Double-wind-wheel wind driven generator capable of capturing maximum wind energy

Abstract

The invention provides a double-wind-wheel wind driven generator capable of achieving maximum wind energy capture. A front wind wheel and a rear wind wheel rotate asynchronously. The front wind wheel is connected to a sun wheel of a planet differential device through an electric control speed change device. The rear wind wheel is connected with a planetary gear system of the planetary differential device through a gear reversing device, and the power of the front wind wheel and the power of the rear wind wheel are integrated through the planetary differential device and transmitted to the generator.

Description

Technical Field

[0001] The present invention relates to the technical field of wind power generation equipment, and in particular to a double-wind-rotor wind power generator capable of achieving maximum wind energy capture. Background Art

[0002] With the rapid growth of society's demand for energy and the decreasing reserves of traditional energy, environmental pollution problems are becoming increasingly serious, which has promoted the rapid development of green energy technology. However, in field projects, the power supply methods of existing technologies still have many shortcomings. At present, although common portable diesel generators are convenient, they are noisy and polluting, and fuel transportation and storage are difficult, especially in remote areas. The fuel supply cost is high and complicated. Although solar power generation devices are environmentally friendly, they are heavily dependent on weather and light conditions and are difficult to provide stable electricity in rainy weather or at night.

[0003] At the same time, although single-rotor wind turbines are widely used, their efficiency is significantly reduced in unstable wind environments due to design limitations. This type of generator is more suitable for scenarios with constant wind power, such as large wind farms, but it is difficult to work efficiently in field environments where wind speeds change frequently. In addition, its adaptability to wind speeds is limited, and it is difficult to meet the demand for continuous power supply under low wind speed conditions. In contrast, dual-rotor wind turbines show higher adaptability, not only can they better adapt to random wind directions, but also show higher power generation efficiency when wind speeds are low or wind power is unstable. They are especially suitable for complex field environments and provide a better solution for the application of green energy in field operations.

[0004] Compared with traditional wind turbines, dual-rotor wind turbines have more applicable environments and larger wind sweeping areas. In the single-wind turbine scenario, obtaining a higher efficiency in the conversion of wind energy has become the primary consideration for the field power supply system. At present, how to achieve a more efficient dual-rotor wind turbine structure and control mechanism is a key issue to be studied. Summary of the invention

[0005] Considering how to further improve the efficiency of the dual-rotor wind turbine and ensure the safety during operation is a new topic. To this end, the present invention provides a design of a dual-rotor wind turbine that achieves maximum wind energy capture. By adopting a design with different front and rear wind wheels, combined with an electronically controlled transmission and a planetary differential device, it is possible to improve the utilization of wind energy while minimizing the energy loss caused by the asynchronous rotation of the front and rear wind wheels. In addition, the integrated wind energy capture control device and overload protection system effectively ensure the operating efficiency and safety of the system, overcoming the instability problem of the wind power generation system in the prior art under high wind speed and variable wind speed conditions.

[0006] The present invention specifically adopts the following technical solutions:

[0007] A dual-rotor wind turbine generator that achieves maximum wind energy capture, wherein the front wind wheel and the rear wind wheel rotate asynchronously: the front wind wheel is connected to the sun wheel of the planetary differential device through an electronically controlled speed change device; the rear wind wheel is connected to the planetary gear system of the planetary differential device through a gear reversal device, and the power of the front wind wheel and the rear wind wheel is integrated through the planetary differential device and then transmitted to the generator. The generator is preferably a permanent magnet synchronous generator.

[0008] Furthermore, the controller monitors the current external wind speed in real time through a wind speed sensor, and adjusts the electronically controlled speed change device within a maximum limited speed according to the current external wind speed, so that the speed ratio of the front wind wheel output to the rear wind wheel output is maintained within an optimal range to achieve maximum wind energy capture.

[0009] Furthermore, when the current external wind speed or the load of the dual-rotor wind turbine generator is greater than the corresponding set threshold, the controller adjusts the speed and torque of the generator through the generator's inverter to ensure that the front and rear wind rotors operate within a safe range.

