Wind turbine blade partition ice melting system and method based on multi-mode perception

Abstract

The invention belongs to the technical field of wind power generation equipment safety, and particularly relates to a wind turbine blade partition ice melting system and method based on multi-mode sensing. The system comprises a multi-mode sensing module, an intelligent control module and an execution and energy saving module. The multi-mode sensing module comprises a plurality of infrared array sensors which are arranged along the front edge of the blade in an equidistant array manner, and a vibration accelerometer which is mounted at the root of the blade and is used for monitoring the surface temperature distribution and vibration characteristics of the blade in real time; and the intelligent control module comprises an edge calculation unit integrated with an AI algorithm based on a long short-term memory (LSTM) neural network, and a partition control strategy, and is used for realizing real-time prediction and partition control of ice layer thickness distribution according to the data acquired by the multi-modal sensing module. The method can effectively solve the problems of high energy consumption and high misjudgment rate of a traditional ice melting technology, has the advantages of high detection precision, high response speed, long service life and the like, and is suitable for anti-icing protection of various large wind generating sets in a high-humidity and low-temperature environment.

Description

Technical Field

[0001] The present invention belongs to the technical field of wind power generation equipment safety, and in particular relates to a wind turbine blade zoned de-icing system and method based on multimodal sensing. Background Art

[0002] As the global energy mix accelerates toward cleaner, lower-carbon energy, wind energy, with its renewable and pollution-free nature, has become a crucial energy pillar for achieving the "dual carbon" goals. According to statistics, global wind power capacity will reach 140 GW by 2023, bringing the cumulative installed capacity to over 1,000 GW. Horizontal-axis wind turbines, owing to their high aerodynamic efficiency and mature technology, hold over 98% of the market share and are widely used in wind-rich areas such as coastal areas and high altitudes. Problems with existing technologies: However, in low-temperature, high-humidity environments (such as in Northeast and Northwest my country during winter, where temperatures often drop below -10°C and humidity exceeds 85%), wind turbine blades are highly susceptible to icing. Ice accumulation can distort the blade's aerodynamic shape, reducing the turbine's aerodynamic efficiency by 30%-50% and resulting in an average annual power generation loss of 15%-25%. Furthermore, the added mass of ice (approximately 10 kg per square meter of 1 cm thick ice) exacerbates blade fatigue loads. According to industry statistics, blade breakage caused by icing accounts for 42% of wind turbine structural failures, posing a serious threat to the stable operation of the power grid. Currently, wind turbine icing detection technology has significant flaws: traditional detection devices rely on temperature changes in the natural environment to simulate icing conditions, making it impossible to reproduce complex icing forms such as rime and rime. Laboratory simulations often use single-variable control, making it difficult to reproduce the icing process under the coupled effects of temperature, humidity, and wind speed in actual operation. This detection limitation makes it impossible to accurately grasp the power characteristics of the wind turbine under different icing conditions, making it difficult for the unit control strategy to adapt to complex operating conditions, resulting in energy waste and equipment safety hazards. The shortcomings of existing technologies urgently require innovative detection methods to achieve accurate identification and efficient response to icing conditions. Summary of the Invention

[0003] The purpose of the present invention is to provide a wind turbine blade zoned ice melting system and method based on multimodal sensing, which can effectively solve the problems of high energy consumption and high misjudgment rate of traditional ice melting technology, and has the advantages of high detection accuracy, fast response speed, and long service life. It is suitable for anti-icing protection of various large wind turbine generator sets in high humidity and low temperature environments.

[0004] The technical solutions adopted by the present invention are as follows: A wind turbine blade zoned de-icing system based on multimodal sensing, comprising a multimodal sensing module, an intelligent control module, and an execution and energy-saving module; The multimodal sensing module includes a plurality of infrared array sensors arranged in an equidistant array along the leading edge of the blade, and a vibration accelerometer installed at the root of the blade, which is used to monitor the temperature distribution and vibration characteristics of the blade surface in real time; The intelligent control module includes an edge computing unit that integrates an AI algorithm based on a long short-term memory neural network (LSTM), and a partition control strategy for achieving real-time prediction of ice thickness distribution and partition control based on data collected by the multimodal perception module; The execution and energy-saving module includes a shape memory alloy fixing part for fixing the carbon fiber heating belt to the surface of the blade, and an energy management system with an integrated supercapacitor energy storage device for executing zoned ice melting operations and storing energy through valley electricity.

