Control method and device for telescopic semi-submersible wind power generation system

Abstract

The invention provides a control method and device for a telescopic semi-submersible wind power generation system, and the method comprises the steps: predicting the newly-increased power generation amount of a tower tube of a wind turbine generator set, which is adjusted to each target tower tube height, through time sequence wind field prediction data; predicting adjustment power consumption corresponding to each target tower drum height adjusted by the tower drum through a pre-trained target network model, and adjusting the height of the tower drum according to the target tower drum height in which the newly increased power generation amount is greater than the adjustment power consumption and the difference value between the newly increased power generation amount and the adjustment power consumption reaches a preset difference value. Meanwhile, the ballast water amount of the floating body and the tension of the mooring system are adjusted in a linkage mode, and it is guaranteed that the steady center of the floating type wind power system is in a reliable level. According to the scheme, the adjustment power consumption of the tower drum adjusted to each target tower drum height can be accurately predicted, the tower drum can be accurately adjusted to enable the overall output electric quantity of the wind power generation system to reach a high level, and the overall safety and reliability of the system are guaranteed.

Description

Technical Field

[0001] This invention relates to the field of new energy technology, and in particular to a control method and device for a retractable semi-submersible wind power generation system. Background Technology

[0002] Most existing floating wind power systems use a fixed tower height design. With technological advancements, wind power systems with extendable tower heights have gradually emerged. Among these, the control complexity of offshore extendable semi-submersible wind power systems is significantly higher than that of extendable onshore wind power systems due to the unique characteristics of the marine environment.

[0003] For semi-submersible wind power systems with retractable tower height, existing retraction control methods primarily rely on the distribution characteristics of wind power along the height direction to determine whether to adjust the tower height. Specifically, under non-typhoon conditions, if the wind power at higher altitudes meets the requirements for increasing power generation and is within safe operating conditions, the tower is raised as high as possible to capture more wind energy; while under typhoon conditions, the tower height is lowered to reduce the unit's center of gravity, thereby mitigating the risk of tipping over.

[0004] While the aforementioned telescoping control method can improve the power generation of wind turbines to some extent, it has significant limitations in practical applications. Specifically, because the tower telescoping movement requires simultaneous coordinated control of auxiliary systems such as the ballast water regulation subsystem and the mooring tension regulation subsystem, the overall control energy consumption of the wind power generation system increases significantly. This control energy consumption directly offsets some of the power generation gains obtained by increasing the tower height, ultimately limiting the potential for increasing the overall output power of the wind power generation system. Summary of the Invention

[0005] This invention provides a control method and device for a scalable semi-submersible wind power generation system, in order to solve the problem that the current scalable control method for wind turbine generators restricts the potential for increasing the output power of the wind turbine generators.

[0006] This specification provides a control method for a scalable semi-submersible wind power generation system, comprising: acquiring time-series wind field prediction data for the next adjustment cycle; determining multiple target tower heights after the current height is lowered within the height adjustment range of the wind turbine tower, and predicting the additional power generation corresponding to each target height based on the time-series wind field prediction data for the next adjustment cycle; combining the current tower height, the wind field prediction data at the time of adjustment, the control parameters of the mooring tension regulation subsystem before adjustment, and the control parameters of the ballast water regulation subsystem before adjustment with each target tower height to obtain combined data corresponding to each target tower height, inputting each combined data into a target network model to obtain the adjustment power consumption corresponding to each target tower height; adjusting the tower height according to the target tower height set, and controlling the ballast water regulation subsystem and the mooring tension regulation subsystem to adjust in a timely manner according to the new tower height and wind turbine load; wherein the target tower height set includes target tower heights where the additional power generation is greater than the adjustment power consumption, and the difference between the additional power generation and the adjustment power consumption reaches a predetermined difference.

[0007] In some embodiments, before combining the current tower height, wind field prediction data at the time of adjustment, control parameters of the mooring tension regulation subsystem before adjustment, and control parameters of the ballast water regulation subsystem before adjustment with each target tower height to obtain combined data corresponding to each target tower height, and inputting each combined data into the target network model to obtain the adjustment power consumption corresponding to each target tower height, the method further includes: acquiring the tower height before adjustment, tower height after adjustment, wind field data at the time of adjustment, control parameters of the mooring tension regulation subsystem before and after adjustment, control parameters of the ballast water regulation subsystem before and after adjustment, and power consumption during the adjustment process corresponding to multiple historical adjustments of the wind turbine tower height as sample data; using the tower height before adjustment, tower height after adjustment, wind field data at the time of adjustment, control parameters of the mooring tension regulation subsystem before and after adjustment, and control parameters of the ballast water regulation subsystem before and after adjustment from the sample data as input data to the target network model, and using the power consumption during the adjustment process as output data to train the target network model.

[0008] In some embodiments, the method further includes: acquiring multiple historical adjustment scenarios of the wind turbine, each adjustment scenario including: tower height before adjustment, tower height after adjustment, time-series wind field data at the time of adjustment, control parameters of the mooring tension regulation subsystem before adjustment, control parameters of the ballast water regulation subsystem before adjustment, and adjustment power consumption corresponding to each adjustment scenario; calculating the similarity between each historical adjustment scenario and the current adjustment scenario; for each target adjustment scenario with a similarity threshold, acquiring the adjustment power consumption and adjustment strategy corresponding to each target adjustment scenario; and adjusting the tower of the wind turbine to the target tower height based on the adjustment power consumption and adjustment strategy corresponding to each target adjustment scenario.

[0009] In some embodiments, the similarity between two adjustment scenarios is calculated using a similarity network model. The similarity network model includes: a convolutional network for performing convolution calculations on the time-series wind field data of the two adjustment scenarios; and a fully connected network for mapping the convolution values ​​of the time-series wind field data of the two adjustment scenarios, the difference in tower height before adjustment, the difference in tower height after adjustment, the difference in each control parameter of the mooring tension regulation subsystem before adjustment, the difference in each control parameter of the ballast water regulation subsystem before adjustment, and the similarity between the two adjustment scenarios.

[0010] In some embodiments, adjusting the tower height of a wind turbine to a target tower height based on the adjustment power consumption and adjustment strategy corresponding to each target adjustment scenario includes: obtaining a set of optimization rules for the adjustment strategy; determining whether the adjustment strategy corresponding to the minimum adjustment power consumption conforms to the optimization rules in the set of optimization rules; if it conforms, then using the corresponding optimization rules to optimize the adjustment strategy corresponding to the minimum adjustment power consumption to obtain a target adjustment strategy; and using the target adjustment strategy to adjust the tower height of the wind turbine to the target tower height.

[0011] In some embodiments, the optimization rules in the set of optimization rules include: when there are at least two minimum values, and the first minimum value is less than the second minimum value, controlling the tower height to increase at the position of the first minimum value.

[0012] In some embodiments, the adjustment strategy includes at least one of the following parameters: tower height, control parameters of the mooring tension regulation subsystem, adjustment sequence of control parameters of the ballast water regulation subsystem, and adjustment speed; adjustment target values ​​of control parameters of the mooring tension regulation subsystem and adjustment target values ​​of control parameters of the ballast water regulation subsystem.

[0013] In some embodiments, the method further includes: if the average wind speed of the time-series wind field prediction data in the next adjustment period is greater than the first wind speed and no typhoon warning is received, executing the control method of the scalable semi-submersible wind power generation system described above; and if a typhoon warning is received, adjusting the tower of the wind turbine unit to the lowest height and adjusting the mooring tension regulation subsystem and the ballast water regulation subsystem to typhoon-resistant state.