[0010] Here, the controller directly controls the inverter. Because this is a relatively dangerous working condition, indirect and refined fine-tuning of the electronically controlled speed change device is not safe enough, so the inverter of the generator is adjusted directly from the source to reduce the speed.

[0011] Furthermore, the electronically controlled speed change device is composed of an input sprocket, an output sprocket and a chain, the input sprocket is connected to the output sprocket through the chain, the front wind wheel is connected to the input sprocket, and the output sprocket is connected to the sun gear in the planetary differential device to transmit the power of the front wind wheel to the sun gear in the planetary differential device.

[0012] Furthermore, the planetary differential device includes a planetary gear system and a sleeve; the planetary gear system includes a sun gear and a plurality of planetary gears; the sun gear is meshed with a plurality of planetary gears; the output shaft of the rear wind wheel is connected to the second gear of the gear reversal device, and the first gear is meshed with the second gear of the gear reversal device; the second gear is fixedly connected to the sleeve, and the sleeve is arranged on the outer periphery of the planetary gear system and the inner ring is meshed with a plurality of planetary gears.

[0013] Furthermore, the controller obtains a preliminary range of the optimal speed ratio corresponding to the current external wind speed based on the maximum wind energy capture power estimation, and takes the preliminary range of the optimal speed ratio corresponding to the current external wind speed as a preliminary constraint; within the range of the preliminary constraint, the SVR model obtained by training the current external wind speed input collected in real time is fitted to obtain the optimal range between the speed ratio of the front wind wheel output and the rear wind wheel output under the preliminary constraint conditions.

[0014] Further, the wind speed sensor is specifically a wind direction wind speed sensor; the method for estimating the maximum wind energy capture power to obtain a preliminary range of the optimal rotational speed ratio corresponding to the current external wind speed also includes: based on the maximum wind energy capture power estimation, according to the simulation test empirical formula of the forward wind and the lateral wind and the simulation results, obtaining the initial rotational speed ratio of the best wind energy capture under different wind directions based on the current external wind speed as the preliminary range of the optimal rotational speed ratio;

[0015] Within the scope of the preliminary constraints, the SVR model obtained by training the current external wind speed input collected in real time is fitted to obtain the optimal range of the speed ratio of the front wind wheel output to the rear wind wheel output under the preliminary constraints, including: taking the meteorological characteristics of wind speed, wind direction and initial speed ratio as input, combining the discretized feature labels input into the SVR model obtained by training to fit the optimal range of the speed ratio of the front wind wheel output to the rear wind wheel output under the preliminary constraints.

[0016] Furthermore, the training process of the SVR model is specifically as follows:

[0017] The K-means clustering algorithm is used to classify the historical wind data, and a data set for SVR model training is constructed based on the classification results;

[0018] The SVR model is trained using the data set. Each category of data input features in the data set corresponds to a meteorological feature term, which includes wind speed, wind direction, and initial speed ratio. The training goal of the SVR model is to fit the nonlinear relationship between the meteorological feature terms including wind speed, wind direction, and initial speed ratio and the target speed ratio by minimizing the following loss function of the SVR model:

[0019]

[0020] Among them, ω is the model weight in vector form. According to the mathematical definition of support vector regression (SVR), it describes the mapping relationship of the SVR model to the input data in the feature space; y i is the real output, corresponding to the target speed ratio, f(x i ) is the prediction output, corresponding to the meteorological feature term, ∈ is the tolerance error, and C is the penalty coefficient;

[0021] The kernel function of the SVR model uses the radial basis kernel function RBF.

[0022] Furthermore, K-means clustering is used to classify the historical wind condition data to obtain various categories, and the classification label of each category corresponds to the gear position of the stepped transmission in the electronically controlled transmission device.

[0023] Furthermore, the wind condition data, power generation and efficiency accumulated during the operation of the dual-rotor wind turbine are used as incremental data and added to the training set of the SVR model to perform incremental learning.

[0024] Compared with the prior art, the outstanding advantages of the present invention and its preferred embodiment include at least: integrating the power output of the front wind wheel and the rear wind wheel through a planetary differential device, achieving maximum energy conversion, and solving the energy loss problem caused by asynchronous rotation of the wind wheel in traditional wind power generation systems.

