Fan blade icing monitoring system and method

Abstract

The invention discloses a fan blade icing monitoring system and method, and relates to the technical field of wind power generation, and the system collects temperature, humidity and vibration signals in real time through a multi-sensor module installed on a fan blade, fuses multi-parameter data through an edge calculation unit, and calculates a blade icing risk factor # imgabs0 #. The risk factors are calculated by adopting a dynamic weighting model, and the icing state of the blade is accurately evaluated in combination with normalized characteristic values # imgabs1 #, # imgabs2 # and # imgabs3 #. And when the risk factor exceeds a set threshold value, the system triggers an alarm and transmits data to the cloud through the wireless communication module. The cloud platform predicts the future icing risk based on a long-short term memory (LSTM) neural network model in combination with historical data and real-time monitoring data, the accuracy and real-time performance of icing monitoring are improved, the method is suitable for operation in cold regions and high-altitude wind fields, and reliable guarantee is provided for safe and efficient operation of fans.

Description

Technical Field

[0001] The present invention relates to the technical field of wind power generation, and in particular to a system and method for monitoring icing of wind turbine blades. Background Art

[0002] As an important part of clean energy, wind power generation has been widely promoted around the world in recent years. However, with the continuous increase in wind power installed capacity, the complexity of the wind turbine operating environment has become increasingly apparent, especially in cold regions and high-altitude areas. The problem of wind turbine blade icing has become an important factor affecting wind power efficiency and safety.

[0003] Impact of icing on fan blades: When fan blades are running in a cold environment, ice is easily formed because supercooled water droplets in the air hit the blade surface and freeze. This phenomenon not only affects the normal operation of the fan, but also may cause the following serious threats to the safety and economy of the fan: Reduced power generation efficiency: Blade icing will change the aerodynamic shape, causing airflow separation and reduced lift, which in turn leads to a significant reduction in power generation efficiency. Studies have shown that in severe icing conditions, the power generation of wind turbines may drop by 20%-50%.

[0004] Increased mechanical loads: Uneven ice formation on the blades can disrupt the dynamic balance of the rotating system, increase vibration amplitude, and cause excessive wear on core components such as the main shaft, gearbox, and bearings.

[0005] Safety hazards: When the blades are frozen, ice may fall off due to centrifugal force during rotation, threatening the safety of nearby equipment and personnel. In addition, the structural strength of the blades may also be damaged due to excessive vibration caused by ice.

[0006] Existing icing monitoring technologies: In response to the problem of wind turbine blade icing, a variety of technical solutions have been proposed to monitor the blade icing status, mainly including the following categories: Meteorological data monitoring method: By monitoring the ambient temperature, humidity, wind speed and other meteorological parameters, combined with empirical formulas, predict whether the blades are frozen. This method is simple to implement and has low cost, but it can only indirectly infer the icing state and lacks accurate reflection of the actual icing condition of the blades.

[0007] Visual monitoring method: Use a camera or infrared imaging device to monitor the state changes on the blade surface. The visual method can clearly capture the icing conditions on the blade surface in an ideal environment, but its monitoring effect will be significantly reduced in extreme weather conditions such as low light and heavy snowfall.

[0008] Vibration signal analysis method: by installing a vibration sensor, the vibration signal changes of the blades are analyzed to determine whether ice has formed. This method is sensitive to dynamic balance damage and can detect a certain degree of uneven icing, but the detection effect on initial mild icing is limited.

[0009] Surface temperature monitoring method: Use a temperature sensor to directly measure the temperature of the blade surface and combine it with the ambient temperature to determine the possibility of icing. This method is easy to implement, but a single temperature parameter cannot reflect the complex icing state and may result in false alarms or missed alarms.

[0010] Deficiencies of existing technologies: Although the above methods can monitor blade icing to a certain extent, there are still many technical limitations, which are mainly manifested in the following aspects: Single parameter limitation: Most current monitoring methods rely on a single parameter (such as temperature or vibration signal) to judge icing, and fail to comprehensively consider the various physical properties of the blade surface, especially the icing state under the influence of multiple variables.

[0011] Inaccurate meteorological monitoring methods: Prediction models based on ambient temperature and humidity are susceptible to fluctuations in meteorological conditions and cannot truly reflect the icing conditions on the blade surface.

[0012] Missed and false alarms: The recognition rate for slight icing or initial icing is low, and the monitoring effect for complex icing (such as slight damage to dynamic balance but no obvious temperature abnormality) is poor.

