A wind turbine tower tilt monitoring method, device and system

Abstract

The present invention relates to the field of exploration and measurement technology, and proposes a method, device and system for monitoring the inclination of a wind turbine tower, including: collecting tower inclination data and wind speed data of multiple dimensions of a wind turbine tower; dividing each day into several time periods, obtaining the wind speed correlation and abnormal possibility of the tower inclination data of each dimension in each time period; obtaining the tower abnormality influencing factor of the basic inclination angle in each time period; obtaining the tower abnormality influencing factor of the tower shaking amplitude in each time period, and the tower abnormality influencing factor of the absolute displacement of the tower top in each time period; analyzing the tower inclination data of different dimensions in each time period, obtaining the overall inclination coefficient at each moment, and completing the inclination monitoring of the wind turbine tower. The present invention aims to solve the problem that the monitoring data of the wind turbine tower inclination monitoring process is relatively single, which affects the accuracy of safety warning.

Description

Technical Field

[0001] The present invention relates to the field of exploration and measurement technology, and in particular to a method, device and system for monitoring the inclination of a wind power tower. Background Art

[0002] With the rapid development of wind power, communications and other industries, the tower, as a supporting component of the wind turbine, has a direct impact on the normal operation of the wind turbine and the safe production of the entire wind farm. Therefore, it is particularly important to establish an efficient and accurate tower safety monitoring and early warning system. During the operation of wind turbine towers, due to the influence of various factors such as basic geological conditions, temperature, wind load, sun exposure, rain, corrosion, etc., it is easy to tilt. Traditional tower tilt monitoring methods often rely on manual operation, which makes it difficult to achieve real-time monitoring and remote monitoring, and cannot meet the needs of modern wind farms for intelligence and automation.

[0003] With the continuous development of technologies such as the Internet of Things, big data, and cloud computing, intelligent monitoring has become an important trend in wind turbine tower tilt monitoring; by installing sensors and other monitoring equipment, real-time monitoring of tower tilt can be achieved. Existing tower safety monitoring generally monitors the vibration or tilt of the tower itself, and does not monitor vibration, tilt, and shaking at the same time. The monitoring data is relatively single, and it is impossible to make an overall judgment on the cause of tower tilt, and the accuracy of early warning is poor.

[0004] In actual operation of wind turbine towers, the foundation tilt angle, tower sway amplitude and tower top absolute displacement interact and influence each other, jointly determining the stability and safety of the tower. Therefore, the overall tilt coefficient of the tower is obtained through data fusion processing and analysis, and the abnormality of a single type of data also reflects the problem of the tower in a certain aspect. Even if the overall tilt coefficient is within the safe range, the abnormality of a single type of data may have an adverse effect on the overall safety of the tower. Therefore, it is also necessary to jointly analyze and adjust the overall tilt coefficient of multiple types of data to ensure the accuracy of the safety warning of wind turbine tower tilt monitoring. Summary of the invention

[0005] The present invention provides a wind turbine tower tilt monitoring method, device and system to solve the problem that the monitoring data of the existing wind turbine tower tilt monitoring process is relatively single and affects the accuracy of safety warning. The technical solution adopted is as follows:

[0006] The present invention proposes a method for monitoring the inclination of a wind turbine tower, which comprises the following steps:

[0007] Collecting tower tilt data of multiple dimensions of wind turbine towers at each moment for several days, as well as wind speed data at each moment; the tower tilt data includes the base tilt angle, tower sway amplitude and tower top absolute displacement;

[0008] Each day is divided into several time periods. Based on the similarity of the changing trends between the tower tilt data and the wind speed data in each dimension in each time period, the wind speed correlation of the tower tilt data in each dimension in each time period is obtained. Based on the data distribution on the left and right sides of each maximum point in the tower tilt data in each dimension of each day, and the difference between the wind speed data corresponding to the left and right sides, the abnormal possibility of the tower tilt data in each dimension of each day is obtained.

[0009] Based on the abnormal possibility and wind speed correlation of the tower tilt data in each dimension every day, the tower abnormality influencing factor of the basic tilt angle in each time period is obtained; combined with the similarity relationship between the change of the tower sway amplitude and the absolute displacement of the tower top in each time period, the similarity relationship between the change of the absolute displacement of the tower top and the basic tilt angle, and the wind speed correlation of the tower tilt data in each dimension in each time period, the tower abnormality influencing factor of the tower sway amplitude in each time period and the tower abnormality influencing factor of the absolute displacement of the tower top in each time period are obtained respectively;

[0010] According to the tower abnormality influencing factors of the tower tilt data in each dimension of each time period, the tower tilt data of different dimensions in each time period are analyzed, and the overall tilt coefficient at each moment is obtained by weighted analysis to complete the tilt monitoring of the wind turbine tower.