[0025] The wind energy capture control system and overload protection functions effectively improve the stability and safety of the system. In addition, by capturing the real-time wind speed to calculate the optimal wind rotor speed ratio, it can maintain efficient operation under different wind speed conditions and is suitable for a wide range of wind power generation application scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] The present invention is further described in detail below with reference to the accompanying drawings and specific embodiments:

[0027] Figure 1 It is a partial enlarged schematic diagram of the overall external structure of a wind turbine generator and a wind energy capture control device according to an embodiment of the present invention;

[0028] Figure 2 This is a front view of the interior of a wind turbine generator compartment according to an embodiment of the present invention;

[0029] Figure 3 This is an oblique view of the interior of a wind turbine generator compartment according to an embodiment of the present invention;

[0030] Figure 4 Schematic diagram of the control principle of an embodiment of the present invention. DETAILED DESCRIPTION

[0031] Hereinafter, the specific embodiments of the present application will be described in detail with reference to the accompanying drawings. According to these detailed descriptions, those skilled in the art can clearly understand the present application and can implement the present application. Without violating the principles of the present application, the features in the various embodiments can be combined to obtain new implementations, or certain features in certain embodiments can be replaced to obtain other preferred implementations.

[0032] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, it indicates the presence of features, steps, operations, devices, components and / or combinations thereof.

[0033] In order to make the features and advantages of this patent more obvious and easy to understand, the following embodiments are specifically described in detail as follows:

[0034] like Figure 1 As shown, this embodiment provides a dual-wind-wheel wind turbine generator system, including: a front wind wheel 2 and a rear wind wheel 3 with wind blades 1, a generator system 4, a tower 5 and a base 6. The main body of the wind turbine generator system is installed on the top of the tower 5, and the front wind wheel 2 and the rear wind wheel 3 are respectively installed on both sides of the generator system 4, and are connected to the generator system 4 through a rotating shaft to form a power transmission structure. The rotational power of the wind wheel is transmitted to the generator system 4 for wind energy conversion and power output. The generator system 4 is fixed on the top of the tower 5 and connected to the tower 5 through a supporting structure. The tower 5 supports the entire wind power generation device, and the generator system 4 and its upper components are installed at a certain height to capture wind energy. The tower is fixed to the ground through the base 6 to ensure the stability and wind resistance of the system.

[0035] The dual-wind-rotor wind turbine system also includes a wind energy capture control device. The wind energy capture control device integrates a wind direction and wind speed sensor 8 and a controller 7. The wind direction and wind speed sensor 8 is used to monitor the external wind speed in real time. The controller 7 is connected to the electric control speed change device inside the generator system 4. After the wind speed acquisition system obtains real-time data, it processes the data and adjusts the system parameters according to the wind speed acquisition and subsequent calculation results to adjust the speed ratio of the front wind wheel and the rear wind wheel.

[0036] like Figure 2 , Figure 3 As shown, in this embodiment, the rotation of the wind blades drives the front wind wheel and the rear wind wheel to rotate asynchronously, the front wind wheel is connected to the planetary differential device through an electronically controlled speed change device, and the rear wind wheel is connected to the sleeve of the planetary differential device through a gear reversal device. The planetary differential device integrates the power of the front wind wheel and the rear wind wheel and transmits it to the permanent magnet synchronous generator 10, achieving maximum wind energy capture and efficient electric energy conversion.

[0037] Specifically, Figure 2 The electronically controlled speed change device shown therein is connected to the output shaft 15 of the front wind wheel and is connected to the controller 7 to adjust the rotation speed output by the front wind wheel 2 so that the rotation speed ratio of the output of the front wind wheel 2 and the rear wind wheel 3 is maintained within the optimal range, thereby ensuring maximum wind energy capture under different wind speed conditions.

[0038] Specifically, the electronically controlled speed change device is composed of an input sprocket 14, an output sprocket 16 and a chain 19. The input sprocket 14 is connected to the output sprocket 16 via the chain 19 to transmit the power of the front wind wheel to the sun gear 11 in the planetary differential device.

