A wind turbine concrete tower structure health online monitoring system and method

Abstract

The present invention provides a system and method for online monitoring of the health of a concrete tower structure of a wind turbine generator set. The system includes a data acquisition module, a data transmission module, a data processing module and a user interface; the data acquisition module is used to collect monitoring data of various parts of the concrete tower, the data transmission module is used to transmit the monitoring data to the data processing module, and the data processing module is used to process and analyze the monitoring data, thereby realizing an assessment of the local and overall health status of the concrete tower structure of the wind turbine generator set; the user interface is used to visually display the assessment results of the local and overall health status of the concrete tower structure; the present invention analyzes the local health status of the concrete tower structure through the monitoring results of multiple sensors, and can also assess the overall health status of the concrete tower.

Description

Technical Field

[0001] The present invention relates to the field of equipment health assessment, and in particular to an online monitoring system and method for the health of a concrete tower structure of a wind turbine generator set. Background Art

[0002] In the equipment management of wind farms, the safe operation of wind turbines is the primary consideration. Since wind turbines are located in the wild, real-time appearance inspection of the equipment cannot be achieved. Especially for areas with geological hazards, such as mudslides, mining, earthquakes and other natural disasters, pre-emptive control cannot be achieved on site. In addition, the main equipment of wind turbines are all operated above the tower dozens of meters high. Under the influence of wind speed, gravity, and blade torque, whether the bolts and welds of the wind turbine tower can withstand the design load, especially to achieve the stability of the wind turbine and tower safety under extreme wind speeds, has become a topic of great importance to the wind power industry.

[0003] Consulting relevant disclosed technical solutions, such as CN212360043U prior art discloses a wind turbine concrete tower crack monitoring system, which divides the concrete tower into multiple sections along its length direction, and each tower section is installed with a concrete block collection unit, and each concrete block collection unit is connected to a trigger unit below. When the weight of the concrete blocks in the concrete block collection unit reaches a specific value, it will trigger the trigger unit below it. The controller can detect which trigger unit is triggered, thereby controlling the unit to shut down in real time and issuing an alarm signal, allowing maintenance personnel to detect the problematic tower section in a targeted manner; Another typical prior art with publication number CN115478981A discloses a wind farm tower tilt monitoring system A system and a precise wind-facing method, comprising a wind turbine and a ring-shaped base, wherein the wind turbine is provided with an auxiliary plate, the auxiliary plate is provided with a tilt sensor, and the tower of the wind turbine is provided with an anti-drift mechanism, by providing the tilt sensor, the tilt angle of the tower when the wind speed is stable can be monitored in real time, thereby accurately identifying and calibrating the wind direction, so that the wind turbine can quickly adjust the direction of the fan blades according to the wind direction, so as to achieve precise wind-facing, and by providing the anti-drift mechanism, the tilted tower can be supported, and with the cooperation of a first electric telescopic rod and a second electric telescopic rod, the length and angle of the supporting square plate can be fine-tuned so that the inclination of the tower can be corrected; the above scheme can only monitor the local part of the concrete tower, and cannot evaluate the overall health status of the concrete tower. Summary of the Invention

[0004] The purpose of the present invention is to address the current deficiencies and propose an online monitoring system and method for the health of a wind turbine concrete tower structure.

[0005] The present invention adopts the following technical solutions:

[0006] An online monitoring system for the health of a concrete tower structure of a wind turbine generator set is characterized in that the system comprises a data acquisition module, a data transmission module, a data processing module and a user interface;

[0007] The data acquisition module is used to collect monitoring data of various parts of the concrete tower, the data transmission module is used to transmit the monitoring data to the data processing module, and the data processing module is used to process and analyze the monitoring data, thereby realizing the evaluation of the local and overall health status of the concrete tower structure of the wind turbine generator set; the user interface is used to visually display the local and overall health status evaluation results of the concrete tower structure.

[0008] Furthermore, the data acquisition module includes a tilt sensor, an acceleration sensor, a strain sensor and a data sorting module. The tilt sensor is used to collect the tilt angle information of the top and bottom of the concrete tower, the acceleration sensor is used to collect the vibration information of the concrete tower, and the strain sensor is used to collect the deformation information of the steel strand in the concrete tower structure; the data sorting module is used to filter and compress the data collected by each sensor to generate local monitoring data.

