Synchronous monitoring method and system for dynamic response and static deformation of wind driven generator tower column

Abstract

The invention provides a wind driven generator tower column dynamic response and static deformation synchronous monitoring method and system, and the method comprises the steps: arranging an acceleration sensor, a strain sensor and a tilt angle sensor on a wind driven generator tower column, collecting the dynamic response data and static deformation data of the tower column respectively, and collecting environment data at the same time; comprise wind speed, wind direction and temperature; filtering and denoising processing is carried out on the collected dynamic response data, effective vibration signals are extracted, calibration and compensation processing is carried out on the collected static deformation data, and influences of sensor errors and environmental factors are eliminated. Through cooperation of algorithm combination, high-precision synchronous monitoring of dynamic response and static deformation of the wind driven generator tower column is achieved, monitoring comprehensiveness and precision are improved, structural abnormity can be found earlier, a scientific basis is provided for maintenance decision, and the method is suitable for popularization and application. And the reliability and practicability of health monitoring of the tower column of the wind driven generator are remarkably improved.

Description

Technical Field

[0001] The invention relates to the technical field of wind turbine tower monitoring, and in particular to a method and system for synchronously monitoring dynamic response and static deformation of a wind turbine tower. Background Art

[0002] At present, the health monitoring of wind turbine towers usually adopts the separation of dynamic response monitoring and static deformation monitoring. Dynamic response monitoring mainly captures the vibration characteristics of the tower through acceleration sensors, while static deformation monitoring measures the deformation state of the tower through strain sensors and inclination sensors. However, this separation monitoring method is difficult to fully reflect the complex mechanical behavior of the tower in actual operation, especially under the combined action of dynamic wind loads and static gravity loads, there is a significant coupling effect between the dynamic response and static deformation of the tower. The existing technology lacks an effective synchronous monitoring method, resulting in one-sided monitoring results and difficulty in accurately evaluating the structural health of the tower. At the same time, in the sensor layout of wind turbine towers, the existing technology usually relies on experience or simple mechanical analysis and lacks scientific layout methods. This blind layout may cause the sensor to fail to capture key modes or strain distribution, affecting the accuracy and comprehensiveness of the monitoring data.

[0003] Therefore, in order to solve the above problems, a method and system for synchronously monitoring the dynamic response and static deformation of a wind turbine tower is developed. Summary of the invention

[0004] In order to overcome the disadvantages that the prior art usually adopts the method of separating dynamic response monitoring and static deformation monitoring, lacks an effective synchronous monitoring method, resulting in one-sided monitoring results and difficulty in accurately evaluating the structural health status of the tower column, and in the arrangement of sensors on the tower column of a wind turbine, the prior art usually relies on experience or simple mechanical analysis and lacks a scientific arrangement method. This blind arrangement may cause the sensor to be unable to capture key modes or strain distribution, affecting the accuracy and comprehensiveness of the monitoring data. The present invention provides a method and system for synchronously monitoring the dynamic response and static deformation of a wind turbine tower column.

[0005] The technical solution of the present invention is: a method for synchronously monitoring the dynamic response and static deformation of a wind turbine tower, comprising: Step 1: Arrange acceleration sensors, strain sensors and inclination sensors on the wind turbine tower to collect dynamic response data and static deformation data of the tower respectively, and collect environmental data at the same time, including wind speed, wind direction and temperature; Step 2: Filter and denoise the collected dynamic response data to extract effective vibration signals, calibrate and compensate the collected static deformation data to eliminate the influence of sensor errors and environmental factors; Step 3: Extract frequency, amplitude and damping ratio features from the preprocessed dynamic response data, and extract strain and inclination features from the preprocessed static deformation data; Step 4: Based on the extracted dynamic response features, the dynamic response of the tower column is predicted using a time series analysis algorithm. The static deformation features extracted are predicted using a regression analysis algorithm. The dynamic response prediction results and the static deformation prediction results are used to calculate the interaction between the two through a coupling analysis algorithm. The dynamic response features and the static deformation features are input into the multi-task learning model. By sharing the underlying network and the independent upper network, the prediction accuracy of the dynamic response and the static deformation is optimized synchronously. A joint loss function is designed. The dynamic response prediction error and the static deformation prediction error are combined. The model parameters are optimized through the gradient descent method to achieve coordinated monitoring of dynamic response and static deformation. Step 5: Based on the interactive analysis results of dynamic response and static deformation, the structural health status of the tower column is evaluated, and an early warning message is issued when the dynamic response or static deformation exceeds the preset threshold.

