lift gradient by delaying separation, resulting in the highest suction benefit. In angle-of-attack intervals with stable gradients in the lift gradient distribution, the marginal benefit of suction control is lower because the flow has not yet shown a separation tendency, and the improvement space for suction is limited. When a suction sail encounters crosswinds at sea, it needs to increase the angle of attack to obtain greater lateral thrust. The lift gradient distribution shows that as the angle of attack increases, the lift gradient on the intake side begins to decline from the peak region, indicating that the boundary layer flow is approaching the critical separation state. At this point, applying suction control can effectively delay separation and restore the lift gradient. The combined sensitivity at this angle of attack is calculated as a high level based on the ratio of the improvement in lift gradient before and after suction to the suction flow rate. However, under stable conditions at small angles of attack, the lift gradient is stable with no separation tendency, and the improvement in lift gradient after applying the same suction flow rate is very small, resulting in a low level of combined sensitivity. The angle-of-attack-flow rate joint sensitivity was calculated at each angle of attack using data on the impact of pump flow rate on the lift gradient distribution. The joint sensitivity is defined as the improvement in lift gradient caused by a unit change in pump flow rate, calculated as S = ΔC_Lα / ΔQ, where S is the joint sensitivity (unit: 1 / (°·m³ / s); ΔC_Lα is the improvement in lift gradient (unit: 1 / °); and ΔQ is the change in pump flow rate (unit: m³ / s). The joint sensitivity values ​​for each angle of attack were arranged in order of angle of attack to form a sensitivity distribution sequence. This sequence describes the variation of joint sensitivity with angle of attack, exhibiting a typical trend of initially low, then high, and then decreasing.

[0036] Inflection point identification of the sensitivity distribution sequence determines the efficient control angle of attack range. The curve shape of the sensitivity distribution sequence exhibits characteristic points where the sensitivity values ​​change significantly. The first and second derivatives of the sensitivity distribution sequence are calculated. The first derivative reflects the rate of change of sensitivity, while the second derivative reflects the change in the rate of change, i.e., the concavity / convexity of the curve. The inflection point is the location on the sensitivity distribution sequence curve where the sign of the second derivative changes, marking a turning point in the sensitivity change trend. The sensitivity distribution sequence of the suction sail shows that the joint sensitivity is stable within the range of 0.008 to 0.012 from the angle of attack between 4° and 8°. From 8°, the sensitivity rapidly increases to 0.034 at 10°, reaches a peak of 0.041 at 12°, and then begins to decline. Calculation of the second derivative identifies 8° as the lower boundary inflection point and 12° as the upper boundary inflection point, defining 8° to 12° as the efficient control angle of attack range. The inflection point where the sensitivity increases from a slow rise to a rapid rise corresponds to the lower boundary of the efficient control angle of attack range. The inflection point where the sensitivity reaches its peak and then begins to decline corresponds to the upper boundary of the efficient control angle of attack range. The efficient control angle of attack range is the operating range where the pumping control has the optimal cost-effectiveness. When the sail's operating angle of attack enters the efficient control angle of attack range, the control layer increases the pumping flow rate; when the angle of attack exits this range, the flow rate is appropriately reduced to save energy.

[0037] The corresponding flow rate adjustment coefficient is determined based on the efficient control angle of attack range. Statistical characteristics of the sensitivity distribution sequence are extracted within the efficient control angle of attack range, including the average sensitivity, maximum sensitivity, and sensitivity variation range. The average sensitivity of the range is used as the representative sensitivity index for that angle of attack range. When the suction sail is sailing at sea, changes in the angle between wind direction and heading require the sail to operate at different angles of attack. When the angle of attack enters the efficient control angle of attack range, the sail is in a critical separation state, at which point the cost-effectiveness ratio of suction control is optimal. Based on the average sensitivity within this range and the preset lift recovery target, the required flow rate compensation per unit angle of attack deviating from the optimal value is calculated; this is the flow rate adjustment coefficient. This coefficient guides the control layer to quickly adjust the suction flow rate to maintain lift stability during angle of attack fluctuations. Based on representative sensitivity indicators and preset control gain requirements, the required flow regulation amplitude to achieve the target control effect is calculated. The flow regulation coefficient is defined as the flow compensation required when the angle of attack deviates from the optimal value by a unit angle. The calculation formula is K_Q = ΔQ_opt / Δα, where K_Q is the flow regulation coefficient, in (m³ / s) / °; ΔQ_opt is the optimal flow change, in m³ / s; and Δα is the angle of attack change, in °. The flow regulation coefficient has a larger value within the high-efficiency control angle of attack range and a smaller value or zero outside the range. This zoning setting allows the suction control to respond actively in the high-efficiency range and operate conservatively in the low-efficiency range. The correspondence between the flow regulation coefficient and the angle of attack is stored in the form of a lookup table or function for real-time access by the control layer.

[0038] Based on the flow regulation coefficient, the sail is divided into leading-edge, mid-chord, and trailing-edge regions to form a regional flow configuration table. According to the distribution characteristics of the flow regulation coefficient at different chordal positions, the sail is divided into three control zones along the chord direction, based on the differences in flow characteristics and suction effects in each zone. The leading-edge zone, located at approximately 0 to 30% of the chord length at the front of the sail, exhibits accelerated flow, a thin and stable boundary layer, and a low flow regulation coefficient, requiring minimal suction and primarily used to control the transition point of the laminar boundary layer. The mid-chord zone, located at 30% to 70% of the chord length, shows a gradually thickening boundary layer that may begin to be affected by the adverse pressure gradient. It has a moderate flow regulation coefficient and requires moderate suction to maintain boundary layer stability and prevent premature separation. The trailing-edge zone, located at 70% to 100% of the chord length, has the strongest adverse pressure gradient, the highest risk of separation, and the largest flow regulation coefficient, making it the key area for suction control. The suction sail allocates baseline flow rates to three zones based on the chordal distribution of the flow rate adjustment coefficients. The baseline flow rate for the leading edge zone is set at 0.08 m³ / s, with an adjustment range of 0 to 0.15 m³ / s; for the middle chord zone, it is set at 0.25 m³ / s, with an adjustment range of 0.1 to 0.4 m³ / s; and for the trailing edge zone, it is set at 0.45 m³ / s, with an adjustment range of 0.2 to 0.7 m³ / s. These parameters are organized into a tabular form to form a zone flow rate configuration table. The zone flow rate configuration table includes fields such as zone number, chordal range, baseline flow rate, minimum flow rate, maximum flow rate, and priority. The control layer quickly determines the flow rate allocation scheme for each zone by querying the zone flow rate configuration table based on the current operating conditions.

