Wind turbine generator blade infrared radiation deicing system and intelligent control method thereof

Abstract

The invention relates to the technical field of wind power generation, and discloses a wind turbine generator blade infrared radiation deicing system and an intelligent control method thereof, and the wind turbine generator blade infrared radiation deicing system comprises a multi-mode state sensing module, an intelligent deicing control module and an infrared radiation module. The multi-mode state sensing module is used for collecting multi-source state data such as icing thickness, temperature field, unit operation and environment; the intelligent deicing control module analyzes data, when the icing thickness exceeds a preset threshold value or the power deviation rate exceeds a threshold value and the infrared image confirms a low-temperature area, a deicing decision is triggered, and the diagnosis accuracy is high. During deicing, the module cooperatively controls the heating area, the power and the time of the infrared radiation module according to the width and the icing grade of the target area of the blade, so that the partitioned precise heating is realized, and the energy utilization rate is improved. The invention further discloses a corresponding automatic control method, blade safety is guaranteed through real-time temperature closed-loop control, and full-process automatic operation from monitoring, decision making, execution to resetting is achieved.

Description

Technical Field

[0001] This invention relates to the field of wind power generation technology, specifically to an infrared radiation de-icing system for wind turbine blades and its intelligent control method. Background Technology

[0002] When wind turbines operate in cold and humid regions, their blades are highly susceptible to icing. This phenomenon alters the aerodynamic shape of the blades, leading to a decrease in power generation efficiency. Simultaneously, the aerodynamic and mass imbalances caused by icing can induce severe vibrations in the turbine, accelerate fatigue damage to critical components, and even pose safety risks due to ice shedding. Existing technologies have explored solutions to address this issue. However, these solutions still have shortcomings in terms of the accuracy of icing diagnosis, the energy efficiency of de-icing operations, and the automation and safety of operation.

[0003] First, existing technologies generally lack the ability to accurately and reliably diagnose the icing status of blades in real time. The decision to start the de-icing system often relies on a single or indirect parameter, which is prone to misjudgment. Premature start-up can cause unnecessary downtime and power generation loss, or failure to respond in time can worsen the icing problem, making it impossible to guarantee that de-icing operations are performed at the appropriate time.

[0004] Secondly, regarding energy efficiency, traditional internal heating methods, such as gas-fired or electric-fired de-icing, generally suffer from low thermal efficiency and high energy consumption because they require heating the entire blade structure to melt the surface ice layer. Furthermore, they cannot provide differentiated treatment based on the actual distribution of ice accretion. As for infrared radiation non-contact heating technology, existing applications also lack mechanisms for fine-tuning the heating energy, making it difficult to achieve the energy-saving goal of supplying energy on demand for different icing areas.

[0005] Finally, regarding automation and safety, most existing de-icing solutions fail to achieve fully automated operation from condition monitoring to task completion, increasing the cost and uncertainty of manual intervention. Furthermore, during rapid de-icing using high-intensity energy, existing technologies lack a real-time closed-loop control mechanism for monitoring blade surface temperature. This exposes the blade composite materials to the risk of overheating damage, posing a potential threat to the structural safety of the equipment itself. Summary of the Invention

[0006] To address the shortcomings of existing technologies, this invention provides an infrared radiation de-icing system for wind turbine blades and its intelligent control method. This system solves the problem that existing technologies lack precise icing state perception, refined energy regulation methods, and fully automated safety closed-loop control processes, making it difficult to achieve efficient, energy-saving, and safe automated de-icing of wind turbine blades.

[0007] To solve the above-mentioned technical problems, the present invention provides the following technical solution: The first aspect of the present invention provides an infrared radiation de-icing system for wind turbine blades, the system comprising: a multimodal state sensing module, an intelligent de-icing control module, and an infrared radiation module.

[0008] The multimodal state sensing module is configured to collect multi-source state data related to the icing state of the blades and output the multi-source state data.

[0009] The intelligent de-icing control module is configured to be electrically connected to the multimodal state sensing module, and is used to receive multi-source state data output by the multimodal state sensing module, analyze the multi-source state data to perform icing diagnosis and de-icing decision, and generate and output control commands based on the decision results of the icing diagnosis and de-icing decision.

[0010] The infrared radiation module is configured to be electrically connected to the intelligent de-icing control module, and is used to receive the control command from the intelligent de-icing control module, and perform non-contact heating de-icing operation on the surface of the wind turbine blades according to the control command.

[0011] Preferably, the multimodal state perception module includes: at least one icing flexibility monitoring unit, one infrared thermal imaging unit, one unit operation data monitoring unit, and one environmental unit. Accordingly, the multi-source state data includes: icing thickness data collected by the icing flexibility monitoring unit, temperature field data collected by the infrared thermal imaging unit, unit operation data collected by the unit operation data monitoring unit, and environmental data collected by the environmental unit.

[0012] In one specific embodiment, the specific process of the intelligent de-icing control module performing icing diagnosis and de-icing decision-making is configured to include: Extract the actual output power and real-time wind speed from the unit operation data from the multi-source status data; Based on the real-time wind speed, the theoretical output power is calculated using a theoretical power linear interpolation formula; The power deviation rate between the theoretical output power and the actual output power is calculated using a power deviation rate calculation formula. A decision to initiate de-icing is triggered when any of the following conditions are met: the ice thickness data acquired by the multimodal state sensing module exceeds a preset thickness threshold; or, the power deviation rate exceeds a preset deviation rate threshold, and the temperature field data acquired by the infrared thermal imaging unit indicates the presence of a low-temperature region.

[0013] The deviation rate threshold and thickness start-up threshold are preset based on historical unit operation data and / or experimental test data of the wind turbine.

[0014] Preferably, the intelligent de-icing control module is further configured to control the heating area, and its specific process includes: A target location for heating a blade is determined based on an internally preset scanning de-icing operation program, and the blade width at the target location is calculated using a blade width interpolation formula. A heating area control command is generated and sent based on the calculated blade width to control the heating area of ​​the infrared radiation module to match the calculated blade width.