[0039] The planetary differential device provided in this embodiment includes a sun gear 11, a plurality of planetary gears 13 and a sleeve 12. The power of the front wind wheel 2 is transmitted to the sun gear 11 through the output shaft 15 of the front wind wheel via the electronically controlled speed change device. The sun gear 11 is meshed with a plurality of planetary gears 13 through the internal structure, and is connected to the output sprocket 16 to form a power transmission path. With the support of the sleeve 12, the planetary gears 13 complete power integration to achieve efficient transmission and energy distribution. The power of the rear wind wheel 3 is transmitted to the sleeve 12 through the output shaft 9 of the rear wind wheel, through the gear reversing device 18 directly connected thereto, and then through the meshingly connected gear reversing device 17. The gear reversing device 17 is tightly connected to the sleeve 12 to ensure the effective transmission of the power of the rear wind wheel. The planetary gear 13 integrates the power of the two and transmits it to the main shaft of the permanent magnet synchronous generator 10 to ensure efficient power transmission and conversion.

[0040] In a planetary transmission, multiple planetary gears rotate around a central gear (sun gear), and these planetary gears can also rotate on their own in an internal gear ring or sleeve. Planetary transmissions are compact and efficient, and can provide multiple gear ratios, suitable for scenarios with high torque requirements.

[0041] In this embodiment, the controller 7 adjusts the speed ratio of the electronically controlled speed change device according to the wind speed to ensure that the wind wheel is always in the best operating state. The controller 7 is connected to the generator 10, the wind direction and speed sensor 8 and the electronically controlled speed change device 14, receives data from the wind direction and speed sensor 8, and adjusts the speed ratio of the electronically controlled speed change device 14 and the speed and torque of the generator 10 according to system requirements to ensure efficient and stable operation of the system.

[0042] Preferably, the controller 7 also integrates an overload protection function. When the wind speed is too high or the generator load is too large, the controller 7 automatically adjusts the speed and torque of the generator to ensure that the wind wheel operates within a safe range, avoids equipment damage and ensures system safety.

[0043] The present invention innovatively applies a planetary transmission device to a dual-rotor wind turbine system, with the front and rear rotors installed at different positions of the planetary transmission device. The front rotor is connected to the sun wheel through an electronically controlled speed change device, and the rear rotor is connected to the planetary gear system through a gear reversal device. The planetary gears can rotate around the sun wheel while also rotating on their own, thereby integrating the power of the front and rear rotors and transmitting it to the generator.

[0044] Through the structural design of the planetary transmission device, the present invention effectively solves the problem of asynchronous rotation speed of the front and rear wind wheels due to different wind speeds and wind pressures, reduces the loss of internal energy consumption, and maximizes the conversion of wind energy into mechanical energy. This innovative design improves the overall efficiency of the generator set and ensures that the wind wheel can operate efficiently under different wind conditions.

[0045] This wind turbine uses the coaxial and skewed connection of the front and rear wind wheels to increase the force-bearing area by using the stable structure of the front and rear three-blade impellers, thereby improving the wind energy capture rate. The design of the front and rear two sets of impellers enables the double impeller generator to significantly increase the output power compared to the single impeller generator under medium and low wind speed conditions, ensuring efficient power generation in a single wind turbine scenario.

[0046] In this example, if Figure 1 The internal structure of the wind turbine is shown in Figure 2 , Figure 3 , including generator, planetary differential device, speed change device, wind energy capture control device and sensor and other components. Control method such as Figure 4 As shown, the control panel connects the generator, the transmission and the sensor to comprehensively analyze the actual wind conditions and make corresponding adjustments. The basic algorithm idea of ​​the controller 7 to achieve the adjustment is as follows:

[0047] The formula for maximum wind energy capture power is as follows:

[0048] P=0.5×A×ρ×C p

[0049] Among them, ρ is an objective physical quantity related to air density, A is the swept area of ​​the fan rotor, C p The wind energy utilization coefficient is the degree to which the wind turbine absorbs energy from natural wind energy. The wind energy utilization coefficient indicates the efficiency of the wind turbine in converting wind energy into electrical energy. The characteristics of the wind turbine are often expressed by the power coefficient C. p The dimensionless performance curve is used to express that the wind energy utilization coefficient is a function of the fan tip speed ratio φ and the pitch angle β, that is, Therefore, the wind energy utilization coefficient C p It can be expressed by an approximate formula:

[0050]

[0051] For a twin-rotor wind turbine, when the wind is in the positive direction, the front and rear rotors have different abilities to capture wind energy. After simulation tests, the approximate empirical formula for the wind speed region when the front and rear rotors capture wind energy is summarized as follows:

[0052] V b =V a (1-a) 2

[0053] Where V a is the wind speed at the front wind rotor, V b is the wind speed at the rear wind wheel.

