Wind turbine monitoring method, device, system and computer readable storage medium

Abstract

The application discloses a kind of wind power fan monitoring method, device, system and computer readable storage medium, the method includes: classification step: when detecting starting instruction, target data is collected by pre-installed sensor and data collector, and the target data is classified, and classification result is obtained;Monitoring step: according to the classification result, the target data is input into corresponding monitoring model, and the monitoring result is obtained by the monitoring model in combination with pre-created wind power fan digital twin;Display step: based on the wind power fan digital twin, the monitoring result is displayed.The application collects target data by pre-installed sensor and data collector, classifies target data and inputs corresponding monitoring model, obtains monitoring result by corresponding monitoring model and wind power fan digital twin, and improves the accuracy of abnormal monitoring of wind power fan.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of wind power technology, and particularly relates to a wind turbine monitoring method, device, system and computer readable storage medium. BACKGROUND

[0002] With the increasing awareness of environmental protection, the use of clean energy is also becoming more and more popular, and wind power is the most representative of clean energy; wind power is through wind turbine to convert wind energy into blade kinetic energy, and convert the kinetic energy of the blade into electrical energy, since the wind turbine works outdoors, it is easy to be affected by the environment and appear abnormal, at present, most of the monitoring of wind turbines is carried out by manual or through sensors to obtain some data characteristics of the wind turbine during operation, and then determine whether the wind turbine is abnormal, the wind turbine is a complex structure, and manual or single data cannot accurately monitor whether the wind turbine is abnormal.

[0003] Therefore, how to improve the accuracy of abnormal monitoring of the wind turbine is a problem to be solved. SUMMARY

[0004] The main purpose of the present application is to provide a wind turbine monitoring method, device, system and computer readable storage medium, which aims to solve the problem of how to improve the accuracy of wind turbine monitoring.

[0005] To achieve the above purpose, the present application provides a wind turbine monitoring method, which comprises the following steps:

[0006] The classification step: when the start instruction is detected, the target data is collected by the pre-installed sensor and data collector, and the target data is classified to obtain a classification result;

[0007] The monitoring step: according to the classification result, the target data is input into the corresponding monitoring model, and the monitoring model is combined with the pre-created wind turbine digital twin to obtain a monitoring result;

[0008] The display step: based on the wind turbine digital twin, the monitoring result is displayed.

[0009] Optionally, the sensor comprises: a sound, vibration and temperature integrated sensor installed in the transmission chain of the wind turbine, a sound, vibration and temperature integrated sensor installed at the root of the blade of the wind turbine, a sway sensor and an inclination sensor installed at the top end of the tower of the wind turbine, and an inclination sensor installed at the bottom end of the tower, and the classification step comprises:

[0010] The transmission chain and the blade data acquisition sub-step: collecting the sound data, the vibration data and the temperature data of the transmission chain and the blade through the sound, vibration and temperature integrated sensor;

[0011] The tower cylinder data acquisition sub-step: collecting the swing data of the tower cylinder through the swing sensor and collecting the inclination data of the tower cylinder through the inclination sensor;

[0012] The target data determination sub-step: collecting the working condition data set and the environmental data of the wind turbine through the data collector, and combining the sound data, the vibration data, the temperature data, the swing data and the inclination data to obtain the target data.

[0013] Optionally, the monitoring model includes: a transmission chain monitoring model, a blade monitoring model and a tower cylinder monitoring model, and the monitoring step includes:

[0014] The transmission chain monitoring sub-step: according to the classification result, inputting the environmental data, the transmission chain working condition data in the working condition data set and the sound data, the vibration data and the temperature data of the transmission chain into the transmission chain monitoring model, and combining the pre-created wind turbine digital twin through the transmission chain monitoring model to obtain the transmission chain monitoring result;

[0015] The blade monitoring sub-step: inputting the environmental data, the blade working condition data in the working condition data set and the sound data, the vibration data and the temperature data of the blade into the blade monitoring model, and combining the pre-created wind turbine digital twin through the blade monitoring model to obtain the blade monitoring result;

[0016] The tower cylinder monitoring sub-step: inputting the environmental data, the working condition data set and the swing data and the inclination data of the tower cylinder into the tower cylinder monitoring model, and combining the pre-created wind turbine digital twin through the tower cylinder monitoring model to obtain the tower cylinder monitoring result.