[0011] Optionally, the specific method of obtaining the wind speed correlation of the tower tilt data in each dimension in each time period includes:

[0012] According to the tower tilt data of each dimension in each time period and the wind speed data of each time period, obtain the tower tilt data sequence and tower tilt change sequence of each dimension in each time period, as well as the wind speed sequence and wind speed change sequence of each event segment;

[0013] No. Tiandi Time period Wind speed correlation of tower tilt data in 4 dimensions The calculation method is:

[0014]

[0015] in, Indicates Tiandi Time period The tower tilt change sequence in the first dimension is Tiandi The Pearson correlation coefficient of the wind speed change series in each time period is Indicates Tiandi The time period is The order value in all time periods of the day, Indicates the number of time periods per day.

[0016] Optionally, the tower tilt data sequence and tower tilt change sequence in each dimension of each time period, and the wind speed sequence and wind speed change sequence of each event segment are specifically obtained by:

[0017] Based on the tower tilt data of each dimension in each time period, a tower tilt data sequence of each dimension in each time period is obtained by arranging the data in time sequence, and a wind speed sequence of each time period is obtained based on the wind speed data;

[0018] For each dimension of the tower tilt data sequence and wind speed sequence, the slope of each element is obtained respectively, and the sequence obtained by sorting the slopes of each element according to the time sequence of the corresponding time period is recorded as the tower tilt change sequence and wind speed change sequence of each dimension in each time period respectively.

[0019] Optionally, the specific method of obtaining the abnormal possibility of the tower tilt data in each dimension every day includes:

[0020] For Tiandi The tower tilt data of all dimensions are used to obtain a number of maximum points, and the maximum points are clustered to obtain a number of clusters. The cluster with the largest number of maximum points among all clusters is taken as the first cluster. Tiandi The representative extreme points of each dimension;

[0021] For Tiandi The dimension a representative extreme value point, respectively obtaining the number of tower inclination data between the representative extreme value point and the maximum value points adjacent to the left and right sides, and taking the minimum value of the numbers as the reference number of the representative extreme value point;

[0022] No. Tiandi The abnormal possibility of tower tilt data in each dimension The calculation method is:

[0023]

[0024] in, Indicates Tiandi The number of representative extreme points in the dimension, Indicates Tiandi The dimension The reference number representing the extreme points, and Respectively represent Tiandi The dimension represents the left and right sides of the extreme point Tower tilt data, and Respectively represent Tiandi The dimension represents the left and right sides of the extreme point The wind speed data at the corresponding moment of the tower tilt data, represents the absolute value function.

[0025] Optionally, the specific method of obtaining the tower abnormality influence factor of the foundation tilt angle in each time period includes:

[0026] when At that time, The tower inclination data of the first dimension is used as the basic inclination angle; Tiandi The abnormal possibility of tower tilt data in the first dimension is related to the Tiandi Time period The ratio of the wind speed correlation of the tower tilt data in the first dimension is taken as Tiandi The tower abnormality influencing factor of the foundation inclination angle within a time period.

[0027] Optionally, the tower abnormality influencing factor of the tower sway amplitude in each time period is specifically obtained by:

[0028] For Tiandi Time period The tower tilt data in dimensions At that time, The tower tilt data in the dimension is the tower sway amplitude. Tiandi Tower anomaly influencing factors of tower sway amplitude in a time period The calculation method is:

[0029]

[0030] in, Indicates The possibility of abnormal tower sway amplitude within a day, Indicates Tiandi The wind speed correlation of the tower sway amplitude in a time period is Indicates Tiandi The Pearson correlation coefficient between the tower sway amplitude and the tower tilt data series of the absolute displacement of the tower top in a time period is represents the absolute value function.

[0031] Optionally, the tower abnormality influencing factor of the absolute displacement of the tower top in each time period is specifically obtained by:

[0032] For Tiandi Time period The tower tilt data in dimensions At that time, The tower tilt data in the dimension is the absolute displacement of the tower top. Tiandi Tower abnormality influencing factors of absolute displacement of tower top in a time period The calculation method is:

[0033]

[0034] in, Indicates The possibility of anomalies in the absolute displacement of the tower top within a day, Indicates Tiandi The wind speed correlation of the absolute displacement of the tower top in a time period, Indicates Tiandi The Pearson correlation coefficient between the tower top absolute displacement and the tower inclination data series of the base inclination angle in a time period.