[0014] The second aspect of this specification provides a control device for a scalable semi-submersible wind power generation system, comprising: a first acquisition unit for acquiring time-series wind field prediction data for the next adjustment cycle; a power generation prediction unit for determining multiple target tower heights after the current height is lowered within the height adjustment range of the wind turbine tower, and predicting the additional power generation corresponding to each target height based on the time-series wind field prediction data for the next adjustment cycle; and a power consumption prediction unit for inputting the current tower height, wind field prediction data at the time of adjustment, control parameters of the mooring tension adjustment subsystem before adjustment, and ballast water adjustment data before adjustment. The control parameters of the junction subsystem are combined with the height of each target tower to obtain the combined data corresponding to each target tower height. The combined data are then input into the target network model to obtain the adjustment power consumption corresponding to each target tower height. The first adjustment unit is used to adjust the tower height according to the target tower height set and control the ballast water regulation subsystem and the mooring tension regulation subsystem to adjust in a timely manner according to the new tower height and the wind turbine load. The target tower height set includes the target tower heights where the newly generated power is greater than the adjustment power consumption and the difference between the newly generated power and the adjustment power consumption reaches a predetermined difference.

[0015] In some embodiments, the apparatus further includes: a second acquisition unit, configured to acquire, for historical multiple adjustments to the tower height of the wind turbine tower, the tower height before adjustment, the tower height after adjustment, wind field data during adjustment, control parameters of the mooring tension regulation subsystem before and after adjustment, control parameters of the ballast water regulation subsystem before and after adjustment, and power consumption during the adjustment process as sample data; and a training unit, configured to use the tower height before adjustment, the tower height after adjustment, the wind field data during adjustment, the control parameters of the mooring tension regulation subsystem before and after adjustment, and the control parameters of the ballast water regulation subsystem before and after adjustment from the sample data as input data for a target network model, and use the power consumption during the adjustment process as output data for the target model, to train the target network model.

[0016] In some embodiments, the apparatus further includes: a third acquisition unit, configured to acquire multiple historical adjustment scenarios of the wind turbine, each adjustment scenario including: tower height before adjustment, tower height after adjustment, time-series wind field data at the time of adjustment, control parameters of the mooring tension regulation subsystem before adjustment, control parameters of the ballast water regulation subsystem before adjustment, and adjustment power consumption corresponding to each adjustment scenario; a calculation unit, configured to calculate the similarity between each historical adjustment scenario and the current adjustment scenario; a fourth acquisition unit, configured to acquire the adjustment power consumption and adjustment strategy corresponding to each target adjustment scenario for each target adjustment scenario whose similarity reaches a similarity threshold; and a second adjustment unit, configured to adjust the tower of the wind turbine to the target tower height based on the adjustment power consumption and adjustment strategy corresponding to each target adjustment scenario.

[0017] In some embodiments, the computing unit calculates the similarity between two adjustment scenarios using a similarity network model. The similarity network model includes: a convolutional network for performing convolution calculations on the time-series wind field data of the two adjustment scenarios; and a fully connected network for mapping the convolution values ​​of the time-series wind field data of the two adjustment scenarios, the difference in tower height before adjustment, the difference in tower height after adjustment, the difference in each control parameter of the mooring tension regulation subsystem before adjustment, and the difference in each control parameter of the ballast water regulation subsystem before adjustment, to the similarity between the two adjustment scenarios.

[0018] In some embodiments, the second adjustment unit includes: an acquisition subunit for acquiring a set of optimization rules for the adjustment strategy; a judgment subunit for judging whether the adjustment strategy corresponding to the minimum power consumption meets the optimization rules in the set of optimization rules; an optimization subunit for optimizing the adjustment strategy corresponding to the minimum power consumption using the corresponding optimization rules if it meets the requirements, to obtain a target adjustment strategy; and an adjustment subunit for adjusting the tower of the wind turbine to the target tower height using the target adjustment strategy.

[0019] In some embodiments, the optimization rules in the set of optimization rules include: when there are at least two minimum values, and the first minimum value is less than the second minimum value, controlling the tower height to increase at the position of the first minimum value.

[0020] In some embodiments, the adjustment strategy includes at least one of the following parameters: tower height, control parameters of the mooring tension regulation subsystem, adjustment sequence of control parameters of the ballast water regulation subsystem, and adjustment speed; adjustment target values ​​of control parameters of the mooring tension regulation subsystem and adjustment target values ​​of control parameters of the ballast water regulation subsystem.

[0021] In some embodiments, if the average wind speed of the time-series wind field prediction data in the next adjustment period is greater than the first wind speed and no typhoon warning is received, the device executes the control device of the scalable semi-submersible wind power generation system according to any one of the first aspects; the device further includes: a third adjustment unit, used to adjust the tower of the wind turbine to the lowest height and adjust the mooring tension regulation subsystem and the ballast water regulation subsystem to typhoon-resistant state when a typhoon warning is received.

[0022] A third aspect of this specification provides an electronic device, comprising: a memory and a processor, wherein the processor and the memory are communicatively connected to each other, the memory stores computer instructions, and the processor executes the computer instructions to implement the control method of the scalable semi-submersible wind power generation system according to any one of the first aspects.

[0023] A fourth aspect of this specification provides a computer storage medium storing computer program instructions, which, when executed, implement the steps of the control method for the scalable semi-submersible wind power generation system described in any of the first aspects.

[0024] The fifth aspect of this specification provides a computer program product comprising a computer program that, when executed by a processor, implements the steps of the control method for the scalable semi-submersible wind power generation system described in any of the first aspects.

[0025] The control method and device for the scalable semi-submersible wind power generation system provided in this specification predicts the additional power generation of the wind turbine tower when adjusted to various target tower heights using time-series wind farm prediction data. It also predicts the adjustment power consumption corresponding to each target tower height using a pre-trained target network model. The tower height is adjusted according to the target tower height where "the additional power generation is greater than the adjustment power consumption, and the difference between the additional power generation and the adjustment power consumption reaches a predetermined difference." This scheme can accurately predict the adjustment power consumption when adjusting the tower to various target tower heights, thereby enabling more precise tower adjustment to achieve a higher overall power output level for the wind power generation system and ensuring the overall safety and reliability of the system. Attached Figure Description

[0026] The accompanying drawings described herein are for illustrative purposes only and are not intended to limit the scope of the invention in any way. Furthermore, the shapes and proportions of the components in the drawings are merely illustrative to aid in understanding the invention and do not specifically limit the shapes and proportions of the components. Those skilled in the art, guided by the teachings of this invention, can select various possible shapes and proportions to implement the invention according to specific circumstances.

[0027] Figure 1 This is a schematic diagram of the overall structure of a scalable semi-submersible wind power generation system.

[0028] Figure 2 A schematic diagram of the tower structure of a retractable semi-submersible wind power generation system;

[0029] Figure 3 This is a flowchart illustrating a control method for the scalable semi-submersible wind power generation system provided in this specification.

[0030] Figure 4 This is another flowchart illustrating the control method for the scalable semi-submersible wind power generation system provided in this specification.

[0031] Figure 5 This is a flowchart illustrating the process of adjusting the tower height of a wind turbine to the target tower height based on the power consumption adjustment and adjustment strategy corresponding to each target adjustment scenario.

[0032] Figure 6 This is a structural schematic diagram of the control device for the retractable semi-submersible wind power generation system provided in this manual.