[0017] Optionally, the transmission chain monitoring sub-step includes:

[0018] The transmission chain monitoring model input grandchild step: inputting the environmental data, the transmission chain working condition data and the sound data, the vibration data and the temperature data of the transmission chain into the transmission chain monitoring model;

[0019] The time domain characteristic value calculation grandchild step: generating the corresponding time domain waveform diagram according to the vibration data of the transmission chain through the transmission chain monitoring model, and calculating the time domain characteristic value corresponding to the vibration data of the transmission chain according to the time domain waveform diagram;

[0020] Characteristic value comparison son step: generate simulation time domain characteristic values of the transmission chain according to the environment data and the transmission chain working condition data through the transmission chain monitoring model combined with the wind turbine digital twin, and compare the time domain characteristic values with the simulation time domain characteristic values;

[0021] Transmission chain monitoring result obtaining son step: if the time domain characteristic values are greater than the simulation time domain characteristic values, obtain transmission chain monitoring results through the transmission chain monitoring model combined with the wind turbine digital twin according to the transmission chain sound data and temperature data, and the time domain waveform diagram.

[0022] Optionally, the transmission chain monitoring result obtaining son step comprises:

[0023] Frequency band signal set producing great-grandson step: generate a frequency band signal set through wavelet multi-resolution analysis of the time domain waveform diagram by the transmission chain monitoring model;

[0024] Fault characteristic frequency value obtaining great-grandson step: obtain fault characteristic frequency values through refined spectrum analysis and envelope spectrum analysis of each frequency band signal in the frequency band signal set;

[0025] Transmission chain monitoring result determining great-grandson step: input the fault characteristic frequency values into the wind turbine digital twin for simulation to determine the fault position of the transmission chain, and determine the fault severity according to the fault characteristic frequency values, the transmission chain sound data and temperature data, to obtain the transmission chain monitoring results.

[0026] Optionally, the blade monitoring sub-step comprises:

[0027] Blade monitoring model input son step: align the environment data, the blade working condition data in the working condition data set, and the sound data, vibration data and temperature data of the blade according to time series, and input them into the blade monitoring model;

[0028] Simulation vibration data generation son step: generate simulation vibration data through the blade monitoring model according to the aligned environment data and blade working condition data combined with the wind turbine digital twin;

[0029] Deviation comparison son step: calculate the deviation coefficient of the vibration data of the blade and the simulation vibration data through the blade monitoring model, and compare the deviation coefficient with a preset deviation interval;

[0030] Time domain characteristic set extraction son step: if the deviation coefficient is not within the preset deviation interval, extract a time domain characteristic set of the vibration data of the blade;

[0031] The blade monitoring sub-step includes: inputting the time domain feature set into the wind turbine digital twin, determining the fault position of the blade, and determining the fault severity according to the time domain feature set, sound data and temperature data of the blade to obtain a blade monitoring result.

[0032] Optionally, the tower monitoring sub-step includes:

[0033] The deformation data calculation sub-step includes: inputting the environment data, the working condition data set, the swing data and the inclination data of the tower into the tower monitoring model, and calculating the deformation data of the tower through the tower monitoring model.

[0034] The tower monitoring sub-step includes: obtaining a time domain feature set of the deformation data through the tower monitoring model, inputting the time domain feature set of the deformation data into the wind turbine digital twin, determining the fault position and the fault severity of the tower to obtain a tower monitoring result.

[0035] In addition, to achieve the above-mentioned purpose, the present application also provides a wind turbine monitoring device, which includes:

[0036] The classification module is configured to, when the start instruction is detected, collect target data through the pre-installed sensor and data collector, and classify the target data to obtain a classification result.

[0037] The input module is configured to, according to the classification result, input the target data into a corresponding monitoring model, and obtain a monitoring result through the monitoring model in combination with a pre-created wind turbine digital twin.

[0038] The display module is configured to display the monitoring result based on the wind turbine digital twin.

[0039] Further, the classification module is further configured to:

[0040] The sound, vibration and temperature integrated sensor is configured to collect sound data, vibration data and temperature data of the transmission chain and the blade.

[0041] The swing sensor is configured to collect swing data of the tower, and the inclination sensor is configured to collect inclination data of the tower.

[0042] The data collector is configured to collect a working condition data set and environment data of the wind turbine, and obtain target data in combination with the sound data, the vibration data, the temperature data, the swing data and the inclination data.

[0043] Further, the input module is further configured to:

[0044] According to the classification result, the environmental data, the transmission chain operating condition data in the operating condition data set, and the sound data, vibration data, and temperature data of the transmission chain are input into the transmission chain monitoring model, and the transmission chain monitoring result is obtained by combining the transmission chain monitoring model with a pre-created wind turbine digital twin;

[0045] Inputting the environmental data, the blade operating condition data in the operating condition data set, and the sound data, vibration data, and temperature data of the blade into the blade monitoring model, and combining the blade monitoring model with a pre-created wind turbine digital twin to obtain a blade monitoring result;

[0046] The environmental data, the operating condition data set, and the tower sway data and the tilt data are input into the tower monitoring model, and the tower monitoring model is combined with a pre-created wind turbine digital twin to obtain a tower monitoring result.