[0006] As a preferred embodiment of the present invention, the dynamic response prediction algorithm adopts a long short-term memory network model, and the formula required to calculate the dynamic response prediction function is as follows: ; in, For the dynamic response prediction function, the i-th dynamic response feature needs to be input and environmental parameters , get the dynamic response prediction value, is the weight coefficient of the kth vibration mode, which is used to quantify the contribution of this mode to the overall dynamic response. It represents the nonlinear transfer function in the tower column vibration and is used to describe the nonlinear characteristics in the vibration. is the sensitivity coefficient of the kth neuron to the lth dynamic response feature, reflecting the coupling relationship between the mode and the feature. is the reference vibration level of the kth mode, which is used to adjust the initial state of the mode. The error term in the prediction of the dynamic response represents random vibration or noise not captured by the model.

[0007] As a preferred embodiment of the present invention, the static deformation prediction algorithm adopts a support vector regression model, and the static deformation prediction algorithm formula to be used is as follows: ; in, is the static deformation prediction function, and the jth static deformation feature is input during calculation and environmental parameters , the predicted value of static deformation is obtained, which is the weight coefficient of the pth strain mode, represents the contribution of different strain modes to the static deformation, represents the reference center data of the pth strain mode, indicating the typical position of the strain distribution, represents the diffusion parameter of the pth strain mode, indicating the range of strain distribution, The error term represents the static deformation prediction, which accounts for random deformation and noise not captured by the model.

[0008] As a preferred embodiment of the present invention, the dynamic-static coupling analysis algorithm is based on a finite element model to calculate the interaction relationship between the dynamic response and the static deformation, and the formula is as follows: ; in, is the dynamic response prediction value, is the static deformation prediction value, and is the time rate of change of dynamic response and static deformation, indicating the instantaneous change of vibration and deformation, is the Laplace operator, which is used to describe the coupling relationship between dynamic response and static deformation in space. is the coupling strength coefficient, which is used to adjust the interaction strength between dynamic response and static deformation.

[0009] As a preferred embodiment of the present invention, the joint loss function is a weighted sum of the dynamic response prediction error and the static deformation prediction error, and the weight coefficient is dynamically adjusted according to the structural characteristics of the tower column. The formula used is as follows: ; in, and are the true values ​​of dynamic response and static deformation, respectively, representing the actual vibration and deformation data of the tower column, is the regularization coefficient, which is used to control the complexity of the model and prevent overfitting. is a regularization term used to constrain model parameters and improve generalization ability.

[0010] As a preferred embodiment of the present invention, when arranging the sensors, the method is as follows: Step s1: Establish a finite element model of the tower column, perform modal analysis, extract the first few modal vibration shapes of the tower column, determine the key positions with large vibration amplitudes based on the modal vibration shapes, and select the optimal sensor layout points in the key positions to ensure that the observability of the modal vibration shapes is maximized; Step s2: Perform static analysis on the tower column, calculate the strain energy distribution under different load conditions, determine the high strain energy area according to the strain energy distribution, and arrange strain sensors at representative locations in the high strain energy area; Step s3: Perform force analysis on the tower column, calculate the inclination change under different load conditions, determine the position where the inclination change is significant according to the sensitivity of the inclination change, and select the optimal arrangement point of the inclination sensor; Step s4: Analyze the impact of environmental data on the dynamic response and static deformation of the tower column, dynamically adjust the sensor layout or add temporary monitoring points according to the impact of environmental factors, predict the impact of environmental factors on sensor layout, and optimize the layout plan.

[0011] A synchronous monitoring system for dynamic response and static deformation of a wind turbine tower, comprising: Hardware part, the hardware part includes a sensor module, a data acquisition and transmission module, a central processing unit and a power supply module. The data acquisition and transmission module is used to receive and process sensor data, supports multi-channel synchronous acquisition, and uses wireless transmission to transmit information. The central processing unit is used to run dynamic response prediction, static deformation prediction, coupling analysis and multi-task learning algorithms; The software part includes a data preprocessing module, a dynamic response prediction module, a static deformation prediction module, a coupling analysis module, a joint optimization module and a user interface and visualization module.