[0013] Delayed response: Some monitoring solutions rely on manual data analysis or weather forecasts, making it difficult to respond immediately when icing occurs.

[0014] Limited data processing capabilities: Current systems mostly use simple threshold judgments or single parameter analysis, and fail to utilize multi-sensor fusion data to achieve fast calculations and real-time evaluation.

[0015] Complex environments are difficult to cope with: In high-altitude and cold areas, wind farm environments are mostly characterized by strong winds, high humidity and low temperatures. Visual equipment is limited by light, and vibration signals are easily interfered with, resulting in unreliable monitoring results.

[0016] Weak anti-interference ability: The existing system is easily interfered in wind fields with complex electromagnetic environments, especially the wireless communication equipment suffers from severe signal attenuation during long-distance transmission.

[0017] Lack of intelligent thawing judgment: The current thawing control process is usually judged manually, and lacks an automated thawing completion judgment function based on real-time data, which may cause the thawing device to over-operate or stop prematurely, thus affecting operating efficiency and energy utilization.

[0018] In order to make up for the shortcomings of existing technologies, future blade icing monitoring systems should have the following capabilities: Multi-parameter comprehensive monitoring: Combining multiple parameters such as temperature, humidity and vibration signals, the data fusion algorithm is used to fully reflect the actual icing conditions on the blade surface.

[0019] Intelligent prediction: Using big data and machine learning technology, we can predict future icing risks based on historical and real-time data and improve the system's early warning capabilities.

[0020] Real-time and reliability: Introduce efficient edge computing units to achieve rapid data processing and real-time judgment, and improve transmission reliability through highly stable wireless communication technology.

[0021] Automated thawing determination: Develop a thawing completion determination algorithm based on the dynamic changes of risk factors, monitor the operating effect of the thawing device in real time, and avoid energy waste.

[0022] To this end, the present application aims to design and provide a fan blade icing monitoring system and method to solve the above-mentioned problems.

[0023] The purpose of the present invention is to provide a fan blade icing monitoring system and method to address the deficiencies of the prior art and solve the problems raised in the background technology. Summary of the invention

[0024] To achieve the above object, the present invention provides the following technical solutions: The solution of the present invention provides a fan blade icing monitoring system on one hand, the system comprising: The sensor module is installed at the leading edge, middle and trailing edge of the fan blade. A sensor group is set at each position. The sensor group includes: Surface temperature sensor, used to measure blade surface temperature , the unit is Celsius; Capacitive humidity sensor for measuring leaf surface humidity , the unit is relative humidity percentage; Micro vibration accelerometer, used to collect blade vibration signals , in decibels; Data processing and edge computing unit: The data processing and edge computing unit communicates with the sensor module to perform multi-parameter fusion processing on real-time data and determine the blade icing risk factor through the following formula ,Right now: ; in, , , is the weight coefficient, satisfying ; , , are the normalized functions of temperature, humidity and vibration characteristics respectively; when When the system determines that the blades are frozen and triggers an alarm signal; Wireless communication module: uses LoRa or 5G technology to transmit real-time data to the cloud; Cloud computing platform: Based on regional meteorological data and historical data, combined with the long short-term memory neural network (LSTM) model, the icing risk factor within the next 30 minutes is predicted using the following calculation formula: ; in, , which is the current monitoring data vector; Alarm and control module: When blade icing risk factor or predicted value Exceeding the threshold When the blade heating device is triggered or the machine is shut down for protection.

[0025] Furthermore, the vibration signal The power spectrum Based on wavelet transform calculation, the formula is: ; in, is the wavelet transform result of the vibration signal, is the frequency variable, is the time variable, .

[0026] Furthermore, the temperature The normalization function Calculated by the following formula: ; in, and They are the maximum and minimum values ​​of the temperature monitoring range respectively.

[0027] Furthermore, the humidity The normalization function Calculated by the following formula: ; in, and They are the maximum and minimum values ​​of the humidity monitoring range respectively.

[0028] Furthermore, the vibration The normalization function Calculated by the following formula: ; in, and are the maximum and minimum values ​​of the vibration characteristics respectively.

[0029] Furthermore, the weight coefficient , , Determined by the following optimization formula: , ; in, The first risk factors, is the real icing risk factor in historical data, is the sample size, Indicates that during the optimization process, the weight , , The constraint that the sum of weights must be 1 must be satisfied.