[0009] Furthermore, the data processing module includes a preprocessing module and an analysis module. The preprocessing module is used to remove outliers in the monitoring data. The analysis module includes a local analysis module and an overall analysis module. The local analysis module is used to analyze the health status of each part of the concrete tower based on the monitoring data after removing the outliers. The overall analysis module is used to analyze the overall health status of the concrete tower based on the health status of each part of the concrete tower.

[0010] Furthermore, the present invention provides a monitoring method for an online monitoring system for the health of a concrete tower structure of a wind turbine generator set, the method comprising:

[0011] S1: Obtain monitoring data of each local area within the set time period;

[0012] S2: removing outliers from the monitoring data to obtain local data sequences after removing outliers;

[0013] S3: Analyze the health status of each part of the concrete tower based on the local data series after removing outliers;

[0014] S4: Analyze the overall health status of the concrete tower based on the health status of each part of the concrete tower.

[0015] Furthermore, the monitoring data within the set time period is the monitoring data transmitted by the data transmission module within the set time period from the current time as the end time forward;

[0016] Furthermore, the specific method of analyzing the health status of each part of the concrete tower based on each local data sequence after removing outliers is as follows: each local data sequence is placed into different pre-trained local neural network models according to the local type, and the health status of each part of the concrete tower is output, wherein the health status of each part includes the probability of each part being in a normal state, a caution state, a warning state, and a dangerous state;

[0017] Furthermore, the specific method of analyzing the overall health status of the concrete tower based on the health status of each part of the concrete tower is as follows:

[0018] Take 10 groups of health status probability values ​​for each part of the concrete tower from the current time forward; each group of probability values ​​includes the probability values ​​of each part being in the four health states of normal, caution, warning, and danger;

[0019] The probability values ​​of each local health state in a group being in the four health states of normal, caution, warning and danger are divided into two segments according to the numerical value of each probability value and put into two sets. The two sets are respectively denoted as Y1=(m1, m2, m3, ..., m l ), Y2=(p1, p2, p3,..., p l ); where Y1 is a small probability data set, m i is a value in the smaller part of the probability value of the local health status of the group being in normal, caution, warning or dangerous health status, Y2 is the data set with high probability, p i is a value of the larger part of the probability values ​​of the local health status in the group being normal, caution, warning or dangerous, satisfying 1≤i≤l; the total number of local health status probability values ​​is 2l;

[0020] Draw two scatter plots with the element subscripts in the above two data sets as the horizontal axis and the element values ​​as the vertical axis, and calculate the fitting slopes of the two scatter plots:

[0021]

[0022]

[0023] where m i is an element in Y1, p i is an element in Y2, is the average value of the corresponding values ​​of the elements in Y1, is the average value of the corresponding values ​​of the elements in Y2, k1 is the fitting slope of the scatter plot corresponding to the small probability data set, and k2 is the fitting slope of the scatter plot corresponding to the large probability data set;

[0024] The characteristic coordinates (x, y) of the probability value group are calculated based on the two fitting slopes above, where:

[0025]

[0026]

[0027] Perform the above operation on the probability values ​​of all groups to obtain 10 feature coordinates. Let a feature coordinate be (x j ,y j ), where 1≤j≤10;

[0028] Compute the maximum Chebyshev distance between feature coordinates:

[0029] d max =max i，j [max(|x i -x j |), (|y i -y j |)]; 1≤i≤10, 1≤j≤10;

[0030] Among them, x i and x j is the horizontal coordinate of two feature coordinates in each feature coordinate, y i and y j is the ordinate of two feature coordinates in each feature coordinate, max i，j Indicates enumeration of all i and j, max(|x i -x j |), (|y i -y j |) means calculating the Chebyshev distance between the i-th coordinate and the j-th coordinate;

[0031] Each time the health status of each part is updated, a new d max , by calculating two adjacent d max The absolute value of the difference is calculated and compared with the set threshold to evaluate the overall health status of the concrete tower.