[0012] As a preferred embodiment of the present invention, the sensor module includes: Acceleration sensors are arranged at the top, middle and bottom of the tower column and are used to collect dynamic response data of the tower column; A strain sensor is arranged in a key stress-bearing area of ​​the tower column and is used to collect static deformation data of the tower column; An inclination sensor is arranged at the top and the middle of the tower column and is used to collect inclination angle data of the tower column; The environmental sensors include wind speed sensors, wind direction sensors and temperature sensors, and are used to collect environmental data.

[0013] By adopting the above technical solution, the present invention has the following advantages: 1. The present invention realizes high-precision synchronous monitoring of the dynamic response and static deformation of wind turbine towers through the coordinated cooperation of algorithm combinations. The combination of dynamic response prediction algorithm and static deformation prediction algorithm can comprehensively reflect the structural health status of the tower, further quantify the interaction between dynamic response and static deformation, reveal the complex mechanical behavior of the tower in actual operation, and synchronously optimize the prediction accuracy of dynamic response and static deformation, avoiding the deviation caused by single task optimization. Algorithm collaboration not only improves the comprehensiveness and accuracy of monitoring, but also can detect structural abnormalities earlier, providing a scientific basis for maintenance decisions, and significantly improving the reliability and practicality of wind turbine tower health monitoring.

[0014] 2. The present invention provides operators with a clear sensor layout method through finite element analysis and environmental factor impact analysis, which significantly improves the scientificity and adaptability of sensor installation. This scientific sensor layout method not only improves the accuracy and comprehensiveness of monitoring data, but also can adapt to complex environmental changes, providing strong technical support for the health monitoring of wind turbine towers. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 The present invention is a flowchart of the steps of the method for synchronously monitoring the dynamic response and static deformation of a wind turbine tower.

[0016] Figure 2 Flowchart of steps for the sensor arrangement of the present invention.

[0017] Figure 3 It is a structural schematic diagram of a system for synchronously monitoring the dynamic response and static deformation of a wind turbine tower according to the present invention.

[0018] Figure 4 It is a schematic diagram of the structure of the sensor module of the present invention. DETAILED DESCRIPTION

[0019] Reference to an embodiment herein means that a particular feature, structure, or characteristic described in conjunction with the embodiment may be included in at least one embodiment of the present invention. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0020] A method for synchronously monitoring the dynamic response and static deformation of a wind turbine tower column, such as Figure 1 As shown, including: Step 1: Arrange acceleration sensors, strain sensors and inclination sensors on the wind turbine tower to collect dynamic response data and static deformation data of the tower respectively, and collect environmental data at the same time, including wind speed, wind direction and temperature; Step 2: Filter and denoise the collected dynamic response data to extract effective vibration signals, calibrate and compensate the collected static deformation data to eliminate the influence of sensor errors and environmental factors; Step 3: Extract frequency, amplitude and damping ratio features from the preprocessed dynamic response data, and extract strain and inclination features from the preprocessed static deformation data; Step 4: Based on the extracted dynamic response features, the dynamic response of the tower column is predicted using a time series analysis algorithm. The static deformation features extracted are predicted using a regression analysis algorithm. The dynamic response prediction results and the static deformation prediction results are used to calculate the interaction between the two through a coupling analysis algorithm. The dynamic response features and the static deformation features are input into the multi-task learning model. By sharing the underlying network and the independent upper network, the prediction accuracy of the dynamic response and the static deformation is optimized synchronously. A joint loss function is designed. The dynamic response prediction error and the static deformation prediction error are combined. The model parameters are optimized through the gradient descent method to achieve coordinated monitoring of dynamic response and static deformation. Step 5: Based on the interactive analysis results of dynamic response and static deformation, the structural health status of the tower column is evaluated, and an early warning message is issued when the dynamic response or static deformation exceeds the preset threshold.

[0021] The dynamic response prediction algorithm uses a long short-term memory network model. The formula required to calculate the dynamic response prediction function is as follows: ; in, For the dynamic response prediction function, the i-th dynamic response feature needs to be input and environmental parameters , get the dynamic response prediction value, is the weight coefficient of the kth vibration mode, which is used to quantify the contribution of this mode to the overall dynamic response. It represents the nonlinear transfer function in the tower column vibration and is used to describe the nonlinear characteristics in the vibration. is the sensitivity coefficient of the kth neuron to the lth dynamic response feature, reflecting the coupling relationship between the mode and the feature. is the reference vibration level of the kth mode, which is used to adjust the initial state of the mode. The error term in the prediction of the dynamic response represents random vibration or noise not captured by the model.