[0035] Optionally, the overall tilt coefficient at each moment is obtained by:

[0036] The tower tilt data in three dimensions are analyzed by AHP algorithm and PCA algorithm, and the initial weights of the tower tilt data in three dimensions are obtained respectively;

[0037] For any moment, the time period in which the moment is located The tower anomaly influencing factors of the tower tilt data in the first dimension are The product of the initial weights of the tower tilt data in dimensions is used as the The adjustment weight of the tower tilt data in each dimension;

[0038] Based on the adjustment weights of the tower tilt data in each dimension at that moment, the tower tilt data in each dimension at that moment are weighted and summed, and the result is used as the overall tilt coefficient at that moment.

[0039] The present invention also proposes a wind power tower tilt monitoring device, which comprises:

[0040] A tower tilt data acquisition module, a tower tilt data analysis module and a tower tilt monitoring module. The tower tilt data acquisition module is used to collect multi-dimensional tower tilt data and wind speed data of the wind turbine tower. The tower tilt data analysis module implements the steps of the above method by calling a computer program to obtain the overall tilt coefficient of the wind turbine tower at each moment. The tower tilt monitoring module performs tilt monitoring and early warning of the wind turbine tower based on the overall tilt coefficient.

[0041] The present invention also proposes a wind turbine tower tilt monitoring system, which includes a memory, a processor, and a computer program stored in the memory and running on the processor, and the processor implements the steps of the above method when executing the computer program.

[0042] The beneficial effects of the present invention are as follows: the present invention quantifies the wind speed correlation between the tower tilt data and wind speed data of multiple dimensions of the wind turbine tower, and simultaneously analyzes the anomalies generated by the tower tilt data itself when not affected by the wind speed, thereby obtaining the possibility of anomalies; and further analyzes the mutual influence between the tower tilt data of different dimensions based on the similar relationship of changes therebetween; wherein the change trend of the tower tilt data of each dimension and the change trend of the wind speed data are quantified by similarity, thereby obtaining the wind speed correlation that can reflect the degree to which the tower tilt data of each dimension in each time period is affected by the wind speed; and then, utilizing the characteristics that the wind speed has a short-term impact on the tower tilt data of each dimension and then recovers, analyzes the possibility of anomalies of the tower tilt data of each dimension every day when it is not affected by the wind speed, thereby obtaining the possibility of anomalies of the tower tilt data of each dimension every day; thereby reducing the interference of mutual influence on the overall tilt monitoring quantification process, so that the overall tilt coefficient is analyzed for anomalies based on the tower tilt data of each dimension itself, and ultimately improving the accuracy of wind turbine tower tilt monitoring and the timeliness of early warning. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in 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 described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0044] Figure 1 A schematic flow chart of a method for monitoring wind turbine tower inclination provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0045] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0046] See also Figure 1 , which shows a flow chart of a method for monitoring the inclination of a wind turbine tower provided by an embodiment of the present invention, the method comprising the following steps:

[0047] Step S001: Collect tower tilt data of wind turbine towers in multiple dimensions at each moment for several days, as well as wind speed data at each moment.

[0048] The purpose of this embodiment is to fuse multi-dimensional data of wind turbine towers, so as to realize the monitoring of wind turbine tower inclination. Therefore, it is necessary to obtain the basic inclination angle of the tower separately, which is a key parameter for evaluating the stability of the tower; the tower sway amplitude is used to monitor the sway amplitude of the tower under the action of external forces such as wind and earthquake, which can reflect the dynamic response characteristics and structural strength of the tower; at the same time, it is also necessary to monitor the absolute displacement of the tower top, that is, to monitor the overall deformation of the tower; and the wind speed change has an impact on the data of multiple dimensions of the wind turbine tower, so it is also necessary to obtain wind speed data.

[0049] Specifically, this embodiment collects the basic tilt angle of the tower, the tower sway amplitude and the absolute displacement of the tower top through the inclination sensor, the sway displacement sensor and the Beidou positioning and orientation monitoring unit arranged on the wind turbine tower, and performs numerical normalization processing on the collected data of each dimension (the data of the same dimension is linearly normalized), and uses the processed data as the basic tilt angle, tower sway amplitude and tower top absolute displacement in the tower tilt data, that is, the basic tilt data, the tower sway amplitude and the tower top absolute displacement are the tower tilt data of one dimension respectively, and the tower tilt data of multiple dimensions are obtained; the tower tilt data of the last 30 days are collected, and the sampling interval is set to 10 minutes; the wind speed is obtained at the same time, the sampling time and sampling interval of the wind speed are the same as those of the tower tilt data, and the wind speed is standardized to obtain the wind speed data.