[0030] Furthermore, when the cloud computing platform optimizes the LSTM model, it uses the following loss function : ; in, is the icing risk factor predicted by the model, is the true value, To predict future time steps.

[0031] Furthermore, the alarm and control module determines whether thawing is completed based on the following formula: ; in, Icing risk factor recalculated after blade thawing; is the real-time temperature of the leaf surface after thawing, in degrees Celsius; It is the real-time humidity of the leaf surface after thawing, in relative humidity percentage; is the power spectrum amplitude of the vibration signal after the blade surface is thawed, in decibels; Based on temperature Normalization function of ; Based on humidity Normalization function of ; Based on vibration signal Normalization function of ; , , are the weight coefficients of temperature, humidity and vibration signals respectively, and satisfy ; set up is the thawing completion threshold, when When the blades are judged to be thawed, the system stops heating or releases the alarm signal.

[0032] Furthermore, the wireless communication module adopts LoRa technology, specifically a module based on SemtechSX127x series or SX126x series chips, with a data transmission range of not less than 10 kilometers and anti-interference performance.

[0033] Another aspect of the present invention provides a method for monitoring icing of fan blades, which is implemented based on the above-mentioned fan blade icing monitoring system and includes the following steps: The blade surface temperature is collected through the sensor module ,humidity and vibration signal ; The data processing unit integrates the , , , calculate the risk factor ; set up is the icing risk threshold, when The alarm is triggered and uploaded to the cloud; Cloud platform predicts future risk factors ; When the alarm signal is triggered, the control module starts the blade heating device or shuts down for protection.

[0034] Compared with the prior art, the present invention has the following beneficial effects: 1. The technical solution of the present invention adopts a multi-parameter fusion algorithm, taking temperature, humidity and vibration signals as core monitoring indicators, and calculating the risk factor through dynamic weighting. , comprehensively reflecting the icing condition of the blade surface. Compared with the traditional monitoring method that relies on a single parameter, the solution of the present invention can accurately identify slight icing, local icing and icing conditions in complex environments through normalization and weighted optimization, significantly reducing the false alarm and missed alarm rates; 2. The technical solution of the present invention introduces the LSTM model in the cloud platform, combines historical data and real-time monitoring data, and predicts future icing risk factors This intelligent prediction mechanism breaks through the lag of traditional monitoring, can detect potential risks in advance, and provide efficient early warning for wind farm operation; at the same time, it optimizes the weights of multiple parameters through historical data , , , which realizes the dynamic adjustment of risk factors and further improves the accuracy and flexibility of icing assessment; 3. The technical solution of the present invention is to monitor the real-time data after the thawing device is in operation, combined with the risk factor of dynamic calculation , realizing automatic determination of the thawing completion status, avoiding the drawbacks of manual intervention or traditional fixed time control; the system of the present invention can accurately adjust the operating time of the thawing device according to the actual status of the blades, effectively saving energy, while reducing the loss of equipment caused by excessive operation, and significantly improving the system energy efficiency and control accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 A system block diagram of a fan blade icing monitoring system proposed by the present invention; Figure 2 The present invention provides a flow chart of a method for monitoring icing of fan blades. DETAILED DESCRIPTION

[0036] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0037] The following combination Figure 1-Figure 2 With reference to the accompanying drawings and a number of embodiments, specific embodiments of the present invention are described in detail.

[0038] The implementation scheme of the present invention provides a fan blade icing monitoring system and method, which aims to accurately determine the blade icing status and take necessary protective measures by real-time monitoring of the temperature, humidity and vibration signals of the fan blades, combining a multi-parameter fusion algorithm with cloud-based intelligent prediction.

[0039] 1. System overall framework description The system includes the following modules: Sensor module: installed on the surface of the wind turbine blade, using an integrated multi-parameter sensor (including surface temperature sensor; capacitive humidity sensor; micro-vibration accelerometer) to collect blade surface status data in real time. An electric slip ring is provided at the wind turbine shaft, and the sensor is connected to the data processing and edge computing unit through the electric slip ring; Data processing and edge computing unit: Based on the embedded AI chip, it is deployed in the wind turbine nacelle to pre-process the sensor data and identify anomalies.

[0040] Wireless communication module: transmits data to the cloud through LoRa or 5G technology.

[0041] Cloud-based big data platform: integrates meteorological data and combines machine learning algorithms to make icing predictions and alarm decisions.

[0042] Alarm and control module: triggers alarm signal according to calculation results, starts thawing device or shuts down for protection.