[0047] Furthermore, the input module further includes a transmission chain monitoring module, and the transmission chain monitoring module is used to:

[0048] inputting the environmental data, the transmission chain operating condition data, and the sound data, vibration data, and temperature data of the transmission chain into the transmission chain monitoring model;

[0049] Generate a corresponding time domain waveform diagram based on the vibration data of the transmission chain using the transmission chain monitoring model, and calculate a time domain eigenvalue corresponding to the vibration data of the transmission chain based on the time domain waveform diagram;

[0050] By combining the transmission chain monitoring model with the wind turbine digital twin, a simulated time domain characteristic value of the transmission chain is generated according to the environmental data and the transmission chain operating condition data, and the time domain characteristic value is compared with the simulated time domain characteristic value;

[0051] If the time domain eigenvalue is greater than the simulated time domain eigenvalue, the transmission chain monitoring result is obtained by combining the transmission chain monitoring model with the wind turbine digital twin based on the sound data and temperature data of the transmission chain and the time domain waveform diagram.

[0052] Furthermore, the transmission chain monitoring module is also used to:

[0053] Performing wavelet multi-resolution analysis on the time domain waveform using the transmission chain monitoring model to generate a frequency band signal set;

[0054] Performing a refined spectrum analysis and an envelope spectrum analysis on each frequency band signal in the frequency band signal set to obtain a fault characteristic frequency value;

[0055] The fault feature frequency value is input into the digital twin of the wind turbine for simulation to determine the fault position of the transmission chain, and the fault severity is determined according to the fault feature frequency value, the sound data and the temperature data of the transmission chain, so as to obtain the transmission chain monitoring result.

[0056] Further, the input module further comprises a blade monitoring module, and the blade monitoring module is configured to:

[0057] According to the time sequence, the environment data, the blade working condition data in the working condition data set, and the sound data, vibration data and temperature data of the blade are aligned respectively and input into the blade monitoring model;

[0058] The blade monitoring model generates simulated vibration data according to the aligned environment data and blade working condition data in combination with the digital twin of the wind turbine;

[0059] The blade monitoring model calculates the deviation coefficient of the vibration data of the blade and the simulated vibration data, and compares the deviation coefficient with a preset deviation interval;

[0060] If the deviation coefficient is not within the preset deviation interval, a time domain feature set of the vibration data of the blade is extracted;

[0061] The time domain feature set is input into the digital twin of the wind turbine to determine the fault position of the blade, and the fault severity is determined according to the time domain feature set, the sound data and the temperature data of the blade, so as to obtain the blade monitoring result.

[0062] Further, the input module further comprises a tower monitoring module, and the tower monitoring module is configured to:

[0063] The environment data, the working condition data set, the sway data and the inclination data of the tower are input into the tower monitoring model, and the deformation data of the tower is calculated by the tower monitoring model;

[0064] The time domain feature set of the deformation data is obtained by the tower monitoring model, the time domain feature set of the deformation data is input into the digital twin of the wind turbine to determine the fault position and the fault severity of the tower, so as to obtain the tower monitoring result.

[0065] In addition, to achieve the above-mentioned purpose, the present application further provides a wind turbine monitoring system, which comprises a memory, a processor and a wind turbine monitoring program stored in the memory and executable on the processor, and the wind turbine monitoring program realizes the steps of the wind turbine monitoring method when executed by the processor.

[0066] In addition, to achieve the above-mentioned purpose, the present invention also provides a computer-readable storage medium, which is a computer-readable storage medium. A wind turbine monitoring program is stored on the computer-readable storage medium. When the wind turbine monitoring program is executed by the processor, the steps of the wind turbine monitoring method as described above are implemented.

[0067] The wind turbine monitoring method proposed in the present invention includes: a classification step: when a startup command is detected, target data is collected through pre-installed sensors and data collectors, and the target data is classified to obtain a classification result; a monitoring step: based on the classification result, the target data is input into a corresponding monitoring model, and the monitoring result is obtained by combining the monitoring model with a pre-created wind turbine digital twin; and a display step: based on the wind turbine digital twin, the monitoring result is displayed. The present invention collects target data through pre-installed sensors and data collectors, classifies the target data and inputs it into a corresponding monitoring model, obtains monitoring results through the corresponding monitoring model and the wind turbine digital twin, and improves the accuracy of abnormal monitoring of wind turbines. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] Figure 1 Schematic diagram of the device structure of the hardware operating environment involved in the embodiment of the present invention;

[0069] Figure 2 This is a flow chart of a first embodiment of a wind turbine monitoring method according to the present invention;

[0070] Figure 3 This is a flow chart of a second embodiment of a wind turbine monitoring method according to the present invention;

[0071] Figure 4 This is a flow chart of a third embodiment of a wind turbine monitoring method according to the present invention;

[0072] Figure 5 2 is a flow chart of a fourth embodiment of a wind turbine monitoring method according to the present invention.

