Wind power load test method and system based on multi-source data analysis

Abstract

The invention provides a wind power load test method and system based on multi-source data analysis, and relates to the technical field of data processing, and the method comprises the steps: obtaining a target sector set; acquiring first multi-dimensional feature information of the first sector; if the preset test constraint is met, taking the first sector as a target test sector; performing a local load test to obtain a target local load time sequence, and collecting target load feature information; real-time monitoring information is obtained through dynamic monitoring of the distributed monitoring platform; and introducing a wind power load evaluation function to evaluate and analyze the target load characteristic information and the real-time monitoring information to obtain a target wind power load index. According to the method and the device, the technical problem that the wind power load test is not accurate enough due to the fact that the influence of a complex environment on wind power load evaluation is not comprehensively considered in the prior art can be solved, the load states of the unit in different areas are evaluated by performing distributed monitoring on the wind power unit, and the accuracy of the wind power load test is improved.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a wind power load testing method and system based on multi-source data analysis. Background Art

[0002] Wind power load testing is the process of measuring and analyzing the various forces (such as wind, aerodynamics, structure, etc.) that wind turbines are subjected to during operation, with the purpose of evaluating the load performance of wind turbines under different working conditions. The real-time monitoring system of wind turbines usually collects a large amount of data, including wind speed, wind direction, vibration, unit operating status, etc. Usually, when processing these massive amounts of data, there are often problems such as data processing delays and lags, resulting in large deviations in load assessment results, and unable to reflect the stress state of wind turbines in real time, thus affecting the accuracy of wind power load testing.

[0003] In summary, the prior art has a technical problem that the wind power load test is not accurate enough because the impact of complex environment on wind power load assessment is not fully considered. Summary of the invention

[0004] The purpose of this application is to provide a wind power load testing method and system based on multi-source data analysis, so as to solve the technical problem in the prior art that the wind power load test is not accurate enough due to the failure to fully consider the impact of complex environment on wind power load assessment.

[0005] In view of the above problems, the present application provides a wind power load testing method and system based on multi-source data analysis.

[0006] In the first aspect, the present application provides a wind power load testing method based on multi-source data analysis, and the wind power load testing method based on multi-source data analysis is implemented by a wind power load testing system based on multi-source data analysis, wherein the wind power load testing method based on multi-source data analysis includes: obtaining a target sector set, wherein the target sector set refers to a set of sectors obtained by dividing a target neighborhood of a target wind turbine; obtaining first multidimensional feature information of a first sector, wherein the first sector refers to any sector in the target sector set; if the first multidimensional feature information meets a predetermined test constraint, the first sector is used as a target test sector; performing a local load test on the target test sector to obtain a target local load time series, and collecting target load feature information; obtaining real-time monitoring information of the target wind turbine through dynamic monitoring on a distributed monitoring platform; introducing a wind power load evaluation function to evaluate and analyze the target load feature information and the real-time monitoring information to obtain a target wind power load index.

[0007] In a second aspect, the present application also provides a wind power load testing system based on multi-source data analysis, which is used to execute the wind power load testing method based on multi-source data analysis as described in the first aspect, wherein the wind power load testing system based on multi-source data analysis includes: an area acquisition module, which is used to acquire a target sector set, wherein the target sector set refers to a set of sectors obtained by dividing the target neighborhood of the target wind turbine group; a feature acquisition module, which is used to acquire first multidimensional feature information of a first sector, wherein the first sector refers to any sector in the target sector set; a target area determination module, which is used to use the first sector as a target test sector if the first multidimensional feature information meets a predetermined test constraint; a load feature collection module, which is used to perform a local load test on the target test sector to obtain a target local load time series, and collect target load feature information; a monitoring information acquisition module, which is used to dynamically monitor the target wind turbine group through a distributed monitoring platform; an evaluation and analysis module, which is used to introduce a wind power load evaluation function to evaluate and analyze the target load feature information and the real-time monitoring information to obtain a target wind power load index.

[0008] One or more technical solutions provided in this application have at least the following technical effects or advantages: By obtaining a target sector set, the target sector set refers to a set of sectors obtained by dividing the target neighborhood of the target wind turbine; obtaining the first multidimensional feature information of the first sector, wherein the first sector refers to any sector in the target sector set; if the first multidimensional feature information meets the predetermined test constraints, the first sector is used as the target test sector; a local load test is performed on the target test sector to obtain the target local load time series, and the target load feature information is collected; the real-time monitoring information of the target wind turbine is obtained through dynamic monitoring on a distributed monitoring platform; a wind power load evaluation function is introduced to evaluate and analyze the target load feature information and the real-time monitoring information, and a target wind power load index is obtained. In other words, by dividing the target sector set of the target wind turbine and performing detailed local load tests on the sectors, the multidimensional feature information of the target sector is obtained, the operating status of the wind turbine is monitored in real time, and the load condition of the wind turbine is evaluated, thereby improving the accuracy and efficiency of the wind power load test.

