Multi-dimensional deformation monitoring method and system for wind turbine generator blades based on millimeter wave radar space-time trajectory analysis

Abstract

The invention discloses a wind turbine generator blade multi-dimensional deformation monitoring method and system based on millimeter wave radar space-time trajectory analysis, and belongs to the technical field of wind power monitoring. According to the method, an ultra-wide-angle millimeter-wave radar array on the outer wall of a tower is used for constructing a horizontal monitoring plane, and a continuous space-time trajectory sequence containing distance, angle and time information when blades sweep over is collected. The processor performs two-channel analysis on the trajectory sequence; a distance channel extracts a trajectory distance extreme value, and calculates the clearance distance and deformation of the blade waving direction; the time channel extracts the accurate transit moment of the blade center line through curve fitting, and the shimmy direction deformation is calculated in combination with the blade root reference and the rotating speed. The millimeter wave radar array with the tower wall distributed in the axial direction is reused, the bottom layer radar serves as the time reference, the cabin laser radar is introduced for multi-source fusion calibration, the problem that the shimmy deformation cannot be monitored through the non-contact technology is effectively solved, and all-weather monitoring of blade waving and shimmy two-way deformation is achieved.

Description

Technical Field

[0001] This invention relates to the field of wind power monitoring technology, specifically to a method and system for acquiring the continuous spatiotemporal trajectory of blades sweeping across the tower area on the outer wall of a wind turbine using an ultra-wide-angle millimeter-wave radar array, and obtaining the multidimensional deformation (flapping direction + oscillation direction) of the blades through fitting calculation. Background Technology

[0002] As wind turbine generators develop towards larger capacity and longer blades, the flexibility of the blades has increased significantly. During operation, the blades bear complex dynamic loads, and their deformation mainly occurs in two directions:

[0003] 1. Blade direction: Perpendicular to the plane of rotation, i.e., the blade bends towards the tower. Excessive blade deformation may lead to a "tower sweep" accident.

[0004] 2. Direction of oscillation: parallel to the plane of rotation, i.e., the bending of the blade along the tangent of rotation (leading or lagging).

[0005] Currently, monitoring technologies for the direction of blade flapping are relatively mature. For example, existing tower-mounted clearance monitoring systems use millimeter-wave radar installed on the tower to measure the absolute distance (clearance value) when the blades pass through the tower; or they use nacelle lidar to measure the blade tip distance.

[0006] However, there are still technical bottlenecks in monitoring the direction of oscillation:

[0007] 1. Limitations of contact-type solutions: Although strain gauges or fiber optic grating sensors can measure oscillation strain, they have problems such as difficulty in lightning protection, easy damage, and high installation and maintenance costs.

[0008] 2. Lack of non-contact solutions: Existing radar or laser monitoring mainly uses the "range measurement" function to calculate flapping deformation, ignoring the "time" information of the blade sweeping across the sensor beam. Since flapping deformation manifests as phase lag or lead of the blade in the plane of rotation, simple distance measurement cannot detect this change.

[0009] Therefore, there is an urgent need for a method that can reuse existing non-contact hardware and effectively monitor blade flapping deformation, enabling all-weather monitoring of both flapping and flapping deformation of blades. Summary of the Invention

[0010] The purpose of this invention is to provide a method and system for monitoring multi-dimensional deformation of wind turbine blades based on spatiotemporal trajectory analysis using millimeter-wave radar. This invention utilizes the ultra-wide-angle monitoring characteristics of millimeter-wave radar to collect the complete spatiotemporal trajectory of the blades as they pass overhead. By employing a curve fitting algorithm, the accuracy of time measurement is significantly improved, thereby accurately calculating the oscillation deformation while monitoring the flapping clearance.

[0011] To achieve the above objectives, the present invention provides the following technical solution:

[0012] 1. Dual-channel analysis: This analyzes the raw point cloud data (distance) collected by the radar. ,angle ,time The process is divided into two channels. Channel one utilizes... Extreme value calculation and clear space; Channel 2 utilizes The relationship is calculated based on the hysteresis.