[0054]

[0055] According to the empirical formula and simulation results, when the wind speed is stable, the speed ratio of the front and rear wind wheels for maximum wind energy capture is concentrated in the range of R = 0.6 to 0.7. For side wind, due to the uneven wind speed distribution of the front and rear wind wheels when the wind is from the side, the front wind wheel is subjected to a large direct wind pressure, while the rear wind wheel is partially blocked, and the wind speed is significantly attenuated. In the case of side wind, according to the empirical formula and simulation results, the speed ratio of the front and rear wind wheels for optimal wind energy capture is concentrated in the range of R = 0.6 to 0.8. However, the optimal speed ratio for each specific scenario needs to be determined in combination with specific wind force and speed information.

[0056] In order to further improve the control accuracy of the system, especially to optimize the speed ratio of the front and rear wind rotors under complex wind conditions, the present invention introduces a machine learning model as a supplement based on the above formula calculation. The system collects data such as wind speed, wind direction, front and rear wind rotor speeds, and wind energy capture efficiency in real time through equipment such as wind direction and speed sensors. The data not only comes from monitoring during actual operation, but also includes results obtained through simulation tests. To ensure the accuracy and consistency of the input data, all data must be preprocessed before entering the system, including removing outliers and standardizing, so that subsequent machine learning models can make full use of these data.

[0057] One of the innovations of the present invention is that it combines the machine learning model with the traditional formula calculation. The system first estimates the preliminary range of the optimal speed ratio corresponding to the current wind speed through the traditional formula calculation, and then optimizes the calculation results using the machine learning model. When the wind conditions change, the machine learning model can quickly analyze the new wind speed and wind direction data, provide more accurate adjustment suggestions, and ensure that the system is always in an efficient operating state. The calculated speed ratio range is used as a constraint using a mathematical model and input into the SVR model to further fit the wind rotor speed ratio that achieves maximum wind energy capture under specific conditions.

[0058] The machine learning model system first uses the K-means clustering algorithm to classify historical wind data to build a data set for SVR model training. K-means clustering discretizes continuous meteorological data and generates feature terms, making the structure of the feature set more concise and easier to analyze and optimize. After each sampling, the system inputs the discretized features as labels into the SVR model to ensure that the system can quickly output the optimal front and rear wind rotor speed ratio after detecting new wind conditions.

[0059] The wind speed interval is divided into units of 4 m / s, starting at 0 m / s, and mapped to 0, 1, 2, and so on (such as 0-4 m / s, 4-8 m / s, 8-12 m / s, inclusive), and the wind speed data sampled each time is rounded to reduce the complexity of the feature space.

[0060] The horizontal wind direction component is mapped to 0, 1, and 2, corresponding to "radial positive 0°", "radial lateral 15°-45°", and "radial lateral 45°-75°", respectively.

[0061] The vertical wind direction component is mapped to 0 and 1, corresponding to “longitudinal 15°-75°” and “longitudinal positive 90°” respectively.

[0062] The temperature range is discretized in units of 5°C, starting at -20°C, and mapped to 0, 1, 2, and so on (e.g., -20°C to -15°C, -15°C to -10°C).

[0063] The weather variability is expressed as 0 for stable weather conditions and 1 for significant changes within the next hour.

[0064] In a design example, the system uses a step-variable transmission with four gears: 0.9, 0.8, 0.7, and 0.6. During operation, the K-means clustering labels correspond to the gears of the step-variable transmission. Whenever new meteorological data is detected, the system fits the category through the SVR model and directly outputs the optimal speed ratio of the front and rear wind wheels. This structured strategy avoids recalculating complex models every time the operating conditions change, ensuring the efficiency and accuracy of the system response.