[0009] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented according to the contents of the specification, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are specifically cited below. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easy to understand through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the technical solutions in the present application or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0011] Figure 1 This is a flow chart of a wind power load testing method based on multi-source data analysis in this application; Figure 2 This is a structural schematic diagram of a wind power load testing system based on multi-source data analysis in this application.

[0012] Explanation of reference numerals: region acquisition module 11 , feature acquisition module 12 , target region determination module 13 , load feature collection module 14 , monitoring information acquisition module 15 , evaluation and analysis module 16 . DETAILED DESCRIPTION

[0013] This application solves the technical problem in the prior art that the wind power load test is not accurate enough due to the failure to fully consider the impact of complex environment on wind power load evaluation by providing a wind power load test method and system based on multi-source data analysis. By dividing the target sector set of the target wind turbine and performing detailed local load tests on the sectors, the multi-dimensional feature information of the target sectors is obtained, the operating status of the wind turbine is monitored in real time, and the load status of the wind turbine is evaluated, thereby improving the accuracy and efficiency of the wind power load test.

[0014] Below, the technical solutions in the present application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments of the present application. It should be understood that the present application is not limited to the example embodiments described herein. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application. It should also be noted that, for the convenience of description, only the parts related to the present application are shown in the accompanying drawings, rather than all of them.

[0015] For example, please refer to the attached Figure 1 The present application provides a wind power load test method based on multi-source data analysis, wherein the wind power load test method based on multi-source data analysis is applied to a wind power load test system based on multi-source data analysis, and the wind power load test method based on multi-source data analysis specifically includes the following steps: S100: Acquire a target sector set, where the target sector set refers to a set of sectors obtained by dividing a target neighborhood of a target wind turbine generator set.

[0016] Specifically, the installation location, environmental conditions and wind farm characteristics of the target wind turbine are analyzed to determine its target neighborhood. The target neighborhood refers to an area around the wind turbine, which may include different locations around the wind turbine, including blades, towers, nacelles and other areas. The target neighborhood is divided into several very small fan-shaped parts, each of which is a sector. For example, the neighborhood area can be divided into several fan-shaped areas within a certain angle range by a polar coordinate system. After dividing the target neighborhood of the target wind turbine, a target sector set is obtained, which includes various different test areas of the entire wind turbine, ensuring that the stress conditions of each area can be monitored and analyzed separately. By dividing the target neighborhood into multiple sectors and forming a target sector set, accurate monitoring of different areas of the wind turbine can be achieved, which helps to reveal the local load differences of the wind turbine under different wind conditions.

[0017] S200: Acquire first multi-dimensional feature information of a first sector, where the first sector refers to any sector in the target sector set.

[0018] Further, the present application S200 includes: Obtain historical meteorological records of the target wind turbine, wherein the historical meteorological records include a first main wind direction of the first sector; obtain first environmental obstacle information of the first sector; obtain first terrain information of the first sector; and compose the first multi-dimensional feature information based on the first main wind direction, the first environmental obstacle information and the first terrain information.

[0019] Specifically, a sector is randomly selected as the first sector in the target sector set. The historical meteorological records of the target wind turbine are obtained from the meteorological station, the meteorological monitoring system of the wind farm, or the third-party meteorological data service, including the wind speed, wind direction, temperature, humidity and other information at multiple historical time points, which are used to analyze the performance of the wind turbine under different working conditions. The historical meteorological records include the first main wind direction of the first sector, which is usually obtained by counting the wind direction with the highest frequency in the sector. The wind direction frequency refers to the proportion of the number of wind directions in the sector to the total number of observations during the wind measurement period.

[0020] Determine whether there are environmental obstacles in the first sector, such as high-rise buildings, mountains, trees, etc. Obstacles have a significant impact on wind speed and wind direction, usually causing wind flow separation and turbulence, which in turn affects the working efficiency and structural load of wind turbines. The first environmental obstacle information is any natural or artificial obstacles (such as mountains, buildings, trees, etc.) that may exist in a sector. At the same time, determine the terrain characteristics in the first sector, that is, whether there is a large undulating ground, whether there is a steep slope, etc. The first terrain information is the geographical terrain characteristics of the first sector, including the undulations of the ground, the slope, whether there are mountains, canyons and other geographical structures. For example, mountains may cause wind deflection, and canyons may accelerate wind speed, so terrain information is an important basis for wind turbine performance analysis. Obstacles and terrain can be obtained through geographic information system (GIS) tools, combined with data such as topographic maps and building distribution maps. In other words, through the geographic information system (GIS), obstacles and terrain features can be accurately marked and combined with meteorological data for analysis.

[0021] The first main wind direction, the first environmental obstacle information and the first terrain information are combined to obtain the first multi-dimensional feature information of the first area, which is used to describe the environmental characteristics of the first sector. The first multi-dimensional feature information includes the wind speed distribution of the first sector, the changes in the airflow affected by the terrain and obstacles, and other factors that may affect the load of the wind turbine. It is used to identify the changes in wind force under different working conditions, the impact of obstacles and terrain on the airflow, and how they affect the load distribution of the wind turbine. By obtaining multi-dimensional data such as wind speed, wind direction, obstacles, and terrain, the environmental characteristics of the area where the wind turbine is located can be fully understood, providing a more accurate data basis for subsequent load testing and evaluation.