[0039] Step S140: Perform flow-lift-resistance ratio response surface analysis on the suction demand distribution map to generate lift-resistance ratio gain values ​​for each region. Perform efficiency and power consumption co-optimization analysis on the lift-resistance ratio gain values ​​to generate a flow priority sequence. Based on the flow priority sequence and the regional flow configuration table, construct an adaptive control strategy.

[0040] Specifically, flow rate-lift-drag ratio response surface analysis is performed on the suction demand distribution map to generate lift-drag ratio gain values ​​for each region. The suction demand distribution map is divided into leading edge, mid-sine, and trailing edge regions, and a response relationship between suction flow rate and aerodynamic performance is established for each region. Lift-drag ratio is a core indicator for measuring sail aerodynamic efficiency, defined as the ratio of lift coefficient to drag coefficient. A higher lift-drag ratio indicates that the sail consumes less drag to generate the same lift. Within the flow rate range indicated by the suction demand distribution map, different intensities of suction flow rate are applied to each region, and the corresponding changes in lift coefficient and drag coefficient are recorded. Flow rate is used as the input variable, and lift-drag ratio is used as the output variable to construct the response surface. When a suction sail is sailing at sea, the response characteristics of different regions of the sail to the suction flow rate vary significantly. In the trailing edge region, due to the strong adverse pressure gradient and high risk of separation, increasing the suction flow rate significantly improves the boundary layer adhesion, increasing the lift coefficient while substantially reducing the pressure drag caused by separation. The lift-to-drag ratio increases rapidly with increasing flow rate, and a steeper gradient in the response surface corresponds to a larger lift-to-drag ratio gain. In the leading edge region, the boundary layer is thin and stable, and increasing the suction flow rate provides limited improvement in aerodynamic performance. A gentler gradient in the response surface corresponds to a smaller lift-to-drag ratio gain. The response surface is established using polynomial fitting or radial basis function interpolation methods, and its shape reflects the nonlinear influence of the suction flow rate on aerodynamic performance. The formula for calculating the lift-to-drag ratio gain is as follows: Where G_L / D is the lift-to-drag ratio gain, in units of 1 / (m³ / s); C_L / C_D is the lift-to-drag ratio, dimensionless; and Q is the pump flow rate, in units of m³ / s. This formula represents the gradient of the response surface at the current operating point, i.e., the change in lift-to-drag ratio caused by a unit change in flow rate. The lift-to-drag ratio gain values ​​differ significantly in different regions. The trailing edge region typically has the highest lift-to-drag ratio gain value due to its high separation risk, while the leading edge region has a relatively lower lift-to-drag ratio gain value.

[0041] In some embodiments, the step of generating a flow priority sequence by performing a performance-power co-optimization analysis on the lift-to-drag ratio gain value includes: decomposing the lift gain component and drag reduction component from the lift-to-drag ratio gain value to construct a performance evaluation space; evaluating the aerodynamic performance contribution of the evaluation area within the performance evaluation space to generate a performance ranking; analyzing the suction power consumption of each area based on the performance ranking to generate a power consumption cost index; and performing a co-optimization of the performance ranking and the power consumption cost index to obtain a flow priority sequence.

[0042] The performance evaluation space is constructed by decomposing the lift gain component and drag reduction component from the lift-to-drag ratio gain value. The lift-to-drag ratio gain value is a comprehensive reflection of the increase in lift coefficient and the increase in drag coefficient. Separating the two components allows for a more refined evaluation of the aerodynamic contribution of each region. The lift-to-drag ratio gain value of each region is decomposed, and the change in lift coefficient caused by suction is extracted as the lift gain component, and the change in drag coefficient caused by suction is extracted as the drag reduction component. A positive lift gain value indicates that suction increases lift, and a positive drag reduction value indicates that suction reduces drag. The decomposition results of the lift-to-drag ratio gain value in the three regions of the suction sail are as follows: the lift-to-drag ratio gain value of 1.2 / (m³ / s) in the leading edge region corresponds to a lift gain component of 0.8 and a drag reduction component of 0.4; the lift-to-drag ratio gain value of 2.5 / (m³ / s) in the mid-chord region corresponds to a lift gain component of 1.6 and a drag reduction component of 0.9; and the lift-to-drag ratio gain value of 3.8 / (m³ / s) in the trailing edge region corresponds to a lift gain component of 2.3 and a drag reduction component of 1.5. A two-dimensional coordinate system is established with the lift gain component as the horizontal axis and the drag reduction component as the vertical axis. The component values ​​of each region are marked on the coordinate system to form a scattered distribution. This coordinate system is the performance evaluation space. Iso-performance lines can also be drawn in the performance evaluation space. Points on the same iso-performance line have the same comprehensive performance level. The performance evaluation space presents the relative performance of each region in both lift and drag dimensions in a visual way.

[0043] Within the performance evaluation space, the aerodynamic performance contribution of each evaluation region is used to generate a performance ranking. A method for calculating the performance contribution is defined within the performance evaluation space, where the lift gain component and drag reduction component are weighted and summed to obtain a comprehensive performance index. The weighting coefficients are determined based on the ship's current navigation requirements. When the ship needs to maximize propulsion, the weight of lift gain is higher; when the ship needs to reduce fuel consumption, the weight of drag reduction is higher. The weights can be dynamically adjusted according to the navigation mission. For the suction sail operating in crosswind conditions, the lift gain weight is set to 0.7 and the drag reduction weight to 0.3. Based on the component values ​​of each region in the performance evaluation space, the comprehensive performance index is calculated: the comprehensive performance of the trailing edge region is 2.06, the comprehensive performance of the mid-chord region is 1.39, and the comprehensive performance of the leading edge region is 0.68. Based on this, the performance ranking is generated as follows: trailing edge region first, mid-chord region second, and leading edge region third. The comprehensive performance index of each region in the performance evaluation space is calculated. A higher comprehensive performance index indicates a greater aerodynamic contribution to that region. The regions are then sorted from highest to lowest comprehensive performance index to form a performance ranking list. The performance ranking is dynamically updated as navigation conditions and mission requirements change. The performance ranking of each region may change when the ship changes course or adjusts its speed target.