[0032] The beneficial effects achieved by the present invention are:

[0033] The present invention collects monitoring data of various parts of the concrete tower through a data acquisition module, and generates the health status of each part of the concrete tower in real time through a data processing module, which can timely discover abnormal conditions of the structure, avoid serious accidents of the structure, and improve the safety of the structure; the overall health status of the concrete tower is generated in real time through the health status of each part, so that users can understand the health status of the concrete tower structure more intuitively, thereby improving the user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] The present invention can be further understood from the following description in conjunction with the accompanying drawings. The components in the figures are not necessarily drawn to scale, but rather the emphasis is placed on illustrating the principles of the embodiments. In different views, the same reference numerals designate corresponding parts.

[0035] Figure 1 It is a schematic diagram of the overall module of the present invention.

[0036] Figure 2 It is a flow chart of the online monitoring method for the health of concrete tower structures of the present invention.

[0037] Figure 3 This is a schematic diagram of the process of removing outliers in the present invention. DETAILED DESCRIPTION

[0038] In order to make the objectives, technical solutions and advantages of the present invention more clear, the present invention is further described in detail below in conjunction with its embodiments; it should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention; for those skilled in the art, after reviewing the following detailed description, other systems, methods and / or features of the present embodiment will become apparent; it is intended that all such additional systems, methods, features and advantages are included in this specification; included within the scope of the present invention and protected by the appended claims; additional features of the disclosed embodiments are described in detail below, and these features will be apparent from the following detailed description.

[0039] The same or similar numbers in the drawings of the embodiments of the present invention correspond to the same or similar parts; in the description of the present invention, it should be understood that if there are terms such as "upper", "lower", "left", "right", etc. indicating directions or positional relationships, they are based on the directions or positional relationships shown in the drawings. This is only for the convenience of describing the present invention and simplifying the description, and does not indicate or imply that the device or component referred to must have a specific direction, be constructed and operate in a specific direction. Therefore, the terms describing the positional relationship in the drawings are only used for illustrative purposes and cannot be understood as limiting this patent. For ordinary technicians in this field, the specific meanings of the above terms can be understood according to specific circumstances.

[0040] Example 1.

[0041] like Figure 1 、 Figure 2 As shown, this embodiment provides an online monitoring system for the health of a concrete tower structure of a wind turbine generator set, characterized in that the system includes a data acquisition module, a data transmission module, a data processing module and a user interface;

[0042] The data acquisition module is used to collect monitoring data of various parts of the concrete tower, the data transmission module is used to transmit the monitoring data to the data processing module, and the data processing module is used to process and analyze the monitoring data, thereby realizing the evaluation of the local and overall health status of the concrete tower structure of the wind turbine generator set; the user interface is used to visually display the local and overall health status evaluation results of the concrete tower structure.

[0043] Furthermore, the data acquisition module includes a tilt sensor, an acceleration sensor, a strain sensor and a data sorting module. The tilt sensor is used to collect the tilt angle information of the top and bottom of the concrete tower, the acceleration sensor is used to collect the vibration information of the concrete tower, and the strain sensor is used to collect the deformation information of the steel strand in the concrete tower structure; the data sorting module is used to filter and compress the data collected by each sensor to generate local monitoring data.

[0044] Furthermore, the data processing module includes a preprocessing module and an analysis module. The preprocessing module is used to remove outliers in the monitoring data. The analysis module includes a local analysis module and an overall analysis module. The local analysis module is used to analyze the health status of each part of the concrete tower based on the monitoring data after removing the outliers. The overall analysis module is used to analyze the overall health status of the concrete tower based on the health status of each part of the concrete tower.

[0045] Furthermore, the present invention provides a monitoring method for an online monitoring system for the health of a concrete tower structure of a wind turbine generator set, the method comprising:

[0046] S1: Obtain monitoring data of each local area within the set time period;

[0047] S2: removing outliers from the monitoring data to obtain local data sequences after removing outliers;

[0048] S3: Analyze the health status of each part of the concrete tower based on the local data series after removing outliers;

[0049] S4: Analyze the overall health status of the concrete tower based on the health status of each part of the concrete tower.