[0050] At this point, the tower tilt data and wind speed data in multiple dimensions at each moment are obtained.

[0051] Step S002, divide each day into several time periods, and obtain the wind speed correlation of the tower tilt data in each dimension in each time period based on the similarity between the changing trends of the tower tilt data and the wind speed data in each dimension in each time period; obtain the abnormal possibility of the tower tilt data in each dimension of each day based on the data distribution on the left and right sides of each maximum point in the tower tilt data in each dimension of each day, and the difference between the wind speed data corresponding to the data on the left and right sides.

[0052] It should be noted that the inclination of the foundation will directly affect the overall stability of the tower. If the foundation inclination angle is too large, the tower may not be able to withstand wind loads and other external forces, thereby increasing the risk of collapse; wind force may cause the tower to tilt, especially under extreme weather conditions such as strong winds, typhoons, earthquakes and other severe weather conditions, which will aggravate the inclination of the tower; and the tower sway amplitude reflects the dynamic response of the tower under external forces such as wind loads and mechanical vibrations. By comparing the sway amplitudes at different wind speeds in different time periods, the greater the wind speed and the greater the wind load, the greater the sway amplitude of the tower will be accordingly; when the wind speed increases, the thrust of the wind on the tower and the top of the tower will also increase accordingly, resulting in a greater displacement of the top of the tower; when the basic conditions such as the structure of the tower are good, the fluctuation of wind speed often shows a trend close to or similar to the fluctuation of the absolute displacement of the tower top.

[0053] Preferably, in one embodiment of the present invention, each day is divided into several time periods, and the wind speed correlation of the tower tilt data in each dimension in each time period is obtained according to the similarity relationship between the change trends of the tower tilt data in each dimension and the wind speed data in each time period, including the specific method of:

[0054] In this embodiment, the preset division time length is two hours. Several moments of each day are divided based on the division time length to obtain several time periods of each day. Based on the tower tilt data of each dimension in each time period, the tower tilt data sequence of each dimension in each time period is obtained in time series, and the wind speed sequence of each time period is obtained based on the wind speed data. The slope of each element of the tower tilt data sequence and wind speed sequence in each dimension is obtained respectively. It should be noted that the ratio of the difference obtained by subtracting the previous element from the latter element to the time difference between the two elements is used as the slope of the latter element. The slope is calculated for the first element in each sequence and the last element in the previous sequence, and the slope is not calculated for the first data in the tower tilt data of each dimension and the first data in the wind speed data; the sequence obtained by sorting the slope of each element according to the time series of the corresponding time period is recorded as the tower tilt change sequence of each dimension in each time period and the wind speed change sequence.

[0055] Further, Tiandi Time period Wind speed correlation of tower tilt data in 4 dimensions The calculation method is:

[0056]

[0057] in, Indicates Tiandi Time period The tower tilt change sequence in the first dimension is Tiandi The Pearson correlation coefficient of the wind speed change series in each time period is Indicates Tiandi The time period is The order value in all time periods of the day, Represents the number of time periods in a day; the wind speed correlation of the tower tilt data in each dimension of each time period of each day is obtained according to the above method.

[0058] It should be noted that the greater the correlation between the tower inclination change sequence and the wind speed change sequence, that is, the greater the similarity of the slope changes of the elements, the closer the change trends of the tower inclination data and the wind speed data in the corresponding dimensions are, and the greater the corresponding impact of the wind speed, the greater the wind speed correlation; and in a day, the later the time period, the greater the impact of the accumulated wind speed before, and a larger weight should be given to increase the wind speed correlation.

[0059] It should be noted that, taking the basic tilt angle as an example, changes in wind speed will affect the basic tilt angle of the tower. However, the impact is only for a period of time. If the wind speed decreases, the basic tilt angle will return to its previous state after the change. The analysis is based on the maximum point of the basic tilt angle within a day. The closer the wind speed data changes corresponding to the data on the left and right sides of the maximum point are, the greater the possibility of data changes caused by the influence of wind speed. On the contrary, the greater the difference, the tilt is not caused by the influence of wind speed, and it is more likely to be an abnormality. Similarly, the relationship between the tower sway amplitude and the absolute displacement of the tower top and the wind speed is also the case. This is used to obtain the possibility of abnormalities in the tower tilt data in each dimension every day.