[0022] The static deformation prediction algorithm uses a support vector regression model. The static deformation prediction algorithm formula required is as follows: ; in, is the static deformation prediction function, and the jth static deformation feature is input during calculation and environmental parameters , the predicted value of static deformation is obtained, which is the weight coefficient of the pth strain mode, represents the contribution of different strain modes to the static deformation, represents the reference center data of the pth strain mode, indicating the typical position of the strain distribution, represents the diffusion parameter of the pth strain mode, indicating the range of strain distribution, The error term represents the static deformation prediction, which accounts for random deformation and noise not captured by the model.

[0023] The dynamic-static coupling analysis algorithm is based on the finite element model and calculates the interaction between dynamic response and static deformation. The formula is as follows: ; in, is the dynamic response prediction value, is the static deformation prediction value, and is the time rate of change of dynamic response and static deformation, indicating the instantaneous change of vibration and deformation, is the Laplace operator, which is used to describe the coupling relationship between dynamic response and static deformation in space. is the coupling strength coefficient, which is used to adjust the interaction strength between dynamic response and static deformation.

[0024] The joint loss function is the weighted sum of the dynamic response prediction error and the static deformation prediction error. The weight coefficient is dynamically adjusted according to the structural characteristics of the tower column. The formula used is as follows: ; in, and are the true values ​​of dynamic response and static deformation, respectively, representing the actual vibration and deformation data of the tower column, is the regularization coefficient, which is used to control the complexity of the model and prevent overfitting. is a regularization term used to constrain model parameters and improve generalization ability.

[0025] It should be noted that the workflow of the four algorithm formulas is as follows: Based on the dynamic response characteristics (such as frequency, amplitude, damping ratio) and environmental data (such as wind speed, wind direction, and temperature) collected by the acceleration sensor, the time series analysis model is used to predict the dynamic response of the tower column. The contribution of different vibration modes to the dynamic response is quantified by the modal weight coefficient and sensitivity coefficient, and the nonlinear characteristics in the tower column vibration are described by the nonlinear transfer function. Then, based on the static deformation characteristics (such as strain, inclination) and environmental data collected by the strain sensor and the inclination sensor, the radial basis function model is used to predict the static deformation of the tower column. The strain modal weight coefficient and the reference center are used to describe the influence of different strain distributions on the static deformation, and the range of the strain distribution is adjusted by the diffusion parameter. Based on the output The dynamic response prediction results and the output static deformation prediction results are used to calculate the coupling relationship between the dynamic response and the static deformation, where the instantaneous changes of vibration and deformation are described by the time rate of change, and the interaction between dynamic and static in space and time is quantified by the Laplace operator and the coupling strength coefficient. Finally, the extracted dynamic response features and static deformation features are input into the multi-task learning model, and the prediction accuracy of the dynamic response and static deformation is synchronously optimized by sharing the underlying network and the independent upper network. The model parameters are constrained by the regularization coefficient and the regularization term to prevent overfitting, and the dynamic response prediction error and the static deformation prediction error are balanced by the joint loss function to achieve collaborative monitoring of the dynamic response and static deformation.

[0026] like Figure 2 As shown, when arranging the sensors, the method is as follows: Step s1: Establish a finite element model of the tower column, perform modal analysis, extract the first few modal vibration shapes of the tower column, determine the key positions with large vibration amplitudes based on the modal vibration shapes, and select the optimal sensor layout points in the key positions to ensure that the observability of the modal vibration shapes is maximized; Step s2: Perform static analysis on the tower column, calculate the strain energy distribution under different load conditions, determine the high strain energy area according to the strain energy distribution, and arrange strain sensors at representative locations in the high strain energy area; Step s3: Perform force analysis on the tower column, calculate the inclination change under different load conditions, determine the position where the inclination change is significant according to the sensitivity of the inclination change, and select the optimal arrangement point of the inclination sensor; Step s4: Analyze the impact of environmental data on the dynamic response and static deformation of the tower column, dynamically adjust the sensor layout or add temporary monitoring points according to the impact of environmental factors, predict the impact of environmental factors on sensor layout, and optimize the layout plan.