[0015] In one specific embodiment, the intelligent de-icing control module is also configured to control the heating power, the specific process of which includes: Based on the icing thickness data obtained by the multimodal state sensing module, an icing level is determined and a heating power coefficient is matched. Based on the calculated blade width and the heating power coefficient, the actual heating power is calculated using a segmented heating power calculation formula; A heating power control command is generated and sent based on the calculated actual heating power to control the infrared radiation module to heat according to the calculated actual heating power.

[0016] Furthermore, the intelligent de-icing control module is also configured to control the heating time, the specific process of which includes: For a blade section defined by the two icing flexibility monitoring units, the icing level at both ends of the blade section is obtained, and the larger value of a heating time coefficient is matched. Based on the larger value of the heating time coefficients, the actual heating time of the blade interval is calculated using a range heating time calculation formula. The infrared radiation module is controlled to perform scanning irradiation heating on the blade area, and the total heating time is equal to the calculated actual heating time.

[0017] Preferably, the infrared radiation module is installed at the bottom of the wind turbine tower and mounted on a gimbal. The straight-line distance between its installation position and the center of the tower simultaneously satisfies the geometric constraints defined by both a minimum installation distance formula and a maximum installation distance formula.

[0018] Preferably, multiple icing flexibility monitoring units are installed along the length of the blade. Each icing flexibility monitoring unit directly measures the icing thickness at its location and outputs the icing thickness data to the intelligent de-icing control module, serving as a direct basis for determining the heating power coefficient and the heating time coefficient.

[0019] Preferably, the infrared thermal imaging unit is configured to monitor the temperature field data on the blade surface in real time during the de-icing operation and feed the temperature field data back to the intelligent de-icing control module to achieve closed-loop temperature control of the heating process, so as to prevent the blade from overheating and verify the de-icing effect.

[0020] A second aspect of the present invention provides an intelligent control method for infrared radiation de-icing of wind turbine blades. This method, applied to the infrared radiation de-icing system for wind turbine blades described in any of the foregoing embodiments, includes the following steps: Step S1, Status Monitoring and Diagnosis: The multi-modal status sensing module acquires multi-source status data, including environmental data, unit operation data, icing thickness data, and temperature field data; when the environmental data meets a preset icing condition, the system is switched to an early warning state; and based on the real-time wind speed and actual output power in the unit operation data, the power deviation rate is calculated to continuously monitor unit performance.

[0021] Step S2, Decision and Activation: When the ice thickness data exceeds a thickness activation threshold, or the power deviation rate exceeds a deviation rate threshold and the temperature field data confirms the existence of a low-temperature region, a de-icing decision is triggered; and based on the de-icing decision, the wind turbine is instructed to safely shut down, and one blade to be de-iced is rotated to a preset working position.

[0022] Step S3, Precise De-icing by Zone: The infrared radiation module is controlled to perform precise de-icing of the target area of ​​the blade to be de-iced by zone. This step is coordinated with the following controls: dynamically adjusting the heating area according to the blade width of the target area, adjusting the heating power according to the ice thickness of the target area in stages, and accurately calculating the heating time according to the ice thickness of the target area.

[0023] Step S4, Rotation and Reset: After de-icing a single blade, rotate to the next blade to be de-iced and repeat step S3; after all blades that need de-icing have been processed, instruct the fan to resume operation and return the system to status monitoring mode.

[0024] This invention provides an infrared radiation de-icing system for wind turbine blades and its intelligent control method. It has the following beneficial effects: 1. This invention improves the accuracy and reliability of icing diagnosis by incorporating a multimodal state sensing module and combining it with the multi-condition decision-making logic of the intelligent de-icing control module. The system does not rely on a single data source but comprehensively analyzes icing thickness data directly measured by the icing flexibility monitoring unit, power deviation rate indirectly reflecting performance degradation by the unit operation data monitoring unit, and blade surface temperature field data confirmed by the infrared thermal imaging unit. This cross-validation of multi-source data effectively avoids misjudgments caused by single sensor failure or environmental interference, ensuring that de-icing decisions are triggered at the appropriate time, avoiding unnecessary downtime losses or exacerbated icing due to missed diagnoses.

[0025] 2. This invention achieves precise, zoned de-icing, improving de-icing efficiency and saving energy. The intelligent de-icing control module dynamically matches the heating area of ​​the infrared radiation module based on the blade width at the target location; simultaneously, it finely adjusts the heating power and heating time according to the actual ice thickness data of different areas. This control method precisely applies energy to the area requiring de-icing, avoiding indiscriminate and excessive energy output to the entire blade, thereby shortening the single operation time and reducing the overall energy consumption of the system.

[0026] 3. This invention provides a complete automated operation process, achieving unmanned operation from status monitoring, icing diagnosis, task initiation, zoned de-icing to system reset, thus improving the automation level of wind turbine operation and maintenance. Furthermore, by using an infrared thermal imaging unit to monitor and provide feedback on the blade surface temperature field in real time during de-icing operations, a closed-loop temperature control system for the heating process is established. This mechanism effectively prevents overheating damage to the blade material caused by localized heat accumulation, ensuring both effective de-icing and the structural safety of the equipment itself. Attached Figure Description

[0027] Figure 1 This is a schematic diagram of an infrared radiation de-icing system architecture for wind turbine blades according to an embodiment of the present invention. Figure 2 This is a flowchart of an intelligent control method for infrared radiation de-icing of wind turbine blades according to an embodiment of the present invention. Detailed Implementation

[0028] The specific embodiments of the present invention will be further described in detail below with reference to the accompanying drawings and examples.

[0029] Please see the appendix Figure 1 and 2 This invention provides an infrared radiation de-icing system for wind turbine blades, comprising: an infrared radiation module 100; a multi-modal state sensing module 200; and an intelligent de-icing control module 300.

[0030] The multimodal state perception module 200 is configured to collect multi-source state data related to wind turbine blades in real time.

[0031] The intelligent de-icing control module 300 communicates with the multimodal state perception module 200 to receive and process the aforementioned multi-source state data, and performs icing diagnosis and de-icing decision based on the internally integrated algorithm model, thereby generating control instructions containing specific execution parameters.