[0060] Preferably, in one embodiment of the present invention, the abnormal possibility of the tower tilt data in each dimension of each day is obtained according to the data distribution on the left and right sides of each maximum point in the tower tilt data in each dimension of each day, and the difference between the wind speed data corresponding to the data on the left and right sides, and the specific method includes:

[0061] For Tiandi All the tower tilt data in dimensions are collected, and several maximum points are obtained through the AMPD algorithm based on the time series order, where the AMPD algorithm is a multi-scale peak automatic search algorithm; DBSCAN clustering is performed on all the obtained maximum points, and the distance measurement uses the absolute value of the difference between the data corresponding to the maximum points, and several clusters are obtained; the cluster with the largest number of maximum points in all clusters is taken as the Tiandi The representative extreme point of the dimension; Tiandi The dimension The number of tower inclination data between the representative extreme point and the maximum value points adjacent to the left and right sides is obtained respectively, and the minimum value of the two numbers is used as the reference number of the representative extreme point; in particular, if there is no maximum value point on the left or right side of the representative extreme point, it is the first The first maximum point or the last maximum point of all the maximum points in the tower tilt data of the dimension is used for subsequent comparison based on the number of tower tilt data actually existing on the left or right side.

[0062] Further, Tiandi The abnormal possibility of tower tilt data in each dimension The calculation method is:

[0063]

[0064] in, Indicates Tiandi The number of representative extreme points in the dimension, Indicates Tiandi The dimension The reference number representing the extreme points, and Respectively represent Tiandi The dimension represents the left and right sides of the extreme point Tower tilt data, and Respectively represent Tiandi The dimension represents the left and right sides of the extreme point The wind speed data at the corresponding moment of the tower tilt data, Represents the absolute value function; the abnormal possibility of the tower tilt data in each dimension every day is obtained according to the above method.

[0065] What needs to be explained is that if the tower tilt data of the same reference number on the left and right sides representing the extreme point have the same time interval as the extreme point, then the smaller the difference in the tower tilt data, the greater the reference value for comparing the wind speed data at the corresponding moment; and when the difference in the wind speed data is smaller, the difference in the tower tilt data is smaller or even tends to 0, which indicates that the change in the tower tilt data at this position on the left and right sides is greatly affected by the wind speed, and the corresponding tower itself is less likely to have an abnormality; then, under the premise of limiting the reference value of the wind speed data difference by the difference between the tower tilt data, the greater the difference in the wind speed data, the less likely it is to be affected by the wind speed, and the greater the possibility of an abnormality.

[0066] At this point, we first quantify the similarity between the changing trends of the tower tilt data in each dimension and the changing trends of the wind speed data, so as to obtain the wind speed correlation that can reflect the degree to which the tower tilt data in each dimension in each time period is affected by the wind speed; then, using the characteristics that the wind speed has a short-term impact on the tower tilt data in each dimension and then recovers, we analyze the possibility of abnormalities in the tower tilt data of each dimension every day when it is not affected by the wind speed, so as to obtain the abnormal possibility of the tower tilt data in each dimension every day.

[0067] Step S003, based on the abnormal possibility and wind speed correlation of the tower tilt data in each dimension of each day, obtain the tower abnormality influence factor of the basic tilt angle in each time period; combined with the change similarity relationship between the tower sway amplitude and the absolute displacement of the tower top in each time period, the change similarity relationship between the absolute displacement of the tower top and the basic tilt angle, and the wind speed correlation of the tower tilt data in each dimension of each time period, respectively obtain the tower abnormality influence factor of the tower sway amplitude in each time period, and the tower abnormality influence factor of the absolute displacement of the tower top in each time period.

[0068] Preferably, in one embodiment of the present invention, based on the abnormal possibility and wind speed correlation of the tower tilt data in each dimension every day, the tower abnormality impact factor of the basic tilt angle in each time period is obtained, and the specific method includes:

[0069] when At that time, The tower tilt data in the dimension is the basic tilt angle; At that time, The tower tilt data in the dimension is the tower sway amplitude; At that time, The tower inclination data of the dimension is the absolute displacement of the tower top; Tiandi The abnormal possibility of tower tilt data in each dimension ( Time The probability of abnormal base tilt angle within a day) and Tiandi Time period Wind speed correlation of tower tilt data in 2 dimensions ( Time Tiandi The ratio of the wind speed correlation of the basic tilt angle in the time period is taken as the first Tiandi The tower abnormality influencing factor of the foundation inclination angle within a time period.

[0070] It should be noted that the foundation tilt angle will directly affect the tilt performance of the tower. The ratio of the abnormal possibility and the wind speed correlation is directly used as the tower abnormality impact factor of the foundation tilt angle in each time period. The smaller the wind speed impact and the greater the abnormal possibility, the greater the tower abnormality impact factor.