[0022] S300: If the first multi-dimensional feature information meets the predetermined test constraints, the first sector is used as a target test sector.

[0023] Specifically, according to the specific needs of the power plant, the characteristics of the wind turbines, safety requirements and other factors, predetermined test constraints are set to determine whether a sector is suitable as a target test sector, including factors such as wind speed range, wind direction range, climate conditions, equipment load capacity, regional safety, etc. For example, the predetermined test constraints include whether the main wind direction is within a predetermined range, whether the average wind speed or maximum wind speed is within a predetermined threshold, whether the location of obstacles will affect the test results, and whether the terrain slope will cause abnormal wind flow or wind speed changes.

[0024] The first multi-dimensional feature information obtained is compared with the predetermined test constraints. If the multi-dimensional feature information of the first sector meets the predetermined test constraints, the first sector is determined as the target test sector. If the first sector does not meet the predetermined test constraints, but after evaluating the first sector, it is considered that this area is very critical and can be adjusted, the test constraints are appropriately adjusted to meet the requirements; if the first sector does not meet the constraints and cannot be adjusted, another sector is selected in the target sector, and the above steps are repeated until a sector that meets the predetermined test constraints is found.

[0025] By comparing and analyzing the multi-dimensional feature information of the first sector with the predetermined test constraints, it is ensured that the selection of the target test sector meets the environmental, safety and performance standards. Only by testing in areas that meet specific constraints can the accuracy and reliability of the test results be ensured.

[0026] S400: Performing a local load test on the target test sector to obtain a target local load timing, and collecting target load characteristic information.

[0027] Further, the present application S400 includes: Acquire the time domain characteristic information of the target local load time series; acquire the frequency domain characteristic information of the target local load spectrum, wherein the target local load spectrum refers to the spectrum obtained by fast Fourier transforming the target local load time series; the time domain characteristic information and the frequency domain characteristic information constitute the target load characteristic information.

[0028] Specifically, in the target test sector, local load tests are carried out on the wind turbines to test the loads of specific components (such as wind turbine blades, towers or other load-bearing components) under different wind speeds, wind directions and other conditions. In the target test sector, multi-source sensors (such as strain gauges, etc.) are arranged to collect the loads of wind turbine blades, towers, nacelles and other components in real time to ensure that the installation positions and number of sensors can cover key parts. Strain sensors and accelerometers are installed on the blades, such as the root, middle section and tip of the blades, to measure the effective angle of attack of the blades, relative wind speed and local loads of the blades. Pressure sensors are arranged in the radial direction of the outer wall of the tower structure to measure the load at the bottom of the tower, and sensors are installed at the top and bottom of the tower to measure the bending moments in the pitch and left and right directions.

[0029] After installing the sensor, start the wind turbine and collect load data in real time according to different operating conditions such as wind speed, wind direction, and wind turbine load. Record the load change data of the wind turbine in the sector over time, obtain the target local load time series, and record the entire process of the force change of the wind turbine components (such as blades and towers) in the target test sector over time.

[0030] From the time series of the target local load, extract the time domain characteristic information, including mean, peak, standard deviation, fluctuation amplitude, etc. The average value of the load is obtained by summing all the time series data and dividing by the number of data points; the peak value is obtained by finding the maximum and minimum values ​​in the time series, indicating the maximum and minimum loads that the unit components may encounter under specific working conditions; the standard deviation is obtained by calculating the sum of the squares of the differences between the time series data and the mean and taking the square root. The larger the standard deviation, the more severe the load fluctuation and the more unstable the operating state of the wind turbine may be; the fluctuation amplitude refers to the maximum difference in the load change within a certain period of time. The time domain characteristic information is a statistical feature about time changes extracted from the time series data of the target local load, which is used to describe the regularity and fluctuation of the load change over time.

[0031] Perform a fast Fourier transform (FFT) on the target local load time series to convert the time domain signal to the frequency domain. The target local load spectrum refers to the frequency distribution obtained after transforming the target local load time series data through FFT. The spectrum represents the intensity or amplitude information of the signal at different frequencies, which can reveal the load characteristics of the wind turbine at different frequencies. The characteristics of the spectrum include two main parameters: amplitude and frequency.

[0032] Preprocess the target local load time series, such as removing DC components and filtering noise, to improve the accuracy of spectrum analysis. Determine the sampling frequency of the data, that is, the frequency of data acquisition. This frequency is a value in the FFT algorithm formula. Input the time domain signal (load time series data) into the FFT algorithm. The FFT output result is a complex array, which represents the amplitude and phase information of each frequency component. Extract frequency domain feature information from the target local load spectrum output by FFT, including main frequency, frequency component, amplitude spectrum, etc. By finding the position with the largest amplitude in the spectrum, the main vibration frequency, that is, the main frequency, is obtained, which reflects the main period of force on the wind turbine; by analyzing the spectrum, all the main frequency components in the signal are identified to obtain the frequency components; extract the amplitude of each frequency component to obtain the amplitude spectrum, which represents the intensity of the load at different frequencies. Frequency domain feature information refers to the features extracted from the spectrum, which is used to describe the changes of the load signal in the frequency domain.

