Wind turbine monitoring and early warning method, monitoring and early warning system and readable storage medium

Abstract

The present application discloses a wind turbine monitoring and early warning method, a monitoring and early warning system, and a readable storage medium, which relate to the field of wind power generation, can monitor the performance changes of wind turbines in real time and issue fault early warnings, reduce the fault occurrence rate, improve the working efficiency of wind turbines, and at the same time enhance safety and stability. The method includes: the edge device acquires the real-time monitoring data of the wind turbine and stores it as historical monitoring data, and the ring network switch transmits the real-time monitoring data and the historical monitoring data to the vibration monitoring server; the vibration monitoring server receives the real-time monitoring data and the historical monitoring data, acquires the preset structural parameters of the wind turbine, and establishes a real-time information mapping model; the wind power fault early warning platform acquires meteorological data, calls the real-time information mapping model and calculates the vibration early warning value, and when it detects that the vibration early warning value reaches the preset dangerous early warning threshold, shuts down the wind turbine and issues an alarm.

Description

Technical Field

[0001] The present application relates to the field of wind power generation, and particularly to a method for monitoring and warning of a wind turbine, a monitoring and warning system and a readable storage medium. Background Art

[0002] With the continuous development of the social economy and the continuous penetration of the green concept, this green energy of wind energy has gradually attracted attention. As a very clean energy, wind energy can be converted into electrical energy through a wind energy unit, and then this green energy can be widely applied. However, due to the relatively complex operating conditions and its own structure of the wind turbine generator set, the external excitation force and vibration degrees of freedom it receives are more than those of other large rotating machinery. Therefore, in order to ensure the normal and stable operation of the wind turbine generator set, it is necessary to detect the operating state of the wind turbine generator set and predict faults as much as possible.

[0003] In the related art, most wind power plants use a control system to detect the operating state of the wind turbine generator set, and the control system stores information such as the state information of the wind turbine generator set and the wind speed conditions. However, the applicant has recognized that detecting the operating state of the wind turbine generator set based on static parameters cannot reflect the change process of the performance of the wind turbine generator set in real time, so that the fault has occurred when the control system issues an alarm, resulting in low equipment utilization rate. Moreover, the vibration of the fan of the wind turbine generator set is likely to cause damage to the wind turbine generator set, resulting in a high fault incidence rate, low working efficiency of the wind turbine generator set, and loss of electric energy. Summary of the Invention

[0004] In view of this, the present application provides a method for monitoring and warning of a wind turbine, a monitoring and warning system and a readable storage medium, mainly aiming to solve the problem that detecting the operating state of the wind turbine generator set based on static parameters cannot reflect the change process of the performance of the wind turbine generator set in real time, so that the fault has occurred when the control system issues an alarm, resulting in low equipment utilization rate. Moreover, the vibration of the fan of the wind turbine generator set is likely to cause damage to the wind turbine generator set, resulting in a high fault incidence rate, low working efficiency of the wind turbine generator set, and loss of electric energy.

[0005] According to a first aspect of the present application, there is provided a method for monitoring and warning of a wind turbine, the method comprising:

[0006] The edge device acquires real-time monitoring data of the wind turbine generator set, stores the real-time monitoring data of the wind turbine generator set as historical monitoring data, and transmits the real-time monitoring data and the historical monitoring data to a ring network switch, and the ring network switch transmits the real-time monitoring data and the historical monitoring data to a vibration monitoring server;

[0007] The vibration monitoring server receives the real-time monitoring data and the historical monitoring data transmitted by the ring network switch, obtains the preset structural parameters of the wind turbine, and establishes a real-time information mapping model based on the structural parameters, the real-time monitoring data, and the historical monitoring data;

[0008] The wind power fault warning platform obtains meteorological data, calls the real-time information mapping model established by the vibration monitoring server, calculates a vibration warning value based on the meteorological data and the real-time information mapping model, and when it detects that the vibration warning value reaches a preset dangerous warning threshold, shuts down the wind turbine and issues an alarm.

[0009] According to a second aspect of the present application, a monitoring and warning system is provided, which includes an end-side device, a ring network switch, a vibration monitoring server, and a wind power fault warning platform;

[0010] The end-side device is configured to obtain real-time monitoring data of the wind turbine, store the real-time monitoring data of the wind turbine as historical monitoring data, and transmit the real-time monitoring data and the historical monitoring data to the ring network switch;

[0011] The ring network switch is configured to transmit the real-time monitoring data and the historical monitoring data to the vibration monitoring server;

[0012] The vibration monitoring server is configured to receive the real-time monitoring data and the historical monitoring data transmitted by the ring network switch, obtain the preset structural parameters of the wind turbine, and establish a real-time information mapping model based on the structural parameters, the real-time monitoring data, and the historical monitoring data. Among them, when constructing a vibration mathematical model, a simplified model of the wind turbine is constructed using the geometric structure parameters and the structural material parameters, the displacements and velocities in the body coordinate system of the simplified model of the wind turbine are transformed to the inertial coordinate system according to the Euler angle rotation matrix, and the displacements and velocities in the inertial coordinate system are calculated according to the Lagrangian dynamics equation to obtain the vibration mathematical model;

[0013] The wind power fault warning platform is configured to obtain meteorological data, call the real-time information mapping model established by the vibration monitoring server, calculate a vibration warning value based on the meteorological data and the real-time information mapping model, and when it detects that the vibration warning value reaches a preset dangerous warning threshold, shuts down the wind turbine and issues an alarm.

[0014] According to a third aspect of the present application, a readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the method described in any one of the above first aspects are implemented.

[0015] With the above technical solutions, a wind turbine monitoring and early warning method, a monitoring and early warning system, and a readable storage medium provided by the present application. In the present application, the edge device acquires the real-time monitoring data of the wind turbine, stores the real-time monitoring data of the wind turbine as historical monitoring data, and transmits the real-time monitoring data and the historical monitoring data to the ring network switch. The ring network switch transmits the real-time monitoring data and the historical monitoring data to the vibration monitoring server. The vibration monitoring server receives the real-time monitoring data and the historical monitoring data transmitted by the ring network switch, acquires the preset structural parameters of the wind turbine, and based on the structural parameters, the real-time monitoring data, and the historical monitoring data, establishes a real-time information mapping model. The wind power fault early warning platform acquires meteorological data, calls the real-time information mapping model established by the vibration monitoring server, and calculates the vibration early warning value based on the meteorological data and the real-time information mapping model. When it detects that the vibration early warning value reaches the preset dangerous early warning threshold, it shuts down the wind turbine and issues an alarm, which can monitor the change process of the performance of the wind turbine in real time and issue a fault early warning, reduce the failure rate, and while improving the working efficiency of the wind turbine, it can also improve the safety and stability of the wind turbine.

[0016] The above description is only an overview of the technical solution of the present application. In order to be able to understand the technical means of the present application more clearly, it can be implemented according to the content of the description. And in order to make the above and other purposes, features, and advantages of the present application more obvious and understandable, the following specifically gives the specific implementation manners of the present application. Brief Description of the Drawings

[0017] By reading the following detailed description of the preferred embodiments, various other advantages and benefits will become clear to those of ordinary skill in the art. The drawings are only for the purpose of showing the preferred embodiments and are not considered to be a limitation of the present application. And throughout the drawings, the same reference numerals are used to represent the same components. In the drawings:

[0018] Figure 1 It shows a schematic flow chart of a wind turbine monitoring and early warning method provided by an embodiment of the present application;

[0019] Figure 2A It shows a schematic flow chart of another wind turbine monitoring and early warning method provided by an embodiment of the present application;

[0020] Figure 2B It shows a schematic structural diagram of a monopile foundation NREL 5MW wind turbine provided by an embodiment of the present application;

[0021] Figure 2C It shows a simplified model of a monopile foundation NREL 5MW wind turbine provided by an embodiment of the present application;

[0022] Figure 2DShows a schematic flow chart of establishing and using a real-time information mapping model provided by an embodiment of the present application;

[0023] Figure 2E Shows a schematic flow chart of a wind turbine monitoring and early warning method provided by an embodiment of the present application;

[0024] Figure 3A Shows an interaction schematic diagram between a terminal device and a vibration monitoring server provided by an embodiment of the present application;

[0025] Figure 3B Shows a schematic flow chart of a wind turbine monitoring and early warning method provided by an embodiment of the present application. Detailed implementation manners

[0026] Hereinafter, exemplary embodiments of the present application will be described in more detail with reference to the accompanying drawings. Although the exemplary embodiments of the present application are shown in the drawings, it should be understood that the present application can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided so that the present application can be more thoroughly understood and the scope of the present application can be completely conveyed to those skilled in the art.