[0071] It should be noted that when the foundation of the tower is tilted, due to the integrity and continuity of the tower structure, this tilt will be transmitted upward along the tower, causing corresponding displacement of the tower top. In the process of quantifying the tower abnormality influencing factor of the absolute displacement of the tower top, the greater the similarity between the change in the absolute displacement of the tower top and the foundation inclination angle, the smaller the abnormal impact of the absolute displacement of the tower top itself. The amplitude of the tower sway is also affected by the absolute displacement of the tower top. The larger the absolute displacement of the tower top, the greater the corresponding tower sway. It is also necessary to avoid the impact of this similarity of change on the abnormality.

[0072] Preferably, in one embodiment of the present invention, the tower abnormality influence factor of the tower sway amplitude in each time period and the tower top absolute displacement in each time period are obtained by combining the change similarity relationship between the tower sway amplitude and the tower top absolute displacement in each time period, the change similarity relationship between the tower top absolute displacement and the base tilt angle, and the wind speed correlation of the tower tilt data in each dimension in each time period, respectively, and the tower abnormality influence factor of the tower sway amplitude in each time period and the tower abnormality influence factor of the tower top absolute displacement in each time period are obtained, and the specific method includes:

[0073] For Tiandi Time period The tower tilt data in dimensions At that time, The tower tilt data of the dimension is the tower sway amplitude, then Tiandi Tower anomaly influencing factors of tower sway amplitude in a time period The calculation method is:

[0074]

[0075] in, Indicates The possibility of abnormal tower sway amplitude within a day ( Time Tiandi The abnormal possibility of tower tilt data in each dimension), Indicates Tiandi Wind speed correlation of tower sway amplitude in a time period ( Time Tiandi Time period wind speed correlation of tower tilt data in 2 dimensions), Indicates Tiandi The Pearson correlation coefficient between the tower sway amplitude and the tower tilt data series of the absolute displacement of the tower top in a time period is Represents the absolute value function; according to the above method, the tower abnormality influencing factor of the tower sway amplitude in each time period of each day is obtained.

[0076] Similarly, for Tiandi Time period The tower tilt data in dimensions At that time, The tower tilt data in the dimension is the absolute displacement of the tower top, then Tiandi Tower abnormality influencing factors of absolute displacement of tower top in a time period The calculation method is:

[0077]

[0078] in, Indicates The possibility of abnormal absolute displacement of the tower top within a day ( Time Tiandi The abnormal possibility of tower tilt data in each dimension), Indicates Tiandi Wind speed correlation of absolute displacement of tower top in a time period ( Time Tiandi Time period wind speed correlation of tower tilt data in 2 dimensions), Indicates Tiandi The Pearson correlation coefficient between the absolute displacement of the tower top and the tower inclination data series of the base inclination angle in a time period is Represents the absolute value function; according to the above method, the tower abnormality influencing factor of the absolute displacement of the tower top in each time period of each day is obtained.

[0079] At this point, by analyzing the changes in the abnormal possibility of the tower tilt data in each dimension and the correlation with the wind speed, and eliminating the mutual influence between the tower tilt data in each dimension, the abnormal characteristics of the tower tilt data in each dimension are obtained, and then the tower abnormality influencing factors of the tower tilt data in each dimension in each time period are obtained.

[0080] Step S004: Analyze the tower tilt data of different dimensions in each time period according to the tower abnormality influencing factor of the tower tilt data of each dimension in each time period, obtain the overall tilt coefficient at each moment by weight, and complete the tilt monitoring of the wind turbine tower.

[0081] It should be noted that, in the process of tower tilt monitoring, the foundation tilt angle, tower sway amplitude and tower top absolute displacement interact and influence each other, so in the process of obtaining the overall tilt coefficient, it is necessary to use a data fusion method to obtain a comprehensive safety score; firstly, the AHP hierarchical analysis method is used to construct a judgment matrix based on expert experience and judgment to determine the target weights of tower tilt data in each dimension; then the PCA principal component analysis algorithm is used to analyze the tower tilt data in each dimension to obtain the principal components and their variance contribution rates, which are used as a reference for AHP target weight allocation; if a principal component is highly correlated with a key aspect of tower safety and the variance contribution rate of the principal component is large, then when constructing the judgment matrix in AHP, the tower tilt data related to the principal component can be given a higher relative importance.

[0082] Specifically, the tower tilt data in three dimensions are analyzed by the AHP algorithm and the PCA algorithm to obtain initial weights of the tower tilt data in three dimensions, wherein the initial weight acquisition method adopts the existing technology and will not be described in detail in this embodiment.