[0065] The logic of constructing the feature set is to convert complex meteorological data into simplified category labels, which correspond to the gear positions of the step-by-step transmission, reduce computational overhead, and optimize the response time of the system. The dataset generated by K-means clustering is the input basis of the SVR model, enabling the system to quickly predict the optimal control strategy under different working conditions.

[0066] The solution of the SVR model includes the following steps:

[0067] Model training: The training is performed using the K-means clustered dataset. Each category of data input features corresponds to meteorological feature terms, including wind speed, wind direction, temperature, meteorological variability, and initial speed ratio. Objective function: The SVR model fits the nonlinear relationship between wind speed and speed ratio by minimizing the following loss function:

[0068]

[0069] Among them, y i is the real output, f(x i ) is the predicted output, ∈ is the tolerance error, and C is the penalty coefficient, which is used to control the complexity of the model.

[0070] Kernel function selection: The system uses radial basis kernel function (RBF) to handle complex nonlinear relationships. The form of the radial basis kernel function is:

[0071] K(x i ,y i ) = exp(-γ||x i -x j || 2 )

[0072] Among them, γ controls the learning ability and generalization performance of the model.

[0073] Based on the joint software simulation analysis of OpenFast and Simulink, this system tested the following two typical wind conditions. For each scenario, the system identifies the corresponding category based on the K-means clustering terms, and selects the optimal front and rear wind wheel speed ratio to achieve maximum power output. In this system, the front and rear wind path speed ratios set by the stepless transmission are 0.9, 0.8, 0.7, and 0.6, and the optimal front and rear wind wheel speed ratios are selected to obtain the maximum wind energy capture rate. Since the word bag is used to summarize the typical characteristics of the meteorological environment, various wind speeds and directions are discretized, and the wind path speed ratio that can achieve maximum wind wheel capture in the two scenarios is analyzed.

[0074] like Figure 4 As shown, based on the above preferred design, an optimal control scheme provided by an embodiment of the present invention is:

[0075] First, the controller 7 collects real-time wind direction and wind speed through the wind direction and wind speed sensor 8, and obtains the corresponding data of the meteorological station through wireless communication to evaluate the meteorological stability. The latter data collection and subsequent judgment and calculation process can be completed locally according to the computing power of the hardware itself or realized through edge computing and other methods;

[0076] The intervals for collecting real-time wind direction and wind speed and evaluating the adjustment of the speed ratio are determined according to meteorological stability; and the need to adjust the speed ratio and its value are judged and calculated in turn, and then a judgement is made as to whether the maximum speed needs to be limited, so as to control the electronically controlled speed change device and the generator 10.

[0077] It should be noted that, if conditions permit, the system can obtain meteorological environment status information by adding sensors (such as temperature sensors) and wireless transmission methods (such as meteorological variability), thereby collecting and updating data. However, according to the above design of the present invention, as long as the wind speed related information can be obtained, the optimal speed ratio can be fitted, and other parameters can use default values. If accurate values ​​are used, the fitting results will be more accurate, but it is not an absolutely necessary condition.

[0078] Scenario 1: K-means terms in a stable wind speed scenario:

[0079] Wind speed: 12-16 m / s; horizontal wind direction: radial lateral 15°-45°; vertical wind direction: longitudinal 15°-75°; temperature: 20-25℃; meteorological variability: 0 (meteorological stability).

[0080] Real data collected on site:

[0081] Wind speed (m / s): 13.2, 14.0, 15.8, 12.5, 14.9

[0082] Horizontal wind direction (degrees): 20°, 22°, 35°, 28°, 42°

[0083] Vertical wind direction (degrees): 25°, 25°, 35°, 20°, 25°

[0084] Temperature (℃): 21.5, 22.0, 23.0, 22.5, 22.5

[0085] Weather variability: 0 (weather conditions are stable)

[0086] In this scenario, the wind speed is high and the wind direction changes are limited, and the meteorological conditions are stable. The system selects the best front and rear wind wheel speed ratio in the current wind condition category based on the K-means clustering results. According to simulation analysis, under this condition, the front and rear wheel speed ratio of 0.7 can achieve maximum power output and ensure the optimal wind energy capture rate. Due to "Meteorological Variability: 0", it is necessary to re-collect meteorological information after 1H and determine whether to adjust the speed ratio based on the clustering results.