[0033] Combine the time domain characteristic information and the frequency domain characteristic information to form the target load characteristic information, which fully reflects the load characteristics of the wind turbine in the target test sector. Through the combined analysis of time domain and frequency domain characteristics, the load characteristics of the wind turbine in a specific area can be fully understood. It can not only analyze the time variation law of the load, but also identify the vibration characteristics of different frequency components, and accurately evaluate the load level of different components of the wind turbine (such as blades, towers, etc.) under specific working conditions.

[0034] S500: Real-time monitoring information of the target wind turbine generator set is obtained through dynamic monitoring on a distributed monitoring platform.

[0035] Specifically, wind turbines are monitored in real time through a distributed monitoring platform, which refers to a system consisting of multiple monitoring devices and sensors distributed in different locations, used to collect and analyze various real-time data of wind turbines. The distributed monitoring platform can install multiple monitoring nodes in different geographical locations to collect information from multiple data sources.

[0036] Various monitoring equipment are deployed in multiple key locations, including tower top sensors, meteorological equipment, operation monitoring equipment, etc., covering all working areas of the wind turbine. Each monitoring equipment (such as meteorological monitoring, operation monitoring, tower top monitoring, etc.) dynamically collects real-time data during the operation of the wind turbine. For example, the tower top monitor can obtain tower top bending moment and tower top torque data; meteorological monitoring equipment provides real-time wind speed and wind force information; operation monitoring equipment provides the unit's active power, yaw azimuth and other data. The real-time monitoring information of the target wind turbine is obtained through the monitoring equipment, covering various status information of the wind turbine during operation, including but not limited to wind speed, wind direction, unit output power, load condition, mechanical operation status, etc.

[0037] Through the distributed monitoring platform, data from various parts of the wind turbine are collected in real time, helping operation and maintenance personnel to understand the working status of the wind turbine at any time, discover abnormalities in time, reduce the probability of failure, and ensure the safe operation of the wind turbine.

[0038] S600: Introducing a wind power load evaluation function to evaluate and analyze the target load characteristic information and the real-time monitoring information to obtain a target wind power load index.

[0039] Specifically, the purpose of the wind power load evaluation function is to integrate data from different sources (such as target load characteristic information and real-time monitoring information) to calculate a comprehensive indicator, namely the target wind power load index, which reflects the load level of the wind turbine in its current state, and then helps to judge whether the unit is safe and whether it can withstand the current wind force and load.

[0040] The target load characteristic information (time domain and frequency domain) and real-time monitoring information (including tower top bending moment, tower top torque, wind force, wind speed, active power and yaw azimuth) are input into the wind power load evaluation function for evaluation and analysis, and the target wind power load index is calculated. The evaluation and analysis specifically refers to the use of the wind power load evaluation function to calculate and analyze the load characteristic information and real-time monitoring information of the wind turbine to obtain a quantitative indicator, namely the target wind power load index. The target wind power load index obtained by analysis and calculation is used to evaluate the load status of the wind turbine, determine whether the load is within the safe range, identify potential risk points and faults, and comprehensively consider different data sources to accurately evaluate the load status of the wind turbine in actual operation, which can not only help the daily maintenance and safety inspection of the wind turbine, but also provide early warning of possible risks under different working conditions.

[0041] Evaluation and analysis specifically refers to the use of wind power load evaluation functions to calculate and analyze the load characteristic information and real-time monitoring information of wind turbines to obtain a quantitative indicator, namely the target wind power load index. The target wind power load index calculated by analysis and calculation can evaluate the load status of wind turbines, determine whether the load is within a safe range, identify potential risk points and faults, comprehensively consider different data sources, and accurately evaluate the load status of wind turbines in actual operation, which can not only help with daily maintenance and safety inspections of wind turbines, but also provide early warning of possible risks under different working conditions.

[0042] Furthermore, the present application S500 includes: The distributed monitoring platform includes a tower top monitor, a meteorological monitoring device and an operation monitoring device; tower top monitoring information is obtained by dynamic monitoring through the tower top monitor, wherein the tower top monitoring information includes tower top bending moment and tower top torque; meteorological monitoring information is obtained by dynamic monitoring through the meteorological monitoring device, wherein the meteorological monitoring information includes wind force and wind speed; operation monitoring information is obtained by dynamic monitoring through the operation monitoring device, wherein the operation monitoring information includes active power and yaw azimuth; the tower top bending moment, the tower top torque, the wind force, the wind speed, the active power and the yaw azimuth constitute the real-time monitoring information.

[0043] Furthermore, the present application also includes the following steps: The meteorological monitoring equipment at least includes a laser wind measuring radar, and the laser wind measuring radar is arranged upwind of the first main wind direction.