[0050] Furthermore, the monitoring data within the set time period is the monitoring data transmitted by the data transmission module within the set time period from the current time as the end time forward;

[0051] Furthermore, the specific method of analyzing the health status of each part of the concrete tower based on each local data sequence after removing outliers is as follows: each local data sequence is placed into different pre-trained local neural network models according to the local type, and the health status of each part of the concrete tower is output, wherein the health status of each part includes the probability of each part being in a normal state, a caution state, a warning state, and a dangerous state;

[0052] Furthermore, the specific method of analyzing the overall health status of the concrete tower based on the health status of each part of the concrete tower is as follows:

[0053] Take 10 groups of health status probability values ​​for each part of the concrete tower from the current time forward; each group of probability values ​​includes the probability values ​​of each part being in the four health states of normal, caution, warning, and danger;

[0054] The probability values ​​of each local health state in a group being in the four health states of normal, caution, warning and danger are divided into two segments according to the numerical value of each probability value and put into two sets. The two sets are respectively denoted as Y1=(m1, m2, m3, ..., m l ), Y2=(p1, p2, p3,..., p l ); where Y1 is a small probability data set, m i is a value in the smaller part of the probability value of the local health status of the group being in normal, caution, warning or dangerous health status, Y2 is the data set with high probability, p i is a value of the larger part of the probability values ​​of the local health status in the group being normal, caution, warning or dangerous, satisfying 1≤i≤l; the total number of local health status probability values ​​is 2l;

[0055] Draw two scatter plots with the element subscripts in the above two data sets as the horizontal axis and the element values ​​as the vertical axis, and calculate the fitting slopes of the two scatter plots:

[0056]

[0057]

[0058] where m i is an element in Y1, p i is an element in Y2, is the average value of the corresponding values ​​of the elements in Y1, is the average value of the corresponding values ​​of the elements in Y2, k1 is the fitting slope of the scatter plot corresponding to the small probability data set, and k2 is the fitting slope of the scatter plot corresponding to the large probability data set;

[0059] The characteristic coordinates (x, y) of the probability value group are calculated based on the two fitting slopes above, where:

[0060]

[0061]

[0062] Perform the above operation on the probability values ​​of all groups to obtain 10 feature coordinates. Let a feature coordinate be (x j ,y j ), where 1≤j≤10;

[0063] Compute the maximum Chebyshev distance between feature coordinates:

[0064] d max =max i，j [max(|x i -x j |), (|y i -y j |)]; 1≤i≤10, 1≤j≤10;

[0065] Among them, x i and x j is the horizontal coordinate of two feature coordinates in each feature coordinate, y i and y j is the ordinate of two feature coordinates in each feature coordinate, max i，j Indicates enumeration of all i and j, max(|x i -x j |), (|y i -y j |) means calculating the Chebyshev distance between the i-th coordinate and the j-th coordinate;

[0066] Each time the health status of each part is updated, a new d max , by calculating two adjacent d max The absolute value of the difference is calculated and compared with a set threshold to evaluate the overall health status of the concrete tower. This embodiment uses a data acquisition module to collect monitoring data from various parts of the concrete tower, and a data processing module to generate the health status of each part of the concrete tower in real time. This allows for timely detection of structural abnormalities, avoids serious structural accidents, and improves structural safety. The overall health status of the concrete tower is generated in real time based on the health status of each part, allowing users to more intuitively understand the health status of the concrete tower structure, thereby improving user experience.

[0067] Example 2.

[0068] This embodiment should be understood to include at least all the features of any of the aforementioned embodiments and be further improved thereon;

[0069] This embodiment provides a system and method for online monitoring of the health of a concrete tower structure of a wind turbine generator set, characterized in that the system includes a data acquisition module, a data transmission module, a data processing module and a user interface;

[0070] The data acquisition module is used to collect monitoring data of various parts of the concrete tower, the data transmission module is used to transmit the monitoring data to the data processing module, and the data processing module is used to process and analyze the monitoring data, thereby evaluating the local and overall health status of the wind turbine concrete tower structure; the user interface is used to visually display the evaluation results of the local and overall health status of the concrete tower structure;