[0032] The infrared radiation module 100 communicates with the intelligent de-icing control module 300 to respond to the aforementioned control commands and adjusts its angle via its onboard gimbal to project infrared radiation energy onto the target area of ​​the blades, thereby performing physical de-icing operations.

[0033] In one specific embodiment of the present invention, the infrared radiation module 100 is deployed at the bottom of the wind turbine tower. To ensure that the infrared radiation it projects can effectively cover the working range of the blades, the straight-line distance between its installation position and the center of the tower is [not specified]. The following geometric constraints must be met: The formula for minimum installation distance is: ; The formula for the maximum installation distance is: ; In the formula: The straight-line distance between the infrared radiation module 100 and the center point of the unit tower; This refers to the tower height of the wind turbine. This refers to the blade length of the wind turbine. The elevation angle of the infrared radiation module 100; This refers to the fixed elevation angle between the wind turbine blades and the tower.

[0034] When performing icing diagnosis, the intelligent de-icing control module 300 will rely on the real-time wind speed collected by the multi-modal state sensing module 200. The theoretical output power at the current wind speed is calculated using the theoretical power linear interpolation formula. The theoretical power linear interpolation formula is: ; In the formula: To provide real-time wind speed Theoretical output power at the following level; Standard wind speed point The corresponding theoretical power; Standard wind speed point The corresponding theoretical power; For power curves less than And closest Standard wind speed point; For power curve greater than And closest The standard wind speed point.

[0035] Subsequently, the intelligent de-icing control module 300 calculates the theoretical output power. The actual output power obtained from the multimodal state sensing module 200 By comparison, the power deviation rate is obtained through the power deviation rate calculation formula. The formula for calculating the power deviation rate is: ; In the formula: The theoretical power deviation rate is the wind speed. Current wind speed Theoretical output power at the following level; Current wind speed The actual output power of the wind turbine unit.

[0036] When generating control commands for the infrared radiation module 100, the intelligent de-icing control module 300 first calculates the blade width of the target area to be de-iced using the blade width interpolation formula. This is used to control the area of ​​the heating spot. The formula for calculating the blade width interpolation is: ; In the formula: The blade width at the target location to be calculated; This is the distance from the target location to the blade tip; and Two known measurement points on the blade and Length from the leaf tip; and To measure at the respective points and The known blade width at the location.

[0037] Next, the intelligent de-icing control module 300 determines the heating power coefficient based on the ice thickness level. The actual heating power for the target area is calculated using a segmented heating power calculation formula. The formula for calculating the power of segmented heating is: when hour, ; when hour, ; In the formula: For width of The actual heating power applied to the blade area; The heating power coefficient is determined by the icing level; The width of the blade in the currently irradiated area; This is the maximum chord length of the blade; This is the rated total power of the infrared radiation module 100.

[0038] Meanwhile, the intelligent de-icing control module 300 determines the heating time coefficient based on the icing level. The required heating time for the target area can be calculated using the interval heating time calculation formula. The formula for calculating the interval heating time is: ; In the formula: This refers to the actual heating time of the blade section to be heated; Heating time coefficient corresponding to the icing levels at both ends of an interval The larger value in; The maximum base heating time per unit location is set.

[0039] The infrared radiation module 100 is a terminal device for performing physical de-icing operations. In one embodiment of the invention, the infrared radiation module 100 includes one or more infrared radiation units, preferably mid- to far-infrared radiators. Specifically, the infrared radiation units can be carbon fiber infrared heating tubes or ceramic infrared heating plates. These units are arranged in an array to form a controllable heating spot with a specific size and shape.

[0040] The infrared radiation emitted by the infrared radiation unit is set within the range of 8μm to 14μm. The principle behind this wavelength range is that ice and water have a high absorption rate in this band, which allows the infrared radiation energy to be effectively absorbed by the ice or melted water layer on the blade surface, thereby efficiently converting the radiation energy into heat energy and causing the interface layer between the ice layer and the blade surface to melt.

[0041] The infrared radiation module 100 is deployed at the bottom of the wind turbine tower, preferably in the upwind position of the prevailing winter wind direction. This deployment method is designed to use natural wind to blow melted water or detached ice from the tower and the infrared radiation module 100 body during the de-icing process, avoiding secondary icing or equipment damage.

[0042] The infrared radiation module 100 is fixedly mounted on a multi-axis gimbal. For example, the gimbal can be a two-dimensional gimbal, which includes a horizontal rotation mechanism for controlling the horizontal azimuth angle and a vertical pitch mechanism for controlling the vertical pitch angle. The mechanisms are driven by independent motors and receive precise angle control signals from the intelligent de-icing control module 300, thereby driving the infrared radiation module 100 to point, achieving precise aiming and scanning illumination of different areas on the stationary blade.

[0043] Considering that there is usually a fixed elevation angle of 2° to 8° between the wind turbine blades and the tower, While balancing heating effect and safety, the infrared radiation module 100 itself has a pitch angle It is set within a reasonable range.

[0044] Specifically, to ensure that the highest point of the leaf is illuminated, its pitch angle... The maximum angle should not exceed 82°, and preferably not exceed 80°. To ensure safety and effective irradiation distance, the elevation angle should be... The minimum temperature should not be lower than 20°.

[0045] Based on the above angle constraints, the straight-line distance can be derived. The constraint formula.

[0046] To ensure that the infrared radiation module 100 can completely cover the entire working surface of the blade, its installation position is at a straight-line distance from the center of the tower. Specific geometric constraints must be followed. These constraints are defined by the following two formulas: The formula for minimum installation distance is: ; This formula defines the closest installation distance between the infrared radiation module 100 and the center of the tower. This constraint ensures that when the infrared radiation module 100 is at its maximum pitch angle, its radiation beam can still illuminate the upper region of the blade at its highest point, avoiding the creation of a blind spot.

[0047] The formula for the maximum installation distance is: ; This formula defines the farthest installation distance between the infrared radiation module 100 and the center of the tower. This constraint ensures that when the infrared radiation module 100 is at its minimum pitch angle, its radiation beam can effectively hit the tip of the blade without wasting energy or failing to cover the lower part of the blade due to excessive distance.