[0033] Figure 7 This is a schematic diagram of the electronic device provided in this specification. Detailed Implementation

[0034] The details of the present invention can be more clearly understood by referring to the accompanying drawings and the description of specific embodiments. However, the specific embodiments of the present invention described herein are for illustrative purposes only and should not be construed as limiting the invention in any way. Under the teachings of this invention, those skilled in the art can conceive of any possible modifications based on the invention, all of which should be considered within the scope of the invention. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the application. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0035] A retractable semi-submersible wind power generation system includes a generator set, a retractable tower, a semi-submersible floating foundation, a ballast system, a mooring system, a sensing system, and a control system.

[0036] like Figure 1 and Figure 2 As shown, the telescopic tower A includes an upper section A1, a lower section A2, and a telescopic section A3, the height of which can be extended or retracted. The telescopic section can be composed of a high-precision gear and rack lifting mechanism, a hydraulic pin locking device, and a magnetic fluid sealing structure.

[0037] like Figure 1 The semi-submersible floating foundation B comprises three columns arranged in an equilateral triangle. The columns are connected by horizontal trusses, and each column is divided into three independent ballast tanks (upper / middle / lower tanks), with different volumes for each ballast tank.

[0038] The ballast system is equipped with high-pressure water pumps and seawater valve groups to support independent water injection and drainage in each compartment; the liquid level sensor (accuracy ±0.5%) provides real-time feedback on the water volume in the compartment.

[0039] The mooring system uses chain mooring (C), for example. Figure 2 The mooring system employs a 3×3 chain mooring system, which consists of three sets of three anchor chains per set. The length of the anchor chains can be dynamically adjusted via the anchor winch.

[0040] The sensing system is responsible for comprehensively collecting system operating status and external environmental data, specifically including laser wind radar, attitude sensors, displacement sensors, liquid level sensors, and tension sensors. The laser wind radar is installed at the top of the tower to scan and acquire wind speed profile information within a height range of 100–200m; the attitude sensors are used to monitor the levelness and tilt angle of the floating foundation in real time; the displacement sensors are used to monitor the expansion and contraction displacement and locking status of each section of the tower; the liquid level sensors are arranged in each ballast tank to accurately measure the water level inside the tank; and the tension sensors are integrated at the mooring anchor chains to monitor the real-time tension of the anchor chains.

[0041] The control system includes a tower extension / retraction control subsystem, a mooring tension regulation subsystem, and a ballast water regulation subsystem. The tower extension / retraction control subsystem adjusts the tower height to the target height. The mooring tension regulation subsystem includes high-pressure water pumps, distributed valve groups, and independent compartment piping, used for rapid filling and emptying of each ballast tank. The mooring tension regulation subsystem dynamically controls the anchor chain length via the anchor winch, adjusting mooring stiffness and platform positioning.

[0042] Typically, the tower telescopic control subsystem, mooring tension regulation subsystem, and ballast water regulation subsystem are controlled collaboratively. For example, tower height, ballast water, and anchor chain length are not independent parameters—they are coupled together through mechanical transmission (center of gravity, stress) and environmental response (wind, wave, and current loads). The core of coordinated control is to "balance power generation efficiency and system load under the premise of safety and stability," avoiding cascading risks caused by adjusting a single parameter. The relationship between these three parameters is essentially the "transmission of force and balance": tower height determines the location of wind load, ballast water determines the center of gravity distribution, and anchor chain length determines the horizontal constraint force. All three jointly affect the wind turbine platform's "attitude stability," "structural stress," and "power generation efficiency." The specific relationships can be divided into three categories:

[0043] 1. Mechanical coupling: The chain effect of center of gravity-attitude-force is the most direct connection. Adjusting one parameter will change the requirements of the other two parameters through "center of gravity shift" or "load change".

[0044] Specifically, (1) the relationship between tower height and ballast water is as follows: when the tower is raised, the center of gravity of the wind turbine will shift upward (the weight of the tower + nacelle + blades is concentrated at the high position), which will reduce the platform's anti-overturning ability - even a slight tilt may amplify the sway amplitude due to the high center of gravity. At this time, it is necessary to increase the weight of ballast water (especially the ballast tank at the bottom of the platform) or adjust the distribution of ballast water (such as injecting water in the opposite direction of tilt) to lower the overall center of gravity and offset the instability caused by the tower raising. Conversely, if the tower is lowered (such as to reduce wind load in windy weather), the center of gravity will shift downward, and the ballast water can be appropriately reduced to lower the platform's draft (to avoid surge impact). (2) the relationship between anchor chain length and ballast water is as follows: when the anchor chain is shortened, the horizontal tension of the anchoring system on the platform increases, which may cause the platform to tilt in the direction of tension (such as when the anchor chain on one side is too short, the tension is concentrated). At this time, it is necessary to balance the torque generated by the horizontal tension and maintain the platform's attitude by dynamically distributing ballast water (injecting water in the opposite direction of tension). If the anchor chain is too long, the horizontal constraint force will be weakened, and the platform will be easily pushed by the current / surge and drift. At this time, it is necessary to increase the weight of the bottom ballast water to improve the platform's "anti-drift inertia" (the greater the weight, the more difficult it is to be pushed). (3) The relationship between the tower height and the anchor chain length is as follows: the tower height will increase the torque of the wind load (the thrust of the wind on the higher point × the longer lever arm), which will cause the platform to have a tendency to rotate around the anchor chain fixing point (such as tilting on the downwind side). At this time, it is necessary to shorten the downwind side anchor chain and increase the horizontal tension to offset the wind load torque; or adjust the anchor chain tension distribution (such as appropriately loosening the upwind side anchor chain and tightening the downwind side) to optimize the wind load force in conjunction with the tower height.

[0045] 2. Environmental response coupling: The combined effect of wind, waves and current amplifies the correlation.

[0046] In actual marine environments, the simultaneous action of wind, waves, and currents makes the relationship between the three parameters more complex. Specifically, (1) In strong wind weather, the higher the tower is, the greater the wind load. Two things need to be done at the same time: ① shorten the anchor chain (to enhance horizontal fixation and avoid wind-driven drift); ② increase the bottom ballast water (to lower the center of gravity and prevent overturning). All three need to be adjusted synchronously and none can be omitted (shortening the anchor chain alone may cause tilting due to the high center of gravity, and adding ballast water alone may cause drifting due to insufficient anchor chain restraint). (2) In the weather of giant waves, the surge will cause the platform to periodically sway up and down. If the anchor chain is too short, the sway may cause the anchor chain to be overloaded instantly (risk of breakage). At this time, the anchor chain needs to be lengthened appropriately (to increase buffering). However, after lengthening, the platform is prone to drifting. It is necessary to increase the weight of ballast water (to improve platform stability and reduce the drift amplitude) and at the same time reduce the tower height (to reduce the amplification of wind load on swaying).

[0047] 3. Coupling of control objectives: a trade-off between efficiency and safety.

[0048] There are potential conflicts in the adjustment objectives of the three parameters, which need to be coordinated and balanced to achieve the goals of power generation efficiency (i.e., raising the tower to capture higher wind speeds by utilizing wind shear effects) and safety and stability (i.e., raising the tower requires increasing ballast water and shortening the anchor chain). For example, at medium wind speeds (12-25 m / s, the optimal power generation range), the tower can be raised first, while the bottom ballast water is increased appropriately, and the anchor chain is kept at a "moderate length" (to meet the constraint force and avoid excessive tension). However, at high wind speeds (>25 m / s, exceeding the rated wind speed), the tower should be lowered first (to reduce wind load), the anchor chain should be loosened (to avoid wind thrust causing tension overload), and the ballast water should be increased (to resist surges). At this time, power generation efficiency must give way to safety.

