Tower Barrel Health Status Monitoring Method Based on Fixed Detection and Mobile Detection

Abstract

The present invention discloses a method for monitoring the health status of a wind turbine tower based on fixed detection and mobile detection. A fixed sensor for detecting the sway at the top of the tower is fixedly installed at the top of the tower of the wind turbine, and a mobile sensor for detecting the vibration of the tower is installed on the tower and can move up and down along the tower. By comparing the monitoring results of the fixed sensors on different towers, the suspected faulty tower is determined. By comparing the monitoring results of the mobile sensors on different towers, the suspected fault point of the suspected faulty tower is determined. The present invention introduces mobile detection on the basis of fixed detection, can troubleshoot the overall fault problems of the tower and accurately determine the possible location of the fault, improves the accuracy of identifying the potential risks of the tower, and using equipment for pre-troubleshooting can avoid the workload of inspectors climbing the wall, improve the detection work efficiency and reduce the need for personnel to work at heights.

Description

Technical Field

[0001] The present invention relates to the field of health state monitoring of wind power generation equipment, and particularly to a method for monitoring the health state of a tower barrel based on fixed detection and mobile detection. Background Art

[0002] Wind power generation is a power generation method in which wind turbines are deployed in the offshore or mountainous areas, and the wind energy is converted into electric energy by the rotation of the wind turbines. Wind turbines are generally deployed outdoors and are greatly affected by the surrounding environment. The wind turbines interact violently with the wind for a long time, which may cause hidden faults in the tower barrel structure. If not eliminated in time, it may further cause structural damage to the tower barrel, and even the tower barrel may collapse under the dual action of its own gravity and wind force. In view of this, it is necessary to monitor the health state of the tower barrel in real time, quickly and accurately locate the hidden faults, and timely maintain the health of the tower barrel to ensure the continuous and stable operation of the wind turbine.

[0003] The tower barrel is an important part of the wind power generation equipment, which is mainly used to support the generator set and absorb the vibration of the wind blades and the generator set at the same time. The tower barrel is generally formed by connecting multiple hollow cylinders end to end in series, and the cylinders are fixedly connected through flanges. Due to the need to bear the self-weight of the wind blades and the generator set as well as the pressure brought by the strong wind to the wind turbine, there may be structural damage to each cylinder itself, or after long-term use, the structure at the connection between the cylinders ages, and the connection between the flanges becomes loose, which further causes the tower barrel to be unstable in support. The unstable support of the tower barrel will further aggravate the vibration of the wind turbine and further accelerate the structural aging of the wind turbine. Therefore, the health monitoring of the tower barrel is particularly important.

[0004] In order to monitor the health status of a wind turbine in real time, the existing tower health monitoring technologies mainly install angle sensors, displacement sensors, etc. at the top and bottom of the tower to monitor the inclination angle of the tower and the offset of the top of the tower respectively. This monitoring method cannot obtain the overall deformation of the tower, nor can it monitor the displacement curve of the tower under external forces at various frequencies. Therefore, the health status of the tower cannot be comprehensively and accurately monitored and evaluated. To overcome this problem, Patent ZL201711008222.8 discloses a method for monitoring the health status of a tower. It fixedly installs a biaxial acceleration sensor inside the tower to collect the vibration data of the tower, and at the same time uses a biaxial inclination sensor to collect the inclination data of the tower base. By comprehensively analyzing the vibration data and the inclination data, it is determined whether the tower is healthy. If there is an abnormality in the tower, an automatic alarm will be issued. Obviously, through the health detection of the tower, whether there is a connection fault in the tower can be found in time. However, the occurrence of tower faults is random, and the fault occurrence points are often impossible to be determined in advance. Therefore, the above method can only be used to check the connection strength at the joints of each column of the tower, and is not easy to be used to find the structural damage caused by the wind force and the self-weight of the fan to the tower column itself. There are problems such as inaccurate health monitoring of the tower or waste of installing a large number of devices while increasing the burden on the tower. On the other hand, when a strong wind acts on the wind turbine, the tower will sway. Monitoring only with a biaxial acceleration sensor may vary due to different acting wind forces, which is very likely to cause problems such as false alarms or missed alarms. The above monitoring method needs to be further improved. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide a method for monitoring the health status of a tower based on fixed detection and mobile detection that can accurately locate the fault position of the tower.