[0027] An embodiment of the present application provides a wind turbine monitoring and early warning method, as Figure 1 shown, the method includes:

[0028] 101. The terminal device acquires real-time monitoring data of the wind turbine, stores the real-time monitoring data of the wind turbine as historical monitoring data, and transmits the real-time monitoring data and the historical monitoring data to a ring network switch, and the ring network switch transmits the real-time monitoring data and the historical monitoring data to a vibration monitoring server.

[0029] Among them, the environment of the wind power plant is harsh, and the structure of the wind turbine itself is also very complex, so that some parts of the wind turbine vibrate continuously during operation. The damage of components such as bearings caused by this vibration is the main reason for the wind turbine to fail and stop. However, at present, the fault diagnosis of wind turbines is based on the method of static detection of state parameters. For example, the vibration signal, oil parameter, temperature parameter or the current and voltage at the generator output end at certain moments are detected by the control system for diagnosis. However, these methods cannot reflect the performance change process of the wind turbine, so that the fault has occurred when the control system issues an alarm. Therefore, to solve this problem, the present application proposes a wind turbine monitoring and early warning method.

[0030] In the embodiment of the present application, the edge device collects relevant status data of the operation of the wind turbine in real time. Among them, the edge device can be a mobile phone, a tablet computer, a portable mobile communication device, etc. The edge device collects real-time monitoring data used to indicate the operation status of the wind turbine and transmits it to the ring network switch together with the historical monitoring data. Subsequently, the ring network switch transmits the real-time monitoring data and the historical monitoring data to the vibration monitoring server. Among them, the ring network switch can complete data exchange between the edge device and the vibration monitoring server in a complex industrial environment, so as to achieve information sharing. By collecting and transmitting the real-time monitoring data and the historical monitoring data, the change process of the operation status of the wind turbine is obtained, providing a data basis for subsequent monitoring of the operation status of the wind turbine and for fault early warning.

[0031] 102. The vibration monitoring server receives the real-time monitoring data and the historical monitoring data transmitted by the ring network switch, obtains the preset structural parameters of the wind turbine, and establishes a real-time information mapping model based on the structural parameters, the real-time monitoring data, and the historical monitoring data.

[0032] The vibration monitoring server can remotely monitor the operation status of the wind turbine in real time, build a real-time information mapping model for monitoring the operation status of the wind turbine, and provide reliable data for analyzing the cause of the fault and for fault early warning. Among them, the real-time information mapping model can combine the real-time monitoring data and the historical monitoring data with the structural parameters of the wind turbine to more accurately and detailedly monitor the operation status of the wind turbine. In the embodiment of the present application, the vibration monitoring server obtains the preset structural parameters and receives the real-time monitoring data and the historical monitoring data transmitted by the ring network switch, and then establishes a real-time information mapping model of the wind turbine based on the parameter structure, the real-time monitoring data, and the historical monitoring data. In the actual application process, since the real-time monitoring data is constantly updated, the real-time information mapping model of the wind turbine can be continuously optimized and updated by using the obtained real-time monitoring data, so that the information predicted by the real-time information mapping model is more accurate, while improving the data accuracy and ensuring timely response to emergencies.

[0033] 103. The wind power fault early warning platform obtains meteorological data, calls the real-time information mapping model established by the vibration monitoring server, calculates the vibration early warning value based on the meteorological data and the real-time information mapping model, and when it detects that the vibration early warning value reaches the preset dangerous early warning threshold, shuts down the wind turbine and issues an alarm.

[0034] Among them, the wind power fault warning platform adopts a B / S (Browser / Server) architecture with WebGIS (Web Geographic Information System) technology as the core. It obtains numerical meteorological forecast data including air pressure, wind speed, and wind direction data in the area where the wind farm is located through the meteorological private network, Internet technology, and database technology, and quickly analyzes the obtained numerical meteorological forecast data and completes real-time data sharing between the wind power fault warning cloud platform and the vibration monitoring server. In addition, the wind power fault warning platform can also calculate the vibration warning value based on the real-time information mapping model established by the vibration monitoring server. Among them, the vibration warning value is used to describe the strength of the vibration of the wind turbine unit, and the wind power fault warning platform divides the operating state of the wind turbine unit into 5 risk levels according to the vibration warning value, namely, no risk level, vibration risk level I, vibration risk level II, vibration risk level III, and vibration danger level.

[0035] In the embodiment of the present application, the wind power fault warning platform obtains the meteorological data in the area where the wind farm is located and calls the real-time information mapping model established by the vibration monitoring server. Among them, the meteorological data includes air pressure, wind speed, and wind direction data at a future time. The wind power fault warning platform can predict the operating state of the wind turbine unit under this weather condition and whether a fault will occur under this operating state according to the weather condition at a future time, so as to achieve the purpose of wind turbine unit fault warning. Correspondingly, when it is detected that the vibration warning value reaches the preset dangerous warning threshold, it means that the operating risk level of the wind turbine unit is high in the future time period, and it is predicted that the wind turbine unit will malfunction. Therefore, the wind turbine unit is shut down and an alarm is sent to the maintenance personnel. The maintenance personnel can perform maintenance on the wind turbine unit in advance to avoid major faults and reduce the working efficiency of the wind turbine unit. In the actual application process, the preset dangerous warning value can be the upper limit value of the vibration risk level III. By real-time monitoring of the vibration warning value, the wind farm can reduce the failure rate of the wind turbine unit even in the face of extreme weather, improve the reliability of equipment operation, and improve the working efficiency of the wind turbine unit.

[0036] The method provided by the embodiment of the present application, which establishes a real-time information mapping model, the wind power fault warning platform calculates the vibration warning value based on the meteorological data and the real-time information mapping model, and when it is detected that the vibration warning value reaches the preset dangerous warning threshold, shuts down the wind turbine unit and issues an alarm, can real-time monitor the change process of the performance of the wind turbine unit and issue a fault warning, reduce the failure rate, and while improving the working efficiency of the wind turbine unit, can also improve the safety and stability of the wind turbine unit.

[0037] Furthermore, as a refinement and extension of the specific implementation manner of the above embodiment, in order to completely illustrate the specific implementation process of this embodiment, the embodiment of the present application provides another method for monitoring and warning of wind turbine units, as Figure 2AAs shown, the method includes:

[0038] 201. The sensor cluster of the edge device monitors the wind turbine in real time and collects data to obtain real-time monitoring data, and stores the real-time monitoring data in the storage unit of the edge device.

[0039] In the embodiment of the present application, the sensor cluster of the edge device includes multiple data acquisition units. These data acquisition units are installed at positions such as the main shaft of the wind turbine, the gearbox, the input shaft of the generator set, and the nacelle housing, etc., and are used to monitor the vibration data and inclination data of the main shaft of the wind turbine and the input shaft of the generator set in real time. Among them, the data acquisition unit is composed of a vibration sensor and an inclination sensor. By installing the data acquisition unit at some key parts of the operation of the wind turbine, the operation data of the wind turbine collected is more accurate and more detailed, and the operation data such as the vibration data and inclination data collected can more comprehensively reflect the real-time operation state of the wind turbine, providing a data basis for the subsequent establishment of a real-time information mapping model. The data acquisition unit takes the vibration data and inclination data and other operation data collected as the real-time monitoring data of the wind turbine, and stores the real-time monitoring data in the storage unit of the edge device. Among them, the storage unit is in the memory of the edge device and is used to store the historical monitoring data and real-time monitoring data of the wind turbine. A large amount of historical monitoring data and real-time monitoring data can comprehensively reflect the operation change process of the wind turbine, and the subsequent use of these stored monitoring data can more accurately predict the operation state of the wind turbine.