[0049] Given the complexity of the coordinated control of the tower telescopic control subsystem, mooring tension regulation subsystem, and ballast water regulation subsystem, existing technologies simply divide the power generation operation into low wind speed, medium wind speed, and typhoon (i.e., high wind speed) conditions, and set optimal regulation target values ​​for each condition. It is only necessary to adjust the tower height, ballast water regulation subsystem, and mooring tension regulation subsystem to the corresponding regulation target values ​​based on the predicted wind speed.

[0050] While this adjustment method can increase the power generation of wind turbines to some extent, it is difficult to increase the overall output power of the wind power generation system.

[0051] To address this issue, this specification provides a control method for a scalable semi-submersible wind power generation system, such as... Figure 3 As shown, this includes the following steps S10 to S40.

[0052] S10: Obtain the time-series wind field forecast data for the next adjustment cycle.

[0053] The power generation time of a scalable semi-submersible wind power generation system can be divided into various adjustment cycles, and the tower height can be adjusted once in each cycle. The duration of each adjustment cycle can be the same (i.e., the adjustment cycle is divided according to a fixed duration) or different (for example, the adjustment cycle is divided according to the wind field change cycle, and a continuous period of relatively small changes in wind field data is taken as an adjustment cycle).

[0054] Time-series wind farm forecast data is predictive data organized in a time series (i.e., in a fixed time interval sequence) to describe wind conditions and related environmental factors over a future period. It deeply binds wind farm forecast data to the time dimension, presenting the dynamic trend of wind conditions through continuous time nodes (such as every 10 minutes, every hour, and every day), ultimately providing continuous time-dimensional data for accurate prediction of future wind turbine power generation, wind turbine operation control, and wind farm scheduling.

[0055] S20: Within the height adjustment range of the wind turbine tower, determine multiple target tower heights after the current height is lowered, and predict the additional power generation corresponding to each target height based on the time-series wind farm forecast data of the next adjustment cycle.

[0056] For example, if the tower height adjustment range is 140m to 160m, and the adjustment step is 5m, then four target tower heights can be determined: 145m, 150m, 155m, and 160m. For each target tower height, the corresponding increase in power generation is calculated separately. The increase in power generation is the difference between the predicted power generation at the target tower height and the predicted power generation at the current tower height (before height adjustment).

[0057] The predicted power generation at the current tower height and the predicted power generation at the target tower height can be calculated separately, and then the difference between the two can be calculated. The predicted power generation at each tower height can be calculated using formulas or using a pre-trained network model.

[0058] In some embodiments, the additional power generation can be calculated using the following formula:

[0059] The additional power generation can be calculated using the following formula:

[0060] ,

[0061] Where △E represents the newly generated electricity, and P 变 Let P0 be the wind power density at the current tower height, f(v) be the Weibull distribution function, and v be the wind power density as the tower rises from height H0 to H. 变 The wind speed at height h during the process. Wind power density can be calculated using the following formula:

[0062] ,

[0063] Where P is the wind power density; ρ is the air density; A=πr 2 r is the impeller radius; v is the wind speed; C p This refers to the wind energy conversion efficiency of a wind turbine.

[0064] As the tower extends from height H0 to h, the wind speed increases logarithmically, as expressed by the following formula:

[0065] ,

[0066] v h Let h be the wind speed at the height h of the tower, where h is the distance from H0 to H. 变 During the process, a height h is given, v0 is the wind speed at the tower height H0, H0 is the current tower height, and z0 is the surface roughness length.

[0067] S30: Combine the current tower height, wind field prediction data during adjustment, control parameters of the mooring tension regulation subsystem before adjustment, and control parameters of the ballast water regulation subsystem before adjustment with each target tower height to obtain the combined data corresponding to each target tower height. Input each combined data into the target network model to obtain the adjustment power consumption corresponding to each target tower height.

[0068] The power consumption adjustment includes the power consumption for tower extension, ballast water pump pumping, and anchor chain movement.

[0069] As analyzed above, the tower height, ballast water volume, and anchor chain length of a retractable semi-submersible wind power generation system require multi-variable coordinated control. When power generation or power consumption is the sole optimization objective, determining the coordinated control target value is relatively simple; however, the wind farm operating conditions are highly time-varying, causing the adjustment energy consumption and power generation gain of the above-mentioned adjustment parameters to exhibit significant nonlinear characteristics under different wind farm conditions, greatly increasing the optimization complexity of multi-variable coordinated control.

[0070] To address this issue, the control method provided in this manual adopts a data-driven modeling approach. It uses actual energy consumption data from historical adjustment conditions as training samples to supervise the training of the target network model. This allows the model to fully learn the nonlinear mapping relationship between various adjustment parameters and adjustment energy consumption under time-varying conditions, thereby achieving accurate prediction of energy consumption during the adjustment process from the current tower height to the target height.

[0071] The control parameters of the mooring tension regulation subsystem refer to the control parameters of the execution unit of the mooring tension regulation subsystem, which may specifically include anchor chain length, mooring tension, mooring angle, and multi-cable synchronous deployment and retraction ratio.

[0072] The control parameters of the ballast water conditioning subsystem refer to the control parameters of the execution unit of the ballast water conditioning subsystem. Specifically, they can refer to the ballast water injection or discharge volume, center of gravity height, platform attitude, draft, and ballast water compartment allocation ratio.

[0073] S40: Adjust the tower height according to the target tower height set, and control the ballast water regulation subsystem and mooring tension regulation subsystem to adjust in a timely manner according to the new tower height and wind turbine load.

[0074] There may be one or more criteria for a target tower height that meet the condition that "the newly generated power output is greater than the adjusted power consumption, and the difference between the newly generated power output and the adjusted power consumption reaches the predetermined difference." If none of these criteria are met, there is no need to adjust the tower height.

[0075] S40 can select the optimal target tower height from the set of target tower heights where the difference between the new power generation and the adjusted power consumption is the largest, and adjust the tower to the optimal target tower height.

[0076] Due to the variability of wind farm data and the optimizability of adjustment strategies, there may be other target tower heights in the target tower height set that make the overall output power of the wind power generation system higher than the above-mentioned optimal target tower height. Based on this, S40 can also be to adjust the tower to any target tower height in the target tower height set.

[0077] During the process of adjusting the tower height based on the target tower height set, the ballast water regulation subsystem and the mooring tension regulation subsystem also need to be adjusted in a coordinated manner. Existing methods can be used for this coordinated adjustment.

[0078] For example, the regulation method of the ballast water conditioning subsystem can be described as follows: when the tower extends to H... 变 Subsequently, the center of gravity of the retractable semi-submersible wind power generation system rises, requiring a recalculation of the center of gravity height GM.

[0079] GM = KB + BM - KG, where KB represents the height of the center of buoyancy (from the baseline to the center of buoyancy); BM represents the radius of the transverse metacenter (BM = I / ▽, where I is the moment of inertia at the waterline and ▽ is the volume of water displaced); and KG represents the height of the center of gravity (from the baseline to the center of gravity).

[0080] ,

[0081] m0 is the total mass before the tower height is adjusted, KG0 is the height of the center of gravity before the tower height is adjusted, and Δm is the total mass before the tower height is adjusted. i For the water injection quality of the i-th compartment, z ci Let be the centroid height of the i-th compartment.

[0082] If the recalculated center height GM is ≥ 1.8m, then the ballast water regulation subsystem will be controlled to regulate the ballast.