[0013] 2. Spatiotemporal Trajectory Fitting: Addressing the high time accuracy requirements of vibration monitoring, this invention abandons the traditional "single-point triggering" logic and instead records the entire trajectory of the blade as it sweeps across the radar field of view. By performing least-squares fitting on dozens of sampling points, a smooth motion curve is reconstructed, thereby finding the blade's alignment with the tower centerline. This method effectively eliminates random errors in single-point measurements and achieves sub-sampling level time accuracy.

[0014] 3. Reference Differential: Using a radar installed at the blade root as a "clock reference", the influence of wind turbine speed fluctuations is eliminated, and only the elastic deformation hysteresis of the blade itself is extracted.

[0015] 4. Optional Laser Fusion: The system design is compatible with lidar. In embodiments equipped with lidar, the high-frequency data from the laser can be used to calibrate the millimeter-wave radar parameters online; when not equipped or the laser fails, the millimeter-wave radar operates independently, ensuring all-weather monitoring capability.

[0016] The beneficial effects of this invention are as follows:

[0017] 1. Hardware reuse and low cost: Existing tower wall-mounted airspace monitoring radar hardware can be used directly without adding new equipment. "Waving + swinging" bidirectional monitoring can be achieved simply by upgrading the algorithm.

[0018] 2. All-weather operation: Utilizing the strong penetrating power of millimeter-wave radar, it can still monitor normally even in extreme weather conditions such as rain, fog, and sandstorms where optical sensors fail (which are often the conditions with the highest load).

[0019] 3. Controllable precision: Online calibration is performed by combining the cabin lidar, taking into account both the all-weather capability of millimeter waves and the high precision advantage of lasers. Attached Figure Description

[0020] Figure 1 This is a diagram of the monitoring structure of the present invention (waving direction and swinging direction);

[0021] Figure 2 This is an overall architecture diagram of an embodiment of the present invention; Figure 3 is a flowchart of the calculation process of the present invention; Detailed Implementation

[0022] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0023] Example 1: Monitoring based on millimeter-wave radar array on tower wall

[0024] This embodiment reuses a set of "tower wall type millimeter wave 11 cross section deformation monitoring system".

[0025] 1. Hardware Layout: Eleven millimeter-wave radars are vertically arranged on the outer wall of the tower. The radars use 77GHz millimeter-wave frequency and have a horizontal field of view of [missing information]. The first group of radars is located near the blade root (such as close to the center of the hub) and serves as the reference radar; the second to eleventh groups of radars are distributed in the area from the blade body to the blade tip and serve as the measurement radar.

[0026] 2. Trajectory Acquisition: As the blades rotate and sweep across the tower, the radar continuously outputs a point cloud sequence. .

[0027] 3. Wave direction monitoring: The processor extracts the sequence... minimum value .like <Set a threshold to trigger an airspace warning.>

[0028] 4. Monitoring of oscillation direction:

[0029] The processor selects segments with high signal-to-noise ratios from the sequence and establishes... The quadratic fitting model: .

[0030] make Solving the equation yields the precise transit time. .

[0031] Perform the above operations on the first group of radars (blade root) and the eleventh group of radars (blade tip) respectively, and obtain the results. and .

[0032] Calculate the time difference .

[0033] Calculate pendulum deformation ,in This refers to the real-time rotational speed.

[0034] Example 2: LiDAR-assisted calibration

[0035] 1. Auxiliary calibration logic: Based on the millimeter-wave radar array on the tower wall, a multi-line lidar is installed at the bottom of the cabin to project a virtual light curtain downwards.

[0036] 2. Online Calibration: Under clear weather conditions, the processor simultaneously receives data from both lidar and millimeter-wave radar. Due to its extremely narrow beam, lidar measures the transit time. Treated as true. Calculate the bias. And update the fitting compensation parameters of the millimeter-wave radar.

[0037] 3. All-weather switching: When the rain and fog sensor detects rainfall or low visibility, the confidence level of the lidar data decreases, and the system automatically switches to the "millimeter-wave radar + correction factor" mode, ignoring the lidar data to ensure uninterrupted monitoring under harsh conditions.