[0071] The data acquisition module includes a tilt sensor, an acceleration sensor, a strain sensor, and a data sorting module. The tilt sensor is used to collect tilt angle information of the top and bottom of the concrete tower, the acceleration sensor is used to collect vibration information of the concrete tower, and the strain sensor is used to collect deformation information of the steel strands in the concrete tower structure; the data sorting module is used to filter and compress the data collected by each sensor to generate monitoring data for each local area;

[0072] The monitoring data of each part includes tower top inclination data, tower bottom inclination data, tower vibration data and steel strand deformation data;

[0073] The tilt sensor is a common dual-axis tilt sensor on the market;

[0074] The data processing module includes a preprocessing module and an analysis module. The preprocessing module is used to remove abnormal values ​​in the monitoring data. The analysis module includes a local analysis module and an overall analysis module. The local analysis module is used to analyze the health status of each part of the concrete tower according to the monitoring data after the abnormal values ​​are removed. The overall analysis module is used to analyze the overall health status of the concrete tower according to the health status of each part of the concrete tower.

[0075] The various parts of the concrete tower include the tower top inclination angle, tower bottom inclination angle, tower vibration and m steel strands, where m is the number of all steel strands in the concrete tower;

[0076] This embodiment provides a method for online monitoring of the health of a concrete tower structure of a wind turbine generator set, the method comprising:

[0077] S1: Obtain monitoring data of each local area within the set time period;

[0078] S2: removing outliers from the monitoring data to obtain local data sequences after removing outliers;

[0079] S3: Analyze the health status of each part of the concrete tower based on the local data series after removing outliers;

[0080] S4: Analyze the overall health status of the concrete tower based on the health status of each part of the concrete tower;

[0081] The monitoring data within the set time period is the monitoring data transmitted by the data transmission module within the set time period forward from the current time as the deadline;

[0082] like Figure 3 As shown, the method of removing abnormal values ​​in the monitoring data in step S2 is as follows:

[0083] S21: Generate multiple data sequences arranged in order of data size based on the monitoring data within the set time period according to the differences in each local area. Let (x1, x2, x3, ..., x n ) is a local data sequence, x i is the monitoring data of the local area at a certain moment in the set time period, where 1≤i≤n; n is the total number of monitoring data of the local area in the set time period, that is, the data in the data sequence is the monitoring data of the local area in the set time period, such as the tower top inclination data;

[0084] S22: Calculate the upper quartile and the lower quartile of the data series;

[0085] The upper quartile and the lower quartile are calculated as follows:

[0086] S221: Find the median Q2 of the data sequence, and divide the data in the data sequence into two parts, one part is data less than or equal to Q2, and the other part is data greater than or equal to Q2; when the number of data in the data sequence is an even number, then the median Q2 is the average of the two middle numbers;

[0087] S222: For data in the data sequence that is less than or equal to Q2, find the median Q1; for data in the data sequence that is greater than or equal to Q2, find the median Q3, where Q1 is the lower quartile of the data sequence and Q3 is the upper quartile of the data sequence;

[0088] S23: Calculate the interquartile range Q of the data series r :

[0089] Q r =Q3-Q1;

[0090] S24: Determine the filtering edge of the data sequence:

[0091] Q 上 =Q3+1.5Q r ;

[0092] Q 下 =Q1-1.5Q r ;

[0093] Among them, Q 上 is the upper filter edge of the data sequence, Q 下 The lower filter edge of the data sequence is used to compare the data in the data sequence with the filter edge in turn, and the values ​​in the data sequence that are greater than the upper filter edge and less than the lower filter edge are removed;

[0094] S25: Execute the above steps S21 to S24 on all local data sequences to obtain local data sequences after removing outliers;

[0095] The specific method of analyzing the health status of each part of the concrete tower based on the data sequence after removing outliers is as follows:

[0096] Each local data sequence is placed into a different pre-trained local neural network model according to the local type, and the health status of each local part of the concrete tower is output. The health status of each local part includes the probability of each local part being in a normal state, a caution state, a warning state, and a dangerous state. For example, the data sequence corresponding to the tower top inclination angle is placed into the corresponding tower top inclination angle neural network model, and the output of the neural network model is the probability of the tower top inclination angle being in a normal state, a caution state, a warning state, and a dangerous state.