[0083] For example, the adjustment method of the mooring tension regulation subsystem can be described as follows: based on the determined target tower height H... 变 Based on the wind speed at that location and the stability adjustment, calculate and determine in real time whether to adjust the anchor chain tension using the following formula. Tighten the anchor chain when T is greater than the safety threshold (the safety threshold is 0.7 times the breaking force of the anchor chain) to suppress platform offset. The anchor chain tension calculation formula is:

[0084] T = T0 + k·x + △T wind ,

[0085] Where T is the anchor chain tension; T0 is the pretension; k is the anchor chain stiffness, inversely proportional to the anchor chain length L, k=EA / L, where EA is the axial stiffness; x is the axial elongation of the anchor chain; ΔT wind The additional tension caused by wind load is approximately:

[0086] ,

[0087] ρ is the density of air; C d ρ is the drag coefficient; A(h) is the wind-receiving area, which varies with the tower height h; v is the wind speed; H hub L is the hub height when the tower height is h. arm This is the length of the horizontal lever arm for mooring.

[0088] The aforementioned ballast water regulation subsystem and mooring tension regulation subsystem can be adjusted simultaneously with the tower height adjustment, or they can be adjusted after the tower height adjustment is completed. The specific adjustment can be determined based on the future trend of the wind field (including the wind speed and direction on the tower lifting path) and the overall stability conditions of the retractable semi-submersible wind power generation system.

[0089] The control method for the aforementioned scalable semi-submersible wind power generation system predicts the additional power generation when the wind turbine tower is adjusted to each target tower height using time-series wind farm prediction data. It also predicts the adjustment power consumption corresponding to each target tower height using a pre-trained target network model. The tower height is adjusted based on the condition that "the additional power generation exceeds the adjustment power consumption, and the difference between the additional power generation and the adjustment power consumption reaches a predetermined difference." This scheme can accurately predict the adjustment power consumption when the tower is adjusted to each target tower height, thereby enabling more precise tower adjustment to achieve a higher overall power output level for the wind power generation system and ensuring the overall safety and reliability of the system.

[0090] In some embodiments, before S30, the following data can be obtained: the tower height before adjustment, the tower height after adjustment, the wind field data at the time of adjustment, the control parameters of the mooring tension regulation subsystem before and after adjustment, the control parameters of the ballast water regulation subsystem before and after adjustment, and the power consumption during the adjustment process, corresponding to multiple historical adjustments of the wind turbine tower height. These data are then used as input data for the target network model, and the power consumption during the adjustment process is used as output data for the target model.

[0091] The parameters mentioned above, such as "tower height before adjustment, tower height after adjustment, wind field data during adjustment, control parameters of the mooring tension regulation subsystem before and after adjustment, and control parameters of the ballast water regulation subsystem before and after adjustment," actually describe an adjustment scenario for tower height adjustment.

[0092] By training the target network model, an empirical model can be formed to represent the complex relationships between historical tower height adjustment parameters, wind field data, control parameters of the mooring tension adjustment subsystem, control parameters of the ballast water adjustment subsystem, and adjustment power consumption. This facilitates subsequent prediction of power consumption for tower height adjustment using the empirical model.

[0093] By training the target network model, the complex coupling relationship between historical tower height adjustment parameters, wind field data, control parameters of the two major regulation subsystems, and power consumption during adjustment was successfully transformed into a directly reusable empirical model. This provides efficient and reliable technical support for subsequent power consumption prediction in tower height adjustment. The input parameter settings and training data settings of the above network model enable it to achieve the following technical effects:

[0094] 1. Accurately capture multi-dimensional correlations and build a highly adaptable experience model. Specifically, (1) integrate the full-dimensional input of "tower height before and after adjustment - environmental data - system control parameters" to fully cover the core influencing factors of power consumption for tower height adjustment, accurately capture the nonlinear correlation and interaction law between variables, and avoid the law deviation caused by single variable analysis; (2) train based on historical actual adjustment data, the network model directly precipitates the real operation experience in the engineering scenario, without relying on idealized assumptions, and the formed experience model is highly adapted to the actual working conditions of wind turbine operation and maintenance, with stronger generalization ability.

[0095] 2. Achieve rapid power consumption prediction and support pre-decision optimization. Specifically, (1) the empirical model simplifies the complex correlation calculation logic. Afterwards, only the preset adjustment parameters, real-time wind field data and the proposed control parameters need to be input to quickly output the power consumption prediction results without complex physical modeling or on-site trial and error; (2) the prediction function allows maintenance personnel to quantify the energy consumption differences of different schemes before adjusting the tower height, which makes it easier to prioritize the low energy consumption adjustment scheme and avoid energy waste caused by blind operation from the source.

[0096] 3. Reduce operation and maintenance costs and risks, and improve operational efficiency. Specifically, (1) there is no need to test energy consumption through actual adjustments, which reduces the power consumption and time costs caused by physical trial and error, while avoiding structural safety risks that may be caused by improper adjustments (such as insufficient tower stability, mooring cable overload, etc.); (2) the experience model transforms energy consumption assessment from post-event statistics to pre-event prediction, which simplifies the operation and maintenance decision-making process, helps operation and maintenance personnel to respond quickly to changes in wind farm conditions, and improves the decision-making efficiency and execution accuracy of tower height adjustment.

[0097] In some embodiments, such as Figure 4 As shown, the method further includes the following steps S50 to S80.

[0098] S50: Acquire multiple adjustment scenarios of wind turbines in history. Each adjustment scenario includes: tower height before adjustment, tower height after adjustment, time-series wind field data at the time of adjustment, control parameters of the mooring tension regulation subsystem before adjustment, control parameters of the ballast water regulation subsystem before adjustment, and the corresponding power consumption for each adjustment scenario.

[0099] S60: Calculate the similarity between each historical adjustment scenario and the current adjustment scenario.

[0100] In some embodiments, the similarity between two adjustment scenarios is calculated using a similarity network model. The similarity network model includes a convolutional network and a fully connected network. The convolutional network is used to perform convolution calculations on the time-series wind field data of the two adjustment scenarios; the fully connected network is used to map the similarity between the convolution values ​​of the time-series wind field data of the two adjustment scenarios, the difference in tower height before adjustment, the difference in tower height after adjustment, the difference in control parameters of the mooring tension regulation subsystem before adjustment, and the difference in control parameters of the ballast water regulation subsystem before adjustment, to the similarity between the two adjustment scenarios.

[0101] Convolutional networks are specifically designed to process time-series wind field data. They can extract dynamic features such as wind field distribution characteristics, wind speed fluctuation trends, peak frequency, and duration (instead of just matching wind speed values ​​at a single time point). This solves the problem that traditional static parameter matching cannot cope with dynamic changes in the wind field (such as accurately distinguishing the differences between "continuous high wind speed + slow increase" and "instantaneous high wind speed + rapid decline").

[0102] The fully connected network integrates "temporal wind field convolutional features + tower height difference before and after adjustment + mooring / ballast water control parameter difference" to achieve full-dimensional matching of "dynamic operating conditions (wind field) + static initial state (equipment parameters)", avoiding "local similarity, overall mismatch" caused by single-dimensional matching (such as strategy reuse failure caused by only matching tower height and ignoring the difference in initial state of mooring tension).

[0103] Similarity network models fit the nonlinear relationship between parameter differences and similarity through multiple layers of neurons. Compared with traditional linear matching algorithms (such as Euclidean distance and cosine similarity), they are better able to capture the adaptation rules under complex working conditions (such as the weight of mooring tension parameter differences in high wind speed scenarios is higher than that in low wind speed scenarios).

[0104] The above analysis shows that the similarity network model can find historical adjustment scenarios that are highly similar to the current adjustment scenario.