[0043] 2. System operation process Data acquisition: The sensor module consists of the following components: Surface temperature sensor: collects temperature data on the blade surface , the unit is degrees Celsius (°C).

[0044] Capacitive humidity sensor: collects humidity data on the leaf surface , the unit is relative humidity percentage (%).

[0045] Micro vibration accelerometer: collects vibration signals of blades , the unit is decibel (dB).

[0046] Vibration signal The power spectrum is calculated by wavelet transform , the formula is as follows: ; in, is the vibration signal at frequency The power spectrum under is the wavelet transform result of the vibration signal.

[0047] Risk factor calculation: The data processing and edge computing unit normalizes the collected data and calculates the icing risk factor of the blade .

[0048] The specific steps are as follows: Temperature normalization: ; in, is the temperature normalized result; and are the maximum and minimum temperatures within the monitoring range, respectively.

[0049] Humidity normalization: ; in, is the humidity normalized result; and are the maximum and minimum values ​​of humidity respectively.

[0050] Vibration Normalization: ;

[0051] in, is the normalized result of vibration characteristics; and are the maximum and minimum values ​​of the vibration characteristics respectively.

[0052] Risk factor calculation: ; in, , , is the multi-parameter weight coefficient, satisfying ; is the icing risk factor of the blade.

[0053] Determine whether it is frozen: If , the system determines that the blades are frozen and triggers an alarm signal.

[0054] Cloud prediction and optimization: The cloud computing platform optimizes weights based on historical data using the following formula: , ; In addition, the LSTM model is used to predict the icing risk factor within the next 30 minutes: ; in, , which is the current monitoring data.

[0055] Thawing completion judgment: After the thawing device is started, the system recalculates the risk factor using the following formula: ; when , the system determines that the thawing is completed and releases the alarm signal.

[0056] 3. Implementation Method Example 1: Detailed calculation process for real-time monitoring of a single wind turbine Background conditions: Wind farm environment: Northern high-altitude cold area, winter temperature -20°C to 5°C, humidity range 10%-95%, vibration characteristic range 0-60dB.

[0057] Current sensor data collection: temperature ;humidity ; Vibration power spectrum .

[0058] Calculation process: Temperature normalization calculation: ; Set temperature range , substitute the data: ; Humidity normalization calculation: ; Humidity range , substitute the data: ; Vibration normalization calculation: ; Set vibration range , substitute the data: ; Risk factor calculation: ; Set weight , , , substitute the normalized result: ; Determine whether there is icing: set icing risk threshold .because , the system determines that the blades are frozen.

[0059] Example 2: Status monitoring of blades after thawing Background conditions: After the heating device has been running for 20 minutes, the sensor collects data when thawing is completed: temperature ;humidity ; Vibration power spectrum .

[0060] Calculation process: Temperature normalization calculation: ; Substitute the temperature range and data: ; Humidity normalization calculation: ; Substitute the humidity range and data: .

[0061] Example 3: Calculation process for large-scale monitoring of wind farms: Background conditions: Wind farm scale: 50 wind turbines, each with 3 blades, totaling 150 monitoring points.

[0062] Real-time risk factor for each blade: Wind turbine 1: , , ; Fan 2: , , .

[0063] Calculation process: Calculation of risk factor for a single wind turbine: ; Risk factors for wind turbine 1: ; Risk factors for wind turbine 2: ; Determine whether to alarm: set the fan alarm threshold ; Fan 1: , triggering an alarm signal; Fan 2: , no alarm was triggered.

[0064] Example 4: Application of weight optimization based on historical data: Background conditions: The actual value of risk factors collected during the past month of operation of a wind turbine and system calculated values The data is as follows: , , ; System calculated value , , .

[0065] The system uses historical data to optimize risk factor weights , , .

[0066] Calculation process: Optimization goal: The system aims to minimize the square error: , ; Set the initial weights: , , .

[0067] Error calculation: For each sample The squared error is: error ; Substitute data: Error ; Optimization process: The system adjusts weights through the gradient descent method , , , updated to: , , ; Optimization effect: Calculate risk factors using new weights , the error is reduced to 0.0015, and the optimization is completed.

[0068] Example 5: Application of thawing completion judgment: Background conditions: The thawing device is triggered by the icing of the fan blades. The thawing device stops after running for 20 minutes. The sensor collects the data after thawing: ; ; .

[0069] The system needs to recalculate the risk factor after the thawing is completed .