[0044] The power consumption index is generated by analyzing the suction power consumption of each region based on the performance ranking. The performance ranking only reflects the quality of aerodynamic performance and does not consider the energy cost required to achieve that performance. Different regions may require different power consumption to achieve the same performance. For each region in the performance ranking, the suction power consumption required to reach the current lift-to-drag ratio gain value is calculated. The power consumption calculation formula is P=Q×Δp, where P is the suction power consumption in W; Q is the suction flow rate in m³ / s; and Δp is the pressure difference inside and outside the suction orifice in Pa. When a suction sail is sailing at sea, the suction power consumption in different regions varies significantly due to the pressure distribution on the sail surface. The leading edge region, located near the suction peak, has low external pressure and a large pressure difference between the inside and outside of the suction orifice, resulting in higher power consumption per unit flow rate. However, the required flow rate in this region is small, so the total power consumption remains relatively low. The trailing edge region, close to the incoming static pressure zone, has relatively high external pressure and a smaller pressure difference between the inside and outside of the suction orifice, resulting in lower power consumption per unit flow rate. However, the required flow rate in this region is large, so the total power consumption is relatively high. Normalization is performed using the region with the highest power consumption as the benchmark to obtain a power consumption cost index. This index reflects the relative magnitude of suction energy consumption in each region. The power consumption cost index ranges from 0 to 1; a higher value indicates higher suction energy consumption in that region. Using the power consumption cost index in conjunction with performance ranking allows for prioritizing regions with lower power consumption among regions with similar performance.

[0045] For example, the step of collaboratively optimizing the performance ranking and the power consumption cost index to obtain the traffic priority sequence includes: constructing a joint performance and power consumption evaluation vector based on the performance ranking and the power consumption cost index; generating a weighted evaluation vector by weighting the joint performance and power consumption evaluation vector by contribution through the performance ranking; obtaining a candidate solution set by screening the weighted evaluation vector for balanced solutions; and determining the traffic priority sequence by fusing the candidate solution set.

[0046] A joint performance-power consumption evaluation vector is constructed based on performance ranking and power consumption cost indicators. The ranking position of each region in the performance ranking is converted into a numerical performance score. A reciprocal transformation or normalization mapping is used to ensure the region with the highest ranking receives the highest score. Performance scores and power consumption cost indicators are combined into tuples, with each region corresponding to a tuple representing its performance and power consumption evaluation results. The suction sail converts the ranking positions of the three regions into performance scores based on the performance ranking: the trailing edge region is ranked first with a score of 1.0, the mid-sine region is ranked second with a score of 0.67, and the leading edge region is ranked third with a score of 0.33. These scores, combined with the power consumption cost indicator, are then arranged into tuples by region number to form the joint performance-power consumption evaluation vector V = [1.0, 1.0, 0.67, 0.9, 0.33, 0.45]. All region tuples are arranged by region number to form the joint performance-power consumption evaluation vector. The length of the joint performance-power consumption evaluation vector is equal to twice the number of regions. Odd-numbered positions store performance scores, and even-numbered positions store power consumption costs. The joint performance and power consumption evaluation vector carries both performance and power consumption information in a compact data structure, which facilitates subsequent weighted processing and optimization calculations.

[0047] A weighted evaluation vector is generated by weighting the joint performance and power consumption evaluation vector based on contribution. The contribution weight of each region is determined according to its ranking position in the performance ranking, with higher-ranked regions having higher weights and lower-ranked regions having lower weights. An exponential decay function is used in the weight allocation to ensure that the weight of the head region is significantly higher than that of the tail region. The contribution weight calculation formula is w_i=e^(-λ×r_i) / Σe^(-λ×r_j), where w_i is the contribution weight of the i-th region (dimensionless); r_i is the ranking position of that region in the performance ranking; λ is a decay coefficient controlling the concentration of weight distribution (dimensionless); and the denominator is a normalization factor to ensure that the sum of all weights is 1. When the suction sail is navigating at sea, the contribution weight of each region is determined based on its effectiveness ranking. The trailing edge region, due to its greatest aerodynamic performance contribution, ranks first and receives the highest weight. Its effectiveness score remains high after weighting, while its power consumption cost is relatively amplified, indicating that although this region has outstanding effectiveness, its power consumption investment requires close attention. The leading edge region, due to its smaller aerodynamic performance contribution, ranks last and receives the lowest weight. Its effectiveness score further decreases after weighting, while its power consumption cost is relatively reduced, indicating that this region has a limited impact on the overall control effect and resource allocation can be appropriately reduced. The weighted evaluation vector maintains the same structure as the original vector, but its values ​​are adjusted for contribution. The relative magnitudes of the elements in the weighted evaluation vector reflect the regional evaluation results after comprehensively considering the effectiveness ranking priority.

[0048] Candidate solution sets are obtained by screening the weighted evaluation vector for equilibrium solutions. The algorithm seeks regions within the weighted evaluation vector that achieve a balance between efficiency and power consumption. An equilibrium solution is one that achieves relatively high efficiency without significantly increasing power consumption. Simply pursuing high efficiency may lead to excessive power consumption, while simply pursuing low power consumption may sacrifice aerodynamic performance. Efficiency and power consumption thresholds are set as screening criteria. The efficiency threshold excludes regions with excessively low aerodynamic contributions, while the power consumption threshold excludes regions with excessively high energy costs. Regions with weighted efficiency scores higher than the efficiency threshold and weighted power consumption costs lower than the power consumption threshold are extracted from the weighted evaluation vector. The suction sail has an efficiency threshold of 0.15 and a power consumption threshold of 3.0. It iterates through the weighted efficiency scores and weighted power consumption costs of each region in the weighted evaluation vector. The trailing edge region, with a weighted efficiency score of 0.506 higher than the threshold and a weighted power consumption cost of 1.98 lower than the threshold, meets the condition. The mid-sine region, with a weighted efficiency score of 0.206 higher than the threshold and a weighted power consumption cost of 2.93 lower than the threshold, also meets the condition. The leading edge region, with a weighted efficiency score of 0.062 lower than the efficiency threshold, is excluded. The trailing edge and mid-sine regions that meet the conditions are included in the preliminary candidate set. Pareto optimality analysis is performed on the regions in the preliminary candidate set to identify solutions that are non-dominated in the efficiency-power consumption two-dimensional space. A non-dominated solution is one where no other solution is superior in both efficiency and power consumption dimensions. All non-dominated solutions form the Pareto front. The region ranking schemes on the Pareto front are summarized to form a candidate solution set. The candidate solution set contains multiple alternative flow priority ranking schemes, each achieving a different equilibrium point between efficiency and power consumption.