[0006] To solve the above technical problem, the technical solution of the present invention is:

[0007] A method for monitoring the health status of a tower based on fixed detection and mobile detection, which fixedly installs a fixed sensor for detecting the sway at the top of the tower on the top of the tower of the wind turbine, and installs a mobile sensor for detecting the vibration of the tower that can move up and down along the tower. By comparing the monitoring results of the fixed sensors on different towers, the suspected faulty tower is determined. By comparing the monitoring results of the mobile sensors on different towers, the suspected fault point of the suspected faulty tower is determined.

[0008] Preferably, the fixed sensor obtains the top offset information of the top of the tower, performs a Fourier transform on the top offset information to obtain a first frequency domain signal, and determines the suspected faulty tower by comparing the first frequency domain signals of the towers of the wind turbines close to each other.

[0009] Preferably, the mobile sensor obtains the wall offset information on the side wall of the tower barrel, performs Fourier transform on the wall offset information to obtain a second frequency domain signal related to the height of the tower barrel, obtains the offset information function of multiple detection points at different heights on the side walls of the tower barrels of each wind turbine and respectively calculates the stiffness at different heights, and determines the suspected fault points by comparing the stiffness at the same height of different wind turbines.

[0010] Preferably, the fixed sensor or the mobile sensor is a GNSS and an acceleration sensor.

[0011] Preferably, when comparing the first frequency domain signals of the tower barrels of wind turbines close to each other, a risk factor is defined for the tower barrel of each wind turbine. The difference between the first frequency domain signals of the tower barrels of different wind turbines is calculated iteratively. If the difference between the first frequency domain signals meets the first threshold condition, the risk factors of the corresponding two wind turbine tower barrels are accumulated respectively. When the risk factor is greater than the risk threshold, it is determined that the tower barrel of this wind turbine is the suspected faulty tower barrel.

[0012] Preferably, the stiffness is

[0013]

[0014] where M(x) is the bending moment function, and the offset information function F(x) can be generated by fitting the function relationship through a large amount of data monitored by the mobile monitoring points.

[0015] Preferably, when determining the suspected fault points by comparing the stiffness at the same height of different wind turbines, first determine the suspected faulty tower barrel, define respective problem factors for different heights on the suspected faulty tower barrel, calculate iteratively the difference in stiffness at the same height between the suspected faulty tower barrel and the comparison tower barrel compared with it at each height. If the detected difference in stiffness meets the second threshold condition, accumulate the problem factors of the heights that meet the second threshold condition. When the problem factor of a certain height is greater than the problem threshold, determine the point at this height as the suspected fault point.

[0016] Preferably, when determining the suspected fault points by comparing the stiffness at the same height of different wind turbines, first determine the suspected faulty tower barrel and the comparison tower barrel, calculate the difference between the mean value of the stiffness at the same height of the suspected faulty tower barrel and all comparison tower barrels. If the detected difference in stiffness meets the second threshold condition, determine the point at this height as the suspected fault point.

[0017] Preferably, densely move the monitoring points near the suspected fault points to obtain a more accurate fault point location.

[0018] Preferably, the fixed sensor and the mobile sensor adopt the same set of sensor devices. When the sensor device is moved up to the top of the tower barrel and remains stationary at the top, it serves as the fixed sensor, and when the sensor device moves up and down, it serves as the mobile sensor.

[0019] After adopting the above solution, since the present invention introduces mobile detection on the basis of fixed detection, it can troubleshoot the overall fault problems of the tower barrel and accurately determine the possible location of the fault, improving the accuracy of identifying potential risks of the tower barrel. Using equipment for pre-fault troubleshooting can avoid the workload of inspectors climbing the wall, improve the detection work efficiency, and reduce the need for personnel to work at heights. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It is a schematic diagram of directly using a rope to control the up and down movement of the mobile sensor;

[0021] Figure 2 It is a schematic diagram of using a winch to control the up and down movement of the mobile sensor;

[0022] Figure 3 It is a schematic diagram of using a drive motor to control the up and down movement of the mobile sensor;

[0023] Figure 4 It is a combined top view schematic diagram of the mobile device and the barrel wall structure;