[0040] In another optional implementation scheme, some wind turbines work at sea. Due to the complex and changeable sea environment, there will be phenomena such as single data loss or data abnormality when the sensor monitors data. Therefore, the most similar filling method can be used to process the situation of single data loss, and the missing data can be supplemented to ensure that the data included in each data group is complete, which is convenient for subsequent participation in calculations. The specific process of processing the situation of single data loss is as follows:

[0041] First, the edge device will identify the real-time monitoring data, and when it identifies that a single data is missing in the real-time monitoring data, it will determine the missing array where the single data is missing in the real-time monitoring data. In the actual application process, since some single variables in the array may be abnormal, such as having a very large value, which is completely inconsistent with the actual situation, etc., these abnormal variables actually also need to be corrected. Therefore, in the embodiment of the present application, when identifying whether a single data is missing in the real-time monitoring data, the Mahalanobis distance metric can also be used to identify whether there are abnormal variables in the real-time monitoring data, and when it is determined that there are abnormal variables, determine the array where the abnormal variables are located, remove the abnormal variables in the array, and use the array after removal as the missing array. Subsequently, appropriate values that are no longer abnormal are supplemented for the array, thereby reducing the numerical error and improving the accuracy of prediction.

[0042] Subsequently, the edge device extracts multiple sample arrays from the multiple historical monitoring data included in the storage unit. Among them, the multiple sample arrays and the missing array are collected by the same sensor. Considering that the data collected by the same sensor has the highest similarity, which can make the filled missing array closer to the normal array without data loss, so multiple sample arrays collected by the same sensor are obtained here, and subsequently, the multiple sample arrays are used to supplement the missing data. After obtaining the multiple sample arrays, the edge device reads the correlation between each sample array and the missing array in the multiple sample arrays respectively, and extracts the target sample array with the highest correlation with the missing array in the multiple sample arrays, and determines the target sample array most similar to the missing array in the multiple sample arrays through the correlation, so that the target sample array is closest to the missing array without abnormality, ensuring the accuracy of the real-time monitoring data sample.

[0043] Next, the edge device sorts the multiple sample values included in the target sample array in descending order to obtain a numerical sorting result, extracts the numerical numbers marked for each sample value in the multiple sample values, and sorts the multiple numerical numbers of the multiple sample values according to the order of the multiple sample values in the numerical sorting result to obtain a number sorting result. Then, the edge device determines the missing numerical number of the single data that is missing in the missing array, determines the target numerical number that is the same as the missing numerical number in the number sorting result, and determines the first numerical number and the second numerical number adjacent to the target numerical number in the number sorting result. Specifically, the first numerical number and the second numerical number can be the previous numerical number and the next numerical number adjacent to the target numerical number.

[0044] Finally, the edge device queries the first value with the same numerical number as the first numerical number, the second value with the same numerical number as the second numerical number in the missing array, calculates the average value of the first value and the second value, and supplements the average value as the value corresponding to the missing numerical number in the missing data group to the missing array. Since the target sample array has the highest correlation with the missing array, the value of the target numerical number is the same as the missing value. Therefore, the missing value is also between the first value and the second value, making the calculated average value of the first value and the second value closer to the normal value without anomalies, providing more accurate and complete real-time monitoring data for subsequent prediction of the operating state of the wind turbine.

[0045] The following uses a specific example to describe in detail the process of handling a single data missing situation above: Assume that the edge device identifies a set of sensor monitoring vibration array samples as active power a, active power b, active power c, active power d, and active power e respectively, as shown in Table 1, where the edge device identifies the missing data as active power e6.

[0046] Active power a Active power b Active power c Active power d Active power e a1 b1 c1 d1 e1 a2 b2 c2 d2 e2 a3 b3 c3 d3 e3 a4 b4 c4 d4 e4 a5 b5 c5 d5 e5 a6 b6 c6 d6 e6

[0047] Table 1

[0048] When processing, first find the target sample array with the highest correlation with the missing array of active power e (such as active power c), sort the six sample values c1, c2, c3, c4, c5, and c6 in active power c in descending order, and extract the numerical numbers marked for each sample value among the six sample values. Then, sort the numerical numbers of these six sample values according to the order of the six sample values in the numerical sorting result, and the obtained number sorting result is 3-1-6-2-4-5. Similarly, sort the six values e1, e2, e3, e4, e5, and e6 in the missing array of active power e according to the number sorting result, and the sorting result is e3-e1-e6-e2-e4-e5, as shown in Table 2.

[0049] Active power a Active power b Active power c Active power d Active power e a3 b3 c3 d3 e3 a1 b1 c1 d1 e1 a6 b6 c6 d6 e6 a2 b2 c2 d2 e2 a4 b4 c4 d4 e4 a5 b5 c5 d5 e5

[0050] Table 2

[0051] As can be seen from Table 2 above, the number of the missing value e6 is 6, which is also the target numerical number described in the above process. The two numbers adjacent to 6 are 1 and 2 (i.e., the first numerical number and the second numerical number) respectively. Therefore, extract e1 with the number 1 and e2 with the number 2 in active power e, calculate the average value of e1 and e2, (e1 + e2) / 2, and supplement (e1 + e2) / 2 as the missing value e6 to active power e.

[0052] It should be noted that if the target numerical ID is at the beginning or end of the ID sorting result, it is impossible to obtain the two numerical IDs adjacent to the target numerical ID, and there is only one numerical ID. For example, e5 in Table 2 above. If e5 is missing, only one adjacent e4 can be obtained. Therefore, in the embodiments of the present application, for this situation, a target numerical ID adjacent to the target numerical ID can be determined in the ID sorting result, and the value in the missing array with the same numerical ID as the target numerical ID is used as the value corresponding to the missing numerical ID in the missing data group and supplemented to the missing array. That is, when e5 is missing, the value of its adjacent e4 can be directly supplemented to e5, and the same applies to the first one.

[0053] 202. The edge device obtains historical monitoring data from the storage unit and transmits the collected real-time monitoring data and historical monitoring data to the ring network switch, and the ring network switch transmits the real-time monitoring data and historical monitoring data to the vibration monitoring server.

[0054] In the embodiments of the present application, both the edge device and the vibration monitoring server are connected to the ring network switch to facilitate the transmission of real-time monitoring data and historical monitoring data. Among them, the ring network switch is used for data exchange between the edge device and the vibration monitoring server, can perform data transmission between multiple port pairs at the same time, and the ring network switch can execute data sending operations and data receiving operations in parallel to improve the data transmission efficiency in the industrial environment.

[0055] The edge device obtains historical monitoring data from the storage unit of the memory and transmits the real-time monitoring data and historical monitoring data collected by the sensor cluster to the ring network switch together. The real-time monitoring data and historical monitoring data are transmitted to the vibration monitoring server through the ring network switch, quickly transmitting the real-time monitoring data and historical monitoring data from the edge device to the vibration monitoring server, improving the data transmission efficiency and ensuring the accuracy of the data at the same time.

[0056] 203. The vibration monitoring server receives the real-time monitoring data and historical monitoring data transmitted by the ring network switch, obtains the preset structural parameters of the wind turbine, and establishes a real-time information mapping model based on the structural parameters, real-time monitoring data, and historical monitoring data.