[0083] Furthermore, for any moment, the time period in which the moment is located is The tower anomaly influencing factors of the tower tilt data in the first dimension are The product of the initial weights of the tower tilt data in dimensions is used as the The adjustment weights of the tower tilt data of each dimension are calculated; based on the adjustment weights of the tower tilt data of each dimension at that moment, the weighted sum of the tower tilt data of each dimension at that moment is performed, and the result is taken as the overall tilt coefficient at that moment; an abnormal tilt threshold is preset, and the abnormal tilt threshold in this embodiment is described as 0.5. If the overall tilt coefficient at a certain moment is greater than or equal to the abnormal tilt threshold, then the tilt degree of the wind turbine tower exceeds the normal range at this moment, and the early warning mechanism is immediately triggered, and an early warning message is sent to relevant personnel, and the tower is maintained and inspected, including checking whether the connecting parts of the tower are loose, whether the coating is peeling off, etc., to ensure that the tower is in good working condition.

[0084] At this point, by quantifying the wind speed correlation between the tower tilt data and wind speed data in multiple dimensions of the wind turbine tower, and analyzing the anomalies generated by the wind turbine tower itself without being affected by wind speed, the possibility of anomalies is obtained; and the mutual influence between the tower tilt data in different dimensions is further analyzed based on the similarity of their changes, thereby reducing the interference of the mutual influence on the overall tilt monitoring quantification process, so that the overall tilt coefficient is based on the tower tilt data of each dimension itself for anomaly analysis, ultimately improving the accuracy of wind turbine tower tilt monitoring and the timeliness of early warning.

[0085] Another embodiment of the present invention provides a wind turbine tower tilt monitoring device, the device comprising:

[0086] A tower tilt data acquisition module, a tower tilt data analysis module and a tower tilt monitoring module. The tower tilt data acquisition module is used to collect multi-dimensional tower tilt data and wind speed data of the wind turbine tower. The tower tilt data analysis module implements steps S001 to S004 of the above method by calling a computer program to obtain the overall tilt coefficient of the wind turbine tower at each moment. The tower tilt monitoring module performs tilt monitoring and early warning of the wind turbine tower based on the overall tilt coefficient.

[0087] Another embodiment of the present invention provides a wind turbine tower tilt monitoring system, which includes a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, the above method steps S001 to S004 are implemented.

[0088] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for monitoring the inclination of a wind turbine tower, characterized in that: The method comprises the following steps: Collecting tower tilt data of multiple dimensions of wind turbine towers at each moment for several days, as well as wind speed data at each moment; the tower tilt data includes the base tilt angle, tower sway amplitude and tower top absolute displacement; Each day is divided into several time periods. Based on the similarity of the changing trends between the tower tilt data and the wind speed data in each dimension in each time period, the wind speed correlation of the tower tilt data in each dimension in each time period is obtained. Based on the data distribution on the left and right sides of each maximum point in the tower tilt data in each dimension of each day, and the difference between the wind speed data corresponding to the left and right sides, the abnormal possibility of the tower tilt data in each dimension of each day is obtained. Based on the abnormal possibility and wind speed correlation of the tower tilt data in each dimension every day, the tower abnormality influencing factor of the basic tilt angle in each time period is obtained; combined with the similarity relationship between the change of the tower sway amplitude and the absolute displacement of the tower top in each time period, the similarity relationship between the change of the absolute displacement of the tower top and the basic tilt angle, and the wind speed correlation of the tower tilt data in each dimension in each time period, the tower abnormality influencing factor of the tower sway amplitude in each time period and the tower abnormality influencing factor of the absolute displacement of the tower top in each time period are obtained respectively; According to the tower abnormality influencing factors of the tower tilt data in each dimension of each time period, the tower tilt data of different dimensions in each time period are analyzed, and the overall tilt coefficient at each moment is obtained by weighting to complete the tilt monitoring of the wind turbine tower; The specific method of obtaining the wind speed correlation of the tower tilt data in each dimension in each time period includes: According to the tower tilt data of each dimension in each time period and the wind speed data of each time period, obtain the tower tilt data sequence and tower tilt change sequence of each dimension in each time period, as well as the wind speed sequence and wind speed change sequence of each event segment; No. Tiandi Time period Wind speed correlation of tower tilt data in 4 dimensions The calculation method is: in, Indicates Tiandi Time period The tower tilt change sequence in the first dimension is Tiandi The Pearson correlation coefficient of the wind speed change series in each time period is Indicates Tiandi The time period is The order value in all time periods of the day, Indicates the number of time periods per day; The specific method for obtaining the abnormal possibility of the tower tilt data in each dimension every day includes: For Tiandi The tower tilt data of all dimensions are used to obtain a number of maximum points, and the maximum points are clustered to obtain a number of clusters. The cluster with the largest number of maximum points among all clusters is taken as the first cluster. Tiandi The representative extreme points of each dimension; For Tiandi The dimension a representative extreme value point, respectively obtaining the number of tower inclination data between the representative extreme value point and the maximum value points adjacent to the left and right sides, and taking the minimum value of the numbers as the reference number of the representative extreme value point; No. Tiandi The abnormal possibility of tower tilt data in each dimension The calculation method is: in, Indicates Tiandi The number of representative extreme points in the dimension, Indicates Tiandi The dimension The reference number representing the extreme points, and Respectively represent Tiandi The dimension represents the left and right sides of the extreme point Tower tilt data, and Respectively represent Tiandi The dimension represents the left and right sides of the extreme point The wind speed data at the corresponding moment of the tower tilt data, represents the absolute value function; The specific method of obtaining the tower abnormality influencing factor of the foundation tilt angle in each time period includes: when At that time, The tower inclination data of the first dimension is used as the basic inclination angle; Tiandi The abnormal possibility of tower tilt data in the first dimension is related to the Tiandi Time period The ratio of the wind speed correlation of the tower tilt data in the first dimension is taken as Tiandi Abnormal tower influence factor of foundation tilt angle in a time period; The tower abnormality influencing factor of the tower sway amplitude in each time period is specifically obtained as follows: For Tiandi Time period The tower tilt data in dimensions At that time, The tower tilt data in the dimension is the tower sway amplitude. Tiandi Tower anomaly influencing factors of tower sway amplitude in a time period The calculation method is: in, Indicates The possibility of abnormal tower sway amplitude within a day, Indicates Tiandi The wind speed correlation of the tower sway amplitude in a time period is Indicates Tiandi The Pearson correlation coefficient between the tower sway amplitude and the tower tilt data series of the absolute displacement of the tower top in a time period is represents the absolute value function; The tower abnormality influencing factor of the absolute displacement of the tower top in each time period is specifically obtained as follows: For Tiandi Time period The tower tilt data in dimensions At that time, The tower tilt data in the dimension is the absolute displacement of the tower top. Tiandi Tower abnormality influencing factors of absolute displacement of tower top in a time period The calculation method is: in, Indicates The possibility of anomalies in the absolute displacement of the tower top within a day, Indicates Tiandi The wind speed correlation of the absolute displacement of the tower top in a time period, Indicates Tiandi The Pearson correlation coefficient between the tower top absolute displacement and the tower inclination data series of the base inclination angle in a time period.