Claims

1. A blade multi-dimensional deformation monitoring method based on millimeter wave radar space-time trajectory analysis, characterized in that, Comprising the following steps: S1 Wide-area trajectory capture: using ultra-wide-angle millimeter wave radar installed on the outer wall of the wind turbine tower, a horizontal monitoring plane is constructed; during the entire process of the blade sweeping across the monitoring plane, a space-time trajectory sequence containing multiple sampling points is continuously collected; each sampling point in the space-time trajectory sequence contains the radial distance, azimuth angle and corresponding timestamp of the target; S2 Flap deformation calculation: the minimum value of the radial distance is extracted from the space-time trajectory sequence, or the distance value when the azimuth angle is zero degrees is extracted, and the minimum distance from the blade surface to the outer wall of the tower is calculated as the clearance monitoring value of the blade in the flap direction; S3 Shake deformation solution: curve fitting is performed on the azimuth-time relationship in the space-time trajectory sequence to construct a blade motion trajectory model; and the accurate transit time of the blade geometric feature center through the vertical plane where the tower center axis is located is solved based on the model ; S4 Deformation synthesis: obtain the reference transit time of the blade root in the same rotation period ; calculate the time difference between the accurate transit time and the reference transit time , and combine the real-time rotation speed and the rotation radius of the target section to calculate the deformation value of the blade in the pitching direction.

2. The method of claim 1, wherein, In step S3, the curve fitting feature is: using the long trajectory data obtained by the ultra-wide-angle characteristics of the millimeter wave radar, the random noise of single-point sampling is eliminated by least squares method or polynomial interpolation algorithm, and sub-sampling level time accuracy better than the physical sampling period of the radar is obtained; The precise transit time To fit the curve azimuth The corresponding time .

3. The method of claim 1, wherein, The method is applied to a monitoring system comprising a plurality of millimeter wave radars distributed along the axial direction of the tower drum; wherein the first millimeter wave radar located in the upper part to the top of the tower drum near the blade root region is used to obtain the reference transit time ; and the remaining millimeter wave radars distributed along the blade span direction to the blade tip region are used to obtain the accurate transit time of each cross section of the blade, thereby constructing the flapwise deformation curve of the blade along the spanwise direction.

4. The method of claim 1, wherein, The method further comprises an optional auxiliary calibration step: the system is also configured with a nacelle laser radar as an optional sensor: Under the preset weather conditions, the reference transit time is obtained using the high-frequency scanning data of the laser radar, the system deviation of the millimeter wave radar trajectory fitting is calculated, and the calibration parameters are updated; Under rain, fog, snow or sand weather conditions, or in a system without a laser radar, only the space-time trajectory sequence of the millimeter wave radar is used to perform monitoring.

5. A wind turbine blade multi-dimensional deformation monitoring system, characterized in that, Comprising: Radar detection array: comprising a plurality of ultra-wide-angle millimeter wave radars installed at different heights on the outer wall of the tower, each radar having a horizontal field of view angle greater than 90 degrees; Clock synchronization module: configured to unify the time reference of each radar in the radar detection array; Processor: communicatively connected to the radar detection array and configured to perform dual-channel data analysis: based on the distance information of the radar echo, step S2 of claim 1 is performed to output the flap direction clearance monitoring data (distance channel); based on the space-time trajectory sequence of the radar echo, steps S3 to S4 of claim 1 are performed to output the edgewise direction deformation monitoring data (time channel); thereby realizing bidirectional decoupling monitoring of blade flap and edgewise using the same set of radar detection array.

6. The system of claim 5, wherein, The layout structure of the radar detection array is: Reference detection site: located in the middle and upper part of the tower to the top, configured to capture the rigid transit signal of the blade root; Deformation detection site: located in the middle and lower part of the tower, configured to capture the transit signal of the flexible part of the blade; the radars in the deformation detection site are distributed in a multi-ring or single wide-angle coverage manner to ensure that complete transit trajectories can be captured when the blade has a large edgewise deformation.

7. The system of claim 5, wherein, The system further comprises: Laser radar interface: the processor has an interface for communication with the nacelle laser radar, configured to receive laser radar data to perform online calibration of the time parameters of the millimeter wave radar.

Images