[0005] The interval between two adjacent infrared array sensors is 10 cm.

[0006] It is characterized in that: the edge computing unit takes infrared temperature field data and vibration signals as input, and predicts the ice thickness distribution through the LSTM model.

[0007] The zoning control strategy is to automatically trigger the carbon fiber heating belt in the corresponding area when the system detects that the ice thickness is ≥2mm.

[0008] The carbon fiber heating belt has a power density of 500W / m², a response time of less than 30s, and adopts a gradient heating mode with decreasing power distribution from the center to the edge. The temperature of the central heating area is set at 5°C and the edge area is set at 3°C.

[0009] The shape memory alloy wire has a phase transition temperature of 5° C., and when the temperature exceeds 5° C., it automatically stretches and releases the mechanical constraint of the heating belt.

[0010] The energy management system is configured to utilize the low electricity price period of the night power grid to store energy, and use the stored energy to supply power during the daytime ice melting operation.

[0011] The infrared array sensors are arranged along the leading edge of the blade to form a high-density temperature monitoring network for acquiring blade surface temperature distribution data in real time. The vibration accelerometer identifies abnormal aerodynamic performance of the blade by detecting changes in vibration characteristics caused by the added mass of the ice layer.

[0012] In the execution and energy-saving module, the carbon fiber heating belt is dynamically connected to the blade through a shape memory alloy fixing piece.

[0013] A wind turbine blade zoned de-icing method based on multimodal sensing includes the following steps: S1, start detection. After the system is initialized, it determines whether the ambient temperature is lower than 0℃ and the humidity is greater than 85%. If the conditions are not met, it enters the dormant state. If the conditions are met, it starts multi-sensor data collection. S2, data processing, performs Kalman filtering and denoising on the data collected by the infrared sensor and vibration sensor, and extracts characteristic parameters of the temperature anomaly area ΔT < -2°C and the vibration frequency offset Δf > 5Hz; S3, ice layer prediction, inputs the processed data into an AI model based on the long short-term memory neural network (LSTM) to predict the ice layer thickness distribution. When the predicted thickness is ≥ 2mm, the zone heating instruction is triggered; otherwise, the process returns to the data collection stage. S4, heating control, triggers the carbon fiber heating belt in the corresponding area to start gradient heating, monitors data in real time, and turns off the heating belt if the ice thickness is less than 1mm. Otherwise, the heating power is dynamically adjusted, and data collection and ice layer prediction are executed cyclically.

[0014] The technical effects achieved by the present invention are: The present invention integrates an infrared array sensor and a vibration accelerometer to monitor the temperature distribution and vibration characteristics of the blade surface in real time, and uses the LSTM neural network algorithm to accurately predict the thickness of the ice layer.

[0015] The present invention adopts a zoned heating strategy to activate the carbon fiber heating belt only in the icing area, and realizes the dynamic connection of the heating belt through the shape memory alloy fixing piece.

[0016] The present invention can effectively solve the problems of high energy consumption and high misjudgment rate of traditional ice melting technology. It has the advantages of high detection accuracy, fast response speed and long service life. It is suitable for anti-icing protection of various large wind turbines in high humidity and low temperature environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a schematic diagram of the overall system architecture of the present invention.

[0018] Figure 2 This is the AI model training process flow chart of the present invention DETAILED DESCRIPTION In order to make the purpose and advantages of the present invention more clearly understood, the present invention is described in detail below with reference to the following examples. It should be understood that the following text is only used to describe one or more specific embodiments of the present invention and does not strictly limit the scope of protection of the present invention.