[0024] Figure 5 It is a schematic diagram of the method steps involved in the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0025] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0026] What the present invention discloses is a method for monitoring the health status of a tower barrel based on fixed detection and mobile detection. The following is a preferred embodiment of the present invention. As Figures 1-5 shown, the method is as follows: A fixed sensor 2 for detecting the sway at the top of the tower barrel is fixedly installed at the top of the tower barrel 1 of the wind turbine. A mobile sensor 3 for detecting the vibration of the tower barrel is installed on the tower barrel and can move up and down along the tower barrel. Compare the monitoring results of the fixed sensors 2 on different tower barrels of wind turbines to determine the suspected faulty tower barrel. Compare the monitoring results of the mobile sensors 3 on different tower barrels of wind turbines, especially the monitoring results of the mobile sensors of the suspected faulty tower barrel and its surrounding tower barrels, to determine the suspected fault points of the suspected faulty tower barrel. Since this method involves data comparison between different tower barrels, it is required that the wind farm has at least three wind turbine tower barrels of the same model and size. The comparison of data between three or more wind turbine tower barrels can accurately determine the suspected faulty wind turbine and the exact location of the fault.

[0027] Specifically, first install fixed sensors on the wind turbine tower barrel, and obtain the top offset information of the tower barrel top through the fixed sensors to overall judge whether each tower barrel is operating normally. Since the base of the tower barrel is fixedly installed on the ground surface, obvious characteristics will occur at the top of the tower barrel when it works abnormally. Therefore, it is better to install the fixed sensors at a high place, such as in the nacelle inside the tower barrel top or on the nacelle surface. The detection of the top offset information can be directly obtained through GNSS (Global Navigation Satellite System) such as GPS (Global Positioning System) and Beidou. However, since the offset information obtained by GNSS may have certain deviations with the change of the tower barrel vibration frequency, when introducing GNSS, an acceleration sensor will also be installed. After the detection results obtained by the acceleration sensor are fused and reconstructed with the GNSS detection results, the detection accuracy of the tower barrel top offset information is improved. The specific fusion method can refer to the article "Research on the Information Extraction Model of GPS and Accelerometer for Bridge Deformation Fusion" on pages 549-556 of the 44th volume, third issue of the Journal of China University of Mining and Technology. In this embodiment, a GNSS for capturing displacement information and an acceleration sensor for capturing sway acceleration information are installed at the same time. The GNSS and the acceleration sensor are both fixedly installed on the tower barrel top by bolts, strong glue, strong magnets or other installation and fixation methods. The applications of the GNSS and the acceleration sensor are prior arts, and the specific circuit connections thereof are not described in detail herein. To ensure better signal transmission, the sensors can be installed on the outer bottom surface of the nacelle. Since the fixed sensor is directly connected to the nacelle, the power supply can be led out from the nacelle for power supply, and its signal transmission can directly lead out a communication line from the nacelle for transmission, or it can also be reported to the server through a wireless transmission module, such as a GPRS chip or a satellite communication module, etc. At present, a power supply system and a signal transmission system including equipment such as switch cabinets are generally arranged in the wind turbine nacelle, and the above systems can be directly used. The power supply system, the power consumption of the sensor terminal, and the signal transmission are all prior arts and will not be elaborated herein.