[0070] Calculation process: Temperature normalization: ; Set temperature range , substitute: ; Humidity normalization: The formula is as follows: ; Humidity range , substitute: ; Vibration Normalization: ; Set vibration range , substitute: ; Calculation of risk factor after thawing: ; Set weight , , , after substituting: .

[0071] Determine whether thawing is complete: set a thawing completion threshold.

[0072] Because the system determines that thawing is complete, it stops the heating device and releases the alarm signal.

[0073] The contents not described in detail in this specification belong to the prior art known to professional and technical personnel in the field. Although the present invention has been described in detail with reference to the aforementioned embodiments, it is still possible for those skilled in the art to modify the technical solutions described in the aforementioned embodiments, or to make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A fan blade icing monitoring system, characterized in that: The system includes: The sensor module is installed at the leading edge, middle and trailing edge of the fan blade. A sensor group is set at each position. The sensor group includes: Surface temperature sensor, used to measure blade surface temperature , the unit is Celsius; Capacitive humidity sensor for measuring leaf surface humidity , the unit is relative humidity percentage; Micro vibration accelerometer, used to collect blade vibration signals , in decibels; Data processing and edge computing unit: The data processing and edge computing unit communicates with the sensor module to perform multi-parameter fusion processing on real-time data and determine the blade icing risk factor through the following formula ,Right now: ； in, , , is the weight coefficient, satisfying ; , , are the normalized functions of temperature, humidity and vibration characteristics respectively; when When the system determines that the blades are frozen and triggers an alarm signal; Wireless communication module: uses LoRa or 5G technology to transmit real-time data to the cloud; Cloud computing platform: Based on regional meteorological data and historical data, combined with the long short-term memory neural network model, the icing risk factor within the next 30 minutes is predicted. The calculation formula is: ； in, , which is the current monitoring data vector; Alarm and control module: When blade icing risk factor or predicted value Exceeding the threshold When the blade heating device is triggered or the machine is shut down for protection; The vibration signal The power spectrum Based on wavelet transform calculation, the formula is: ； in, is the wavelet transform result of the vibration signal, is the frequency variable, is the time variable, ; The weight coefficient , , Determined by the following optimization formula: ， ； in, The first risk factors, is the real icing risk factor in historical data, is the sample size, Indicates that during the optimization process, the weight , , The constraint that the sum of weights is 1 must be satisfied; When the cloud computing platform optimizes the LSTM model, it is based on the following loss function : ； in, is the icing risk factor predicted by the model, is the true value, To predict future time steps; The alarm and control module determines that the thawing is complete based on the following formula: ； in, Icing risk factor recalculated after blade thawing; is the real-time temperature of the leaf surface after thawing, in degrees Celsius; It is the real-time humidity of the leaf surface after thawing, in relative humidity percentage; is the power spectrum amplitude of the vibration signal after the blade surface is thawed, in decibels; Based on temperature Normalization function of ; Based on humidity Normalization function of ; Based on vibration signal Normalization function of ; , , are the weight coefficients of temperature, humidity and vibration signals respectively, and satisfy ; set up is the thawing completion threshold, when When the blades are judged to be thawed, the system stops heating or releases the alarm signal; The wireless communication module adopts LoRa technology, specifically a module based on SemtechSX127x series or SX126x series chips, with a data transmission range of not less than 10 kilometers and anti-interference performance.

2. The wind turbine blade icing monitoring system according to claim 1, characterized in that: The temperature The normalization function Calculated by the following formula: ； in, and They are the maximum and minimum values ​​of the temperature monitoring range respectively.

3. The wind turbine blade icing monitoring system according to claim 1, characterized in that: The humidity The normalization function Calculated by the following formula: ； in, and They are the maximum and minimum values ​​of the humidity monitoring range respectively.

4. The wind turbine blade icing monitoring system according to claim 1, characterized in that: The vibration The normalization function Calculated by the following formula: ； in, and are the maximum and minimum values ​​of the vibration characteristics respectively.

5. A method for monitoring icing of wind turbine blades, which is implemented based on the wind turbine blade icing monitoring system according to any one of claims 1 to 4, characterized in that: The following steps are involved: The blade surface temperature is collected through the sensor module ,humidity and vibration signal ; The data processing unit integrates the , , , calculate the risk factor ; set up is the icing risk threshold, when The alarm is triggered and uploaded to the cloud; Cloud platform predicts future risk factors ; When the alarm signal is triggered, the control module starts the blade heating device or shuts down for protection.

Images