[0049] The candidate solution set is fused to determine the flow priority sequence. The candidate solution set contains multiple balanced solutions, from which one needs to be selected as the final flow allocation basis. Different solutions are suitable for different navigation scenarios and energy supply conditions. Each solution in the candidate solution set has different characteristics: efficiency-priority solutions offer better aerodynamic performance but higher power consumption, suitable for use when power is sufficient; power consumption-priority solutions have lower energy consumption but slightly sacrifice aerodynamic performance, suitable for long-range energy-saving navigation; balanced solutions strike a compromise between the two, suitable for normal navigation conditions. The appropriate solution type is selected based on the current navigation mission and energy supply status. When performing ocean freight missions, fuel economy is prioritized, so the power consumption-priority solution is selected; when maneuvering in port or making emergency avoidance, maximum propulsion is required, so the efficiency-priority solution is selected. Under normal navigation conditions, the suction sail selects a balanced solution from the candidate solution set. This solution comprehensively considers the efficiency contribution and power consumption cost of each region, prioritizing the trailing edge region first, the mid-chord region second, and the leading edge region third. The selected solution is fused to output the flow priority sequence [trailing edge region, mid-chord region, leading edge region]. The candidate solution sets in the selected type are fused. When multiple similar solutions exist, their average ranking is taken as the final result. When the solutions differ significantly, the solution with the highest comprehensive score is selected. The fused region ranking results are output as a traffic priority sequence, which is represented as an ordered list of region numbers. Regions ranked higher have higher priority in traffic allocation.

[0050] An adaptive control strategy is constructed based on the integration of a flow priority sequence and a regional flow allocation table. The flow priority sequence serves as the priority basis for flow allocation between regions, while the regional flow allocation table serves as the reference benchmark for flow values ​​in each region. The two are integrated to form a complete control scheme. When the pump power is sufficient, each region is supplied with the benchmark flow rate from the regional flow allocation table, ensuring that each region receives the suction flow rate required for separation and suppression. When the suction sail encounters a sudden gust of wind at sea, the total pump power is limited to 600W. Based on the flow priority sequence, the trailing edge region is prioritized to receive the benchmark flow rate of 0.45 m³ / s from the regional flow allocation table, consuming 360W of power. The remaining 240W of power is allocated to the mid-south edge region according to the flow priority sequence, receiving 0.2 m³ / s. The leading edge region is temporarily not supplied. Once the pump power recovers to 1000W after the gust, the benchmark flow rate supply to each region is restored according to the regional flow allocation table. When the pump power is limited, flow is allocated sequentially according to the flow priority sequence. Regions ranked higher receive the benchmark flow rate first, while regions ranked lower receive partial flow or no flow within the allowable range of remaining power. A dynamic power allocation adjustment mechanism is established, prioritizing the flow stability of regions with higher flow priority when total power is redistributed among different regional demands. The merged allocation rules, adjustment mechanism, and execution logic are integrated and encoded into an adaptive control strategy. This adaptive control strategy is stored in the form of a rule base or state machine, supporting automatic selection of execution paths based on real-time operating conditions. It can autonomously adjust the flow allocation scheme to maintain the stability of sail aerodynamic performance under conditions such as air pump power fluctuations or sudden wind changes.

[0051] Step S150: The adaptive control strategy is decomposed into time-varying differentiated suction sequences by region, and adjacent region flow gradient constraints are applied to the time-varying differentiated suction sequences to generate smooth transition suction sequences. Real-time flow regulation commands are generated based on the smooth transition suction sequences.

[0052] In some embodiments, the step of generating time-varying differentiated suction sequences by decomposing the adaptive control strategy by region includes: establishing a flow reference set by defining the reference suction flow of each region according to the adaptive control strategy; performing feedforward-feedback decoupling analysis on the flow reference set to determine the dynamic adjustment range; performing time-domain expansion of the dynamic adjustment range to generate regional flow adjustment amounts; and integrating the regional flow adjustment amounts by region to form a time-varying differentiated suction sequence.

[0053] A flow reference set is established by defining the baseline suction flow rate for each region based on the adaptive control strategy. The adaptive control strategy includes flow configuration parameters for each region, including baseline flow rate, minimum flow rate, maximum flow rate, and priority coefficient. The target suction flow rate for each region under standard operating conditions is determined according to the configuration rules of the adaptive control strategy. The baseline flow rate is jointly determined by the region flow configuration table and the flow priority sequence, reflecting the region's normal suction demand. In the adaptive control strategy, the baseline flow rate for the trailing edge region is typically set to the highest, and the baseline flow rate for the leading edge region is set to the lowest. After the suction sail enters the stable navigation phase, baseline flow rates are defined for the three regions according to the adaptive control strategy: 0.08 m³ / s for the leading edge region, 0.25 m³ / s for the mid-chord region, and 0.45 m³ / s for the trailing edge region. The trailing edge region receives the maximum allowance due to the highest separation risk, while the leading edge region requires only a smaller flow rate to maintain transition control due to boundary layer stability. The baseline flow rates for each region and their allowable adjustment ranges are summarized to form the flow reference set. The flow reference set stores the reference flow values ​​and their allowable adjustment ranges for each region using the region number as an index. The flow reference set also records the source basis of the reference flow for each region, including key parameters such as separation risk level, lift-to-drag ratio, and efficiency-to-power ratio. The flow reference set remains relatively stable during ship navigation and is only recalculated and updated when wind conditions change significantly or navigation tasks are adjusted.

[0054] For example, the step of performing feedforward-feedback decoupling analysis on the flow reference set to determine the dynamic adjustment range includes: extracting reference flow values ​​for each region from the flow reference set; performing condition identification analysis on the reference flow values ​​to generate a state-flow correspondence; performing gain scheduling analysis on the state-flow correspondence to generate flow adjustment requirements; and determining the dynamic adjustment range based on the flow adjustment requirements.

[0055] The baseline flow rate values ​​for each region are extracted from the flow rate baseline set. The data structure of the flow rate baseline set stores the baseline flow rate values ​​for each region in order of region number. The baseline flow rate values ​​for each region in the flow rate baseline set are read in order of region number and stored in the working buffer for subsequent calculations. The unit of the baseline flow rate value is usually cubic meters per second or liters per minute. The value reflects the suction intensity requirement of each region. The baseline flow rate value of the trailing edge region is usually the largest, and the baseline flow rate value of the leading edge region is usually the smallest. At the beginning of each control cycle, the suction sail reads the baseline flow rate values ​​of the three regions from the flow rate baseline set, which are 0.08, 0.25, and 0.45 m³ / s respectively. The validity of the values ​​is verified to confirm that the maximum value in the trailing edge region and the minimum value in the leading edge region are consistent with the expected relationship. Simultaneously, the current baseline flow rate value is compared with the historical average to determine whether the deviation exceeds the 5% alarm threshold. Verifying the validity of the baseline flow rate value confirms that the value is within a reasonable range and that the values ​​of each region conform to the expected relationship. The extraction of the baseline flow rate value is the starting point for dynamic adjustment calculations. Subsequent operating condition identification and gain scheduling are all adjusted based on the baseline flow rate value.