[0057] In the embodiment of the present application, the structural parameters of the wind turbine are pre-set in the vibration monitoring server. The structural parameters of the wind turbine include geometric structure parameters and structural material parameters. The geometric structure parameters include the size parameters, mass, moment of inertia, transmission ratio of the transmission chain, sectional moment of inertia, etc. of the main components such as the fan main shaft, gearbox, and doubly-fed generator in the wind turbine; the structural material parameters include the connection stiffness, material type, bending stiffness, density, elastic modulus of the main components in the wind turbine and the components connected and fixed to them, and the electromagnetic characteristics of the doubly-fed generator. The vibration monitoring server regards the operating wind turbine as a mass-stiffness-damping mechanical vibration structure subjected to external excitation. Under the action of external excitation and electromagnetic force, this mechanical vibration structure undergoes complex mechanical vibrations. Therefore, when establishing a real-time information mapping model, the vibration monitoring server will comprehensively consider the geometric structure parameters and structural material parameters of the wind turbine, and establish a real-time information mapping model based on the structural parameters, real-time monitoring data, and historical monitoring data. The specific process of establishing the real-time information mapping model is as follows:

[0058] First, the vibration monitoring server obtains the pre-set geometric structure parameters and structural material parameters of the wind turbine, performs modeling processing on the pre-set geometric structure parameters and structural material parameters according to the Lagrangian dynamics equation, obtains the vibration mathematical model of the wind turbine vibration structure, and uses the vibration mathematical model as the initial mathematical model of the wind turbine. The initial mathematical model established based on the geometric structure parameters and structural material parameters of the wind turbine is more in line with the actual structural situation of the wind turbine, providing a data basis for subsequent prediction of the operating state of the wind turbine and fault warning.

[0059] Among them, when constructing the vibration mathematical model, first use the geometric structure parameters and structural material parameters to construct a simplified fan model, then transform the displacements and velocities in the body coordinate system of the simplified fan model to the inertial coordinate system according to the Euler angle rotation matrix, and then calculate the displacements and velocities in the inertial coordinate system according to the Lagrangian dynamics equation to obtain the vibration mathematical model. The specific process is as follows:

[0060] As Figure 2B shown, the wind turbine model of the present application can adopt the NREL 5MW wind turbine with a monopile foundation. In the simplified model, the monopile foundation wind turbine is modeled as a two-stage inverted pendulum, and the structural stiffness and damping are modeled as a rotary spring and a rotary damper at the bottom of the rigid body. The tower and the foundation are simplified as a rigid connection.

[0061] As Figure 2CAs shown in the figure, in the NREL 5MW wind turbine with a monopile foundation, the inertial coordinate system Op is fixed on the seabed, and its origin Op is located at the hinge between the wind turbine foundation and the seabed in the simplified model of the wind turbine. The OX axis is along the mean wind direction, and the OZ axis is vertically upward; the body coordinate system Ot is fixed on the wind turbine foundation, and its origin Ot is located at the hinge between the foundation and the tower in the simplified model of the wind turbine. The OZ axis is on the same straight line as the foundation axis, and the projection of the OX axis on the horizontal plane is along the mean wind direction; the body coordinate system Or is fixed on the wind turbine tower, and its origin Or is located at the connection between the nacelle and the rotor. The OX axis is on the same straight line as the axis of the rotor main shaft, and the OZ axis is parallel to the tower axis. The simplified model has 7 degrees of freedom, which are θ px , θ py , θ pz , θ tx , θ ty , θ tz , Ω, where θ px , θ py , θ pz are the rotations of the monopile foundation relative to the seabed, representing the rotations of the monopile foundation in the inertial coordinate system Op, θ tx , θ ty , θ tz are the rotations of the tower relative to the foundation, and Ω is the rotation of the rotor relative to the nacelle. The external load of the simplified model is the air load F, where F = [F x F y F z M] T .

[0062] First, according to the Euler angle rotation matrix, the displacements and velocities in the body coordinate system are transformed to the inertial coordinate system. The angular displacements θ tx , θ ty , θ tz of the tower in the body coordinate system Ot are transformed to the inertial coordinate system, and the calculation process is as shown in Equation 1:

[0063] Equation 1:

[0064] where, represents the rotation angle of the tower in the inertial coordinate system Op, represents the rotation angle of the tower in the body coordinate system Ot, and represents the Euler rotation matrix for transforming the body coordinate system Ot to the inertial coordinate system Op.

[0065] Similarly, the angular displacement Ω of the rotor in the body coordinate system On is transformed to the inertial coordinate system, and the calculation process is as shown in Equation 2:

[0066] Equation 2:

[0067] where, represents the rotation angle of the rotor in the inertial coordinate system Op, represents the rotation angle of the wind turbine in the body coordinate system Ot, and represents the Euler rotation matrix for transforming the body coordinate system Ot to the inertial coordinate system Op, represents the rotation angle of the wind turbine in the body coordinate system Or.

[0068] Secondly, a multi-rigid-body dynamics model is established. The specific process is as follows:

[0069] The kinetic energy is calculated. The total kinetic energy of the system T = T p + T t + T r , where T p , T t , T r are the rotational kinetic energies of the monopile foundation, tower barrel, and wind turbine respectively. The calculation process is as shown in Equations 3, 4, and 5:

[0070] Equation 3:

[0071]

[0072] where, represents the inertia tensor of the monopile foundation about the origin Op of the inertial coordinate system.

[0073] Equation 4:

[0074] where, represents the inertia tensor of the tower barrel about the origin Op of the inertial coordinate system.

[0075] Equation 5:

[0076] where, represents the inertia tensor of the wind turbine about the origin Op of the inertial coordinate system.

[0077] Then, the potential energy is calculated. The total potential energy of the system U = U p + U t + U r + U1 + U2, where U p , U t , U r , U1, and U2 represent the gravitational potential energies of the monopile foundation, tower barrel, and wind turbine, and the rotational spring potential energies at the origin of the inertial coordinate system Op and the origin of the body coordinate system Ot respectively. The calculation process is as shown in Equations 6, 7, 8, 9, and 10:

[0078] Equation 6: U p = m p (z p - L p )

[0079] where, z p = Lp · cosθ px · cosθ py and Z p represents the height of the centroid of the single - pile foundation, m p represents the mass of the single - pile foundation, L p represents the length from the centroid of the single - pile foundation to the origin of the inertial system Op

[0080] Formula 7: U t = m t [(z ot + z t1 ) - L t

[0081] where z ot = L1· cosθ px · cosθ py , Z ot represents the height of the centroid of the tower barrel relative to the origin of the body coordinate system Ot and L1 represents the total length of the single - pile foundation, z t1 = L t1 · cosθ tx · cosθ ty , Z t1 represents the height of the centroid of the tower barrel relative to the origin of the body coordinate system Ot and L t1 represents the length from the centroid of the tower barrel to the body coordinate system Ot, Z t = Z ot + Z t1 and Z t represents the height of the centroid of the tower, m t represents the mass of the tower barrel

[0082] Formula 8: U r = m r [(z ot + z t2 ) - L r

[0083] where z t2 = L r1 · cosθ tx · cosθ ty , Z t2 represents the height of the centroid of the wind turbine relative to the origin of the coordinate system Ot and L r1 represents the length from the centroid of the wind turbine to the origin of the body coordinate system Ot, L r represents the length from the centroid of the wind turbine to the origin of the inertial system Op, Z r = Z ot + Z t2 and Z r represents the height of the centroid of the wind turbine, m r represents the mass of the wind turbine ​​

[0084] Equation 9: U1 = K 1x ·θ 2 px +K 1y ·θ 2 py

[0085] where K 1x represents the rotational stiffness about the x-axis of the rotational spring at the hinge at the origin of the inertial frame Op, and K 1y represents the rotational stiffness about the y-axis of the rotational spring at the hinge at the origin of the inertial frame Op.

[0086] Equation 10: U2 = K 2x ·(θ tx -θ px ) 2 +K 2y ·(θ ty -θ ty ) 2

[0087] where K 2x represents the rotational stiffness about the x-axis of the rotational spring at the hinge at the origin of the body coordinate system Ot, and K 2y represents the rotational stiffness about the y-axis of the rotational spring at the hinge at the origin of the body coordinate system Ot.

[0088] Then, the non-conservative force calculation is performed. The non-conservative forces are the external load and the damping force. The external load is the aerodynamic load F = [F x F y F z M] T , where F x , F y , F z represent the concentrated forces in the x, y, and z directions generated by the aerodynamic load acting on the wind turbine, and M is the torque generated by the aerodynamic load acting on the wind turbine. The damping force calculation is as shown in Equation 11, Equation 12, Equation 13, and Equation 14:

[0089] Equation 11:

[0090] Equation 12:

[0091] where F 1x , F 1y represent the damping forces generated by the rotational damping at the hinge at the origin of the inertial frame Op, d 1x represents the damping coefficient about the x-axis of the rotational damping at the hinge at the origin of the inertial frame Op, and d 1y represents the damping coefficient about the y-axis of the rotational damping at the hinge at the origin of the inertial frame Op.