[0027] A synchronous monitoring system for dynamic response and static deformation of wind turbine tower, such as Figure 3 As shown, including: Hardware: The hardware includes sensor module, data acquisition and transmission module, central processing unit and power supply module. The data acquisition and transmission module is used to receive and process sensor data, supports multi-channel synchronous acquisition, and uses wireless transmission to transmit information. The central processing unit is used to run dynamic response prediction, static deformation prediction, coupling analysis and multi-task learning algorithms. The software part includes data preprocessing module, dynamic response prediction module, static deformation prediction module, coupling analysis module, joint optimization module and user interface and visualization module.

[0028] like Figure 4 As shown, the sensor module includes: Acceleration sensors are arranged at the top, middle and bottom of the tower column to collect dynamic response data of the tower column; Strain sensors are arranged in key stress-bearing areas of the tower column to collect static deformation data of the tower column; Inclination sensors are arranged at the top and middle of the tower column to collect the inclination angle data of the tower column; Environmental sensors, including wind speed sensors, wind direction sensors and temperature sensors, are used to collect environmental data.

[0029] In summary, through the coordinated cooperation of algorithm combinations, high-precision synchronous monitoring of the dynamic response and static deformation of wind turbine towers is achieved. The combination of dynamic response prediction algorithm and static deformation prediction algorithm can comprehensively reflect the structural health status of the tower, further quantify the interaction between dynamic response and static deformation, reveal the complex mechanical behavior of the tower in actual operation, and simultaneously optimize the prediction accuracy of dynamic response and static deformation, avoiding the deviation caused by single task optimization. Algorithm collaboration not only improves the comprehensiveness and accuracy of monitoring, but also can detect structural abnormalities earlier, providing a scientific basis for maintenance decisions, and significantly improving the reliability and practicality of wind turbine tower health monitoring. At the same time, through finite element analysis and environmental factor impact analysis, a clear sensor layout method is provided for operators, which significantly improves the scientificity and adaptability of sensor installation. This scientific sensor layout method not only improves the accuracy and comprehensiveness of monitoring data, but also can adapt to complex environmental changes, providing strong technical support for the health monitoring of wind turbine towers.

[0030] The above embodiments are provided for persons familiar with the art to implement or use the present invention. Personnel familiar with the art can make various modifications or changes to the above embodiments without departing from the concept of the present invention. Therefore, the protection scope of the present invention is not limited to the above embodiments, but should be the maximum scope of the innovative features mentioned in the claims.

Claims

1. A method for synchronously monitoring the dynamic response and static deformation of a wind turbine tower, characterized in that: Included are: Step 1: Arrange acceleration sensors, strain sensors and inclination sensors on the wind turbine tower to collect dynamic response data and static deformation data of the tower respectively, and collect environmental data at the same time, including wind speed, wind direction and temperature; Step 2: Filter and denoise the collected dynamic response data to extract effective vibration signals, calibrate and compensate the collected static deformation data to eliminate the influence of sensor errors and environmental factors; Step 3: Extract frequency, amplitude and damping ratio features from the preprocessed dynamic response data, and extract strain and inclination features from the preprocessed static deformation data; Step 4: Based on the extracted dynamic response features, the dynamic response of the tower column is predicted using a time series analysis algorithm. The static deformation features extracted are predicted using a regression analysis algorithm. The dynamic response prediction results and the static deformation prediction results are used to calculate the interaction between the two through a coupling analysis algorithm. The dynamic response features and the static deformation features are input into the multi-task learning model. By sharing the underlying network and the independent upper network, the prediction accuracy of the dynamic response and the static deformation is optimized synchronously. A joint loss function is designed. The dynamic response prediction error and the static deformation prediction error are combined. The model parameters are optimized through the gradient descent method to achieve coordinated monitoring of dynamic response and static deformation. Step 5: Based on the interactive analysis results of dynamic response and static deformation, the structural health status of the tower column is evaluated, and an early warning message is issued when the dynamic response or static deformation exceeds the preset threshold.

2. A method for synchronously monitoring the dynamic response and static deformation of a wind turbine tower according to claim 1, characterized in that: The dynamic response prediction algorithm adopts a long short-term memory network model, and the formula required to calculate the dynamic response prediction function is as follows: ； in, For the dynamic response prediction function, the i-th dynamic response feature needs to be input and environmental parameters , get the dynamic response prediction value, is the weight coefficient of the kth vibration mode, which is used to quantify the contribution of this mode to the overall dynamic response. It represents the nonlinear transfer function in the tower column vibration and is used to describe the nonlinear characteristics in the vibration. is the sensitivity coefficient of the kth neuron to the lth dynamic response feature, reflecting the coupling relationship between the mode and the feature. is the reference vibration level of the kth mode, which is used to adjust the initial state of the mode. The error term in the prediction of the dynamic response represents random vibration or noise not captured by the model.