[0056] A state-flow correspondence is generated by analyzing the baseline flow rate value based on operating conditions. The operating condition category of the system is identified based on the current incoming wind speed, sail angle of attack, and separation risk index. Operating condition categories include low-wind-speed stable conditions, medium-wind-speed transitional conditions, high-wind-speed extreme conditions, and gust disturbance conditions. The flow rate demand patterns differ across regions under different operating conditions. Flow rate demand is low in all regions under low-wind-speed conditions, while it significantly increases in the trailing edge region under high-wind-speed conditions. When the suction sail encounters the monsoon belt during ocean voyages, the incoming wind speed increases to 15 m / s, and the sail angle of attack is forced to increase to 11°. Based on the baseline flow rate value and current operating condition parameters, the system is identified as entering a high-wind-speed extreme condition. The corresponding adjustment coefficient vector for this condition is [1.1, 1.25, 1.4], indicating that the baseline flow rate value for each region needs to be amplified by different proportions, with the highest amplification ratio in the trailing edge region to cope with the rapidly increasing separation risk. The operating condition identifier is associated with the adjustment coefficient vector to form a state-flow correspondence. A mapping rule is established between operating condition categories and flow rate adjustment directions. The adjustment coefficients of the baseline flow rate value under different operating conditions are quantified, with a positive adjustment coefficient corresponding to high wind speed conditions and a negative adjustment coefficient corresponding to low wind speed conditions. The currently identified operating condition category and its corresponding adjustment coefficient are associated with the baseline flow rate value to form a state-flow rate correspondence. The state-flow rate correspondence is stored in key-value pair format, where the key is the operating condition status identifier and the value is the flow rate adjustment coefficient vector for each region.

[0057] Gain scheduling analysis is performed on the state-flow correspondence to generate flow regulation requirements. The state-flow correspondence records the regulation coefficient corresponding to the current operating condition. The specific regulation amount is calculated by multiplying the regulation coefficient by the deviation degree of the operating condition. The deviation degree of the operating condition is defined as the difference between the current state parameter and the center value of the operating condition. The deviation degree is positive when the wind speed is higher than the center value of the operating condition, and negative when it is lower than the center value. When the suction sail passes through the outer area of ​​the storm, the wind speed fluctuates continuously. The current wind speed of 15 m / s has a positive deviation from the center value of 14 m / s of the high wind speed operating condition. According to the regulation coefficient in the state-flow correspondence and the gain scheduling strategy, the flow regulation requirements for the three regions are calculated to be +0.012, +0.038, and +0.072 m³ / s, respectively. The trailing edge region has the largest positive regulation requirement due to its high separation risk. According to the gain scheduling strategy, a smaller regulation gain is used near the operating condition boundary to avoid frequent switching, while a larger regulation gain is used in the core area of ​​the operating condition to accelerate the response speed. The adjusted amounts after gain scheduling are aggregated by region to form the flow adjustment requirements of each region. The flow adjustment requirements represent the flow values ​​that each region needs to increase or decrease relative to the baseline value under the current operating conditions. The values ​​of the flow adjustment requirements may be different for each region. The flow adjustment requirements of the trailing edge region under high-risk operating conditions are usually greater than those of other regions.

[0058] The dynamic adjustment range is determined based on the flow regulation requirements. The flow regulation requirements of each region are compared with the allowable adjustment range of the flow reference set. Flow regulation requirements exceeding the range are limited, using a saturation function to ensure that the actual flow does not exceed the pump capacity or fall below the minimum effective flow. Under extreme gust conditions, the flow regulation requirement of the suction sail in the trailing edge region is +0.072 m³ / s. The reference flow plus the regulation requirement is 0.522 m³ / s, still within the allowable range of 0.2 to 0.7 m³ / s, requiring no limitation. If the adjusted flow exceeds the maximum value of 0.7 m³ / s, the adjustment is limited to the range corresponding to that boundary. Simultaneously, the oscillation phenomenon of frequent directional reversals in the regulation requirements over multiple consecutive control cycles is checked. The dynamic adjustment range is calculated using the formula ΔQ_dyn=sat(Q_req,Q_min,Q_max), where ΔQ_dyn is the dynamic adjustment range, Q_req is the flow regulation requirement, Q_min is the minimum allowable adjustment amount, and Q_max is the maximum allowable adjustment amount. This formula limits the regulation requirement within a safe range. The stability of the flow regulation demand after the limit is checked, and the regulation demand after the limit and stability processing is output as the dynamic regulation amplitude. The dynamic regulation amplitude is the actual flow regulation amount executed by each region at the current moment.

[0059] The dynamic adjustment amplitude is expanded in the time domain to generate the zoned flow regulation amount. The dynamic adjustment amplitude represents the regulation amount at the current moment and needs to be expanded in the time domain to form a continuous regulation trajectory. The continuous time is discretized according to the sampling period of the control system, and the corresponding dynamic adjustment amplitude value is calculated at each sampling moment. Considering the requirement of temporal continuity of the regulation amount, the change of the regulation amount between adjacent sampling moments should not be too drastic. The abrupt dynamic adjustment amplitude is sloped to make it transition smoothly over several sampling periods. When the suction sail is sailing at sea, a sudden change in wind direction causes a rapid change in angle of attack, and the dynamic adjustment amplitude will then undergo a step jump. If this step signal is directly sent to the actuator, it will cause a sudden change in the suction flow. The instantaneous change in the airflow state on the sail surface may cause lift fluctuations or even induce flow separation. After first-order inertial filtering, the step change of the dynamic adjustment amplitude is smoothed into a gradual transition, and the output zoned flow regulation amount gradually approaches the target value, and the aerodynamic performance of the sail surface remains stable during the adjustment process. The time-domain expansion employs a first-order inertial filter. The formula for calculating the regional flow regulation is Q_k(t) = Q_k(t-1) + α × (ΔQ_dyn,k - Q_k(t-1)), where Q_k(t) is the regional flow regulation at time t for the k-th region, in m³ / s; α is a filter coefficient that controls the balance between response speed and smoothness, is dimensionless, and ranges from 0 to 1. The regulation values ​​at each sampling time are arranged in chronological order to form independent time series of regulation values ​​for each region. The time series of regulation values ​​for all regions are then aggregated into a unified data structure to form the regional flow regulation values. The regional flow regulation values ​​are stored in a two-dimensional array, with rows corresponding to time sampling points and columns corresponding to region numbers.