[0019] like Figure 1 As shown, a wind turbine blade partitioned ice melting system based on multimodal perception includes a multimodal perception module, an intelligent control module, and an execution and energy-saving module; The multimodal sensing module includes multiple infrared array sensors arranged in an equidistant array along the leading edge of the blade, with a spacing of 10 cm between adjacent infrared array sensors, and a vibration accelerometer installed at the root of the blade for real-time monitoring of the blade surface temperature distribution and vibration characteristics. The infrared array sensors are arranged along the leading edge of the blade to form a high-density temperature monitoring network for real-time acquisition of blade surface temperature distribution data. The infrared array sensor uses the MLX90640 sensor and has a temperature resolution of 0.1°C and a measurement accuracy of ±0.5°C. The vibration accelerometer identifies abnormalities in the blade's aerodynamic performance by detecting changes in vibration characteristics caused by the added mass of the ice layer. The vibration accelerometer uses the ADXL345 sensor, and its monitoring range covers vibration frequency and amplitude changes from 0 to 200 Hz. The intelligent control module includes an edge computing unit that integrates an AI algorithm based on the long short-term memory neural network (LSTM). The edge computing unit takes infrared temperature field data and vibration signals as input, and predicts the ice thickness distribution through the LSTM model, as well as a zoning control strategy, which is used to achieve real-time prediction and zoning control of the ice thickness distribution based on the data collected by the multimodal perception module. The zoning control strategy is that when the system detects that the ice thickness is ≥2mm, it automatically triggers the carbon fiber heating belt in the corresponding area. The power density of the carbon fiber heating belt is 500W / m², and the response time is less than 30s, achieving rapid heating. A gradient heating mode with decreasing power distribution from the center to the edge is adopted. The temperature of the central heating area is set to 5°C, and the edge area is set to 3°C, effectively reducing the damage to the blade material caused by thermal stress caused by the temperature gradient. The execution and energy-saving module includes a shape memory alloy fixing for fixing the carbon fiber heating belt to the surface of the blade. The carbon fiber heating belt is dynamically connected to the blade through a shape memory alloy wire. The phase change temperature of the shape memory alloy fixing is 5°C. When the temperature exceeds 5°C, it automatically stretches to release the mechanical constraint of the heating belt, avoiding material aging problems caused by long-term pressure, and an energy management system with an integrated supercapacitor energy storage device, which is used to perform zoned ice melting operations and store energy through valley electricity; the energy management system is configured to use the low electricity price period of the night power grid to store energy, and use the stored energy to supply power during the daytime ice melting operation, so as to achieve efficient energy utilization.

[0020] According to the above structure, a zoned heating strategy is adopted to carry out targeted heating only on the iced area, which is more energy-efficient than the traditional full-blade heating method; multimodal data fusion technology effectively reduces the misjudgment rate of icing and significantly improves system reliability; the heating belt is dynamically connected through shape memory alloy fixings to avoid damage to the blade coating caused by long-term mechanical stress and increase the service life of key components of the system; the time control of the entire process from detecting the icing state to the completion of ice melting is relatively short, meeting the real-time protection needs of wind turbines.

[0021] like Figure 2As shown, a wind turbine blade zoned de-icing method based on multimodal sensing includes the following steps: S1, start detection. After the system is initialized, it determines whether the ambient temperature is lower than 0℃ and the humidity is greater than 85%. If the conditions are not met, it enters the dormant state. If the conditions are met, it starts multi-sensor data collection. S2, data processing, performs Kalman filtering and denoising on the data collected by the infrared sensor and vibration sensor, and extracts characteristic parameters of the temperature anomaly area ΔT < -2°C and the vibration frequency offset Δf > 5Hz; S3, ice layer prediction, inputs the processed data into an AI model based on the long short-term memory neural network (LSTM) to predict the ice layer thickness distribution. When the predicted thickness is ≥ 2mm, the zone heating instruction is triggered; otherwise, the process returns to the data collection stage. S4, heating control, triggers the carbon fiber heating belt in the corresponding area to start gradient heating, monitors data in real time, and turns off the heating belt if the ice thickness is less than 1mm. Otherwise, the heating power is dynamically adjusted, and data collection and ice layer prediction are executed cyclically.

[0022] Example 1 Experimental conditions: ambient temperature -5°C, humidity 88%, wind speed 10m / s, and ice formation on some areas of the blades.