3. A method for synchronously monitoring the dynamic response and static deformation of a wind turbine tower according to claim 1, characterized in that: The static deformation prediction algorithm adopts a support vector regression model, and the static deformation prediction algorithm formula required to be used is as follows: ； in, is the static deformation prediction function, and the jth static deformation feature is input during calculation and environmental parameters , the predicted value of static deformation is obtained, which is the weight coefficient of the pth strain mode, represents the contribution of different strain modes to the static deformation, represents the reference center data of the pth strain mode, indicating the typical position of the strain distribution, represents the diffusion parameter of the pth strain mode, indicating the range of strain distribution, The error term represents the static deformation prediction, which accounts for random deformation and noise not captured by the model.

4. A method for synchronously monitoring the dynamic response and static deformation of a wind turbine tower according to claim 1, characterized in that: The dynamic-static coupling analysis algorithm is based on the finite element model and calculates the interaction between dynamic response and static deformation. The formula is as follows: ； in, is the dynamic response prediction value, is the static deformation prediction value, and is the time rate of change of dynamic response and static deformation, indicating the instantaneous change of vibration and deformation, is the Laplace operator, which is used to describe the coupling relationship between dynamic response and static deformation in space. is the coupling strength coefficient, which is used to adjust the interaction strength between dynamic response and static deformation.

5. A method for synchronously monitoring the dynamic response and static deformation of a wind turbine tower according to claim 1, characterized in that: The joint loss function is the weighted sum of the dynamic response prediction error and the static deformation prediction error. The weight coefficient is dynamically adjusted according to the structural characteristics of the tower column. The formula used is as follows: ； in, and are the true values ​​of dynamic response and static deformation, respectively, representing the actual vibration and deformation data of the tower column, is the regularization coefficient, which is used to control the complexity of the model and prevent overfitting. is a regularization term used to constrain model parameters and improve generalization ability.

6. A method for synchronously monitoring the dynamic response and static deformation of a wind turbine tower according to claim 1, characterized in that: When arranging the sensors, the method is as follows: Step s1: Establish a finite element model of the tower column, perform modal analysis, extract the first few modal vibration shapes of the tower column, determine the key positions with large vibration amplitudes based on the modal vibration shapes, and select the optimal sensor layout points in the key positions to ensure that the observability of the modal vibration shapes is maximized; Step s2: Perform static analysis on the tower column, calculate the strain energy distribution under different load conditions, determine the high strain energy area according to the strain energy distribution, and arrange strain sensors at representative locations in the high strain energy area; Step s3: Perform force analysis on the tower column, calculate the inclination change under different load conditions, determine the position where the inclination change is significant according to the sensitivity of the inclination change, and select the optimal arrangement point of the inclination sensor; Step s4: Analyze the impact of environmental data on the dynamic response and static deformation of the tower column, dynamically adjust the sensor layout or add temporary monitoring points according to the impact of environmental factors, predict the impact of environmental factors on sensor layout, and optimize the layout plan.

7. A synchronous monitoring system for dynamic response and static deformation of wind turbine tower, characterized in that: Included are: The hardware part includes a sensor module, a data acquisition and transmission module, a central processing unit and a power supply module. The data acquisition and transmission module is used to receive and process sensor data, supports multi-channel synchronous acquisition, and uses wireless transmission to transmit information. The central processing unit is used to run dynamic response prediction, static deformation prediction, coupling analysis and multi-task learning algorithms.

8. Software part, which includes a data preprocessing module, a dynamic response prediction module, a static deformation prediction module, a coupling analysis module, a joint optimization module and a user interface and visualization module.

9. A wind turbine tower dynamic response and static deformation synchronous monitoring system according to claim 7, characterized in that: The sensor module includes: Acceleration sensors are arranged at the top, middle and bottom of the tower column and are used to collect dynamic response data of the tower column; A strain sensor is arranged in a key stress-bearing area of ​​the tower column and is used to collect static deformation data of the tower column; An inclination sensor is arranged at the top and the middle of the tower column and is used to collect inclination angle data of the tower column; The environmental sensors include wind speed sensors, wind direction sensors and temperature sensors, and are used to collect environmental data.

Images