[0073] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings and in conjunction with the embodiments. DETAILED DESCRIPTION

[0074] It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0075] like Figure 1 As shown, Figure 1 It is a schematic diagram of the device structure of the hardware operating environment involved in the embodiment of the present invention.

[0076] The device of the embodiment of the application can be a PC or a server device.

[0077] As shown in Figure 1 , the device can include a processor 1001, such as a CPU, a network interface 1004, a user interface 1003, a memory 1005, and a communication bus 1002. The communication bus 1002 is used to realize the connection and communication between the components. The user interface 1003 can include a display screen (Display), an input unit such as a keyboard (Keyboard), and the optional user interface 1003 can further include a standard wired interface, a wireless interface. The network interface 1004 can optionally include a standard wired interface, a wireless interface (such as a WI-FI interface). The memory 1005 can be a high-speed RAM memory, or a stable memory (non-volatile memory), such as a magnetic disk memory. The memory 1005 can also be an independent storage device from the aforementioned processor 1001.

[0078] Those skilled in the art can understand that Figure 1 the device structure shown in the above description does not constitute a limitation on the device, and can include more or fewer components than those shown, or combine certain components, or different component arrangements.

[0079] As shown in Figure 1 , the memory 1005 as a computer storage medium can include an operating system, a network communication module, a user interface module, and a wind turbine monitoring program.

[0080] The operating system is a program that manages and controls the portable storage device and software resources, supports the operation of the network communication module, the user interface module, the wind turbine monitoring program, and other programs or software; the network communication module is used to manage and control the network interface 1002; and the user interface module is used to manage and control the user interface 1003.

[0081] In the storage device shown in Figure 1 , the storage device calls the wind turbine monitoring program stored in the memory 1005 through the processor 1001, and performs the operations in each embodiment of the wind turbine monitoring method described below.

[0082] Based on the above hardware structure, embodiments of the wind turbine monitoring method of the application are proposed.

[0083] Referring to Figure 2 , Figure 2 , the flowchart of the first embodiment of the wind turbine monitoring method of the application is shown. The method includes:

[0084] The classification step: when the start instruction is detected, the target data is collected by the pre-installed sensor and data collector, and the target data is classified to obtain a classification result;

[0085] The monitoring step: according to the classification result, the target data is input into a corresponding monitoring model, and a monitoring result is obtained by combining the monitoring model with a pre-created digital twin of a wind turbine;

[0086] The display step: based on the digital twin of the wind turbine, the monitoring result is displayed.

[0087] The wind turbine monitoring method of the embodiment is applied to a monitoring system of a wind power generation mechanism, and is used to monitor whether the running state of the wind turbine is abnormal. For the convenience of description, the monitoring system is taken as an example for illustration. When the monitoring system detects a start instruction, the sound data, vibration data and temperature data of the transmission chain and the blades are collected by a sound, vibration and temperature integrated sensor. The shaking data of the tower drum is collected by a shaking sensor, and the inclination data of the tower drum is collected by an inclination sensor. The working condition data set and the environmental data of the wind turbine are collected by a data collector, and the target data is obtained by combining the sound data, vibration data, temperature data, shaking data and inclination data. The target data is classified to obtain a classification result. According to the classification result, the environmental data, the working condition data set of the transmission chain, and the sound data, vibration data and temperature data of the transmission chain are input into a transmission chain monitoring model. The transmission chain monitoring result is obtained by combining the transmission chain monitoring model with a pre-created digital twin of the wind turbine. The environmental data, the working condition data set of the blades, and the sound data, vibration data and temperature data of the blades are input into a blade monitoring model. The blade monitoring result is obtained by combining the blade monitoring model with the pre-created digital twin of the wind turbine. The environmental data, the working condition data set, and the shaking data and inclination data of the tower drum are input into a tower drum monitoring model. The tower drum monitoring result is obtained by combining the tower drum monitoring model with the pre-created digital twin of the wind turbine.

[0088] When the monitoring system detects a start instruction, the target data is collected by the pre-installed sensor and data collector, and the target data is classified to obtain a classification result. According to the classification result, the target data is input into a corresponding monitoring model, and a monitoring result is obtained by combining the monitoring model with a pre-created digital twin of a wind turbine. The monitoring result is displayed based on the digital twin of the wind turbine. The target data is collected by the pre-installed sensor and data collector, the target data is classified and input into the corresponding monitoring model, the monitoring result is obtained by the corresponding monitoring model and the digital twin of the wind turbine, and the accuracy of the abnormal monitoring of the wind turbine is improved.

[0089] The following describes each step in detail:

[0090] The classification step: when the start instruction is detected, the target data is collected by the pre-installed sensor and data collector, and the target data is classified to obtain a classification result.