[0028] In order to achieve precise positioning of the fault location, a mobile sensor is further installed on the tower barrel. Specifically, a mobile device can be first installed on the outer wall of the tower barrel. For example, Figures 1-4As shown, a guide rail 4 is fixedly welded on the outer surface of the tower. The guide rail 4 can be two parallel steel rails vertically arranged along the outer wall of the tower. The cross-section of each steel rail is an "I" shape. The moving device is a sliding trolley 5 matching the guide rail 4. The side of the sliding trolley 5 close to the outer wall of the tower is provided with a positioning roller 51 embedded in the "I" groove from the side of the steel rail. The sliding trolley 5 can be moved by pulling with a rope 7, and the sliding trolley 5 can also be driven by an automatic control device. For example, in this embodiment, a roller 6 can be fixedly installed on the top surface of the tower, a rope 7 is passed around the roller 6 and one end of the rope is fixedly tied to the sliding trolley 5. At the same time, a winch 8 is provided at the base of the tower, and the other end of the rope is tied to the output shaft of the winch 8. The sliding trolley 5 is located at different heights outside the tower by controlling the winch 8. Of course, the power device that drives the sliding trolley 5 to move can also be directly fixed on the sliding trolley 5. Specifically, a driving motor 9 is fixedly installed on the sliding trolley 5, and a rope 7 is wound around the output shaft of the driving motor 9. The other end of the rope 7 is fixedly tied to the bottom surface of the cabin at the top of the tower 1. The sliding trolley 5 is raised and lowered by the driving motor 9. The sliding trolley 5 is equipped with a battery to provide power for the driving motor 9 so that it can drive the sliding trolley 5 to move up and down along the tower with the mobile sensor. When equipped with a battery, the mobile device can be powered by a vehicle-mounted photovoltaic panel or wireless charging. If wireless charging is used, a wireless charging receiving terminal is provided on the side of the sliding trolley 5 close to the tower, and a wireless charging transmitting terminal is provided on the surface of the tower. The wireless charging transmitting terminal is connected to the wind turbine electrical system to obtain power. An automatic climbing elevator can also be used to transport the sliding trolley 5 to transport the mobile sensor. The specific structure of the automatic climbing device is a prior art and will not be described here. The sliding trolley 5 can also be controlled in the form of magnetic levitation. A track that can generate a magnetic field is set on the surface of the tower. The moving device is limitedly installed on the track. An induction coil that can generate an alternating electromagnetic field is provided on the sliding trolley 5. By adjusting the induction coil, the relationship between the electromagnetic field and the track magnetic field is used to drive the sliding trolley 5 to move up and down.

[0029] The sliding trolley 5 is equipped with a mobile sensor 3, which is similar to the fixed sensor and also includes a GNSS for capturing displacement information and an acceleration sensor for capturing shaking acceleration information. After capturing the data, the mobile sensor sends the data to the server in real time via wireless transmission for signal reporting. Alternatively, a data storage module 52 may be provided on the sliding trolley 5 to collect the data and then report it to the server. The data acquisition of the mobile sensor can be powered by a battery 53 on the mobile device. The specific setting of the power supply circuit is a prior art and will not be described in detail. The battery power supply is powered by a vehicle-mounted photovoltaic panel or wireless charging, and will not be described in detail. Furthermore, in order to ensure that the mobile sensor can obtain accurate offset information of the tower wall, it can be designed that after the mobile device arrives at the specified position, the mobile sensor is tightly attached to the outside of the tower wall to detect the tower wall. Specifically, Figure 4As shown in the figure, a resetable mobile sensor carrier plate 54 can be designed on the side of the mobile device close to the tower barrel. A mobile sensor 3 is fixedly installed on the mobile sensor carrier plate 54. An electromagnet 55 is fixed on the side of the mobile sensor carrier plate 54 close to the tower barrel 1, and the other side is connected to the mobile device through a spring 56. When the mobile device reaches the specified position, the electromagnet 55 is charged to generate magnetism, so that the mobile sensor carrier plate 54 drives the mobile sensor 3 to be adsorbed on the wall of the tower barrel 1, and detection data can be obtained in real time. When the detection is completed, the electromagnet 55 is demagnetized, and the mobile sensor carrier plate resets under the action of the spring 56 and no longer contacts the wall of the barrel.

[0030] Furthermore, the fixed sensor and the mobile sensor can adopt the same set of sensor devices. The same set of sensor devices act as the fixed sensor and the mobile sensor respectively in different usage scenarios. When the sensor device moves up to the top of the tower barrel and stops at the top, the sensor device is the fixed sensor. When the sensor device moves up and down, it is the mobile sensor. In application, the sensor device acts as the fixed sensor for a long time to perform fixed detection on the health status of each wind turbine tower barrel. Once a suspected faulty tower barrel is found, the sensor device starts to slide up and down to obtain data for mobile detection. Each wind turbine can perform synchronous operations by remotely controlling the motor or the winch, and then finally determine the location of the suspected fault point.