[0048] In the formula: The straight-line distance between the infrared radiation module 100 and the center point of the unit tower; This refers to the tower height of the wind turbine. This refers to the blade length of the wind turbine. The elevation angle of the infrared radiation module 100; This refers to the fixed elevation angle between the wind turbine blades and the tower.

[0049] The multimodal state perception module 200 is configured to collect state information related to blade icing from multiple dimensions, and may specifically include: an infrared thermal imaging unit, an icing flexibility monitoring unit, an environmental unit, a wind turbine operation data monitoring unit, and an image recognition unit.

[0050] In one specific embodiment, the infrared thermal imaging unit is a long-wave infrared thermal imager. This infrared thermal imaging unit can be integrated with the infrared radiation module 100, for example, mounted on the same multi-axis gimbal, so that its field of view is synchronized with the illumination direction of the infrared radiation module 100.

[0051] The infrared thermal imaging unit operates by passively receiving infrared radiation emitted from the surface of wind turbine blades in a non-contact manner and converting it into thermal map data characterizing temperature distribution based on the radiation intensity. Since the emissivity and temperature of ice are typically lower than those of the blade substrate, the iced area will appear as a low-temperature zone with a significant temperature difference compared to the non-iced area on the thermal map.

[0052] The heat map data, essentially a two-dimensional data matrix containing temperature values ​​at various points on the blade surface, is transmitted in real time to the intelligent de-icing control module 300. The processing algorithm within the intelligent de-icing control module 300 analyzes this data matrix, for example, by setting a temperature threshold (e.g., 0°C) or a relative temperature difference threshold with the normal surface temperature of the blade, to identify and label all pixels in the heat map that are below that threshold.

[0053] By clustering and contour extraction of these low-temperature pixels, the spatial distribution, geometry, and coverage area of ​​ice on the blade surface can be quickly determined. This temperature field information is one of the high-priority input data for the intelligent de-icing control module 300 to make de-icing decisions and plan the scanning path. Furthermore, it can be used again to collect the blade surface temperature after the de-icing operation is completed to verify the de-icing effect.

[0054] In one specific embodiment, the icing flexible monitoring unit is a flexible sensor that can be directly attached to the blade surface (such as the FB-SA03 flexible icing monitoring sensor manufactured by Warwick Corporation). Multiple icing flexible monitoring units are deployed at different chord length positions on a single blade, for example, along the blade span, on multiple preset cross sections from the blade root to the blade tip, to obtain icing thickness information in different regions of the blade.

[0055] The operating principle of an icing flexibility monitoring unit can be based on changes in capacitance, strain, or resonant frequency. For example, in a capacitance-based embodiment, the sensor includes staggered finger electrodes. As ice accumulates on its surface, it changes the dielectric constant between the electrodes, resulting in a change in capacitance. The sensor's internal measurement circuitry then converts the detected capacitance change into precise icing thickness data according to a pre-calibrated relational model.

[0056] The ice thickness data, as a quantitative and direct parameter of the ice accretion status, is transmitted to the intelligent de-icing control module 300 via wired or wireless means.

[0057] The ice thickness data has two main uses in the intelligent de-icing control module 300: First, it serves as a direct trigger condition for initiating the de-icing task; that is, when the thickness measured by any monitoring unit exceeds the preset trigger threshold, the system immediately starts the de-icing process. Second, it serves as the core basis for calculating heating parameters; its specific value is directly used to determine the heating power coefficient in the segmented heating power calculation formula. and the heating time coefficient in the formula for calculating the interval heating time .

[0058] In one specific embodiment, the environmental unit comprises a set of sensors for measuring key meteorological parameters. This set of sensors may include: a temperature sensor for measuring ambient temperature, a humidity sensor for measuring ambient relative humidity, an anemometer for measuring wind speed, and a wind vane for measuring wind direction.

[0059] The sensor array of the environmental unit can be installed outside the nacelle of the wind turbine or on a dedicated weather tower within the wind farm to ensure that the data it collects accurately reflects the real environmental conditions at the turbine's location.

[0060] Each sensor in the environmental unit converts the physical quantities it measures in real time into electrical signals. After being processed by the data acquisition module, the environmental data, including parameters such as ambient temperature, ambient humidity, wind speed, and wind direction, is transmitted to the intelligent de-icing control module 300 via wired or wireless communication.

[0061] After receiving environmental data, the intelligent de-icing control module 300 performs a preliminary assessment of the risk of icing based on a preset environmental condition judgment logic. This logic is as follows: When the received ambient temperature data is lower than or equal to a preset temperature threshold (e.g., 0°C) and the ambient humidity data is higher than or equal to a preset humidity threshold (e.g., 85%), the intelligent de-icing control module 300 puts the system into an "early warning" state. In this state, the system can increase the frequency of data collection from other sensors to prepare for possible icing events and subsequent de-icing decisions.

[0062] In one specific embodiment, the wind turbine operation data monitoring unit is a data interface module whose function is to establish a communication connection with the wind turbine's own monitoring and data acquisition (SCADA) system.

[0063] The data interface module connects to the wind turbine's internal network via a standard industrial communication protocol. Once the connection is established, the wind turbine's operating data monitoring unit is configured to periodically request or subscribe to specific data tags from the SCADA system. These data tags include at least: real-time wind speed measured by the nacelle anemometer, real-time active power output by the generator, and real-time rotor speed.

[0064] The wind turbine operation data monitoring unit will format the acquired turbine operation data and transmit it to the intelligent de-icing control module 300.

[0065] The intelligent de-icing control module 300 uses the received real-time wind speed data as the input parameter in the theoretical power linear interpolation formula. The received real-time active power data is used as the input parameter in the power deviation rate calculation formula. In this way, the actual operating performance of the wind turbine is quantified and used to calculate the wind speed-theoretical power deviation rate. This deviation rate serves as a key performance indicator, used to indirectly diagnose aerodynamic performance degradation caused by blade icing.

[0066] In one specific embodiment, the image recognition unit consists of one or more high-definition visible light cameras and associated image processing software.