[0044] Specifically, the distributed monitoring platform is a monitoring system composed of multiple devices. By deploying multiple monitoring devices (such as tower top monitors, meteorological monitoring equipment and operation monitoring equipment, etc.) at different locations, the operating data of wind turbines are collected in real time. The distributed monitoring platform provides all-round monitoring through multi-point data collection to ensure that the performance, load and environmental factors of the unit can be comprehensively evaluated. The distributed monitoring platform includes tower top monitors, meteorological monitoring equipment and operation monitoring equipment. Among them, the tower top monitor is a monitoring device installed on the top of the wind turbine tower, which is used to monitor important mechanical parameters such as bending moment and torque of the tower top; the meteorological monitoring equipment is used to monitor the meteorological conditions in the area where the wind turbine is located, especially data such as wind force and wind speed, usually including laser wind radar and other wind speed measurement tools; the operation monitoring equipment is used to monitor the working status of the wind turbine, including parameters such as active power and yaw azimuth.

[0045] Bending moment and torque sensors are installed on the top of the wind turbine tower to monitor the stress conditions on the top of the tower in real time, including the bending moment and torque on the top of the tower. The bending moment on the top of the tower is used to measure the degree of bending of the tower top structure under the action of wind, reflecting the impact of the torque caused by wind on the structure of the unit; the torque on the top of the tower is used to measure the degree of torsion of the tower top structure, and to evaluate whether the wind turbine has rotation or uneven stress caused by wind changes.

[0046] LiDAR is a high-precision meteorological monitoring device that uses laser technology to measure the speed of particles in the air (wind speed) to accurately measure wind speed, wind direction, and wind change trends. LiDAR is usually deployed upwind of wind turbines to accurately capture wind flow patterns and wind speed. Meteorological monitoring equipment, such as a LiDAR deployed upwind of the first main wind direction of the target test sector, is used to monitor wind speed and wind force data. LiDAR measures the speed of tiny particles in the air by reflecting a laser beam and then receives the reflected light waves. By comparing the frequency changes of the transmitted and reflected beams, the wind speed and wind direction can be calculated. LiDAR can usually collect wind speed data at a high frequency (such as 1-10 times per second) with an accuracy of several centimeters per second (cm / s), which is very important for capturing rapid changes in wind speed, especially in environments where wind turbines frequently face strong wind changes. By measuring wind speed in real time with a LiDAR, wind speed data for a specific time period can be obtained. At the same time, through multi-point measurement and comparison, the radar can infer the size of the wind force. Assume that a laser wind radar is deployed in a wind farm and the wind speed is measured to be 12.5m / s and the wind force is 50N / m².

[0047] Various working parameters of wind turbines are collected in real time through operation monitoring equipment, such as power meters, sensors, angle measuring instruments, etc. The active power of wind turbines is measured in real time through the power sensors of the operation monitoring equipment, and the active power is calculated by measuring the changes in current and voltage. In wind turbines, the change in active power is one of the important performance indicators, which can reflect the power generation capacity of wind turbines at different wind speeds and is the actual output power of wind turbines converting wind energy into electrical energy. For example, assuming that the rated power of a wind turbine is 2MW, when the wind speed reaches 12m / s, the active power output of the unit may be stable between 1.8MW and 2MW. If the wind speed fluctuates or the blade angle is improperly adjusted, the active power of the unit will also fluctuate.

[0048] The yaw angle refers to the angle between the direction of the wind turbine blades and the wind direction, and the blade angle is adjusted according to the wind direction. The yaw angle monitoring equipment will track the angle between the blades and the wind in real time and transmit the data to the control system. The yaw angle sensor is generally installed on the rotating part of the tower to measure the current yaw angle of the wind turbine. For example, when the wind speed increases from 10m / s to 15m / s, the control system of the wind turbine may adjust the yaw angle to ensure the best wind energy capture effect.

[0049] Through the distributed monitoring platform, the data of tower top bending moment, tower top torque, wind force, wind speed, active power and yaw azimuth are integrated into real-time monitoring information, which comprehensively reflects the stress state, environmental conditions and operating performance of the wind turbine. It can evaluate the operating status of the wind turbine in real time, promptly discover and solve potential problems, and improve the operation and maintenance efficiency and safety of the wind farm.

[0050] Furthermore, the present application S600 includes: The expression of the wind power load evaluation function is as follows: ;in, The target wind turbine The target wind load index, is the activation function, is the weight, and , refers to the time domain feature information, refers to the frequency domain feature information, is the weight, and , Characterizing the real-time monitoring information, It refers to bias.

[0051] Specifically, the target local load time series of the target test sector is collected, the corresponding time domain characteristic information and frequency domain characteristic information are obtained, and the target wind power load index is calculated by the wind power load evaluation function. The expression of the wind power load evaluation function is as follows: . The target wind turbine The target wind power load index combines the multi-dimensional data of time domain characteristic information, frequency domain characteristic information and real-time monitoring information, performs weighted calculation through the evaluation function, and finally outputs a quantitative load index. Through the distributed monitoring platform, the real-time monitoring data of the wind turbine group is dynamically obtained, including tower top bending moment, tower top torque, wind force, wind speed, active power and yaw azimuth.