[0105] Iterative training feature: When new historical scenarios are added (including new operating conditions and new adjustment strategies), the model can be fine-tuned (such as updating convolutional layer weights and fully connected layer parameters) to allow the model to learn new adaptation rules and avoid rigid matching logic. As the historical scenario library expands, the model matching accuracy can be continuously improved. With each addition of some effective scenarios, the matching accuracy can be further improved, and it can automatically adapt to long-term changes such as equipment aging and wind field environment changes (such as the annual average wind speed increase).

[0106] S70: For each target adjustment scenario where the similarity reaches the similarity threshold, obtain the adjustment power consumption and adjustment strategy corresponding to each target adjustment scenario.

[0107] In some embodiments, the adjustment strategy includes at least one of the following parameters: tower height, control parameters of the mooring tension regulation subsystem, adjustment sequence of control parameters of the ballast water regulation subsystem, and adjustment speed; adjustment target values ​​of control parameters of the mooring tension regulation subsystem and adjustment target values ​​of control parameters of the ballast water regulation subsystem.

[0108] Adjustment speed refers to the adjustment rate of the control parameters of the mooring tension regulation subsystem or the ballast water regulation subsystem. Specifically, it can be the adjustment rate of the ballast water or the adjustment rate of the anchor chain. Adjustment speed can also be the adjustment rate of the tower.

[0109] The adjustment speed can be the ratio of the adjusted amount to the target adjustment amount. The target adjustment amount can be the difference between the target value after adjustment and the parameter value before adjustment. The adjusted amount refers to the difference between the parameter value that has been adjusted to and the parameter value before adjustment.

[0110] S80: Adjust the tower height of the wind turbine to the target tower height based on the power consumption and adjustment strategy corresponding to each target adjustment scenario.

[0111] In some embodiments, such as Figure 5 As shown, S80 includes the following S81 to S84.

[0112] S81: Obtain the set of optimization rules for the adjustment strategy.

[0113] In some embodiments, the optimization rules in the set of optimization rules include: when the wind speed sequence (which can be obtained from the wind field sequence) has at least two minimum values, and the first minimum value is less than the second minimum value, the tower height is controlled to be increased at the first minimum wind speed value.

[0114] By prioritizing the initiation of adjustments within the time window of the first minimum value, lower wind loads can be utilized to shorten adjustment time and complexity, further reducing energy consumption. The example of the optimization rule above could further include a duration where the duration of the first minimum value and its adjacent ranges exceeds the required adjustment time.

[0115] The optimization rules in the above set of optimization rules can also be adjustment constraints added based on actual working conditions.

[0116] S82: Determine whether the adjustment strategy corresponding to the minimum power consumption meets the optimization rules in the set of optimization rules.

[0117] The optimization rules include optimizable conditions and optimization measures. If the adjustment strategy corresponding to the minimum power consumption meets the optimizable condition of a certain optimization rule in the set of optimization rules, then even if the adjustment strategy with the minimum power consumption is selected from the similar adjustment scenarios matched from the set of historical adjustment scenarios, its power consumption is not actually the minimum, and the power consumption of the current adjustment scenario can be even smaller.

[0118] S83: If the conditions are met, the corresponding optimization rules are used to optimize the adjustment strategy corresponding to the minimum power consumption, and the target adjustment strategy is obtained.

[0119] S84: Adjust the tower of the wind turbine to the target tower height using the target adjustment strategy.

[0120] The above solution does not directly determine the adjustment strategy based on the conditions of the current adjustment scenario. Instead, it first finds a similar adjustment scenario with the lowest power consumption, and then optimizes based on this similar scenario to obtain an adjustment strategy applicable to the current adjustment scenario. This method of determining the current adjustment strategy avoids the blindness and high cost of directly designing the current adjustment strategy, and solves the problem of insufficient adaptability when simply reusing similar scenarios, achieving low power consumption, accuracy, security, and high efficiency in multi-dimensional optimization. The technical effects are described in detail below:

[0121] 1. Significantly reduce adjustment energy consumption and lock in the optimal energy consumption range. Specifically, prioritize locking in similar scenarios with "minimum adjustment power consumption". This strategy has been historically verified and has the optimal energy consumption base, avoiding the trial-and-error energy consumption of "designing a strategy from scratch". Subsequent optimizations are only minor adjustments based on the "low-consumption strategy" (such as adapting to the current minimum wind speed window and adjusting mooring tension parameters), without deviating from the core of low consumption, ensuring that energy consumption does not rebound.

[0122] 2. Improve the adaptability of the strategy to the current operating conditions and reduce the risk of adjustment failure. Specifically, similar scenarios have ensured a high degree of matching between "operating conditions (wind field), initial equipment state, and adjustment target", laying the foundation for adaptability; the optimization steps correct the strategy according to the uniqueness of the current scenario (such as the minimum value characteristics of the current wind speed sequence and the real-time status of the equipment) to solve the deviation of "similar ≠ completely consistent" (such as the similar scenario is a steady wind, while the current one is a gust, and the adjustment speed is slowed down after optimization).

[0123] 3. Enhance the safety and stability of the adjustment process and mitigate engineering risks. Specifically, reused similar scenarios are valid cases of "successfully completed adjustments," whose strategies (such as adjustment order and target values) have been verified to comply with safety constraints, avoiding potential safety vulnerabilities in "directly designed strategies." Optimization rules (such as adjustments based on wind speed minimum window and mooring tension threshold constraints) further incorporate the safety requirements of the current scenario, mitigating current risks not covered by similar scenarios (such as the current minimum window being too short, leading to the splitting of the adjustment process after optimization).

[0124] 4. Shorten the total time for strategy design and adjustment, and improve overall efficiency. Specifically, low-cost strategies from similar scenarios can be directly reused, saving a lot of time spent on "remodeling, simulating, and iteratively optimizing strategies based on the current operating conditions"; the optimization steps only fine-tune the core differences (such as adjusting the timing window and correcting the adjustment speed), without the need for overall strategy reconstruction, further reducing preparation time.

[0125] 5. Reduce reliance on real-time computing resources and adapt to complex operation and maintenance scenarios. Specifically, the strategy for similar scenarios is to use the results of "historical pre-calculation + verification", which does not require complex energy consumption modeling and multi-parameter optimization calculations in real time; the optimization steps only involve rule matching (such as judging the characteristics of wind speed minimums) and local parameter correction, with low computational complexity and low hardware requirements.

[0126] 6. Enhanced strategy robustness to adapt to diverse operating conditions and equipment states. Specifically, the similar scenario library covers different operating conditions and equipment states, making the "basic adaptability" of low-power strategies stronger; optimization rules can flexibly superimpose multi-dimensional constraints (wind speed, safety, efficiency), and can be dynamically adjusted according to the specific changes in the current scenario, avoiding adaptation bottlenecks caused by strategy rigidity.

[0127] 7. Form a data closed loop to achieve continuous iterative optimization of strategies. Specifically, the "similar strategies + optimized strategies + actual energy consumption / effect data" of the current scenario can be added back to the historical database to enrich the coverage and accuracy of similar scenarios; optimization rules can be continuously improved based on new data (such as adding "extreme wind speed minimum value adaptation rules"), so that the optimization of subsequent strategies is more in line with actual needs.