2. A wind turbine tower tilt monitoring method according to claim 1, characterized in that: The tower tilt data sequence and tower tilt change sequence in each dimension of each time period, as well as the wind speed sequence and wind speed change sequence of each event segment, are specifically obtained by: Based on the tower tilt data of each dimension in each time period, a tower tilt data sequence of each dimension in each time period is obtained by arranging the data in time sequence, and a wind speed sequence of each time period is obtained based on the wind speed data; For each dimension of the tower tilt data sequence and wind speed sequence, the slope of each element is obtained respectively, and the sequence obtained by sorting the slopes of each element according to the time sequence of the corresponding time period is recorded as the tower tilt change sequence and wind speed change sequence of each dimension in each time period respectively.

3. A method for monitoring wind turbine tower inclination according to claim 1, characterized in that: The overall tilt coefficient at each moment is specifically obtained as follows: The tower tilt data in three dimensions are analyzed by AHP algorithm and PCA algorithm, and the initial weights of the tower tilt data in three dimensions are obtained respectively; For any moment, the time period in which the moment is located The tower anomaly influencing factors of the tower tilt data in the first dimension are The product of the initial weights of the tower tilt data in dimensions is used as the The adjustment weight of the tower tilt data in each dimension; Based on the adjustment weights of the tower tilt data in each dimension at that moment, the tower tilt data in each dimension at that moment are weighted and summed, and the result is used as the overall tilt coefficient at that moment.

4. A wind power tower tilt monitoring device, the device comprising: A tower tilt data acquisition module, a tower tilt data analysis module and a tower tilt monitoring module, wherein the tower tilt data acquisition module is used to collect multi-dimensional tower tilt data and wind speed data of a wind turbine tower, and the tower tilt data analysis module implements the steps of a wind turbine tower tilt monitoring method as described in any one of claims 1 to 3 by calling a computer program to obtain the overall tilt coefficient of the wind turbine tower at each moment, and the tower tilt monitoring module performs tilt monitoring and early warning of the wind turbine tower based on the overall tilt coefficient.

5. A wind power tower tilt monitoring system, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the steps of a wind turbine tower inclination monitoring method as described in any one of claims 1 to 3 are implemented.

Images