[0091] In this embodiment, the monitoring system collects target data through pre-installed sensors and data collectors when a start instruction is detected. Optionally, the monitoring system can collect target data through sensors and data collectors in real time or according to a certain period according to the instructions of relevant operating personnel. After collecting the target data, the monitoring system classifies the target data to obtain a classification result. Optionally, since the target data collected by the monitoring system includes target data corresponding to the transmission chain, the blade, and the tower, it is necessary to classify the target data according to the transmission chain, the blade, and the tower, so as to subsequently input the target data corresponding to different components into the corresponding monitoring model.

[0092] Specifically, the classification step includes:

[0093] The transmission chain and blade data collection sub-step: collecting sound data, vibration data, and temperature data of the transmission chain and the blade through the sound, vibration, and temperature integrated sensor;

[0094] In this step, the monitoring system collects sound data, vibration data, and temperature data of the transmission chain through the sound, vibration, and temperature integrated sensor installed in the transmission chain of the wind turbine, and collects sound data, vibration data, and temperature data of the blade through the sound, vibration, and temperature integrated sensor installed at the root of the blade of the wind turbine. It should be noted that the transmission chain includes the main shaft, the gearbox, and the generator, so one sound, vibration, and temperature integrated sensor is installed at the front and rear ends of the main shaft, one sound, vibration, and temperature integrated sensor is installed in the gearbox, and one sound, vibration, and temperature integrated sensor is installed in the generator. The blade of the wind turbine generally has three, so one sound, vibration, and temperature integrated sensor is installed at the root of each blade. By installing the sound, vibration, and temperature integrated sensor, the number of sensors is reduced, thereby reducing the influence of the sensors on the transmission chain and the blade. Moreover, the sound, vibration, and temperature integrated sensor can collect multiple types of data, making the subsequent analysis of the operating state of the wind turbine more accurate.

[0095] The tower data collection sub-step: collecting the tower sway data through the sway sensor and collecting the tower inclination data through the inclination sensor;

[0096] In this step, the monitoring system collects the tower sway data of the tower through the sway sensor installed at the top end of the tower, and collects the tower inclination data through the inclination sensor installed at the top end of the tower and the inclination sensor installed at the bottom end of the tower.

[0097] The target data determination sub-step: collecting the working condition data set and the environmental data of the wind turbine through the data collector, and combining the sound data, the vibration data, the temperature data, the sway data and the inclination data to obtain the target data.

[0098] In this step, the monitoring system collects the working condition data set and the environmental data of the wind turbine through the data collector, and combines the sound data, the vibration data, the temperature data, the sway data and the inclination data of the tower to obtain the target data.

[0099] The monitoring step: according to the classification result, inputting the target data into the corresponding monitoring model, and combining the pre-created wind turbine digital twin through the monitoring model to obtain the monitoring result;

[0100] Specifically, the monitoring step includes:

[0101] The transmission chain monitoring sub-step: according to the classification result, inputting the environmental data, the transmission chain working condition data in the working condition data set, and the sound data, vibration data and temperature data of the transmission chain into the transmission chain monitoring model, and combining the pre-created wind turbine digital twin through the transmission chain monitoring model to obtain the transmission chain monitoring result;

[0102] The blade monitoring sub-step: inputting the environmental data, the blade working condition data in the working condition data set, and the sound data, vibration data and temperature data of the blade into the blade monitoring model, and combining the pre-created wind turbine digital twin through the blade monitoring model to obtain the blade monitoring result;

[0103] The tower monitoring sub-step: inputting the environmental data, the working condition data set, and the sway data and the inclination data of the tower into the tower monitoring model, and combining the pre-created wind turbine digital twin through the tower monitoring model to obtain the tower monitoring result.

[0104] In the embodiment, the monitoring system classifies the target data, inputs the target data and the environmental data corresponding to the transmission chain into the transmission chain monitoring model, inputs the target data and the environmental data corresponding to the blade into the blade monitoring model, and inputs the target data and the environmental data corresponding to the tower drum into the tower drum monitoring model. The transmission chain monitoring model, the blade monitoring model, and the tower drum monitoring model are combined with the wind turbine digital twin created in advance to obtain the transmission chain detection result, the blade monitoring result, and the tower drum monitoring result. It should be noted that the wind turbine digital twin is a digital model of the wind turbine created by the digital twin technology. The wind turbine digital twin contains all parameters and mechanical structures of the wind turbine and can be used to simulate the working state of the wind turbine. The monitoring system processes the corresponding target data by combining the corresponding monitoring model with the wind turbine digital twin. Since the wind turbine digital twin can simulate according to the corresponding data, it can completely replace manual analysis of the data and has higher accuracy than manual processing of the data. Therefore, the cost of manual work can be reduced, and the accuracy of abnormal monitoring of the transmission chain, the blade, and the tower drum of the wind turbine can be improved.