[0031] After the server obtains the data, it determines the suspected faulty tower barrel and the suspected fault point through processing and comparison. First, the suspected faulty tower barrel is determined through the data obtained from the fixed detection. A large number of instantaneous values of the vibration displacement and acceleration at the top of the tower barrel can be directly obtained by using GNSS and the acceleration sensor. A series of top offset information is obtained through data fusion or the top offset information is directly obtained through GNSS. The top offset information can be a time-domain signal or a frequency-domain signal. The time-domain signal represents the offset amount of the top of the tower barrel at different time points, and the frequency-domain signal represents the offset amount of the top of the tower barrel at different vibration frequencies. For better comparison of different wind turbines, in this embodiment, the frequency-domain signal containing the offset amount is used for analysis. Therefore, during operation, the system will perform Fourier transform on the obtained top offset information to obtain the corresponding first frequency-domain signal S i , where i is the number of the wind turbine in the wind farm, i = 1, 2.....n, and n is the total number of wind turbines in the wind farm. Compare the first frequency-domain signals S of the tower barrels of the wind turbines close to each other i to determine the suspected faulty tower barrel r, r ∈ {i|i = 1, 2.....n}. For the generality of this embodiment, it is selected to compare the first frequency-domain signals S of all the tower barrels of the wind turbines i, where \(i = 1, 2, \cdots, n\). In actual operation, the judgment criteria for being close to each other can also be determined according to the actual situation. For example, a distance threshold is set. If the distance between the two is greater than the distance threshold, it is considered not close to each other; if it is less than or equal to the distance threshold, it is considered close to each other. Then, for the research target wind turbine, only the first frequency domain signals of the specified number of wind turbine towers around it are compared. If only some wind turbines are used for comparison, according to the actual application situation, each wind turbine should be given an equal opportunity for comparison. The specific selection is designed according to the actual wind farm distribution. The corresponding selection method is not the focus of this case and will not be elaborated here. When comparing the first frequency domain signal \(S\) i Specifically, each wind turbine tower is defined with its own risk factor \(\delta\) i . After the tower values detected by all fixed sensors are reported, the differences between the first frequency domain signals of different wind turbine towers are calculated by traversing: \(S\) n - \(S\) n-1 , \(S\) n - \(S\) n-2 ,......, \(S\) n - \(S1\), \(S\) n-1 - \(S\) n-2 , \(S\) n-1 - \(S\) n-3 ,......, \(S2 - S1\). The specific comparison scheme for the differences between the first frequency domain signals can be determined according to the actual business needs. For example, representative frequencies can be selected to view and compare the offsets of each wind turbine at these representative frequencies, or the offsets can be compared sequentially for different frequencies and then the differences in the offsets can be accumulated, etc. This specific comparison scheme is not the focus of this case and will not be further elaborated here. If the difference between the detected first frequency domain signals meets the first threshold condition \(\alpha\), then the risk factors of the corresponding two wind turbine towers are accumulated respectively. For example, if it is detected that \(S\) i - \(S1\in\alpha\), where \(\alpha\) is the first threshold condition, then \(\delta\) i = \(\delta\) i + 1, \(\delta1=\delta1 + 1\). When the risk factor \(\delta\) of a wind turbine tower i is greater than the risk threshold \(\omega1\), the tower of this wind turbine is determined to be a suspected faulty tower \(r\).

[0032] After determining the suspected faulty tower barrel r, it is further necessary to accurately determine the suspected fault point, which is determined by the barrel wall offset information on the side wall of the tower barrel obtained by the mobile sensor. Similarly, the detection of the barrel wall offset information can be directly obtained through GNSS such as GPS and Beidou, or an acceleration sensor can be installed while introducing GNSS. After the detection results obtained by the acceleration sensor are fused and reconstructed with the GNSS detection results, higher-precision offset information at different heights of the tower barrel is obtained. Therefore, in this embodiment, the mobile sensor also selects GNSS and the acceleration sensor. After the mobile sensor obtains the barrel wall offset information, it sends it to the server, and the server calculates the stiffness of the tower barrel at different heights from the obtained barrel wall offset information, and compares the stiffness of the suspected faulty tower barrel with that of the tower barrels at the same height around the suspected faulty tower barrel to determine the fault location. The specific method is as follows.