[0023] System working process: Start detection. After the system is initialized, it determines whether the ambient temperature is below 0°C and the humidity is greater than 85%. If the conditions are not met, it enters the dormant state. If the conditions are met, it starts multi-sensor data collection. The infrared array sensor monitors the temperature distribution on the blade surface in real time and detects a temperature anomaly in the area 30-50 cm from the leading edge of the blade, with a ΔT < -2°C. The vibration accelerometer detects a vibration frequency offset of Δf = 6 Hz, indicating the presence of icing in this area. The edge computing unit inputs temperature and vibration data into an AI model based on a long short-term memory (LSTM) neural network, predicting that the ice thickness in the area is 2.5 mm, triggering a zone heating instruction. The carbon fiber heating belts in the corresponding area are started, using a gradient heating mode with a center temperature of 5°C and an edge temperature of 3°C; the shape memory alloy fasteners extend when the temperature exceeds 5°C, releasing mechanical constraints; the energy management system uses stored energy at night to power the heating belts.

[0024] Ice-melting effect: It takes 12 minutes from ice detection to complete ice melting. The heating belt is turned off when the ice thickness drops to 0.8mm. Compared with traditional full-blade heating, this ice-melting process saves 43.2% energy. There is no sign of coating damage on key components of the system.

[0025] Example 2 Experimental conditions: ambient temperature -10°C, humidity 90%, wind speed 15m / s, and large area of ice on the leading edge of the blade.

[0026] System working process: Start detection. After the system is initialized, it determines whether the ambient temperature is below 0°C and the humidity is greater than 85%. If the conditions are not met, it enters the dormant state. If the conditions are met, it starts multi-sensor data collection. The infrared array sensor showed that the temperature in the 0-80cm area around the leading edge of the blade was generally below -2°C, and the vibration accelerometer detected a vibration frequency offset of Δf = 8Hz, confirming that large areas of ice had formed. An AI model based on a long short-term memory (LSTM) neural network predicted ice thickness of 3mm, triggering heating zones in multiple areas to activate simultaneously, using a gradient heating mode. Multi-zone carbon fiber heating belts operate synchronously, with a power density of 500W / m² and a response time of 25s; shape memory alloy fixtures dynamically adjust constraints; and supercapacitor energy storage devices provide continuous power supply.

[0027] Ice-melting effect: The entire process takes 15 minutes, the ice layer is completely melted, and the heating belt is turned off when the thickness is less than 1mm; after testing, the false positive rate is 4.5%, which is 23.5% lower than that of traditional single sensors, and the life of key components of the system is expected to be extended by 32%.

[0028] Example 3 Experimental conditions: ambient temperature -3°C, humidity 95%, wind speed 20m / s, thin ice appears on the blade surface and continues to condense.

[0029] System working process: Start detection. After the system is initialized, it determines whether the ambient temperature is below 0°C and the humidity is greater than 85%. If the conditions are not met, it enters the dormant state. If the conditions are met, it starts multi-sensor data collection. The infrared sensor monitored the dynamic changes of the abnormal temperature area on the blade surface in real time, and the vibration accelerometer detected frequent deviations of the vibration frequency, Δf>5Hz; The AI model based on the long short-term memory (LSTM) neural network predicts ice thickness fluctuations between 2 and 2.5 mm based on dynamic data, continuously triggering heating zones in corresponding areas and dynamically adjusting heating power. The heating belt adjusts its power based on real-time data to avoid overheating; the shape memory alloy fasteners release constraints in real time as the temperature changes; and the energy management system optimizes energy storage distribution to ensure continuous power supply.

[0030] Ice-melting effect: It takes 13 minutes from ice detection to complete ice melting, effectively preventing the continued condensation of ice. Energy consumption is reduced by 41.8% compared with traditional methods. In high-humidity and high-wind speed environments, the system reliability is verified, and no blade material damage caused by temperature gradients occurs.

[0031] The foregoing is merely a preferred embodiment of the present invention. It should be noted that those skilled in the art may make various improvements and modifications without departing from the principles of the present invention, and such improvements and modifications are also within the scope of protection of the present invention. Structures, devices, and operating methods not specifically described or explained herein shall, unless otherwise specified or limited, be implemented in accordance with conventional means in the art.