[0105] Displaying: based on the wind turbine digital twin, the monitoring result is displayed.

[0106] In the embodiment, after obtaining the monitoring result, the monitoring system displays the monitoring result to the relevant personnel based on the wind turbine digital twin. For example, the monitoring system determines the positions of the transmission chain, the blade, and the tower drum based on the wind turbine digital twin created in advance, and displays the transmission chain monitoring result in the transmission chain position, the blade monitoring result in the blade position, and the tower drum monitoring result in the tower drum position. Further, the monitoring system can also determine the abnormal condition of the wind turbine according to the monitoring result, and display the abnormal part and the fault severity of the abnormal condition through the wind turbine digital twin. This enables the relevant personnel to intuitively and clearly determine the current condition of the wind turbine, which is beneficial to improving the efficiency of determining whether the wind turbine is abnormal and effectively avoiding further deterioration of the abnormality.

[0107] When the monitoring system of this embodiment detects the start-up instruction, it collects the sound data, vibration data and temperature data of the transmission chain and blades through the integrated sound, vibration and temperature sensor; collects the shaking data of the tower through the shaking sensor, and collects the tilt data of the tower through the tilt sensor; collects the working condition data set and environmental data of the wind turbine through the data collector, and obtains the target data by combining the sound data, vibration data, temperature data, shaking data and tilt data, and classifies the target data to obtain the classification result; according to the classification result, the monitoring system classifies the environmental data, the transmission chain working condition data in the working condition data set and the sound data of the transmission chain The system inputs environmental data, operating condition data, and tower sway and tilt data into the tower monitoring model, which is then combined with the tower monitoring model to obtain tower monitoring results. Target data is collected through pre-installed sensors and data collectors, classified, and input into the corresponding monitoring model. Monitoring results are obtained through the corresponding monitoring model and the wind turbine digital twin, improving the accuracy of wind turbine abnormality monitoring.

[0108] Reference Figure 3 , Figure 3 This is a flow chart of a second embodiment of a wind turbine monitoring method according to the present invention. The transmission chain monitoring sub-step includes:

[0109] The transmission chain monitoring model input step includes: inputting the environmental data, the transmission chain operating condition data, and the sound data, vibration data, and temperature data of the transmission chain into the transmission chain monitoring model;

[0110] The time domain eigenvalue calculation step includes: generating a corresponding time domain waveform diagram according to the vibration data of the transmission chain through the transmission chain monitoring model, and calculating the time domain eigenvalue corresponding to the vibration data of the transmission chain according to the time domain waveform diagram;

[0111] In the transmission chain monitoring model input step and the time domain eigenvalue calculation step, the monitoring system inputs the environmental data, the transmission chain working condition data, and the sound data, vibration data, and temperature data of the transmission chain into the transmission chain monitoring model, generates a corresponding time domain waveform diagram according to the vibration data of the transmission chain through the transmission chain monitoring model, and calculates the time domain eigenvalues ​​corresponding to the vibration data of the transmission chain according to the time domain waveform diagram; it can be understood that when the transmission chain is working, the rotation of each mechanical structure will generate corresponding vibration data, the collision of each mechanical structure will generate sound data, and the friction between each mechanical structure will generate temperature data, and the vibration data has the greatest correlation with whether the working state of the transmission chain is abnormal, so the time domain eigenvalues ​​of the vibration data are calculated first;

[0112] Optionally, the time domain characteristic values ​​include dimensional indicators and dimensionless indicators. The dimensional indicators include: peak-to-peak value, mean value, root mean square value, etc., and the dimensionless indicators include: kurtosis, waveform index, peak index, pulse index and margin index, etc. The monitoring system can calculate one or more of the dimensional indicators and dimensionless indicators as the time domain characteristic values ​​of the vibration data of the transmission chain through the transmission chain monitoring model.

[0113] Preferably, the monitoring system calculates the root mean square value in the dimensional index and the kurtosis and peak index in the dimensionless index as the time domain characteristic value of the vibration data of the transmission chain through the transmission chain monitoring model, wherein the formula for calculating the root mean square value is:

[0114]

[0115] The formula for calculating kurtosis is:

[0116]

[0117] in,

[0118]

[0119] The formula for calculating peak index is:

[0120]

[0121] In the above formula, N is the number of sample points in the time domain waveform, x i The value corresponding to each sample point.