[0060] A time-varying differentiated suction sequence is formed by weighted integration of zoned flow regulation amounts by region. The zoned flow regulation amounts are superimposed with the baseline flow rates in the flow benchmark set to calculate the target suction flow rate for each region at each time point. The target flow rate equals the baseline flow rate plus the corresponding regulation amount in the zoned flow regulation amounts. The target flow rates are validated to ensure that the values ​​are non-negative and do not exceed the maximum flow limit for each region; invalid values ​​are corrected. After superimposing the zoned flow regulation amounts with the baseline flow rates, the target flow rates for the suction sail are 0.507 m³ / s in the trailing edge region, 0.282 m³ / s in the mid-edge region, and 0.089 m³ / s in the leading edge region. The sum of the target flow rates for the three regions is 0.878 m³ / s, which does not exceed the total pump supply capacity of 1.0 m³ / s and does not require reduction. When extreme operating conditions cause the total capacity to exceed the limit, reductions are made sequentially according to the flow priority order, with the leading edge region being the lowest priority and being reduced first, and the trailing edge region being the highest priority and being reduced last. The target flow rate is weighted and adjusted according to the priority weight of each region. When the sum of the target flow rates of all regions exceeds the total supply capacity of the air pump, the flow rate of lower priority regions is reduced sequentially according to priority until the total amount meets the constraint. The weighted and adjusted target flow rates are organized into a standard data format according to time and region to form a time-varying differentiated suction sequence. The time-varying differentiated suction sequence fully describes the suction flow rate plan of each region over a future period. The flow waveforms of each region in the time-varying differentiated suction sequence are independent and reflect differentiated characteristics.

[0061] In some embodiments, applying adjacent region flow gradient constraints to the time-varying differentiated suction sequence to generate a smooth transition suction sequence includes: extracting the flow difference at the boundary of adjacent regions from the time-varying differentiated suction sequence; determining the flow gradient constraint boundary based on the flow distribution characteristics of the time-varying differentiated suction sequence; performing gradient limiting processing based on the flow difference and the flow gradient constraint boundary to generate a flow compensation amount; and superimposing the flow compensation amount onto the time-varying differentiated suction sequence to form a smooth transition suction sequence.

[0062] The flow difference at the boundary between adjacent regions is extracted from the time-varying differentiated pumping sequence. The location of the region boundaries in the time-varying differentiated pumping sequence is identified. The sail is divided along the chord direction into a leading edge region, a middle chord region, and a trailing edge region, with two interfaces: leading edge-middle chord and middle chord-tailing edge. For each time step, the flow difference between the regions on both sides of the interface in the time-varying differentiated pumping sequence is calculated. The flow difference at the leading edge-middle chord boundary equals the flow in the middle chord region minus the flow in the leading edge region, and the flow difference at the middle chord-tailing edge boundary equals the flow in the trailing edge region minus the flow in the middle chord region. The time-varying differential suction sequence of the suction sail under high angle-of-attack conditions shows a flow rate of 0.089 m³ / s in the leading edge region, 0.282 m³ / s in the mid-chord region, and 0.507 m³ / s in the trailing edge region. The calculated flow rate difference at the leading edge-mid-chord boundary is 0.193 m³ / s, and the flow rate difference at the mid-chord-trail boundary is 0.225 m³ / s. Significant flow rate differences exist at both boundaries. Direct execution without constraints would lead to a step-like abrupt change in the suction distribution on the sail surface. The flow rate differences at each boundary at each time point are summarized into a flow rate difference dataset. The flow rate differences are stored with time as the index and boundary as the column. A positive flow rate difference indicates an increase in flow rate along the chord, and the absolute value reflects the degree of abrupt change in flow rate between adjacent regions.

[0063] The flow gradient constraint boundary is determined based on the flow distribution characteristics of the time-varying differentiated pumping sequence. The overall distribution range of flow in each region of the time-varying differentiated pumping sequence determines a reasonable gradient constraint threshold. Statistics such as the global maximum flow, minimum flow, and average flow of the time-varying differentiated pumping sequence are calculated. A reasonable gradient constraint threshold is set based on the flow distribution range; the threshold should effectively limit flow jumps without excessively constraining normal regional differences. Statistical analysis of the time-varying differentiated pumping sequence of the pumping sail shows an average flow of 0.29 m³ / s. The baseline flow gradient constraint boundary is set at 0.073 m³ / s, based on 25% of the average flow. The constraint boundary at the leading edge-mid-chord interface is set at 0.08 m³ / s, which is slightly loose, while the constraint boundary at the mid-chord-tail edge interface is set at 0.065 m³ / s, which is more stringent. The differentiated constraint settings balance smoothing effects with regional control freedom. The flow gradient constraint boundary is defined as the maximum allowable difference in flow between adjacent regions. Different constraint boundaries are set for different interfaces. The constraint boundaries of each interface are summarized to form a flow gradient constraint boundary parameter set. The flow gradient constraint boundary stores the upper and lower limit values ​​with the interface number as the index.

[0064] Flow compensation is generated by gradient limiting based on flow difference and flow gradient constraint boundaries. Each element in the flow difference dataset is compared with the corresponding flow gradient constraint boundary at the interface. No compensation is needed when the absolute value of the flow difference is less than or equal to the flow gradient constraint boundary; otherwise, compensation is calculated to make the adjusted difference equal to the boundary value. The formula for calculating the flow compensation is ΔQ_comp = sign(ΔQ) × (|ΔQ| - ΔQ_lim), where ΔQ_comp is the flow compensation in m³ / s; ΔQ is the original flow difference in m³ / s; and ΔQ_lim is the flow gradient constraint boundary in m³ / s. When a suction sail navigates at sea, different regions receive different target flow rates due to varying separation risks. The flow rate in the trailing edge region is significantly higher than that in the mid-chord region, which in turn is higher than that in the leading edge region. If the flow rate difference at the boundary between regions is too large, it will cause a step-like abrupt change in the suction distribution on the sail surface. The airflow near the boundary will create local disturbances, which will affect the stability of the boundary layer. After limiting the flow rate difference exceeding the constraint boundary, a flow rate compensation amount is generated. This compensation amount is distributed between the regions on both sides of the boundary, so that the high-flow side is moderately reduced and the low-flow side is moderately increased. After adjustment, the flow rate transition between regions is smooth and continuous, and the suction distribution on the sail surface exhibits a smooth gradient change. The distribution of the compensation amount follows the principle of minimum disturbance, distributing the flow rate compensation amount proportionally between the regions on both sides of the boundary, with higher-priority regions bearing less and lower-priority regions bearing more. The flow rate compensation amounts at each boundary at each time point are summarized to form a compensation amount dataset.