[0092] Equation 13:

[0093] Formula 14:

[0094] Where, F 2x 、F 2y represent the damping force generated by the rotational damping of the hinge at the origin of the body coordinate system Ot, d 2x represents the damping coefficient of the rotational damping of the hinge at the origin of the body coordinate system Ot about the x-axis, d 2y represents the damping coefficient of the rotational damping of the hinge at the origin of the body coordinate system Ot about the y-axis.

[0095] Then, the Lagrangian dynamics equation is as in Formula 15:

[0096] Formula 15: L = T - U

[0097] Calculate the displacements and velocities in the inertial coordinate system according to the Lagrangian dynamics equation, and list Formulas 16 to 22:

[0098] Formula 16:

[0099] Formula 17:

[0100] Formula 18:

[0101] Formula 19:

[0102] Formula 20:

[0103] Formula 21:

[0104] Formula 22:

[0105] By organizing the above formulas, finally obtain the vibration mathematical model, and this vibration mathematical model is as in Formula 23:

[0106] Formula 23:

[0107] Where, [M0] represents the stiffness matrix, and [C0] represents the damping matrix.

[0108] Subsequently, the vibration monitoring server receives the real-time monitoring data and historical monitoring data transmitted by the ring network switch, screens the real-time monitoring data and historical monitoring data, extracts the effective operating state parameters with non-empty values and numerical errors within the error range, and avoids the influence of abnormal and overly error-prone real-time monitoring data and historical monitoring data on the initial mathematical model, thereby improving the accuracy of the operation state monitoring and early warning of the wind turbine.

[0109] Finally, based on the initial mathematical model, the model parameters of the initial mathematical model are corrected using the effective operating state parameters to obtain a real-time information mapping model of the wind turbine. Through the real-time information mapping model, the development trend of the operating state of the wind turbine can be reflected more accurately and completely. Instead of analyzing the operating state of the wind turbine based on static data, it can monitor the operating state of the wind turbine in real time and perform maintenance on the wind turbine in a timely manner to reduce the failure rate.

[0110] It should be noted that after the real-time information mapping model is established, due to a series of factors such as the complex and changeable operating environment of the wind turbine and the errors in the preset geometric structure parameters and structural material parameters, the established real-time information mapping model is relatively rough and cannot accurately and completely predict the operating state of the wind turbine. Therefore, in the embodiment of the present application, an information mapping method is adopted. After the real-time information mapping model is established, the real-time monitoring data collected and the historical monitoring data stored are continuously used to update the real-time information mapping model in real time. As the real-time monitoring data is continuously updated, the real-time information mapping model is also continuously updated in real time, realizing an accurate mapping from the real physical world wind turbine to the virtual mathematical model and improving the accuracy of the model.

[0111] In summary, the specific process of establishing and using the real-time information mapping model is summarized as follows:

[0112] As Figure 2D shown, obtain the geometric structure parameters and structural material parameters preset in the vibration monitoring server. Subsequently, perform dynamic modeling according to the Lagrangian dynamics equation to obtain the vibration mathematical model of the mechanical vibration structure. Then, obtain the real-time monitoring data and historical monitoring data of the sensor, and use the sensor monitoring data to update in real time, the historical operating state (historical monitoring data), etc., to correct the parameters of the initial mathematical model and obtain a multi-scale and high-precision real-time information mapping model. Realize an accurate mapping from the real physical world wind turbine to the virtual mathematical model. Finally, the vibration monitoring server performs a simulation analysis on the real-time information mapping model, performs an adaptive working condition division according to the simulation results, and based on the working condition division results, controls the pitch system of the wind turbine in real time.

[0113] 204. The wind power fault warning platform obtains meteorological data, calls the real-time information mapping model established by the vibration monitoring server, and calculates the vibration warning value based on the meteorological data and the real-time information mapping model. When it is detected that the vibration warning value reaches the preset dangerous warning threshold, step 205 below is executed; when it is detected that the vibration warning value does not reach the preset dangerous warning threshold, step 206 below is executed.

[0114] In the embodiment of the present application, the wind power fault warning platform is divided into four levels, namely the data meteorological interface layer, the database layer, the service layer, and the visualization interface layer. At the meteorological data interface layer of the wind power fault warning platform, weather numerical information of the wind farm location where the wind turbine is located is obtained from the meteorological private network through the meteorological data interface, and the weather numerical information is stored in the database in the form of a text format message. In addition, weather numerical information within a preset future time range is obtained from the meteorological data stored in the database. The weather numerical information includes air pressure, wind speed, and wind direction data of the area where the wind farm is located. The text message format can be a JSON (Java Script Object Notation, a lightweight data exchange format) format message. It should be noted that in order to avoid the influence of excessive weather numerical information on the fault warning of the wind power fault warning platform, effective weather numerical information needs to be extracted. Therefore, the database layer reads the weather numerical information in the form of a text format message obtained from the data meteorological interface layer, analyzes and screens the weather numerical information using the historical meteorological data stored in the database, and filters out the invalid information in the weather numerical information. The database layer includes data files, data information transmission programs, data information analysis programs, data information screening programs, data information processing programs, data storage, and databases, which can analyze, screen, and process the data transmitted to the database layer to avoid the influence of invalid information or abnormal information on predicting the operating status of wind turbines and fault warning. Subsequently, as the key layer of the wind power fault warning cloud platform, the service layer calls the real-time information mapping model established by the vibration monitoring server, and based on the filtered weather numerical information and the real-time information mapping model, conducts a simulation analysis on the wind turbine to obtain the predicted operating status data within a preset future time range and the corresponding fault risk level of the predicted operating status data. In order to be able to detect possible faults in a timely manner at the initial stage of the fault and repair them, the data fusion control unit of the service layer performs a normalization calculation on the predicted operating status data and the fault risk level to obtain the vibration warning value of the wind turbine. The vibration warning value is used to describe the intensity of the vibration of the wind turbine, and according to the vibration warning value, the operating status of the wind turbine is divided into five risk levels, namely no risk level, vibration risk level I, vibration risk level II, vibration risk level III, and vibration danger level. Through the vibration warning value, the risk degree of the operation of the wind turbine can be more intuitively reflected, and a fault warning can be issued in a timely manner, greatly reducing the incidence of wind turbine faults, thereby improving the operating efficiency and economic benefits of the wind turbine.

[0115] Correspondingly, the wind power fault warning platform monitors the vibration warning value in real time. When it detects that the vibration warning value reaches the preset dangerous warning threshold, it indicates that the wind turbine is about to malfunction and warning operations need to be performed. Therefore, step 205 below is executed; when it detects that the vibration warning value does not reach the preset dangerous warning threshold, it indicates that the wind turbine is in normal operation and warning is not required temporarily. Therefore, step 206 below is executed.

[0116] Furthermore, in another optional implementation, the visualization interface layer of the wind power fault warning platform can achieve more efficient human-machine interaction. Specifically, it can use the image information of the wind turbine to establish a three-dimensional model of the wind turbine, call the device operation status data of the vibration monitoring server and the predicted operation status data of the wind power fault warning platform, and generate a visualization interface based on the three-dimensional model, device operation status data, and predicted operation status data, providing a more intuitive visual interface for the maintenance personnel of the wind power fault warning cloud platform, enabling the maintenance personnel to monitor the operation status of the wind turbine through the visualization interface.

[0117] 205. When the vibration warning value reaches the preset dangerous warning threshold, shut down the wind turbine and issue an alarm.