[0128] In some embodiments, if the average wind speed of the time-series wind field forecast data for the next adjustment period is greater than the first wind speed and no typhoon warning has been received, the following steps are performed: Figure 3 The control method for the scalable semi-submersible wind power generation system is shown. Upon receiving a typhoon warning, the tower of the wind turbine unit is adjusted to its lowest height, and the mooring tension regulation subsystem and ballast water regulation subsystem are adjusted to typhoon-resistant mode. This setting can improve the survival probability under extreme weather (typhoon) conditions.

[0129] This specification provides a control device for a scalable semi-submersible wind power generation system, which can be used to implement the control method of the aforementioned scalable semi-submersible wind power generation system. For example... Figure 6 As shown, the device includes a first acquisition unit 10, a power generation prediction unit 20, a power consumption prediction unit 30, and a first adjustment unit 40.

[0130] The first acquisition unit 10 is used to acquire the time-series wind field prediction data for the next adjustment cycle.

[0131] The power generation prediction unit 20 is used to determine multiple target tower heights after the current height is lowered within the height adjustment range of the wind turbine tower, and to predict the additional power generation corresponding to each target height based on the time-series wind farm prediction data of the next adjustment cycle.

[0132] The power consumption prediction unit 30 is used to combine the current tower height, wind field prediction data during adjustment, control parameters of the mooring tension regulation subsystem before adjustment, and control parameters of the ballast water regulation subsystem before adjustment with each target tower height to obtain the combined data corresponding to each target tower height. The combined data is then input into the target network model to obtain the adjustment power consumption corresponding to each target tower height.

[0133] The first adjustment unit 40 is used to adjust the height of the tower according to the target tower height set, and to control the ballast water regulation subsystem and the mooring tension regulation subsystem to adjust in a timely manner according to the new tower height and the wind turbine load; the target tower height set includes the target tower height where the newly generated power generation is greater than the adjusted power consumption, and the difference between the newly generated power generation and the adjusted power consumption reaches a predetermined difference.

[0134] In some embodiments, the apparatus further includes a second acquisition unit and a training unit.

[0135] The second acquisition unit is used to acquire the tower height before adjustment, tower height after adjustment, wind field data at the time of adjustment, control parameters of the mooring tension regulation subsystem before and after adjustment, control parameters of the ballast water regulation subsystem before and after adjustment, and power consumption during the adjustment process as sample data for multiple historical adjustments of the wind turbine tower height.

[0136] The training unit is used to train the target network model by taking the tower height before adjustment, tower height after adjustment, wind field data during adjustment, control parameters of the mooring tension regulation subsystem before and after adjustment, and control parameters of the ballast water regulation subsystem before and after adjustment from the sample data as input data of the target network model, and the power consumption during the adjustment process as output data of the target model.

[0137] In some embodiments, the apparatus further includes a third acquisition unit, a calculation unit, a fourth acquisition unit, and a second adjustment unit.

[0138] The third acquisition unit is used to acquire multiple adjustment scenarios of the wind turbine in history. Each adjustment scenario includes: tower height before adjustment, tower height after adjustment, time-series wind field data at the time of adjustment, control parameters of the mooring tension regulation subsystem before adjustment, control parameters of the ballast water regulation subsystem before adjustment, and the adjustment power consumption corresponding to each adjustment scenario.

[0139] The calculation unit is used to calculate the similarity between each historical adjustment scenario and the current adjustment scenario.

[0140] The fourth acquisition unit is used to acquire the power consumption and adjustment strategy corresponding to each target adjustment scenario when the similarity reaches the similarity threshold.

[0141] The second adjustment unit is used to adjust the height of the wind turbine tower to the target tower height based on the adjustment power consumption and adjustment strategy corresponding to each target adjustment scenario.

[0142] In some embodiments, the computing unit calculates the similarity between two adjustment scenarios using a similarity network model. The similarity network model includes a convolutional network and a fully connected network. The convolutional network performs convolution calculations on the time-series wind field data of the two adjustment scenarios; the fully connected network maps the convolution values ​​of the time-series wind field data of the two adjustment scenarios, the difference in tower height before adjustment, the difference in tower height after adjustment, the difference in control parameters of the mooring tension regulation subsystem before adjustment, and the difference in control parameters of the ballast water regulation subsystem before adjustment, to the similarity between the two adjustment scenarios.

[0143] In some embodiments, the second adjustment unit includes an acquisition subunit, a judgment subunit, an optimization subunit, and an adjustment subunit.

[0144] The acquisition subunit is used to acquire the set of optimization rules for the adjustment strategy. The judgment subunit is used to determine whether the adjustment strategy corresponding to the minimum power consumption meets the optimization rules in the set of optimization rules. The optimization subunit is used to optimize the adjustment strategy corresponding to the minimum power consumption using the corresponding optimization rules if it meets the requirements, thereby obtaining the target adjustment strategy. The adjustment subunit is used to adjust the wind turbine tower to the target tower height using the target adjustment strategy.

[0145] In some embodiments, the optimization rules in the set of optimization rules include: when there are at least two minimum values, and the first minimum value is less than the second minimum value, controlling the tower height to increase at the position of the first minimum value.

[0146] In some embodiments, the adjustment strategy includes at least one of the following parameters: tower height, control parameters of the mooring tension regulation subsystem, adjustment sequence of control parameters of the ballast water regulation subsystem, and adjustment speed; adjustment target values ​​of control parameters of the mooring tension regulation subsystem and adjustment target values ​​of control parameters of the ballast water regulation subsystem.

[0147] In some embodiments, the device performs the following operation when the average wind speed of the time-series wind field prediction data for the next adjustment period is greater than the first wind speed and no typhoon warning has been received. Figure 3 The control device for the retractable semi-submersible wind power generation system shown; the device further includes: a third adjustment unit, used to adjust the tower of the wind turbine to the lowest height when a typhoon warning is received, and to adjust the mooring tension adjustment subsystem and the ballast water adjustment subsystem to typhoon-resistant state.

[0148] The descriptions and functions of the above-mentioned devices can be understood by referring to the section on control methods for scalable semi-submersible wind power generation systems, and will not be repeated here.

[0149] This invention also provides an electronic device, such as... Figure 7 As shown, the electronic device may include a processor 701 and a memory 702, wherein the processor 701 and the memory 702 may be connected via a bus or other means. Figure 7 Taking the example of a connection between China and Israel via a bus.

[0150] Processor 701 can be a central processing unit (CPU). Processor 701 can also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or combinations of the above types of chips.

[0151] Memory 702, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs, non-transitory computer-executable programs, and modules, such as the program instructions / modules corresponding to the control method of the scalable semi-submersible wind power generation system in this embodiment of the invention (e.g., Figure 6The first acquisition unit 10, power generation prediction unit 20, power consumption prediction unit 30, and first adjustment unit 40 are shown in the diagram. The processor 701 executes various functional applications and data processing by running non-transitory software programs, instructions, and modules stored in the memory 702, thereby realizing the control method of the scalable semi-submersible wind power generation system in the above method embodiments.

[0152] The memory 702 may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store data created by the processor 701, etc. Furthermore, the memory 702 may include high-speed random access memory and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, the memory 702 may optionally include memory remotely located relative to the processor 701, and these remote memories may be connected to the processor 701 via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0153] The one or more modules are stored in the memory 702, and when executed by the processor 701, the control method of the scalable semi-submersible wind power generation system described above is executed.

[0154] The specific details of the above-mentioned electronic device can be understood by referring to the relevant descriptions and effects in the method embodiments, and will not be repeated here.

[0155] This specification also provides a computer storage medium storing computer program instructions, which, when executed, implement the steps of the control method for the scalable semi-submersible wind power generation system described above.

[0156] This specification also provides a computer program product comprising a computer program that, when executed by a processor, implements the steps of the control method for the scalable semi-submersible wind power generation system described above.