[0067] A visible light camera is deployed in a location that can clearly capture the surface of the blade. For example, the camera can be mounted on a tower or on a ground-based pan-tilt unit along with the infrared radiation module 100. The camera is configured to capture visible light images of specific areas of the blade upon receiving instructions from the intelligent de-icing control module 300.

[0068] The visible light image data is transmitted to the intelligent de-icing control module 300. The image processing software within the intelligent de-icing control module 300 analyzes the received visible light image. This analysis process may include the following steps: First, the blade outline is identified using an edge detection algorithm (such as the Canny operator) or an image segmentation algorithm (such as the U-Net model); second, within the blade outline area, areas with significant visual differences from the blade substrate surface are identified based on differences in image features such as color, texture, or gloss. These areas are potential icing areas.

[0069] By analyzing these areas, qualitative information about icing can be obtained, such as the type of icing (e.g., rime or hoarfrost) and its approximate distribution.

[0070] In one embodiment of the present invention, the image recognition unit is configured as an auxiliary verification method. Specifically, when other sensors (such as an icing flexibility monitoring unit or diagnostic logic based on power deviation rate) indicate the possible presence of icing, the intelligent de-icing control module 300 activates the image recognition unit to capture images for intuitive confirmation of the icing situation. This confirmation result is not a direct or necessary condition for initiating the de-icing task, but rather serves as a reference among multiple sources of information to increase the robustness of the system's decision-making. Similarly, after the de-icing operation is completed, the images captured by this unit can also be used to intuitively confirm whether the ice layer has been removed.

[0071] The intelligent de-icing control module 300 is the central data processing and control unit of this invention.

[0072] In one specific embodiment, the hardware of the intelligent de-icing control module 300 can be a programmable logic controller (PLC) or an industrial personal computer (IPC). The hardware includes a central processing unit, a memory for storing programs and data, and multiple communication interfaces. The memory contains a computer program that implements the intelligent control method of this invention.

[0073] Multiple communication interfaces are used to establish data links between the intelligent de-icing control module 300 and other system modules and external devices. Specifically, these communication interfaces may include: One or more input interfaces for connecting the various units in the multimodal state sensing module 200. For example, the input interface may include: I / O terminals for receiving analog or digital signals from the environmental unit and the icing flexibility monitoring unit; and an Ethernet interface for receiving structured data packets from the infrared thermal imaging unit, the image recognition unit, and the wind turbine operation data monitoring unit.

[0074] One or more output interfaces for connecting the infrared radiation module 100. For example, the output interface may include: an Ethernet interface for sending control commands containing target angles or coordinates to the motion controller of the pan-tilt unit mounted on the infrared radiation module 100, and for sending control commands containing power percentages or specific power values ​​to the power controller of the infrared radiation module 100.

[0075] A communication interface for connecting to the wind turbine main control system. This interface enables bidirectional communication with the wind turbine main control system via an industrial fieldbus protocol (e.g., Profinet, EtherCAT, or Modbus TCP). Through this interface, the intelligent de-icing control module 300 can send commands to the wind turbine main control system and receive status feedback information from the wind turbine main control system.

[0076] The intelligent de-icing control module 300 is positioned as follows: by executing the computer program in the memory through its internal central processing unit, it can receive and analyze the data collected by the multimodal state perception module 200, complete the diagnosis of the icing state and the decision of the de-icing task, and finally generate accurate control commands containing timing and parameters, which are sent to the infrared radiation module 100 and the wind turbine main control system respectively to drive and coordinate the entire de-icing operation process.

[0077] The intelligent de-icing control module 300 stores a set of algorithm models that can be executed by its processor. These models collectively constitute the core logic for icing diagnosis and precise heating control. Specifically, the algorithm models include: A theoretical power calculation model is used based on the current real-time wind speed obtained from the multimodal state sensing module 200. Calculate the theoretical output power of the wind turbine. The model is specifically implemented using a linear interpolation formula for theoretical power: ; In the formula: To provide real-time wind speed Theoretical output power at the following level; Standard wind speed point The corresponding theoretical power; Standard wind speed point The corresponding theoretical power; For power curves less than And closest Standard wind speed point; For power curve greater than And closest The standard wind speed point.

[0078] A power deviation calculation model is used to calculate the theoretical output power obtained from the aforementioned model. The actual output power obtained from the multimodal state sensing module 200 By comparing the results, the power deviation rate can be calculated. This deviation rate is used as one of the criteria for judging whether icing causes performance degradation. The model is specifically implemented through the power deviation rate calculation formula: ; In the formula: The theoretical power deviation rate is the wind speed. Current wind speed Theoretical output power at the following level; Current wind speed The actual output power of the wind turbine unit.

[0079] A blade width calculation model is used to calculate the blade span at any target position during de-icing operations. Leaf width at this location This width value will serve as an input parameter for subsequent heating power and area control. The model is specifically implemented using a blade width interpolation calculation formula: ; In the formula: The blade width at the target location to be calculated; This is the distance from the target location to the blade tip; and Two known measurement points on the blade and Length from the leaf tip; and To measure at the respective points and The known blade width at the location.

[0080] Specifically, wind turbine blades have a complex three-dimensional airfoil structure, and their width is not uniform along the span (from the blade root to the blade tip). From the blade tip to the maximum chord length, the blade width increases approximately linearly; from the maximum chord length to the blade root, the width decreases approximately linearly.

[0081] To precisely focus infrared radiation energy onto the blade surface and avoid energy waste caused by radiation into open areas, the infrared radiation module 100 of this invention is designed to dynamically adjust the size of its output light spot. Its core control objective is to match the heated area irradiated onto the blade surface with the actual width of the blade at that location. In a specific embodiment, this area matching is achieved by the intelligent de-icing control module 300 controlling the projection angle of each heating unit within the infrared radiation module 100.

[0082] Therefore, a crucial prerequisite before heating any target location on the blade is to accurately calculate the blade width at that location. This calculation is based on a set of pre-measured baseline data. Specifically, several (e.g., five) key sections are pre-selected along the blade's span, and icing flexibility monitoring units are placed at these locations. The distance from these key sections to the blade tip (e.g., ) and the corresponding blade width ( The known parameters are stored in the intelligent de-icing control module 300.