[0118] In the embodiment of the present application, due to the complex and changeable operating environment of the wind turbine, it often faces extreme weather, and the structure of the wind turbine itself is complex, which will take a long time in fault detection and repair and have a great impact on power production. Therefore, in order to reduce the incidence of wind turbine failures, the upper limit value of vibration risk level III in the present application is defined as the dangerous warning threshold. When it is detected that the vibration warning value reaches the preset dangerous warning threshold, it indicates that the wind turbine is about to malfunction, and it is necessary to control the wind turbine to stop quickly and alarm the maintenance personnel. In this way, the operation status of the wind turbine is detected in real time through the vibration warning value and a fault warning is provided in a timely manner, improving the safety and stability of the wind turbine and at the same time increasing the equipment utilization rate of the wind turbine.

[0119] It should be noted that after the alarm is issued, in order to improve the accuracy of predicting the operation status of the wind turbine, the wind power fault warning platform will store the filtered weather numerical information and predicted operation status data in the database, so as to perform maintenance troubleshooting operations using the filtered weather numerical information and predicted operation status data and update the model of the real-time information mapping model, improving the model accuracy, reducing the incidence of wind turbine failures, and increasing the equipment utilization rate.

[0120] 206. When it is detected that the vibration warning value does not reach the preset dangerous warning threshold, the vibration monitoring server reads the real-time information mapping model. Based on the real-time information mapping model, the wind turbine is simulated and analyzed to obtain the equipment operation status data, and the adaptive working condition division is carried out based on the equipment operation status data to obtain the working condition division result. According to the working condition division result, the pitch system of the wind turbine is controlled in real time.

[0121] In the embodiment of the present application, when it is detected that the vibration warning value does not reach the preset dangerous warning threshold, it indicates that the wind turbine is in a normal operation state and early warning is not temporarily required. Therefore, the vibration monitoring server reads the real-time information mapping model. Based on the real-time information mapping model, the wind turbine is simulated and analyzed to obtain the equipment operation status data, and the adaptive working condition division is carried out based on the equipment operation status data to obtain the working condition division result. Among them, the adaptive working condition is a processing method for the wind turbine to automatically adjust its own state during operation. Through the adaptive working condition division, the wind turbines are classified according to the equipment operation status data, and the wind turbines with similar equipment operation status are determined. Furthermore, the wind turbines with abnormal equipment operation status data are controlled to avoid damage or even fracture of the blades of the wind turbines due to fatigue. Therefore, according to the working condition division result, the pitch system of the wind turbine is controlled in real time, which can reduce the impact of vibration on the wind turbine, improve the working efficiency of the wind turbine, and enhance the safety and stability of the wind turbine.

[0122] In summary, the specific process of the wind turbine monitoring and early warning method proposed in the present application is as follows:

[0123] As Figure 2E shown, the edge device acquires the real-time vibration monitoring data and sends the real-time monitoring data to the ring network switch. The ring network switch sends the real-time monitoring data to the vibration monitoring server through the ring network. Subsequently, the vibration monitoring server establishes a real-time information mapping model according to the initial mathematical model of the wind turbine and the real-time monitoring data. Then, the wind power fault early warning cloud platform acquires the meteorological data of the meteorological special network / numerical value, and conducts a simulation analysis on the real-time information mapping model provided by the vibration monitoring server to obtain the vibration warning value, and evaluates and judges the risk level of the simulation result. Then, the wind power fault early warning cloud platform transmits the simulation result and the meteorological data to the vibration monitoring server to complete the data exchange. Finally, when the wind power fault early warning cloud platform detects that the vibration warning value reaches the vibration warning threshold, it shuts down and alarms the maintenance personnel; when the wind power fault early warning cloud platform detects that the vibration warning value does not reach the vibration warning threshold, it transfers the control right to the vibration monitoring server, and the vibration monitoring server controls the pitch system in real time, thereby affecting the real-time monitoring data collected by the edge device.

[0124] The method provided by the embodiments of the present application can monitor the change process of the performance of a wind turbine in real time and issue a fault warning, reduce the fault occurrence rate, and while improving the working efficiency of the wind turbine, can also enhance the safety and stability of the wind turbine.

[0125] Further, as Figure 1 a specific implementation of the method, the embodiments of the present application provide a monitoring and warning system, as Figure 3A shown, including end-side devices, a ring network switch, a vibration monitoring server, and a wind power fault warning platform:

[0126] The end-side devices include a sensor cluster and a memory, and are used to obtain the real-time monitoring data of the wind turbine, store the real-time monitoring data of the wind turbine as historical monitoring data, and transmit the real-time monitoring data and the historical monitoring data to the ring network switch. The sensor cluster includes a plurality of data acquisition units installed on the main shaft of the wind turbine, the gearbox, the input shaft of the generator set, and the nacelle housing. Each data acquisition unit in the plurality of data acquisition units includes a vibration sensor and an inclination sensor. The sensor cluster is used to monitor the vibration data and inclination data of the main shaft of the wind turbine and the input shaft of the generator set in real time as real-time monitoring data, and store the real-time monitoring data in the storage unit of the memory. The memory is used to store the historical monitoring data and the real-time monitoring data of the wind turbine, and the memory includes a storage unit.

[0127] The ring network switch is used to transmit the real-time monitoring data and the historical monitoring data to the vibration monitoring server.

[0128] The vibration monitoring server is used to receive the real-time monitoring data and the historical monitoring data transmitted by the ring network switch, obtain the preset structural parameters of the wind turbine, and establish a real-time information mapping model based on the structural parameters, the real-time monitoring data, and the historical monitoring data. Among them, when constructing the vibration mathematical model, a simplified model of the wind turbine is constructed using geometric structure parameters and structural material parameters, the displacements and velocities in the body coordinate system of the simplified model of the wind turbine are transformed to the inertial coordinate system according to the Euler angle rotation matrix, and the displacements and velocities in the inertial coordinate system are calculated according to the Lagrangian dynamics equation to obtain the vibration mathematical model.

[0129] The wind power fault warning platform includes a meteorological data interface layer, a database layer, a service layer, and a visualization interface layer, and is used to obtain meteorological data, call the real-time information mapping model established by the vibration monitoring server, calculate the vibration warning value based on the meteorological data and the real-time information mapping model, and when it is detected that the vibration warning value reaches the preset dangerous warning threshold, shut down the wind turbine and issue an alarm.

[0130] The meteorological data interface layer is used to connect to the meteorological private network through the meteorological data interface, obtain the meteorological data of the area where the wind farm is located, store the meteorological data in the database of the database layer, and obtain the meteorological data in the database through the meteorological data interface, convert the meteorological data into a text format message, and transmit the text format message to the wind power fault warning platform.

[0131] The database layer is used to read the text format message of the meteorological data interface layer, extract the weather numerical information within a preset future time range from the text format message, analyze and screen the weather numerical information using the historical meteorological data stored in the database, filter out the invalid information in the weather numerical information, and store the filtered weather numerical information and predicted operation status data transmitted by the service layer into the database. Among them, the database layer includes data files, data information transmission programs, data information analysis programs, data information screening programs, data information processing programs, data storage, and databases.

[0132] The service layer is used to obtain the filtered weather numerical information of the database layer and the real-time information mapping model established by the vibration monitoring server. Based on the filtered weather numerical information and the real-time information mapping model, perform a simulation analysis on the wind turbine to obtain the predicted operation status data within a preset future time range and the corresponding fault risk level of the predicted operation status data. The data fusion control unit of the service layer performs a normalization calculation on the predicted operation status data and the fault risk level to obtain the vibration warning value of the wind turbine, and monitors the vibration warning value in real time. When the vibration warning value reaches the preset danger warning threshold, shut down the wind turbine and issue an alarm. The vibration warning value includes a risk-free level, vibration risk level I, vibration risk level II, vibration risk level III, and vibration danger level. The preset danger warning threshold is the upper limit value when the vibration warning value reaches vibration risk level III.

[0133] Among them, when the service layer detects that the vibration warning value reaches the preset risk threshold, the service layer stores the filtered weather numerical information and the predicted operation status data in the database for performing maintenance troubleshooting operations using the filtered weather numerical information and the predicted operation status data and updating the model of the real-time information mapping model. The preset risk threshold is the lower limit value when the vibration warning value reaches vibration risk level I.