[0052] It is an activation function, which is used to adjust the overall evaluation function. It can be a common mathematical function, such as sigmoid function, ReLU (linear rectification function), etc. Its main function is to adjust or map the target wind power load index according to the input feature information. The activation function is usually used to ensure that the calculation results meet certain nonlinear requirements and enhance the expressiveness of the model. is the weight, which is used to weight the time domain feature information and the frequency domain feature information respectively. The setting of the weight coefficient needs to be adjusted according to the actual situation, and is usually determined through experiments or optimization algorithms. The total weight of the time domain and frequency domain feature information is 1. It is a weight used to weight the real-time monitoring information. It is usually optimized based on the real-time monitoring information of the wind turbine. The sum of the weights of the various parameters of the real-time monitoring information is 1.

[0053] Refers to the bias, which is used to adjust the output of the evaluation function to meet the actual needs. The time domain feature information and the frequency domain feature information are weighted and summed according to their respective weights, and then the real-time monitoring information is weighted and summed according to their respective weights. Finally, the two summation results are added together, and the bias coefficient is added to obtain the final load index.

[0054] Exemplarily, assuming that the time domain characteristic information is calculated to be 0.8, the frequency domain characteristic information is calculated to be 1.2, the real-time monitoring information is the tower top bending moment of 120 kN·m, the tower top torque of 80 kN·m, the wind speed of 10 m / s, the wind force of 50 N / ㎡, the active power of 2.2 MW, and the yaw azimuth of 15°, assuming that the weight of the time domain characteristic information is 0.6, the weight of the frequency domain characteristic information is 0.4, the wind speed weight is 0.2, the wind force weight is 0.2, the active power weight is 0.1, the yaw azimuth is 0.1, the tower top bending moment is 0.2, the vibration data of the wind turbine is 0.2, the bias term is 0.8, the activation function is 1, and the calculation is 39.52, which represents the load level of the wind turbine in the current working state. By integrating time domain, frequency domain and real-time monitoring data, the wind power load evaluation function provides a comprehensive wind turbine performance evaluation method. By optimizing the weight coefficient and activation function, the wind power load evaluation function is more in line with the actual unit operation conditions, which helps to improve the power generation efficiency of the wind farm, reduce fault downtime, and enhance the operating stability of the unit.

[0055] Furthermore, the present application also includes the following steps: Obtain a predetermined operating state, wherein the predetermined operating state corresponds to predetermined operating information, wherein the predetermined operating information includes a predetermined active power and a predetermined yaw azimuth; adjust the real-time monitoring information with the predetermined active power and the predetermined yaw azimuth to obtain predetermined monitoring information; and determine a predetermined wind power load index according to the predetermined monitoring information.

[0056] Furthermore, the present application also includes the following steps: The predetermined operating state includes a predetermined stationary state, a predetermined pitch change state, a predetermined flow change state and a predetermined emergency stop state.

[0057] Specifically, determine the predetermined operating states that the wind turbine needs to execute. These states are usually set according to the operating conditions, environmental factors or emergencies of the wind turbine. The predetermined operating state refers to different working modes preset according to the specific operating requirements or conditions of the wind turbine, which are used to adjust and optimize the unit under specific environments, including a predetermined static state, a predetermined pitch state, a predetermined current-changing state and a predetermined emergency stop state. The predetermined static state means that the wind turbine does not rotate, which is usually used for unit shutdown maintenance or to deal with extreme weather conditions; the predetermined pitch state means that the pitch angle of the wind turbine blades is adjusted to optimize the output power of the wind turbine or reduce the load; the predetermined current-changing state means that the power generation system of the wind turbine is adjusted to adapt to different wind speeds or load conditions; the predetermined emergency stop state means that the wind turbine stops operating quickly due to a fault or safety consideration.

[0058] According to the predetermined operating information corresponding to each predetermined operating state, the predetermined active power and the predetermined yaw azimuth are determined. The predetermined active power is the expected output power of the wind turbine in this specific operating state. The predetermined yaw azimuth is the yaw angle of the blades and tower of the wind turbine, which determines the direction in which the turbine faces the wind.

[0059] According to the predetermined active power and yaw azimuth, the real-time monitoring information is adjusted to obtain the predetermined monitoring information, which reflects the performance of the unit under the predetermined operating state. The adjusted monitoring information is substituted into the wind power load evaluation function to calculate the predetermined wind power load index, which reflects the load level of the wind turbine under the operating state. The wind power load index is the wind power load index recalculated after considering the adjustment of the operating parameters under the predetermined operating state, and is used to evaluate the load level of the wind turbine under these conditions. By obtaining the predetermined operating state and the corresponding predetermined operating information, and adjusting the real-time monitoring information to obtain the predetermined monitoring information, the load condition of the wind turbine under specific conditions can be accurately evaluated, which helps the operation and maintenance personnel to better understand the operating characteristics of the wind turbine, optimize the operating parameters, and improve the power generation efficiency and safety of the wind farm.