[0157] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), random access memory (RAM), flash memory, hard disk drive (HDD), or solid-state drive (SSD), etc.; the storage medium can also include combinations of the above types of memory.

[0158] The various embodiments in this specification are described in a progressive manner. For the same or similar parts between the various embodiments, please refer to each other. The focus of each embodiment is to describe the differences from other embodiments.

[0159] The systems, devices, modules, or units described in the above embodiments can be implemented by computer chips or entities, or by products with certain functions.

[0160] For ease of description, the above devices are described separately by function as various units. Of course, in implementing this application, the functions of each unit can be implemented in one or more software and / or hardware.

[0161] As can be seen from the above description of the embodiments, those skilled in the art can clearly understand that this application can be implemented by means of software plus necessary general-purpose hardware platforms. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute certain parts of the methods of various embodiments of this application.

[0162] This application can be used in a wide variety of general-purpose or special-purpose computer system environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics devices, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, etc.

[0163] Although this application has been described through embodiments, those skilled in the art will know that this application has many modifications and variations without departing from the spirit of this application, and it is intended that the appended claims cover such modifications and variations without departing from the spirit of this application.

Claims

1. A control method for a scalable semi-submersible wind power generation system, characterized in that, include: Obtain the time-series wind field forecast data for the next adjustment cycle; Within the height adjustment range of the wind turbine tower, determine multiple target tower heights after the current height is lowered, and predict the additional power generation corresponding to each target height based on the time-series wind farm forecast data for the next adjustment cycle. The current tower height, wind field prediction data during adjustment, control parameters of the mooring tension regulation subsystem before adjustment, and control parameters of the ballast water regulation subsystem before adjustment are combined with each target tower height to obtain the combined data corresponding to each target tower height. Each combined data is then input into the target network model to obtain the adjustment power consumption corresponding to each target tower height. The tower height is adjusted according to the target tower height set, and the ballast water regulation subsystem and mooring tension regulation subsystem are controlled to adjust in a timely manner according to the new tower height and wind turbine load. The target tower height set includes the target tower height where the newly generated power output is greater than the adjusted power consumption, and the difference between the newly generated power output and the adjusted power consumption reaches a predetermined difference.

2. The method according to claim 1, characterized in that, Before combining the current tower height, wind field prediction data at the time of adjustment, control parameters of the mooring tension regulation subsystem before adjustment, and control parameters of the ballast water regulation subsystem before adjustment with each target tower height to obtain combined data corresponding to each target tower height, and inputting each combined data into the target network model to obtain the adjustment power consumption corresponding to each target tower height, the following steps are also included: The data collected includes the tower height before and after the adjustment, wind field data at the time of adjustment, control parameters of the mooring tension regulation subsystem before and after the adjustment, control parameters of the ballast water regulation subsystem before and after the adjustment, and power consumption during the adjustment process, which were obtained as sample data for the multiple historical adjustments of the wind turbine tower height. The target network model is trained using the tower height before and after adjustment, wind field data during adjustment, control parameters of the mooring tension regulation subsystem before and after adjustment, and control parameters of the ballast water regulation subsystem before and after adjustment from the sample data. The power consumption during the adjustment process is used as the output data of the target model.

3. The method according to claim 1, characterized in that, The method further includes: The system acquires multiple historical adjustment scenarios for wind turbines. Each adjustment scenario includes: tower height before adjustment, tower height after adjustment, time-series wind field data during adjustment, control parameters of the mooring tension regulation subsystem before adjustment, control parameters of the ballast water regulation subsystem before adjustment, and the corresponding power consumption for each adjustment scenario. Calculate the similarity between each historical adjustment scenario and the current adjustment scenario; For each target adjustment scenario where the similarity reaches the similarity threshold, obtain the adjustment power consumption and adjustment strategy corresponding to each target adjustment scenario; Adjust the wind turbine tower to the target tower height based on the power consumption adjustment and adjustment strategy corresponding to each target adjustment scenario.

4. The method according to claim 3, characterized in that, The similarity between two adjusted scenes is calculated using a similarity network model; The similarity network model includes: A convolutional network is used to perform convolutional calculations on time-series wind field data from two adjustment scenarios; A fully connected network is used to map the convolution values ​​of time-series wind field data for two adjustment scenarios, the difference in tower height before adjustment, the difference in tower height after adjustment, the difference in control parameters of the mooring tension regulation subsystem before adjustment, the difference in control parameters of the ballast water regulation subsystem before adjustment, and the similarity between the two adjustment scenarios.

5. The method according to claim 3, characterized in that, Adjusting the tower height of the wind turbine to the target tower height based on the adjustment power consumption and adjustment strategy corresponding to each target adjustment scenario, including: Obtain the set of optimization rules for the adjustment strategy; Determine whether the adjustment strategy corresponding to the minimum power consumption meets the optimization rules in the set of optimization rules; If the conditions are met, the corresponding optimization rules are used to optimize the adjustment strategy corresponding to the minimum power consumption, and the target adjustment strategy is obtained. The target adjustment strategy is used to adjust the tower height of the wind turbine to the target tower height.

6. The method according to claim 5, characterized in that, The optimization rules in the set of optimization rules include: when there are at least two minimum values, and the first minimum value is less than the second minimum value, the tower height is controlled to be increased at the position of the first minimum value.

7. The method according to claim 3, characterized in that, The adjustment strategy includes at least one of the following parameters: tower height, control parameters of the mooring tension adjustment subsystem, adjustment sequence of control parameters of the ballast water adjustment subsystem, and adjustment speed; The target values ​​for the control parameters of the mooring tension regulation subsystem and the target values ​​for the control parameters of the ballast water regulation subsystem.

8. The method according to claim 1, characterized in that, The method further includes: If the average wind speed of the time-series wind field forecast data in the next adjustment period is greater than the first wind speed and no typhoon warning is received, the control method of the scalable semi-submersible wind power generation system according to any one of claims 1 to 7 shall be executed. Upon receiving a typhoon warning, adjust the tower of the duct unit to the lowest possible height and adjust the mooring tension regulation subsystem and ballast water regulation subsystem to typhoon-resistant status.

9. A control device for a retractable semi-submersible wind power generation system, characterized in that, include: The first acquisition unit is used to acquire the time-series wind field prediction data for the next adjustment cycle; The power generation prediction unit is used to determine multiple target tower heights after the current height is lowered within the height adjustment range of the wind turbine tower, and to predict the additional power generation corresponding to each target height based on the time-series wind farm prediction data of the next adjustment cycle. The power consumption prediction unit is used to combine the current tower height, wind field prediction data during adjustment, control parameters of the mooring tension adjustment subsystem before adjustment, and control parameters of the ballast water adjustment subsystem before adjustment with each target tower height to obtain the combined data corresponding to each target tower height. The combined data is then input into the target network model to obtain the adjustment power consumption corresponding to each target tower height. The first adjustment unit is used to adjust the height of the tower according to the target tower height set, and to control the ballast water regulation subsystem and the mooring tension regulation subsystem to adjust in a timely manner according to the new tower height and the wind turbine load; the target tower height set includes the target tower height where the newly generated power generation is greater than the adjusted power consumption, and the difference between the newly generated power generation and the adjusted power consumption reaches a predetermined difference.

10. An electronic device, characterized in that, include: The system includes a memory and a processor, which are interconnected. The memory stores computer instructions, and the processor executes the computer instructions to implement the control method for the scalable semi-submersible wind power generation system as described in any one of claims 1 to 8.

Images