[0134] The visualization interface layer is used to use the image information of the wind turbine to establish a three-dimensional model of the wind turbine, call the device operation status data of the vibration monitoring server and the predicted operation status data of the wind power fault warning platform, generate a visualization interface based on the three-dimensional model, the device operation status data, and the predicted operation status data, and the maintenance personnel monitor the operation status of the wind turbine through the visualization interface. And when it is detected that the vibration warning value reaches the preset danger warning threshold, an alarm is issued through the visualization interface and assistance is provided to the maintenance personnel for maintaining the wind turbine.

[0135] Specifically, the interaction between the edge device and the vibration monitoring server is as follows Figure 3A As shown, the vibration sensor and the inclination sensor form a collection unit, and multiple collection units are combined to form a sensor cluster. In this way, the sensor cluster obtains real-time monitoring data, stores the real-time monitoring data as historical monitoring data in the memory, and transmits the real-time monitoring data to the vibration monitoring server. Subsequently, the memory transmits the historical monitoring data to the vibration monitoring server. In the actual application process, the ring network switch is set between the edge device and the vibration monitoring server, and the vibration monitoring server is docked with the wind power fault warning platform.

[0136] In addition, since the wind power fault warning platform includes multiple layers, the specific process of the wind turbine monitoring and warning method proposed in this application will be described below in combination with the functions of each layer:

[0137] As Figure 3B shown, the meteorological data API interface layer obtains meteorological data from the meteorological private network, converts the meteorological data into a JSON format message, and transmits the JSON format message to the database layer. Then, the database layer performs data analysis, screening, and processing on the JSON format message, and transmits the processed meteorological data to the service layer. Subsequently, the service layer obtains the real-time information mapping model of the vibration monitoring server, performs simulation analysis and risk level assessment on the real-time information mapping model according to the processed meteorological data, sends the simulation results and risk assessment levels to the visualization interface layer, and stores the meteorological data and simulation operation status data in the database of the database layer. Finally, the service layer processes the simulation results to obtain the vibration warning value, and monitors the vibration warning value in real time. When it is detected that the vibration warning value reaches the vibration warning threshold, the machine is stopped and an alarm is sent to the maintenance personnel; when it is detected that the vibration warning value does not reach the vibration warning threshold, the control right is transferred to the vibration monitoring server. In addition, the visualization interface layer visualizes the operation status and provides a visual interface for the maintenance personnel.

[0138] The monitoring and warning system provided by the embodiment of the present application can monitor the change process of the performance of the wind turbine in real time and issue a fault warning, reduce the fault occurrence rate, and while improving the working efficiency of the wind turbine, it can also improve the safety and stability of the wind turbine.

[0139] It should be noted that for other corresponding descriptions of each functional unit involved in the monitoring and warning system provided by the embodiment of the present application, reference can be made to the corresponding descriptions in Figure 1 and Figures 2A to 2C and will not be elaborated here.

[0140] In an exemplary embodiment, a computer-readable storage medium is further provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of the wind turbine monitoring and early warning method described above are implemented.

[0141] Through the description of the above embodiments, those skilled in the art can clearly understand that the present application can be implemented by hardware or by means of software plus a necessary general hardware platform. Based on such an understanding, the technical solution of the present application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.), including several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in various implementation scenarios of the present application.

[0142] Those skilled in the art can understand that the drawings are only schematic diagrams of a preferred implementation scenario, and the modules or processes in the drawings are not necessarily essential for implementing the present application. Those skilled in the art can understand that the modules in the device in the implementation scenario can be distributed in the device in the implementation scenario according to the description of the implementation scenario, or can be correspondingly changed and located in one or more devices different from the present implementation scenario. The modules in the above implementation scenario can be combined into one module, or further split into multiple sub-modules. The above serial numbers of the present application are only for description and do not represent the advantages or disadvantages of the implementation scenarios. The above-disclosed are only several specific implementation scenarios of the present application. However, the present application is not limited thereto, and any changes that can be thought of by those skilled in the art should fall within the protection scope of the present application.

Claims

1. A monitoring and early warning method for a wind turbine unit, characterized in that, Including: The edge device acquires the real-time monitoring data of the wind turbine, stores the real-time monitoring data of the wind turbine as historical monitoring data, and transmits the real-time monitoring data and the historical monitoring data to the ring network switch, and the ring network switch transmits the real-time monitoring data and the historical monitoring data to the vibration monitoring server; The vibration monitoring server receives the real-time monitoring data and the historical monitoring data transmitted by the ring network switch, acquires the preset structural parameters of the wind turbine, and based on the structural parameters, the real-time monitoring data and the historical monitoring data, establishes a real-time information mapping model, including: the vibration monitoring server acquires the preset geometric structure parameters and structural material parameters of the wind turbine, performs modeling processing on the preset geometric structure parameters and structural material parameters according to the Lagrangian dynamics equation, obtains the vibration mathematical model of the wind turbine vibration structure, and uses the vibration mathematical model as the initial mathematical model of the wind turbine. The vibration monitoring server receives the real-time monitoring data and the historical monitoring data transmitted by the ring network switch, screens the real-time monitoring data and the historical monitoring data, and extracts the effective operating state parameters. The value of the effective operating state parameters is non-empty and the numerical error is within the error range. The vibration monitoring server corrects the model parameters of the initial mathematical model by using the effective operating state parameters based on the initial mathematical model, and obtains the real-time information mapping model of the wind turbine. Among them, when constructing the vibration mathematical model, a simplified model of the wind turbine is constructed by using the geometric structure parameters and the structural material parameters, the displacement and velocity in the body coordinate system of the simplified model of the wind turbine are transformed to the inertial coordinate system according to the Euler angle rotation matrix, and the displacement and velocity in the inertial coordinate system are calculated according to the Lagrangian dynamics equation to obtain the vibration mathematical model. After establishing the real-time information mapping model, the real-time information mapping model is continuously updated in real time by using the collected real-time monitoring data and the stored historical monitoring data; The wind power fault warning platform acquires meteorological data, calls the real-time information mapping model established by the vibration monitoring server, calculates the vibration warning value based on the meteorological data and the real-time information mapping model, and when it detects that the vibration warning value reaches the preset danger warning threshold, shuts down the wind turbine and issues an alarm.

2. The method according to claim 1, characterized in that, The edge device acquires the real-time monitoring data of the wind turbine, stores the real-time monitoring data of the wind turbine as historical monitoring data, and transmits the real-time monitoring data and the historical monitoring data to the ring network switch, including: The sensor cluster of the edge device monitors the wind turbine in real time and collects data to obtain the real-time monitoring data; The edge device stores the real-time monitoring data in the storage unit of the edge device, and the storage unit is used to store the historical monitoring data and the real-time monitoring data of the wind turbine; The edge device obtains the historical monitoring data in the storage unit and transmits the collected real-time monitoring data and the historical monitoring data to the ring network switch.

3. The method according to claim 2, wherein After the edge device stores the real-time monitoring data in the storage unit of the edge device, the method further includes: The edge device identifies the real-time monitoring data, and when it is identified that a single data is missing in the real-time monitoring data, determines a missing array in which a single data is missing in the real-time monitoring data. Wherein, when identifying whether a single data is missing in the real-time monitoring data, the Mahalanobis distance metric is simultaneously used to identify whether there are abnormal variables in the real-time monitoring data, and when it is determined that there are abnormal variables, determines the array where the abnormal variable is located, removes the abnormal variable in the array, and uses the array after removal as the missing array; The edge device extracts a plurality of sample arrays from a plurality of historical monitoring data included in the storage unit, wherein the plurality of sample arrays are collected by the same sensor as the missing array; The edge device respectively reads the correlation between each sample array in the plurality of sample arrays and the missing array, and extracts a target sample array with the highest correlation with the missing array from the plurality of sample arrays; The edge device sorts the multiple sample values included in the target sample array in descending order to obtain a numerical sorting result, extracts the numerical numbers marked for each sample value in the multiple sample values, and sorts the multiple numerical numbers of the multiple sample values according to the order of the multiple sample values in the numerical sorting result to obtain a number sorting result; The edge device determines the missing numerical number of the single data that is missing in the missing array, determines the target numerical number that is the same as the missing numerical number in the number sorting result, and determines the first numerical number and the second numerical number adjacent to the target numerical number in the number sorting result; The edge device queries the first numerical value with the same numerical number as the first numerical number and the second numerical value with the same numerical number as the second numerical number in the missing array, and calculates the average value of the first numerical value and the second numerical value, and supplements the average value as the numerical value corresponding to the missing numerical number in the missing array to the missing array; Wherein, if the target numerical number is at the first or last position in the number sorting result, determines a target numerical number adjacent to the target numerical number in the number sorting result, and uses the numerical value with the same numerical number as the target numerical number in the missing array as the numerical value corresponding to the missing numerical number in the missing array, and supplements it to the missing array.