[0097] The input of each local neural network model is the data in each local data sequence, and the number of output layer nodes of each local neural network model is 4, corresponding to the four health states of normal, attention, warning, and danger respectively;

[0098] The pre-training data of each local neural network model can be obtained through laboratory testing or historical data;

[0099] The specific method of analyzing the overall health status of the concrete tower based on the health status of each part of the concrete tower is as follows:

[0100] Take 10 groups of health status probability values ​​for each part of the concrete tower from the current time forward; each group of probability values ​​includes the probability values ​​of each part being in the four health states of normal, caution, warning, and danger;

[0101] The probability values ​​of each local health state in a group being in the four health states of normal, caution, warning and danger are divided into two segments according to the numerical value of each probability value and put into two sets. The two sets are respectively denoted as Y1=(m1, m2, m3, ..., m l ), Y2=(p1, p2, p3,..., p l ); where Y1 is a small probability data set, m i is a value in the smaller part of the probability value of the local health status of the group being in normal, caution, warning or dangerous health status, Y2 is the data set with high probability, p i is a value representing the larger portion of the probability values ​​of the local health status being normal, caution, warning, or dangerous, satisfying 1≤i≤l. The total number of local health status probability values ​​is 2l. In this embodiment, the local parts of the concrete tower are the tower top inclination angle, tower bottom inclination angle, tower vibration, and m steel strands, so l = 2×(3+m).

[0102] Draw two scatter plots with the element subscripts in the above two data sets as the horizontal axis and the element values ​​as the vertical axis, and calculate the fitting slopes of the two scatter plots:

[0103]

[0104]

[0105] where m i is an element in Y1, p i is an element in Y2, is the average value of the corresponding values ​​of the elements in Y1, is the average value of the corresponding values ​​of the elements in Y2, k1 is the fitting slope of the scatter plot corresponding to the small probability data set, and k2 is the fitting slope of the scatter plot corresponding to the large probability data set;

[0106] The characteristic coordinates (x, y) of the probability value group are calculated based on the two fitting slopes above, where:

[0107]

[0108]

[0109] Perform the above operation on the probability values ​​of all groups to obtain 10 feature coordinates. Let a feature coordinate be (x j ,y j ), where 1≤j≤10;

[0110] Compute the maximum Chebyshev distance between feature coordinates:

[0111] d max =maxi，j [max(|x i -x j |), (|y i -y j |)]; 1≤i≤10, 1≤j≤10;

[0112] Among them, x i and x j is the horizontal coordinate of two feature coordinates in each feature coordinate, y i and y j is the ordinate of two feature coordinates in each feature coordinate, max i，j Indicates enumeration of all i and j, max(|x i -x j |), (|y i -y j |) means calculating the Chebyshev distance between the i-th coordinate and the j-th coordinate;

[0113] Each time the health status of each part is updated, a new d max , by calculating two adjacent d max The absolute value of the difference is calculated and compared with the set threshold to evaluate the overall health status of the concrete tower.

[0114] This embodiment collects monitoring data of various parts of the concrete tower through the data acquisition module, and removes outliers in the monitoring data by calculating the quartiles and interquartile ranges to determine the filter edge, thereby ensuring the accuracy of the monitoring data; through the local neural network models in the data processing module, the health status of each part can be intelligently generated through each local monitoring data, ensuring that the user can obtain high-accuracy health status data of each part in real time. The health status of each part of the concrete tower is generated in real time by the data processing module, so that abnormal conditions of the structure can be discovered in time, serious accidents of the structure can be avoided, and the safety of the structure can be improved; the overall health status of the concrete tower is generated in real time through the health status of each local part, so that the user can more intuitively understand the health status of the concrete tower structure, thereby improving the user experience.

[0115] The contents disclosed above are only preferred feasible embodiments of the present invention and do not limit the scope of protection of the present invention. Therefore, all equivalent technical changes made using the contents of the present invention description and drawings are included in the scope of protection of the present invention. In addition, the elements therein can be updated as technology develops.