[0083] A heating power calculation model is used to determine the actual heating power to be applied to a specific area of ​​the blade. The model determines the heating power coefficient based on icing thickness data obtained from the icing flexibility monitoring unit. And combined with the blade width These parameters are calculated using the segmented heating power calculation formula: when hour, ; when hour, ; In the formula: For width of The actual heating power applied to the blade area; The heating power coefficient is determined by the icing level; The width of the blade in the currently irradiated area; This is the maximum chord length of the blade; This is the rated total power of the infrared radiation module 100.

[0084] Specifically, after determining the area that needs to be heated, the intelligent de-icing control module 300 obtains the corresponding ice thickness data from the ice accretion flexibility monitoring unit and makes a judgment based on the preset ice accretion level standard. For example, ice thickness of 0-5mm (inclusive) can be defined as ice level four, 5-10mm (inclusive) as ice level three, 10-20mm (inclusive) as ice level two, and above 20mm as ice level one.

[0085] Based on the determined icing level, the module looks up the corresponding heating power coefficient from a preset mapping table. For example, the heating power coefficient k corresponding to the first, second, third, and fourth stages of icing can be 1, 0.9, 0.8, and 0.7, respectively.

[0086] A heating time calculation model is used to determine the irradiation duration for a specific area of ​​the leaf. The model also determines the heating time coefficient based on ice thickness data. And combined with the maximum foundation heating time per unit location The calculation is performed using the formula for calculating the interval heating time: ; In the formula: This refers to the actual heating time of the blade section to be heated; Heating time coefficient corresponding to the icing levels at both ends of an interval The larger value in; The maximum base heating time per unit location is set.

[0087] Specifically, to achieve precise control of heating time, multiple flexible icing monitoring units deployed along the blade's span physically divide the entire blade's de-icing area into several continuous intervals. During de-icing operations, the intelligent de-icing control module independently calculates and controls the heating time for each interval. For a blade interval defined by two adjacent monitoring units (e.g., monitoring units two and three), the overall heating time is determined based on the combined icing data measured by these two monitoring units.

[0088] This invention also provides an intelligent control method for infrared radiation de-icing of wind turbine blades, comprising the following steps: Step S1: Status Monitoring and Icing Warning. In step S1, the intelligent de-icing control module 300 periodically executes a status monitoring program. This program continuously acquires multi-source status data from the multimodal status sensing module 200 through a communication interface.

[0089] Specifically, step S1 includes: receiving ambient temperature and ambient humidity data from the environmental unit; when the ambient temperature is detected to be lower than or equal to a preset temperature threshold and the ambient humidity is higher than or equal to a preset humidity threshold, the intelligent de-icing control module 300 switches the system's working mode to an early warning state and can increase the frequency of collecting other data.

[0090] Simultaneously, step S1 also includes receiving real-time wind speed and actual output power from the wind turbine operation data monitoring unit. The intelligent de-icing control module 300 uses the theoretical power linear interpolation formula to calculate the theoretical output power at the current wind speed, and further uses the power deviation rate calculation formula to calculate the wind speed-theoretical power deviation rate. This deviation rate is continuously monitored as a parameter characterizing the health of the unit's operating performance.

[0091] Step S2: De-icing Decision and Task Initiation. In step S2, the intelligent de-icing control module 300 determines whether to initiate the de-icing operation based on a multi-condition decision logic.

[0092] In one specific embodiment, the decision logic is configured to: trigger a de-icing decision when any ice thickness data received from the icing flexibility monitoring unit exceeds a preset thickness activation threshold; or, trigger a de-icing decision when the wind speed theoretical power deviation rate calculated in step S1 continuously exceeds a preset deviation rate threshold, and when there is a clear low-temperature area in the thermal map data obtained from the infrared thermal imaging unit.

[0093] Specifically, the deviation rate threshold and thickness start-up threshold are preset based on the historical operating data and / or experimental test data of the wind turbine.

[0094] Once the de-icing decision is triggered, the intelligent de-icing control module 300 sends a control message containing a shutdown command through its communication interface with the wind turbine main control system. After receiving confirmation from the wind turbine main control system that the system has been safely shut down and locked, the intelligent de-icing control module 300 continues to send a blade rotation command containing a target angle value to drive the blade to be de-iced to rotate to a preset initial working position that is convenient for the ground infrared radiation module 100 to irradiate.

[0095] Step S3: Execution of Precise Zoned De-icing Operation. After receiving confirmation that the blades have reached their initial working positions, the intelligent de-icing control module 300 begins executing a scanning de-icing operation program. This program controls the infrared radiation module 100 to perform zoned, parameter-controlled heating of the stationary blade surface. This step can be broken down as follows: Step S3.1: Dynamic control of the heating area. The intelligent de-icing control module 300 drives the pan-tilt unit of the infrared radiation module 100 to aim at the first area to be heated on the blade according to a preset scanning path (e.g., from the leaf root to the leaf tip). At the same time, the module uses the blade width interpolation formula to calculate the blade width of the current target area, and adjusts the output spot size of the infrared radiation module 100 according to the width value to match the heating area with the blade width.

[0096] Step S3.2: Gradual adjustment of heating power. After determining the heating area, the intelligent de-icing control module 300 obtains the ice thickness data corresponding to that area from the ice accretion flexibility monitoring unit, and looks up the corresponding heating power coefficient from a preset mapping table based on the thickness data. Subsequently, the module uses a segmented heating power calculation formula, combined with the blade width of the current area, the maximum chord length of the blade, and the rated total power of the infrared radiation module, to calculate the actual heating power to be applied to that area. This power value is then sent as a command to the power controller of the infrared radiation module 100.

[0097] Step S3.3: Precise calculation of heating time. Simultaneously, the intelligent de-icing control module 300 also retrieves the corresponding heating time coefficient from a preset mapping table based on the ice thickness data for the area. The intelligent de-icing control module 300 divides the entire blade into multiple sections based on the installation location of the icing flexible monitoring unit. For a blade section defined by two monitoring units, the corresponding values ​​of the two monitoring units will be selected. The larger value in the coefficient is used to ensure sufficient heating of the area. Subsequently, the intelligent de-icing control module 300 uses the area heating time calculation formula, combined with this heating time coefficient... and the set maximum base heating time per unit location (For example, 1 minute) The required heating time for the current area is calculated. Based on this time, the intelligent de-icing control module 300 controls the irradiation time of the infrared radiation module 100 on the area.