4. The method according to claim 1, wherein The wind power fault warning platform obtains meteorological data, invokes the real-time information mapping model established by the vibration monitoring server, and calculates the vibration warning value based on the meteorological data and the real-time information mapping model, and when it detects that the vibration warning value reaches a preset danger warning threshold, shuts down the wind turbine and issues an alarm, including: The wind power fault warning platform obtains weather numerical information within a preset future time range from the weather data stored in the database through a weather data interface. The weather numerical information includes air pressure, wind speed, and wind direction data in the area where the wind farm is located. The wind power fault warning platform analyzes and filters the weather numerical information using the historical weather data stored in the database to filter out invalid information in the weather numerical information. The wind power fault warning platform calls the real-time information mapping model established by the vibration monitoring server, and based on the filtered weather numerical information and the real-time information mapping model, performs simulation analysis on the wind turbine to obtain the predicted operation state data within the preset future time range and the fault risk level corresponding to the predicted operation state data. The wind power fault warning platform performs normalization calculations on the predicted operation state data and the fault risk level to obtain the vibration warning value of the wind turbine. When the vibration warning value reaches the preset dangerous warning threshold, the wind turbine is shut down and an alarm is issued. Among them, when it is detected that the vibration warning value reaches the preset risk threshold, the wind power fault warning platform stores the filtered weather numerical information and the predicted operation state data in the database for performing maintenance inspection operations using the filtered weather numerical information and the predicted operation state data and for updating the model of the real-time information mapping model.

5. The method according to claim 4, wherein The method further includes: When it is detected that the vibration warning value does not reach the preset dangerous warning threshold, the vibration monitoring server reads the real-time information mapping model and, based on the real-time information mapping model, performs simulation analysis on the wind turbine to obtain equipment operation state data. The vibration monitoring server performs adaptive working condition division based on the equipment operation state data to obtain a working condition division result, and controls the pitch system of the wind turbine in real time according to the working condition division result.

6. A monitoring and warning system, characterized in that, It includes end-side equipment, a ring network switch, a vibration monitoring server, and a wind power fault warning platform. The end-side equipment is used to obtain real-time monitoring data of the wind turbine, store the real-time monitoring data of the wind turbine as historical monitoring data, and transmit the real-time monitoring data and the historical monitoring data to the ring network switch. The ring network switch is used to transmit the real-time monitoring data and the historical monitoring data to the vibration monitoring server. The vibration monitoring server is used to receive the real-time monitoring data and the historical monitoring data transmitted by the ring network switch, obtain the preset structural parameters of the wind turbine, and establish a real-time information mapping model based on the structural parameters, the real-time monitoring data, and the historical monitoring data, including: the vibration monitoring server obtains the preset geometric structure parameters and structural material parameters of the wind turbine, performs modeling processing on the preset geometric structure parameters and structural material parameters according to the Lagrangian dynamics equation to obtain a vibration mathematical model of the wind turbine vibration structure, and uses the vibration mathematical model as the initial mathematical model of the wind turbine. The vibration monitoring server receives the real-time monitoring data and the historical monitoring data transmitted by the ring network switch, screens the real-time monitoring data and the historical monitoring data, and extracts effective operating state parameters, where the value of the effective operating state parameter is non-empty and the numerical error is within the error range. The vibration monitoring server corrects the model parameters of the initial mathematical model by using the effective operating state parameters based on the initial mathematical model to obtain the real-time information mapping model of the wind turbine. Among them, when constructing the vibration mathematical model, a simplified wind turbine model is constructed by using the geometric structure parameters and the structural material parameters, the displacements and velocities in the body coordinate system of the simplified wind turbine model are transformed to the inertial coordinate system according to the Euler angle rotation matrix, and the displacements and velocities in the inertial coordinate system are calculated according to the Lagrangian dynamics equation to obtain the vibration mathematical model. After establishing the real-time information mapping model, the real-time monitoring data collected and the historical monitoring data stored are continuously used to update the real-time information mapping model in real time; The wind power fault warning platform is used to obtain meteorological data, call the real-time information mapping model established by the vibration monitoring server, calculate a vibration warning value based on the meteorological data and the real-time information mapping model, and when it is detected that the vibration warning value reaches a preset danger warning threshold, shut down the wind turbine and issue an alarm.

7. The system according to claim 6, wherein The edge device includes a sensor cluster and a memory; The sensor cluster includes a plurality of data acquisition units installed on the main shaft of the wind turbine, the gearbox, the input shaft of the generator set, and the nacelle housing. Each data acquisition unit in the plurality of data acquisition units includes a vibration sensor and an inclination sensor. The sensor cluster is used to real-time monitor the vibration data and inclination data of the main shaft of the wind turbine and the input shaft of the generator set as the real-time monitoring data, and store the real-time monitoring data in the storage unit of the memory; The memory is used to store the historical monitoring data and the real-time monitoring data of the wind turbine, and the memory includes a storage unit.

8. The system according to claim 6, wherein The wind power fault warning platform includes a meteorological data interface layer, a database layer, a service layer, and a visualization interface layer; The meteorological data interface layer is used to connect to the meteorological private network through the meteorological data interface, obtain the meteorological data of the area where the wind farm is located, store the meteorological data into the database of the database layer, and convert the meteorological data into a text format message, and transmit the text format message to the wind power fault warning platform; The database layer is used to read the text format message of the meteorological data interface layer, extract the weather numerical information within a preset future time range from the text format message, analyze and screen the weather numerical information by using the historical meteorological data stored in the database, filter out the invalid information in the weather numerical information, and store the filtered weather numerical information and predicted operation status data transmitted by the service layer into the database. Among them, the database layer includes data files, data information transmission programs, data information analysis programs, data information screening programs, data information processing programs, data storage, and databases; The service layer is used to obtain the filtered weather numerical information of the database layer and the real-time information mapping model established by the vibration monitoring server. Based on the filtered weather numerical information and the real-time information mapping model, perform simulation analysis on the wind turbine to obtain the predicted operation status data within the preset future time range, the fault risk level corresponding to the predicted operation status data, and the data fusion control unit of the service layer performs normalization calculation on the predicted operation status data and the fault risk level to obtain the vibration warning value of the wind turbine, and monitors the vibration warning value in real time. When the vibration warning value reaches the preset dangerous warning threshold, the wind turbine is shut down and an alarm is issued. The vibration warning value includes a risk-free level, vibration risk level I, vibration risk level II, vibration risk level III, and vibration danger level. The preset dangerous warning threshold is the upper limit value when the vibration warning value reaches vibration risk level III; Among them, when the service layer detects that the vibration warning value reaches the preset risk threshold, the service layer stores the filtered weather numerical information and the predicted operation status data into the database, so as to perform maintenance inspection operations by using the filtered weather numerical information and the predicted operation status data and update the model of the real-time information mapping model. The preset risk threshold is the lower limit value when the vibration warning value reaches vibration risk level I; The visualization interface layer is used to establish a three-dimensional model of the wind turbine by using the image information of the wind turbine, call the device operation status data of the vibration monitoring server and the predicted operation status data of the wind power fault warning platform, generate a visualization interface based on the three-dimensional model, the device operation status data, and the predicted operation status data, and maintenance personnel monitor the operation status of the wind turbine through the visualization interface. And when it is detected that the vibration warning value reaches the preset dangerous warning threshold, an alarm is issued through the visualization interface and assistance is provided to the maintenance personnel for the maintenance of the wind turbine.

9. A readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 5.

Images