[0065] The flow compensation is superimposed on the time-varying differentiated pumping sequence to form a smooth transition pumping sequence. According to the flow compensation allocation scheme, the flow compensation is added to the flow value at the corresponding time in the corresponding region of the time-varying differentiated pumping sequence. In the high-flow-side region of the interface, a portion of the compensation is subtracted to decrease the flow, while in the low-flow-side region, a portion of the compensation is added to increase the flow. The flow compensation of 0.113 m³ / s at the junction of the pumping sail's leading edge and mid-chord is allocated according to priority. The leading edge region bears 70%, increasing the flow by 0.079 m³ / s to 0.168 m³ / s, while the mid-chord region bears 30%, decreasing the flow by 0.034 m³ / s to 0.248 m³ / s. After adjustment, the flow difference between the two regions narrows to 0.08 m³ / s, satisfying the constraint. The step distribution in the time-varying differentiated pumping sequence is transformed into a gentle gradient distribution. The process iterates through all time points and interfaces requiring compensation in the time-varying differentiated suction sequence, adjusting the flow values ​​sequentially. After adjustment, it re-verifies whether the flow difference meets the constraints, iterating and adjusting locations that still exceed the limits until all constraints are met. The adjusted flow data is then reorganized into a standard format to form a smooth transition suction sequence. This smooth transition suction sequence maintains the same data structure as the time-varying differentiated suction sequence but eliminates flow jumps between regions, resulting in smooth and continuous flow changes between adjacent regions.

[0066] Real-time flow regulation commands are generated based on the smooth transition suction sequence. The target flow rate values ​​in the smooth transition suction sequence are converted into execution commands for each suction orifice array. The commands include the flow rate setpoint, regulation rate, and execution time. The flow rate setpoint is directly derived from the value at the corresponding time in the corresponding region of the smooth transition suction sequence. The regulation rate is set according to the response characteristics of the air pump and valves, and the execution time is aligned with the time index of the smooth transition suction sequence. The suction sail generates real-time flow regulation commands for the current moment based on the smooth transition suction sequence. The commands for the leading edge region are: flow rate setpoint 0.168 m³ / s, regulation rate 0.05 (m³ / s) / s; the mid-chord region: flow rate setpoint 0.248 m³ / s, regulation rate 0.08 (m³ / s) / s; and the trailing edge region: flow rate setpoint 0.462 m³ / s, regulation rate 0.1 (m³ / s) / s. When the ship encounters a sudden gust of wind, an emergency regulation command is automatically inserted into the command queue to prioritize the flow supply to the trailing edge region. Instructions are sorted by execution time to form an instruction queue. The control layer sequentially issues real-time flow regulation instructions to each actuator according to the queue order. Instruction issuance adopts a time-triggered method, sending a control signal to the flow regulation device in the corresponding area at the predetermined time. A feedback mechanism for instruction execution is established, collecting the actual flow in each area and comparing it with the target flow. When deviations exist, correction instructions are generated for closed-loop regulation. Real-time flow regulation instructions drive the suction actuators in each area in the form of a continuous instruction stream to achieve adaptive suppression of flow separation.

[0067] To implement the above method embodiment, a method for adaptive adjustment of suction flow rate in a partitioned suction sail is provided to achieve the corresponding functions and technical effects. See also... Figure 2 , Figure 2 This diagram illustrates a structural block diagram of a suction sail zoned suction flow adaptive adjustment system 200 according to an embodiment of this application. For ease of explanation, only the parts relevant to this embodiment are shown. The suction sail zoned suction flow adaptive adjustment system 200 provided in this embodiment includes: Data acquisition module 201 is used to collect incoming wind speed data and sail angle of attack data in the operating environment of the suction sail, and to perform dynamic feature extraction on the incoming wind speed data and sail angle of attack data to construct a flow field state database. Separation analysis module 202 is used to identify boundary layer flow patterns from the flow field state library, perform flow separation analysis on the boundary layer flow patterns to generate a separation risk index, and convert the separation risk index into a pumping demand distribution map using chord-direction spatial mapping processing. The region division module 203 is used to generate an angle-of-attack-lift coupling response curve based on the dynamic change characteristics of the sail angle-of-attack data, perform sensitivity analysis on the angle-of-attack-lift coupling response curve to obtain the flow regulation coefficient, and divide the sail into leading edge region, mid-chord region and trailing edge region according to the flow regulation coefficient to form a regional flow configuration table. The optimization decision module 204 is used to perform flow-lift-drag ratio response surface analysis on the suction demand distribution map to generate lift-drag ratio gain values ​​for each region, perform efficiency and power consumption co-optimization analysis on the lift-drag ratio gain values ​​to generate a flow priority sequence, and construct an adaptive control strategy based on the flow priority sequence and the regional flow configuration table. The flow control module 205 is used to decompose the adaptive control strategy into time-varying differentiated suction sequences by region, apply adjacent region flow gradient constraints to the time-varying differentiated suction sequences to generate smooth transition suction sequences, and generate real-time flow adjustment commands based on the smooth transition suction sequences.

[0068] The aforementioned suction sail partitioned suction flow adaptive adjustment system 200 can implement one of the suction sail partitioned suction flow adaptive adjustment methods described in the above-described method embodiments. The options in the above method embodiments are also applicable to this embodiment and will not be detailed here. The remaining content of this application embodiment can be referred to the content of the above method embodiments, and will not be repeated in this embodiment.

[0069] The purpose of the above embodiments is to reproduce and derive the technical solution of the present invention by way of example, and to fully describe the technical solution, purpose and effect of the present invention. The purpose is to enable the public to have a more thorough and comprehensive understanding of the disclosure of the present invention, and not to limit the scope of protection of the present invention.

[0070] The above embodiments are not an exhaustive list based on the present invention, and there may be many other embodiments not listed. Any substitutions and improvements made without departing from the concept of the present invention are within the protection scope of the present invention.

Claims

1. A method for adaptive adjustment of suction flow rate in a suction sail partition, characterized in that, include: Collect incoming wind speed data and sail angle of attack data in the operating environment of the suction sail, and perform dynamic feature extraction on the incoming wind speed data and sail angle of attack data to construct a flow field state database; Identify boundary layer flow patterns from the flow field state database, perform flow separation analysis on the boundary layer flow patterns to generate a separation risk index, and use chord-direction spatial mapping processing to convert the separation risk index into a pumping demand distribution map. Based on the dynamic change characteristics of the sail angle of attack data, an angle of attack-lift coupling response curve is generated. Sensitivity analysis is performed on the angle of attack-lift coupling response curve to obtain the flow regulation coefficient. Based on the flow regulation coefficient, the sail is divided into leading edge region, mid-chord region and trailing edge region to form a regional flow configuration table. A flow-lift-drag ratio response surface analysis is performed on the suction demand distribution map to generate the lift-drag ratio gain value for each region. A performance-power consumption co-optimization analysis is performed on the lift-drag ratio gain value to generate a flow priority sequence. An adaptive control strategy is constructed by fusing the flow priority sequence with the region flow configuration table. The adaptive control strategy is decomposed into regions to generate time-varying differentiated suction sequences. Adjacent region flow gradient constraints are applied to the time-varying differentiated suction sequences to generate smooth transition suction sequences. Real-time flow regulation commands are generated based on the smooth transition suction sequences.