[0087] Scenario 2: K-means entries in wind speed fluctuation scenario: Wind speed: 8-12 m / s; Horizontal wind direction: radial positive (0°); Vertical wind direction: longitudinal 15°-75°; Temperature: 15-20°C; Meteorological variability: 1 (meteorological variability is significant)

[0088] Real data collected on site:

[0089] Wind speed (m / s): 8.5, 9.2, 10.8, 11.3, 9.7

[0090] Horizontal wind direction (degrees): 5°, 0°, 10°, 3°, 0°

[0091] Vertical wind direction (degrees): 20°, 30°, 60°, 55°, 50°

[0092] Temperature (℃): 16.5, 18.0, 17.5, 19.0, 15.8

[0093] Weather variability: 1 (weather variability is significant)

[0094] In this scenario, the wind speed fluctuates in the range of 8-12 m / s, and the meteorological conditions are changeable. The system uses the K-means clustering results to identify the wind condition category and dynamically selects the appropriate speed ratio. The simulation results show that under this condition, a speed ratio of 0.8 between the front and rear wind wheels can effectively improve the wind energy capture rate of the system. Due to "weather variability: 1", it is necessary to re-collect meteorological information after 0.5 hours and determine whether to adjust the speed ratio based on the clustering results. The speed ratio obtained by the system through SVR analysis can not only ensure the stability of power generation in wind speed fluctuations, but also avoid energy loss caused by frequent gear switching.

[0095] In addition, the system has online learning and incremental update functions. As the operating data accumulates, the system regularly adds new wind data and the corresponding power generation and converted efficiency to the model training set, and performs incremental learning to ensure that the model is continuously optimized to adapt to new meteorological conditions. When a new wind condition combination is detected, the system can update the model online to ensure that it can still provide accurate optimization strategies in complex and dynamic wind conditions.

[0096] By combining K-means clustering with SVR optimization, the system achieves efficient control of the dual-rotor wind turbine. Even in rapidly changing wind conditions, the system can instantly adjust the speed ratio of the front and rear wind rotors to ensure maximum capture of wind energy and reduce energy loss during mechanical transmission, providing stable protection for the efficient operation of the generator.

[0097] After determining the speed ratio, the controller 7 adjusts the generator speed by connecting to the variable frequency drive (VFD) on the generator. After determining the generator speed, the system can further feedback and adjust the speed input of the front and rear wind rotor fans. Such a design can not only ensure that the wind rotor captures the wind energy to the maximum extent, but also avoid energy loss in the mechanical transmission process.

[0098] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the present invention should be understood by people with ordinary skills in the field to which the present invention belongs. The "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0099] The above is a preferred embodiment of the present invention. It should be pointed out that, for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications are also considered to be within the scope of protection of the present invention.

[0100] This patent is not limited to the above-mentioned optimal implementation mode. Anyone can derive other various forms of a dual-rotor wind turbine that achieves maximum wind energy capture under the inspiration of this patent. All equivalent changes and modifications made within the scope of the patent application of the present invention should be covered by this patent.

Claims

1. A double-rotor wind turbine generator for achieving maximum wind energy capture, characterized in that: The front wind wheel and the rear wind wheel rotate asynchronously: the front wind wheel is connected to the sun wheel of the planetary differential device through an electronically controlled speed change device; the rear wind wheel is connected to the planetary gear system of the planetary differential device through a gear reversal device, and the power of the front wind wheel and the rear wind wheel is integrated through the planetary differential device and then transmitted to the generator.

2. A double-rotor wind turbine generator for achieving maximum wind energy capture according to claim 1, characterized in that: The controller monitors the current external wind speed in real time through the wind speed sensor, and adjusts the electronically controlled speed change device within the maximum limited speed according to the current external wind speed, so that the speed ratio of the output speed of the front wind wheel and the output speed of the rear wind wheel is maintained within the optimal range to achieve maximum wind energy capture.

3. A double-rotor wind turbine generator for achieving maximum wind energy capture according to claim 2, characterized in that: When the current external wind speed or the load of the dual-rotor wind turbine generator is greater than the corresponding set threshold, the controller adjusts the speed and torque of the generator through the generator's inverter to ensure that the front and rear wind rotors operate within a safe range.