[0122] The characteristic value comparison step includes: generating a simulated time-domain characteristic value of the transmission chain according to the environmental data and the transmission chain operating condition data by combining the transmission chain monitoring model with the wind turbine digital twin, and comparing the time-domain characteristic value with the simulated time-domain characteristic value;

[0123] In this step, the monitoring system inputs the environmental data and the transmission chain working condition data into the wind turbine digital twin through the transmission chain monitoring model, simulates the transmission chain of the wind turbine through the wind turbine digital twin according to the environmental data and the transmission chain working condition data, obtains the simulation vibration data of the transmission chain, and calculates the root mean square value, kurtosis and peak value corresponding to the simulation vibration data as the simulation time domain characteristic value through the transmission chain monitoring model, and compares the time domain characteristic value with the simulation time domain characteristic value. It should be noted that the process of calculating the root mean square value, kurtosis and peak value corresponding to the simulation vibration data is the same as the above process, and will not be repeated here. The environmental data includes wind speed, lightning strike, air temperature, rain and snow, etc. The simulation through the digital twin of the wind turbine can determine the simulation time domain characteristic value corresponding to the normal vibration data generated when working according to the transmission chain working condition data under the current environmental data, which is beneficial to improve the accuracy of comparing the time domain characteristic value with the simulation time domain characteristic value.

[0124] The transmission chain monitoring result is obtained in the following step: if the time domain characteristic value is greater than the simulation time domain characteristic value, the transmission chain monitoring result is obtained through the transmission chain monitoring model in combination with the wind turbine digital twin according to the sound data and temperature data of the transmission chain and the time domain waveform.

[0125] In this step, after the monitoring system compares the time domain characteristic value with the simulation time domain characteristic value, if it is determined that the time domain characteristic value is not greater than the simulation time domain characteristic value, it is determined that the transmission chain of the current wind turbine does not exist. Abnormal; if it is determined that the time domain characteristic value is greater than the simulation time domain characteristic value, it is determined that the transmission chain of the current wind turbine exists. Abnormal, and the transmission chain monitoring model is used to perform wavelet multi-resolution analysis, detailed spectrum analysis and envelope spectrum analysis on the time domain waveform, and then the fault position and fault severity of the transmission chain are determined in combination with the sound data and temperature data of the transmission chain, to obtain the transmission chain monitoring result.

[0126] Specifically, the transmission chain monitoring result obtaining step includes:

[0127] The frequency band signal set production step is: performing wavelet multi-resolution analysis on the time domain waveform through the transmission chain monitoring model to generate a frequency band signal set;

[0128] In this step, the monitoring system performs wavelet multi-resolution analysis on the vibration data of the transmission chain in the time domain waveform through the transmission chain monitoring model to generate a frequency band signal set. It can be understood that the frequency band signal set includes low frequency signals and medium frequency signals. The calculation process of wavelet multi-resolution analysis is as follows:

[0129]

[0130] Wherein, the wavelet base wave set in advance in the monitoring system is Z(t), the wavelet base wave is an energy-limited signal space, a is a scaling variable, and b is a translation variable. The monitoring system adjusts a and b according to different frequency bands, and calculates the wavelet sequence Z corresponding to the wavelet base wave for different frequency bands through the above formula a,b .

[0131] Y(a, b) = ∫f(t)Z a，b (t)dt

[0132] Wherein, f(t) is vibration data of the transmission chain, and the monitoring system performs memory integral operation on the vibration data and the wavelet sequence corresponding to different frequency bands according to the above formula to generate a frequency band signal set, so that the vibration data of the transmission chain can be converted into a frequency band signal set containing multiple frequency band signals, which is beneficial to record more detailed data in the vibration data in each frequency band signal and helps to improve the accuracy of analysis.

[0133] The fault characteristic frequency value is obtained through the following steps: performing signal reconstruction on each frequency band signal in the frequency band signal set, and performing detailed spectrum analysis and envelope spectrum analysis on each frequency band signal after the signal reconstruction to obtain the fault characteristic frequency value.

[0134] In this step, after obtaining the frequency band signal set corresponding to the vibration data, the monitoring system performs signal reconstruction on each frequency band signal in the frequency band signal set, and performs detailed spectrum analysis and envelope spectrum analysis on each frequency band signal after the signal reconstruction to obtain the fault characteristic frequency value. For example, the formula for signal reconstruction is:

[0135]

[0136] Wherein, f 1 (t) is the frequency band signal after signal reconstruction, C is a reconstruction constant, which is set in advance in the transmission chain monitoring model of the monitoring system, a is a scaling variable, and b is a translation variable. The monitoring system adjusts a and b according to different frequency bands, that is, different frequency band signals can be reconstructed, so that the reconstructed signal meets the requirements of subsequent detailed spectrum analysis and envelope spectrum analysis, and the processing efficiency is improved.

[0137] The monitoring system performs refined spectrum analysis and envelope spectrum analysis on each frequency band signal after obtaining the frequency band signal after reconstruction. When performing refined spectrum analysis, the monitoring system first determines the analysis frequency interval of each frequency band signal, and analyzes the frequency contained in each frequency band signal according to the corresponding analysis frequency interval. When performing envelope spectrum analysis, the monitoring system first extracts the envelope spectrum of each frequency band signal, and analyzes the frequency contained in each frequency band signal according to the envelope spectrum. Through refined spectrum analysis and envelope spectrum analysis, the fault characteristic frequency value of the vibration data is obtained, which includes the frequency multiplication value and the high-order harmonic frequency value of the vibration data.