Claims

1. An online monitoring system for the health of a wind turbine concrete tower structure, characterized in that: The system includes a data acquisition module, a data transmission module, a data processing module and a user interface; The data acquisition module is used to collect monitoring data of various parts of the concrete tower, and the data transmission module is used to transmit the monitoring data to the data processing module. The data processing module is used to process and analyze the monitoring data, thereby realizing the evaluation of the local and overall health status of the wind turbine concrete tower structure; The user interface is used to visually display the local and overall health status assessment results of the concrete tower structure; The monitoring method of the monitoring system includes: S1: Obtain local monitoring data within a set period; S2: removing outliers from the monitoring data to obtain local data sequences after removing outliers; S3: Analyze the health status of each part of the concrete tower based on the local data series after removing outliers; S4: Analyze the overall health status of the concrete tower based on the health status of each part of the concrete tower; The specific method of analyzing the overall health status of the concrete tower based on the health status of each part of the concrete tower is as follows: Take 10 groups of health status probability values ​​for each part of the concrete tower from the current time forward; each group of probability values ​​includes the probability values ​​of each part being in the four health states of normal, caution, warning, and danger; The probability values ​​of each local health status in a group being in the four health states of normal, caution, warning and danger are divided into two segments according to the numerical value of each probability value and put into two sets. The two sets are respectively recorded as , ;in is a data set with small probability, is a value of the smaller part of the probability values ​​of the local health status of the group being in the normal, caution, warning or dangerous health status, is a data set with high probability, The probability value of the local health status of the group being in the normal, caution, warning or dangerous state is a value with a larger value, satisfying ; The total number of local health status probability values ​​is ; Draw two scatter plots with the element subscripts in the above two data sets as the horizontal axis and the element values ​​as the vertical axis, and calculate the fitting slopes of the two scatter plots: ; ; in for Medium elements, for Medium elements, for The average value of the corresponding values ​​of the elements in , for The average value of the corresponding values ​​of the elements in , is the fitting slope of the scatter plot corresponding to the small probability data set, is the fitting slope of the scatter plot corresponding to the high probability data set; Calculate the characteristic coordinates of this group of probability values ​​based on the two fitting slopes above ,in: ； ； Perform the above operation on the probability values ​​of all groups to obtain 10 feature coordinates. Let a feature coordinate be ,in ; Compute the maximum Chebyshev distance between feature coordinates: ； ， ； in, and is the horizontal coordinate of two characteristic coordinates in each characteristic coordinate, and is the ordinate of two feature coordinates in each feature coordinate, Indicates that all and To enumerate, Indicates calculation of coordinates and Chebyshev distance between coordinates; Each update of the local health status will get a new , by calculating two adjacent The absolute value of the difference is calculated and compared with the set threshold to evaluate the overall health status of the concrete tower; The specific method of analyzing the health status of each part of the concrete tower based on each local data sequence after removing outliers is: each local data sequence is placed into different pre-trained local neural network models according to the local type, and the health status of each part of the concrete tower is output. The health status of each part includes the probability of each part being in a normal state, a caution state, a warning state, and a dangerous state.

2. The wind turbine generator concrete tower structure health online monitoring system according to claim 1, characterized in that: The data acquisition module includes a tilt sensor, an acceleration sensor, a strain sensor and a data sorting module. The tilt sensor is used to collect the tilt angle information of the top and bottom of the concrete tower, the acceleration sensor is used to collect the vibration information of the concrete tower, and the strain sensor is used to collect the deformation information of the steel strands in the concrete tower structure; the data sorting module is used to filter and compress the data collected by each sensor to generate local monitoring data.

3. The wind turbine generator concrete tower structure health online monitoring system according to claim 2, characterized in that: The data processing module includes a preprocessing module and an analysis module. The preprocessing module is used to remove outliers in the monitoring data. The analysis module includes a local analysis module and an overall analysis module. The local analysis module is used to analyze the health status of each part of the concrete tower based on the monitoring data after removing the outliers. The overall analysis module is used to analyze the overall health status of the concrete tower based on the health status of each part of the concrete tower.

4. The wind turbine generator set concrete tower structure health online monitoring system according to claim 3, characterized in that: The monitoring data within the set time period is the monitoring data transmitted by the data transmission module within the set time period before the current time.

Images