2. The method according to claim 1, characterized in that, The step of dynamically extracting features from the incoming wind speed data and the sail angle of attack data to construct a flow field state database includes: Extract wind speed temporal features and turbulence intensity parameters from the incoming wind speed data; The turbulence intensity parameters are subjected to spectral decomposition to identify the dominant frequency of wind speed fluctuations; The wind speed fluctuation frequency is fused with the sail angle of attack data to generate a wind condition-angle of attack coupling matrix; The flow field state library is formed by integrating the wind speed time series characteristics with the wind condition-angle of attack coupling matrix.

3. The method according to claim 1, characterized in that, The process of generating a separation risk index by performing flow separation analysis on the boundary layer flow pattern includes: The wall pressure distribution data is extracted from the boundary layer flow mode to generate the adverse pressure gradient distribution features; Based on the characteristics of the adverse pressure gradient distribution, the separation tendency of the boundary layer flow mode is evaluated to generate a separation tendency coefficient; Boundary layer shape factor analysis is performed on the separation tendency coefficient to generate a separation sensitivity distribution field; The separation risk index is determined by comparing the separation sensitivity distribution field with a preset critical separation threshold.

4. The method according to claim 1, characterized in that, The process of obtaining the flow regulation coefficient by performing sensitivity analysis on the angle-of-attack-lift coupling response curve includes: Extract the lift gradient distribution on the intake and pressure sides from the angle-of-attack-lift coupling response curve; The angle-of-attack-flow combined sensitivity is determined based on the lift gradient distribution to form a sensitivity distribution sequence; Inflection point identification is performed on the sensitivity distribution sequence to determine the efficient control angle of attack range; The corresponding flow rate adjustment coefficient is determined based on the aforementioned high-efficiency control angle of attack range.

5. The method according to claim 1, characterized in that, The step of performing performance-power co-optimization analysis on the boost ratio gain value to generate a flow priority sequence includes: The performance evaluation space is constructed by decomposing the lift gain component and drag reduction component from the lift-to-drag ratio gain value. Within the performance evaluation space, the aerodynamic performance contribution of the evaluation area is used to generate a performance ranking. Based on the performance ranking, the power consumption of each region's suction power is analyzed to generate a power cost index. The performance ranking and the power consumption cost index are optimized together to obtain the traffic priority sequence.

6. The method according to claim 1, characterized in that, The step of generating time-varying differentiated extraction sequences by decomposing the adaptive control strategy into regions includes: Based on the adaptive control strategy, a flow reference set is established by defining the reference suction flow rate for each region. The dynamic adjustment range is determined by performing feedforward-feedback decoupling analysis on the flow reference set. The dynamic adjustment range is expanded in the time domain to generate the partitioned flow adjustment amount; Based on the partitioned flow regulation amount, a time-varying differentiated suction sequence is formed by weighted integration of regions.

7. The method according to claim 1, characterized in that, The step of applying adjacent region flow gradient constraints to the time-varying differentiated pumping sequence to generate a smooth transition pumping sequence includes: Extract the flow difference at the boundary of adjacent regions from the time-varying differential suction sequence; The flow gradient constraint boundary is determined based on the flow distribution characteristics of the time-varying differentiated suction sequence; Based on the flow difference and the flow gradient constraint boundary, gradient limiting processing is performed to generate the flow compensation amount; The flow compensation amount is superimposed on the time-varying differentiated suction sequence to form a smooth transition suction sequence.

8. The method according to claim 5, characterized in that, The step of collaboratively optimizing the performance ranking and the power consumption cost index to obtain the traffic priority sequence includes: A joint performance and power consumption evaluation vector is constructed based on the performance ranking and the power consumption cost index. The performance-power consumption joint evaluation vector is weighted by contribution weighting through the performance ranking to generate a weighted evaluation vector. The weighted evaluation vector is subjected to equilibrium solution screening to obtain a candidate solution set; The candidate solution set is then fused to determine the traffic priority sequence.

9. The method according to claim 6, characterized in that, The step of performing feedforward-feedback decoupling analysis on the flow reference set to determine the dynamic adjustment range includes: Extract the baseline flow values ​​for each region from the aforementioned flow baseline set; The baseline flow rate value is analyzed for operating conditions to generate a state-flow correspondence. Gain scheduling analysis is performed on the state-flow correspondence to generate flow regulation requirements; The dynamic adjustment range is determined based on the stated flow regulation requirements.

10. A suction sail zoned suction flow adaptive adjustment system, characterized in that, include: The data acquisition module is used to collect incoming wind speed data and sail angle of attack data in the operating environment of the suction sail, and to perform dynamic feature extraction on the incoming wind speed data and sail angle of attack data to construct a flow field state database. The separation analysis module is used to identify boundary layer flow patterns from the flow field state library, perform flow separation analysis on the boundary layer flow patterns to generate a separation risk index, and use chord-direction spatial mapping processing to convert the separation risk index into a pumping demand distribution map. The region division module is used to generate an angle-of-attack-lift coupling response curve based on the dynamic change characteristics of the sail angle-of-attack data, perform sensitivity analysis on the angle-of-attack-lift coupling response curve to obtain the flow regulation coefficient, and divide the sail into leading edge region, mid-chord region and trailing edge region according to the flow regulation coefficient to form a regional flow configuration table. The optimization decision module is used to perform flow-lift-drag ratio response surface analysis on the suction demand distribution map to generate lift-drag ratio gain values ​​for each region, perform efficiency and power consumption co-optimization analysis on the lift-drag ratio gain values ​​to generate a flow priority sequence, and construct an adaptive control strategy based on the flow priority sequence and the regional flow configuration table. The flow control module is used to decompose the adaptive control strategy into time-varying differentiated suction sequences by region, apply adjacent region flow gradient constraints to the time-varying differentiated suction sequences to generate smooth transition suction sequences, and generate real-time flow adjustment commands based on the smooth transition suction sequences.

Images