[0060] In summary, the wind power load testing method based on multi-source data analysis provided by this application has the following technical effects: By obtaining a target sector set, the target sector set refers to a set of sectors obtained by dividing the target neighborhood of the target wind turbine; obtaining the first multidimensional feature information of the first sector, wherein the first sector refers to any sector in the target sector set; if the first multidimensional feature information meets the predetermined test constraints, the first sector is used as the target test sector; a local load test is performed on the target test sector to obtain the target local load time series, and the target load feature information is collected; the real-time monitoring information of the target wind turbine is obtained through dynamic monitoring on a distributed monitoring platform; a wind power load evaluation function is introduced to evaluate and analyze the target load feature information and the real-time monitoring information, and a target wind power load index is obtained. In other words, by dividing the target sector set of the target wind turbine and performing detailed local load tests on the sectors, the multidimensional feature information of the target sector is obtained, the operating status of the wind turbine is monitored in real time, and the load condition of the wind turbine is evaluated, thereby improving the accuracy and efficiency of the wind power load test.

[0061] Embodiment 2, based on the same inventive concept as the wind power load test method based on multi-source data analysis in the aforementioned embodiment 1, the present application also provides a wind power load test system based on multi-source data analysis, please refer to the attached Figure 2 , the wind power load test system based on multi-source data analysis includes: A region acquisition module 11 is used to acquire a target sector set, wherein the target sector set refers to a set of sectors obtained by dividing a target neighborhood of a target wind turbine; a feature acquisition module 12 is used to acquire first multidimensional feature information of a first sector, wherein the first sector refers to any sector in the target sector set; a target region determination module 13 is used to use the first sector as a target test sector if the first multidimensional feature information meets a predetermined test constraint; a load feature collection module 14 is used to perform a local load test on the target test sector to obtain a target local load time series, and collect target load feature information; a monitoring information acquisition module 15 is used to obtain real-time monitoring information of the target wind turbine through dynamic monitoring on a distributed monitoring platform; an evaluation and analysis module 16 is used to introduce a wind power load evaluation function to evaluate and analyze the target load feature information and the real-time monitoring information to obtain a target wind power load index.

[0062] Furthermore, the feature acquisition module 12 in the wind power load test system based on multi-source data analysis is also used for: Obtain historical meteorological records of the target wind turbine, wherein the historical meteorological records include a first main wind direction of the first sector; obtain first environmental obstacle information of the first sector; obtain first terrain information of the first sector; and compose the first multi-dimensional feature information based on the first main wind direction, the first environmental obstacle information and the first terrain information.

[0063] Furthermore, the load characteristic collection module 14 in the wind power load test system based on multi-source data analysis is also used for: Acquire the time domain characteristic information of the target local load time series; acquire the frequency domain characteristic information of the target local load spectrum, wherein the target local load spectrum refers to the spectrum obtained by fast Fourier transforming the target local load time series; the time domain characteristic information and the frequency domain characteristic information constitute the target load characteristic information.

[0064] Furthermore, the monitoring information acquisition module 15 in the wind power load test system based on multi-source data analysis is also used for: The distributed monitoring platform includes a tower top monitor, a meteorological monitoring device and an operation monitoring device; tower top monitoring information is obtained by dynamic monitoring through the tower top monitor, wherein the tower top monitoring information includes tower top bending moment and tower top torque; meteorological monitoring information is obtained by dynamic monitoring through the meteorological monitoring device, wherein the meteorological monitoring information includes wind force and wind speed; operation monitoring information is obtained by dynamic monitoring through the operation monitoring device, wherein the operation monitoring information includes active power and yaw azimuth; the tower top bending moment, the tower top torque, the wind force, the wind speed, the active power and the yaw azimuth constitute the real-time monitoring information.

[0065] Furthermore, the monitoring information acquisition module 15 in the wind power load test system based on multi-source data analysis is also used for: The meteorological monitoring equipment at least includes a laser wind measuring radar, and the laser wind measuring radar is arranged upwind of the first main wind direction.

[0066] Furthermore, the evaluation and analysis module 16 in the wind power load test system based on multi-source data analysis is also used for: The expression of the wind power load evaluation function is as follows: ;in, The target wind turbine The target wind load index, is the activation function, is the weight, and , refers to the time domain feature information, refers to the frequency domain feature information, is the weight, and , Characterizing the real-time monitoring information, It refers to bias.

[0067] Furthermore, the wind power load test system based on multi-source data analysis further includes a predetermined monitoring module, and the predetermined monitoring module is used to: Obtain a predetermined operating state, wherein the predetermined operating state corresponds to predetermined operating information, wherein the predetermined operating information includes a predetermined active power and a predetermined yaw azimuth; adjust the real-time monitoring information with the predetermined active power and the predetermined yaw azimuth to obtain monitoring information; and determine a predetermined wind power load index according to the predetermined monitoring information.

[0068] Furthermore, the predetermined monitoring module in the wind power load test system based on multi-source data analysis is also used for: The predetermined operating state includes a predetermined stationary state, a predetermined pitch change state, a predetermined flow change state and a predetermined emergency stop state.

[0069] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. Figure 1The wind power load test method based on multi-source data analysis and the specific examples in the first embodiment are also applicable to the wind power load test system based on multi-source data analysis in this embodiment. Through the above detailed description of the wind power load test method based on multi-source data analysis, those skilled in the art can clearly know the wind power load test system based on multi-source data analysis in this embodiment, so for the sake of brevity of the specification, it will not be described in detail here. For the system disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part description.