[0098] After heating one area, the system repeats steps S3.1 to S3.3 to move to the next area to be heated until the entire blade is scanned.

[0099] Step S4: Operation Closure and System Reset. After completing the de-icing operation of a single blade, the intelligent de-icing control module 300 sends a command to the wind turbine main control system to rotate the next blade that needs de-icing to the initial working position and repeat step S3.

[0100] Once all blades requiring de-icing have been processed, the intelligent de-icing control module 300 sends a task completion and resumption of operation command to the wind turbine main control system. Upon receiving this command, the wind turbine main control system unlocks and resumes normal operation.

[0101] At the same time, the intelligent de-icing control module 300 switches its operating status back to the status monitoring mode of step S1, waiting for the next icing event to occur, thus forming a complete automated operation closed loop.

[0102] To verify the practical effect of the wind turbine blade infrared radiation de-icing system and its intelligent control method according to an embodiment of the present invention, a verification implementation case is provided: Two adjacent 5.5MW wind turbine generators were selected at the Wugongshan Wind Farm in Jiangxi Province for comparative testing. Unit 16 was equipped with the system described in this invention (experimental group), while Unit 17 maintained its original de-icing system based on hot air circulation within the nacelle (control group). The test was conducted on January 15, 2025, under the following environmental conditions: ambient temperature -8℃ to -2℃, relative humidity 95%, average wind speed 8m / s, and continuous drizzle and snow.

[0103] Experimental verification process and data recording: At 08:00 on January 15, 2025, environmental conditions reached the icing threshold, and the SCADA systems of both units showed that the theoretical power generation power and the actual power generation power began to deviate.

[0104] Test verification process and data recording for Unit 17 (control group): At 09:15 on January 15, 2025, the power loss reached 20%, and the unit's main control system activated the nacelle hot air de-icing system according to the preset logic. The hot air boiler started working, blowing heated air into the blade cavities.

[0105] From 09:15 to 12:30 on January 15, 2025, the unit operated continuously at reduced power, with power loss fluctuating between 15% and 25%. Due to the long hot air transfer path, low thermal efficiency, and inability to specifically heat the most severely iced blade tips, de-icing was ineffective. Vibration data showed that the unit experienced continuous unbalanced vibration.

[0106] At 12:30 on January 15, 2025, the ambient temperature rose slightly to -2℃, the external ice layer began to fall off naturally, and the power gradually recovered.

[0107] The total effective de-icing time for Unit 16 is approximately 3 hours and 15 minutes, during which the power generation loss is approximately 5400 kWh.

[0108] Test verification process and data recording for Unit 17 (experimental group, applying this invention): At 08:45 on January 15, 2025, the intelligent control unit of the system of this invention triggered the "icing warning state" based on the performance analysis criteria (power loss reached 12%).

[0109] At 08:47 on January 15, 2025, the controller automatically activated the infrared thermal imager to perform a scan. Image analysis results showed that there were obvious low temperature zones (-6℃ to -4℃) in the tip area and middle leading edge area of ​​leaves A and C, while leaf B had less icing.

[0110] At 08:49 on January 15, 2025, the system confirmed the icing event and immediately activated the personalized de-icing strategy: (1) Blade A & Blade C: The tip and middle heating plates are activated, and the initial power is set to 85% of the rated power. (2) Blade B: Only the middle heating plate is activated, and the initial power is set to 50% of the rated power.

[0111] From 08:49 to 09:40 on January 15, 2025, the system entered the closed-loop de-icing execution phase. Real-time infrared thermal imaging showed a rapid temperature rise in the low-temperature zones of blades A and C. Around 09:25, the temperature in the blade tip area generally rose above 0℃, reducing the adhesion between the ice layer and the skin, causing it to begin to detach in sheets under centrifugal force. The temperature sensors in each zone provided stable feedback data, with no overheating observed.

[0112] At 09:40 on January 15, 2025, the infrared thermal imager showed that the temperature field of the three blades was uniform and all were above 0℃; the unit's power output recovered to 98% of the theoretical value. The system determined that de-icing was complete, automatically stopped all heating elements, and exited the de-icing mode.

[0113] The total effective de-icing time for Unit 17 was approximately 51 minutes, during which approximately 1250 kWh of power generation was lost.

[0114] The experimental results are compared in the table below: Table 1. Comparison of Verification Results Based on the conclusions in Table 1, this verification case fully demonstrates that the wind turbine blade infrared radiation de-icing system and its intelligent control method of this invention have the following advantages: 1. Excellent effectiveness: It can quickly and thoroughly remove ice from leaves.

[0115] 2. Extremely high economic efficiency: It significantly reduces power generation losses while consuming less energy itself.

[0116] 3. Advanced intelligence: Through multi-sensor fusion and image recognition, it achieves accurate diagnosis and zoned control of icing, avoiding unnecessary energy consumption.

[0117] 4. Significant safety: The unit's aerodynamic balance was quickly restored, ensuring safe and stable operation.

[0118] In summary, this invention provides an infrared radiation de-icing system for wind turbine blades and its intelligent control method. This solution systematically integrates a non-contact infrared radiation module 100, a multi-dimensional, multimodal state sensing module 200, and an intelligent de-icing control module 300 as the control center, constructing a fully automated closed-loop operation. This invention uses the multimodal state sensing module to acquire real-time and comprehensive information from multiple sources, including ice thickness, distribution, environmental conditions, and turbine performance degradation. Based on this information, the intelligent de-icing control module performs fusion analysis and intelligent decision-making. After confirming the necessity of de-icing, it coordinates the wind turbine to enter a standby state. Subsequently, by executing the embedded core algorithm model, it drives the infrared radiation module to perform zoned and graded heating operations on different areas of the blade, achieving dynamic and precise control of the heating area, heating power, and heating time. This effectively solves the problems of high energy consumption, low efficiency, uneven heating, and the lack of adaptive control in existing infrared applications of traditional de-icing methods. It realizes a transformation from passive response to active early warning, and from extensive heating to precise energy application, ultimately achieving a highly efficient, energy-saving, safe, and fully automated de-icing effect. Compared with traditional de-icing methods, this invention has significant advancements and competitive advantages in terms of technology, economy, and safety.