4. A double-wind-rotor wind turbine generator for achieving maximum wind energy capture according to claim 1, characterized in that: The electronically controlled speed change device is composed of an input sprocket, an output sprocket and a chain. The input sprocket is connected to the output sprocket through the chain, the front wind wheel is connected to the input sprocket, and the output sprocket is connected to the sun wheel in the planetary differential device to transmit the power of the front wind wheel to the sun wheel in the planetary differential device.

5. A double-wind-rotor wind turbine generator for achieving maximum wind energy capture according to claim 4, characterized in that: The planetary differential device includes a planetary gear system and a sleeve; the planetary gear system includes a sun gear and multiple planetary gears; the sun gear is meshed with multiple planetary gears; the output shaft of the rear wind wheel is connected to the second gear of the gear reversal device, and the first gear is meshed with the second gear of the gear reversal device; the second gear is fixedly connected to the sleeve, and the sleeve is arranged on the outer periphery of the planetary gear system and the inner ring is meshed with multiple planetary gears.

6. A double-wind-rotor wind turbine generator for achieving maximum wind energy capture according to claim 2, characterized in that: The controller obtains a preliminary range of the optimal speed ratio corresponding to the current external wind speed based on the maximum wind energy capture power estimation, and takes the preliminary range of the optimal speed ratio corresponding to the current external wind speed as a preliminary constraint; within the range of the preliminary constraint, the SVR model obtained by training the current external wind speed input collected in real time is fitted to obtain the optimal range between the speed ratio of the front wind wheel output and the rear wind wheel output under the preliminary constraint conditions.

7. A double-rotor wind turbine generator for achieving maximum wind energy capture according to claim 6, characterized in that: The wind speed sensor is specifically a wind direction wind speed sensor; the method for estimating the maximum wind energy capture power is used to obtain a preliminary range of the optimal speed ratio corresponding to the current external wind speed, including: based on the maximum wind energy capture power estimation, according to the simulation test empirical formula of the forward wind and the side wind and the simulation results, obtaining the initial speed ratio of the best wind energy capture under different wind directions based on the current external wind speed as the preliminary range of the optimal speed ratio; Within the scope of the preliminary constraints, the SVR model obtained by training the current external wind speed input collected in real time is fitted to obtain the optimal range of the speed ratio of the front wind wheel output to the rear wind wheel output under the preliminary constraints, including: taking the meteorological characteristics of wind speed, wind direction and initial speed ratio as input, combining the discretized feature labels input into the SVR model obtained by training to fit the optimal range of the speed ratio of the front wind wheel output to the rear wind wheel output under the preliminary constraints.

8. A double-wind-rotor wind turbine generator for achieving maximum wind energy capture according to claim 7, characterized in that: The training process of the SVR model is specifically as follows: The K-means clustering algorithm is used to classify the historical wind data, and a data set for SVR model training is constructed based on the classification results; The SVR model is trained using the data set. Each category of data input features in the data set corresponds to a meteorological feature term, which includes wind speed, wind direction, and initial speed ratio. The training goal of the SVR model is to fit the nonlinear relationship between the meteorological feature terms including wind speed, wind direction, and initial speed ratio and the target speed ratio by minimizing the following loss function of the SVR model: Where ω is the model weight in vector form, y i is the real output, corresponding to the target speed ratio, f(x i ) is the prediction output, corresponding to the meteorological feature term, ∈ is the tolerance error, and C is the penalty coefficient; The kernel function of the SVR model uses the radial basis kernel function RBF.

9. A double-rotor wind turbine generator for achieving maximum wind energy capture according to claim 8, characterized in that: K-means clustering is used to classify the historical wind condition data to obtain various categories, and the classification label of each category corresponds to the gear position of the step-type transmission in the electronically controlled transmission device.

10. A double-wind-rotor wind turbine generator for achieving maximum wind energy capture according to claim 8, characterized in that: The wind condition data, power generation and efficiency accumulated during the operation of the dual-rotor wind turbine are used as incremental data and added to the training set of the SVR model to perform incremental learning.

Images