[0033] Theoretically, under the action of wind at a certain frequency, taking the barrel wall offset information of a certain height of the tower barrel at a certain moment as the research object, the following functional relationship can be obtained:

[0034]

[0035] Among them, this functional relationship takes the base of the tower barrel as the coordinate origin, x represents different height values, F(x) is the barrel wall offset information at the height x of the tower barrel obtained by integration. In the functional relationship, M(x) is a bending moment function that is only related to the height x. The bending moment function can be determined according to its shape and size during the construction of the tower barrel. K(x) represents the stiffness of the tower barrel at different heights x, and C and D are constants. Taking the second derivative of the above functional relationship can obtain:

[0036]

[0037] Then

[0038] The above formula is a functional relationship jointly composed of the second derivative of the offset information function, the bending moment function, and the stiffness function. Among them, the bending moment function M(x) can be obtained by using the basic knowledge of structural mechanics through external forces and the tower barrel structure. For the obtained different bending moment results and height values, a functional relationship between the bending moment function and the height value can be established, and then the bending moment information at any height can be obtained. The offset information function F(x) can be generated by fitting a functional relationship through a large amount of data monitored by the mobile monitoring point. During specific operations, the server obtains the barrel wall offset information at different heights on the tower barrel from the mobile sensor, and the system also performs Fourier transform on the obtained barrel wall offset information to obtain the corresponding second frequency domain signal S’ ij, where i is the number of wind turbines in the wind farm, i = 1, 2.....n, and j is the number of different heights on the same wind turbine, j = 1, 2......m. That is, there are a total of n wind turbines in the wind farm, and m detection points at different heights are selected on the tower barrel of each wind turbine. Since the vibration amplitude varies at different heights, for wind turbine i, the second frequency domain signal S’ ij is data related to the height x, and the offset information function of multiple detection points at different heights on the side wall of the tower barrel of each wind turbine is obtained by fitting:

[0039] S’ ij = F ij (x j ).

[0040] By taking the second derivative of the offset information function at each height of each wind turbine, F″ ij (x j ) can be obtained.

[0041] In summary, when a certain height x = x j of a certain wind turbine i is determined, the bending moment function and the second derivative value F ij ″(x j ) and M(x j ) corresponding to the height can be determined. Substituting them into the above function relationship about stiffness, the stiffness K j of the tower barrel at a certain height x = x ij of any wind turbine i can be obtained as follows:

[0042]

[0043] Similarly, after the suspected faulty tower barrel is determined, by comparing the stiffness K j at the same height x = x ij (x j ) of the tower barrels of different wind turbines, i = 1, 2, 3......n, the position of the sudden change in the stiffness of the tower barrel, that is, the suspected fault point, can be initially determined, and then the operator can be dispatched to the site for further inspection and maintenance of the suspected fault point. During the specific comparison, multiple wind turbine tower barrels located around the suspected faulty tower barrel can be designated as comparison tower barrels. For the generality of this embodiment, all the same wind turbine tower barrels are selected for comparison. Problem factors ε j are defined for different heights on the suspected faulty tower barrel r. Traverse and calculate the difference in stiffness at the same height between the suspected faulty tower barrel r and the comparison tower barrels being compared for each height x = x j , j = 1, 2, 3......m,

[0044] Δ(x j ) = Krj (x j ) - K ij (x j ),

[0045] i = 1, 2.....n, j = 1, 2……m, r ∈ {i|i = 1, 2.....n}。

[0046] If the detected stiffness difference satisfies the second threshold condition, then accumulate the problem factor corresponding to the height x = x j . For example, if Δ(x j ) ∈ β, where β is the second threshold condition, then ε j = ε j + 1. When the problem factor ε j of a certain height is greater than the problem threshold ω2, determine the point at this height as a suspected fault point. Further, the monitoring points for detection can be densely moved near the suspected fault point to obtain a more accurate fault point location, and send the fault point location to the relevant operation and maintenance personnel to remind the operation and maintenance personnel to check and repair the damaged point to avoid major risks.

[0047] To implement the above method, a tower barrel health status monitoring system based on fixed detection and mobile detection needs to be supported. This system includes a server using wireless communication and a detection end installed on the tower barrel. The detection end includes a fixed sensor and a mobile sensor. The mobile sensor is installed on a mobile device, and the mobile device moves up and down along the tower barrel. The mobile device can be powered by a wireless charging method. Both the mobile sensor and the fixed sensor include a GNSS and an acceleration sensor.