Claims

1. A wind turbine blade infrared radiation de-icing system, characterized in that, include: A multimodal state sensing module is configured to collect multi-source state data related to the icing state of the blades and output the multi-source state data. An intelligent de-icing control module is configured to be electrically connected to the multimodal state sensing module, for receiving multi-source state data output by the multimodal state sensing module, analyzing the multi-source state data to perform icing diagnosis and de-icing decision, and generating and outputting control commands based on the decision results of the icing diagnosis and de-icing decision. An infrared radiation module is configured to be electrically connected to the intelligent de-icing control module for receiving control commands from the intelligent de-icing control module and performing non-contact heating de-icing operations on the surface of the wind turbine blades according to the control commands.

2. The wind turbine blade infrared radiation de-icing system according to claim 1, characterized in that, The multimodal state perception module includes: at least one icing flexibility monitoring unit, one infrared thermal imaging unit, one unit operation data monitoring unit, and one environmental unit; The multi-source status data includes: ice thickness data collected by the icing flexibility monitoring unit, temperature field data collected by the infrared thermal imaging unit, unit operation data collected by the unit operation data monitoring unit, and environmental data collected by the environmental unit.

3. The wind turbine blade infrared radiation de-icing system according to claim 2, characterized in that, The specific process of the intelligent de-icing control module for icing diagnosis and de-icing decision-making includes: The actual output power and real-time wind speed are extracted from the unit operation data from the multi-source status data. Based on the real-time wind speed, the theoretical output power is calculated using the theoretical power linear interpolation formula. The power deviation rate between the theoretical output power and the actual output power is calculated using the power deviation rate calculation formula. The decision to initiate de-icing is triggered when any of the following conditions are met: The power deviation rate exceeds a preset deviation rate threshold, and the temperature field data acquired by the infrared thermal imaging unit indicates the presence of a low-temperature region. Alternatively, the icing thickness data acquired by the multimodal state sensing module exceeds a preset thickness threshold. The deviation rate threshold and thickness start-up threshold are preset based on historical unit operation data and / or experimental test data of the wind turbine.

4. The wind turbine blade infrared radiation de-icing system according to claim 1, characterized in that, The intelligent de-icing control module is also configured to control the heating area, and the specific process includes: Based on the target location of the blade to be heated, the blade width at the target location is calculated using the blade width interpolation formula. Based on the calculated blade width, a heating area control command is generated and sent to control the heating area of ​​the infrared radiation module to match the calculated blade width.

5. The wind turbine blade infrared radiation de-icing system according to claim 4, characterized in that, The intelligent de-icing control module is also configured to control heating power, and the specific process includes: Based on the icing thickness data obtained by the multimodal state sensing module, the icing level is determined and a heating power coefficient is matched. Based on the calculated blade width and the heating power coefficient, the actual heating power is calculated using a segmented heating power calculation formula; A heating power control command is generated and sent based on the calculated actual heating power to control the infrared radiation module to heat according to the calculated actual heating power.

6. The wind turbine blade infrared radiation de-icing system according to claim 5, characterized in that, The intelligent de-icing control module is also configured to control the heating time, and the specific process includes: For a blade section defined by the two icing flexibility monitoring units, the icing level at both ends of the blade section is obtained, and a larger value of a heating time coefficient is matched. Based on the larger value of the heating time coefficients, the actual heating time of the blade interval is calculated using the interval heating time calculation formula. The infrared radiation module is controlled to perform scanning irradiation heating on the blade area, and the total heating time is equal to the calculated actual heating time.

7. The wind turbine blade infrared radiation de-icing system according to claim 1, characterized in that, The infrared radiation module is installed at the bottom of the wind turbine tower and mounted on a gimbal. The straight-line distance between the installation position of the infrared radiation module and the center of the tower simultaneously satisfies the geometric constraints defined by the minimum installation distance formula and the maximum installation distance formula.

8. The wind turbine blade infrared radiation de-icing system according to claim 6, characterized in that, Multiple icing flexibility monitoring units are installed along the blade length direction. The icing flexibility monitoring units are used to directly measure the icing thickness data at their location and output the icing thickness data to the intelligent de-icing control module as a direct basis for determining the heating power coefficient and heating time coefficient.

9. The wind turbine blade infrared radiation de-icing system according to claim 2, characterized in that, The infrared thermal imaging unit is configured to monitor the temperature field data on the blade surface in real time during the de-icing operation and feed the temperature field data back to the intelligent de-icing control module to achieve closed-loop temperature control of the heating process, prevent the blade from overheating, and verify the de-icing effect.

10. A method for intelligent control of infrared radiation de-icing of wind turbine blades, characterized in that, An infrared radiation de-icing system for wind turbine blades according to any one of claims 1-9 includes the following steps: Step S1: Acquire multi-source state data, including environmental data, unit operation data, icing thickness data, and temperature field data, through the multi-modal state perception module; when the environmental data meets the icing conditions, switch to the early warning state, and calculate the power deviation rate based on the real-time wind speed and actual output power in the unit operation data to continuously monitor the unit performance. Step S2: When the ice thickness data exceeds the thickness start threshold, or the power deviation rate exceeds the deviation rate threshold and the temperature field data confirms a low temperature region, a de-icing decision is triggered; and based on the de-icing decision, the wind turbine is instructed to safely shut down and one blade to be de-iced is rotated to the working position. Step S3: Control the infrared radiation module to perform precise de-icing of the blades in the target area to be de-iced. This step coordinates the following controls: dynamically adjust the heating area according to the blade width of the target area, adjust the heating power according to the ice thickness of the target area, and accurately calculate the heating time according to the ice thickness of the target area. Step S4: After de-icing a single blade, switch to the next blade to be de-iced and repeat step S3. After all blades have been processed, instruct the fan to resume operation and return to status monitoring mode.

Images