Claims

1. A wind turbine blade zoned de-icing system based on multimodal sensing, characterized by ,including multimodal perception module, intelligent control module and execution and energy saving module; The multimodal sensing module includes a plurality of infrared array sensors arranged in an equidistant array along the leading edge of the blade, and a vibration accelerometer installed at the root of the blade, which is used to monitor the temperature distribution and vibration characteristics of the blade surface in real time; The intelligent control module includes an edge computing unit that integrates an AI algorithm based on a long short-term memory neural network (LSTM), and a partition control strategy for achieving real-time prediction of ice thickness distribution and partition control based on data collected by the multimodal perception module; The execution and energy-saving module includes a shape memory alloy fixing part for fixing the carbon fiber heating belt to the surface of the blade, and an energy management system with an integrated supercapacitor energy storage device for executing zoned ice melting operations and storing energy through valley electricity.

2. The wind turbine blade zoned ice melting system based on multimodal sensing according to claim 1 is characterized by: The interval between two adjacent infrared array sensors is 10 cm.

3. The wind turbine blade zoned ice melting system based on multimodal sensing according to claim 1 is characterized by: The edge computing unit takes infrared temperature field data and vibration signals as input and predicts the ice thickness distribution through the LSTM model.

4. The wind turbine blade zoned ice melting system based on multimodal sensing according to claim 1 is characterized by: The zoning control strategy is to automatically trigger the carbon fiber heating belt in the corresponding area when the system detects that the ice thickness is ≥2mm.

5. The wind turbine blade zoned ice melting system based on multimodal sensing according to claim 1 is characterized by: The carbon fiber heating belt has a power density of 500W / m², a response time of less than 30s, and adopts a gradient heating mode with decreasing power distribution from the center to the edge. The temperature of the central heating area is set at 5°C and the edge area is set at 3°C.

6. The wind turbine blade zoned ice melting system based on multimodal sensing according to claim 1 is characterized by: The shape memory alloy wire has a phase transition temperature of 5° C., and when the temperature exceeds 5° C., it automatically stretches and releases the mechanical constraint of the heating belt.

7. The wind turbine blade zoned ice melting system based on multimodal sensing according to claim 1 is characterized by: The energy management system is configured to utilize the low electricity price period of the night power grid to store energy, and use the stored energy to supply power during the daytime ice melting operation.

8. The wind turbine blade zoned ice melting system based on multimodal sensing according to claim 1 is characterized by: The infrared array sensors are arranged along the leading edge of the blade to form a high-density temperature monitoring network for acquiring blade surface temperature distribution data in real time. The vibration accelerometer identifies abnormal aerodynamic performance of the blade by detecting changes in vibration characteristics caused by the added mass of the ice layer.

9. The wind turbine blade zoned ice melting system based on multimodal sensing according to claim 1, characterized in that: In the execution and energy-saving module, the carbon fiber heating belt is dynamically connected to the blade through a shape memory alloy fixing piece.

10. A wind turbine blade zoned de-icing method based on multimodal sensing, based on the wind turbine blade zoned de-icing system based on multimodal sensing according to any one of claims 1 to 9, characterized in that: The following steps are involved: S1, start detection. After the system is initialized, it determines whether the ambient temperature is lower than 0℃ and the humidity is greater than 85%. If the conditions are not met, it enters the dormant state. If the conditions are met, it starts multi-sensor data collection. S2, data processing, performs Kalman filtering and denoising on the data collected by the infrared sensor and vibration sensor, and extracts characteristic parameters of the temperature anomaly area ΔT < -2°C and the vibration frequency offset Δf > 5Hz; S3, ice layer prediction, inputs the processed data into an AI model based on the long short-term memory neural network (LSTM) to predict the ice layer thickness distribution. When the predicted thickness is ≥ 2mm, the zone heating instruction is triggered; otherwise, the process returns to the data collection stage. S4, heating control, triggers the carbon fiber heating belt in the corresponding area to start gradient heating, monitors data in real time, and turns off the heating belt if the ice thickness is less than 1mm. Otherwise, the heating power is dynamically adjusted, and data collection and ice layer prediction are executed cyclically.

Images