[0048] The above is only a preferred embodiment of the present invention, and does not impose any limitation on the technical scope of the present invention. Therefore, any changes or modifications made according to the claims and the description of the present invention shall fall within the scope covered by the patent of the present invention.

Claims

1. A method for monitoring the health status of a tower barrel based on fixed detection and mobile detection, characterized in that: A fixed sensor for detecting the sway at the top of the tower barrel of a wind turbine is fixedly installed at the top of the tower barrel of the wind turbine. A mobile sensor for detecting the vibration of the tower barrel, which can move up and down along the tower barrel, is installed on the tower barrel. By comparing the monitoring results of the fixed sensors on different tower barrels, the suspected faulty tower barrel is determined. By comparing the monitoring results of the mobile sensors on different tower barrels, the suspected fault point of the suspected faulty tower barrel is determined. The mobile sensor obtains the wall offset information on the side wall of the tower barrel, performs Fourier transform on the wall offset information to obtain a second frequency domain signal related to the height of the tower barrel, obtains the offset information functions of multiple detection points at different heights on the side walls of the tower barrels of each wind turbine from the second frequency domain signal, and respectively obtains the stiffnesses at different heights. By comparing the stiffnesses at the same height of different wind turbines, the suspected fault point is determined.

2. The method for monitoring the health status of a tower barrel based on fixed detection and mobile detection according to claim 1, characterized in that: The fixed sensor obtains the top offset information at the top of the tower barrel, performs Fourier transform on the top offset information to obtain a first frequency domain signal, and determines the suspected faulty tower barrel by comparing the first frequency domain signals of the tower barrels of wind turbines close to each other.

3. The method for monitoring the health status of a wind turbine tower based on fixed detection and mobile detection according to claim 2, characterized in that: The fixed sensor or the mobile sensor is a GNSS and an acceleration sensor.

4. The method for monitoring the health status of a tower barrel based on fixed detection and mobile detection according to claim 2, wherein: When comparing the first frequency domain signals of the tower barrels of wind turbines close to each other, a risk factor is defined for the tower barrel of each wind turbine. The differences between the first frequency domain signals of the tower barrels of different wind turbines are calculated iteratively. If the difference between the first frequency domain signals meets the first threshold condition, the risk factors of the tower barrels of the corresponding two wind turbines are respectively accumulated. When the risk factor is greater than the risk threshold, the tower barrel of this wind turbine is determined as the suspected faulty tower barrel.

5. The method for monitoring the health status of a wind turbine tower based on fixed detection and mobile detection according to claim 1, wherein: The stiffness is where M(x) is the bending moment function, and the offset information function F(x) can be generated by fitting a functional relationship through a large amount of data monitored by moving the monitoring points.

6. The method for monitoring the health status of a tower barrel based on fixed detection and mobile detection according to claim 1, wherein: When determining the suspected fault point by comparing the stiffnesses at the same height of different wind turbines, first determine the suspected faulty tower barrel. A respective problem factor is defined for each different height on the suspected faulty tower barrel. The differences in stiffness at the same height between the suspected faulty tower barrel and the comparison tower barrel being compared with it are calculated iteratively for each height. If the detected difference in stiffness meets the second threshold condition, the problem factors at the heights that meet the second threshold condition are accumulated. When the problem factor at a certain height is greater than the problem threshold, the point at this height is determined as the suspected fault point.

7. The method for monitoring the health status of a wind turbine tower based on fixed detection and mobile detection according to claim 1, characterized in that: When determining the suspected fault point by comparing the stiffnesses at the same height of different wind turbines, first determine the suspected faulty tower barrel and the comparison tower barrel. Calculate the difference between the average stiffness at the same height of the suspected faulty tower barrel and all the comparison tower barrels. If the detected difference in stiffness meets the second threshold condition, the point at this height is determined as the suspected fault point.

8. The method for monitoring the health status of a wind turbine tower based on fixed detection and mobile detection according to claim 6 or 7, characterized in that: Densely move the monitoring points near the suspected fault point to obtain a more accurate fault point position.

9. The method for monitoring the health status of a tower barrel based on fixed detection and mobile detection according to claim 1, wherein: The fixed sensor and the mobile sensor use the same set of sensor devices. When the sensor device moves up to the top of the tower barrel and stops at the top, it is the fixed sensor. When the sensor device moves up and down, it is the mobile sensor.

Images