[0070] The above description of the disclosed embodiments enables professionals in the field to implement or use the present application. Various modifications to these embodiments will be apparent to professionals in the field, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein.

[0071] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application belong to the scope of the present application and its equivalent technology, the present application is also intended to include these modifications and variations.

Claims

1. A wind power load test method based on multi-source data analysis, characterized in that: include: Acquire a target sector set, where the target sector set refers to a set of sectors obtained by dividing a target neighborhood of a target wind turbine generator set; Acquire first multi-dimensional feature information of a first sector, wherein the first sector refers to any sector in the target sector set; If the first multi-dimensional feature information meets the predetermined test constraint, taking the first sector as a target test sector; Performing a local load test on the target test sector to obtain a target local load time sequence, and collecting target load characteristic information; Obtaining real-time monitoring information of the target wind turbine group through dynamic monitoring on a distributed monitoring platform; Introducing a wind power load evaluation function to evaluate and analyze the target load characteristic information and the real-time monitoring information to obtain a target wind power load index; The step of obtaining first multi-dimensional feature information of the first sector includes: Acquire historical meteorological records of the target wind turbine generator set, wherein the historical meteorological records include a first main wind direction of the first sector; Acquire first environmental obstacle information of the first sector; Acquire first terrain information of the first sector; The first multi-dimensional feature information is composed based on the first main wind direction, the first environmental obstacle information and the first terrain information.

2. The wind power load testing method based on multi-source data analysis according to claim 1 is characterized in that: Performing a local load test on the target test sector to obtain a target local load timing, and collecting target load characteristic information, including: Acquiring time domain characteristic information of the target local load time series; Acquire frequency domain characteristic information of a target local load spectrum, wherein the target local load spectrum refers to a spectrum obtained by fast Fourier transforming the target local load time series; The time domain characteristic information and the frequency domain characteristic information constitute the target load characteristic information.

3. The wind power load testing method based on multi-source data analysis according to claim 2 is characterized in that: The distributed monitoring platform includes a tower top monitor, a meteorological monitoring device and an operation monitoring device. The real-time monitoring information of the target wind turbine set obtained by dynamic monitoring via the distributed monitoring platform includes: Dynamically monitoring the tower top monitor to obtain tower top monitoring information, wherein the tower top monitoring information includes tower top bending moment and tower top torque; Dynamically monitoring the meteorological monitoring equipment to obtain meteorological monitoring information, wherein the meteorological monitoring information includes wind force and wind speed; Dynamically monitoring the operation monitoring device to obtain operation monitoring information, wherein the operation monitoring information includes active power and yaw azimuth; The tower top bending moment, the tower top torque, the wind force, the wind speed, the active power and the yaw azimuth constitute the real-time monitoring information.

4. The wind power load testing method based on multi-source data analysis according to claim 3 is characterized in that: The meteorological monitoring equipment at least includes a laser wind measuring radar, and the laser wind measuring radar is arranged upwind of the first main wind direction.

5. The wind power load testing method based on multi-source data analysis according to claim 3 is characterized in that: The expression of the wind power load evaluation function is as follows: ； in, The target wind turbine The target wind load index is is the activation function, is the weight, and , refers to the time domain feature information, refers to the frequency domain feature information, is the weight, and , Characterizing the real-time monitoring information, It refers to bias.

6. The wind power load testing method based on multi-source data analysis according to claim 1 is characterized in that: Also includes: Acquire a predetermined operating state, where the predetermined operating state corresponds to predetermined operating information, wherein the predetermined operating information includes a predetermined active power and a predetermined yaw azimuth angle; The real-time monitoring information is adjusted according to the predetermined active power and the predetermined yaw azimuth to obtain predetermined monitoring information; A predetermined wind power load index is determined according to the monitoring information.

7. The wind power load testing method based on multi-source data analysis according to claim 6 is characterized in that: The predetermined operating state includes a predetermined stationary state, a predetermined pitch change state, a predetermined flow change state and a predetermined emergency stop state.

8. The wind power load test system based on multi-source data analysis is characterized by: The steps for implementing the wind power load testing method based on multi-source data analysis according to any one of claims 1 to 7, the wind power load testing system based on multi-source data analysis comprises: A region acquisition module, used to acquire a target sector set, wherein the target sector set refers to a set of sectors obtained by dividing a target neighborhood of a target wind turbine generator set; A feature acquisition module, used to acquire first multi-dimensional feature information of a first sector, wherein the first sector refers to any sector in the target sector set; a target area determination module, configured to use the first sector as a target test sector if the first multi-dimensional feature information meets a predetermined test constraint; A load characteristic collection module, used for performing a local load test on the target test sector to obtain a target local load timing, and collecting target load characteristic information; A monitoring information acquisition module, used to obtain real-time monitoring information of the target wind turbine group through dynamic monitoring on a distributed monitoring platform; The evaluation and analysis module is used to introduce a wind power load evaluation function to evaluate and analyze the target load characteristic information and the real-time monitoring information to obtain a target wind power load index.

Images