[0138] The transmission chain monitoring result determination step: inputting the fault characteristic frequency value into the digital twin of the wind turbine to perform simulation, determining the fault position of the transmission chain, and determining the fault severity according to the fault characteristic frequency value, the sound data and the temperature data of the transmission chain to obtain the transmission chain monitoring result.

[0139] In this step, after obtaining the fault characteristic frequency value, the monitoring system inputs it into the digital twin of the wind turbine. The digital twin of the wind turbine performs simulation according to the frequency multiplication value and the high-order harmonic frequency value in the fault characteristic frequency value, determines the fault position of the transmission chain, such as the loosening of the support component of the transmission chain, which will cause the superposition of frequency multiplication and high-order harmonics. After simulation according to the input fault characteristic frequency value, the digital twin of the wind turbine can determine that the fault position of the transmission chain is the support component.

[0140] After determining the fault position, the monitoring system determines the fault severity according to the fault characteristic frequency value, the sound data and the temperature data of the transmission chain. For example, the monitoring system performs dimensionless processing on the fault characteristic frequency value, the sound data and the temperature data of the transmission chain, calculates the fault severity coefficient according to the dimensionless processed fault characteristic frequency value, the sound data and the temperature data of the transmission chain, and the formula for calculating the fault severity coefficient is as follows:

[0141]

[0142] Where k is the fault severity coefficient, S i is the dimensionless processed sound data at the i th time point, W i is the dimensionless processed temperature data at the i th time point, f is the dimensionless processed fault characteristic frequency value, a, b and c are the corresponding weight coefficients, and t is the time point of the sound data and the temperature data. Since the fault severity is related to multiple data, the fault severity is calculated by comprehensively considering the fault characteristic frequency value, the sound data and the temperature data of the transmission chain, thereby improving the accuracy of the fault severity.

[0143] The monitoring system compares the fault severity coefficient with a preset fault severity coefficient table after calculating the fault severity coefficient, determines the severity of the fault, and then obtains the transmission chain monitoring result according to the fault position and the fault severity, and displays the fault position and the fault severity to the relevant personnel through the wind turbine digital twin, and simultaneously performs alarm processing when the fault severity is high.

[0144] The monitoring system of the embodiment inputs the environmental data, the transmission chain working condition data, and the sound data, vibration data and temperature data of the transmission chain into the transmission chain monitoring model; calculates the time domain characteristic value corresponding to the vibration data of the transmission chain through the transmission chain monitoring model according to the vibration data of the transmission chain; generates the simulated time domain characteristic value of the transmission chain according to the environmental data and the transmission chain working condition data through the transmission chain monitoring model combined with the wind turbine digital twin, and compares the time domain characteristic value with the simulated time domain characteristic value; if the time domain characteristic value is greater than the simulated time domain characteristic value, the transmission chain monitoring result is obtained through the transmission chain monitoring model combined with the wind turbine digital twin according to the sound data and the temperature data of the transmission chain, and the fault characteristic frequency value in the time domain waveform. By combining the wind turbine digital twin for simulation, the accuracy of the obtained simulated time domain characteristic value is improved, thereby improving the accuracy of judging whether the transmission chain has a fault, and combining the sound data, temperature data and fault characteristic frequency value to determine the fault severity, and improving the accuracy of the alarm.

[0145] Reference Figure 4 , Figure 4 The flowchart of the third embodiment of the wind turbine monitoring method of the application is shown in the figure, and the blade monitoring sub-step includes:

[0146] The blade monitoring model input sub-step: according to the time sequence, respectively align the environmental data, the blade working condition data in the working condition data set, and the sound data, vibration data and temperature data of the blade, and input them into the blade monitoring model;

[0147] The simulated vibration data generation sub-step: through the blade monitoring model, according to the aligned environmental data and blade working condition data, combined with the wind turbine digital twin, generate simulated vibration data;

[0148] In the leaf monitoring model inputting step and the simulated vibration data generating step, after the monitoring system obtains the environmental data, the leaf working condition data in the working condition data set, and the sound data, vibration data and temperature data of the leaf, the monitoring system aligns the environmental data, the leaf working condition data in the working condition data set, and the sound data, vibration data and temperature data of the leaf according to the time sequence of obtaining the above data, and inputs the aligned environmental data and leaf working condition data into the wind turbine digital twin through the leaf monitoring model. The wind turbine digital twin generates simulated vibration data of the leaf based on the generated environmental data and leaf working condition data, and records the simulated vibration data of the leaf generated during the working process. It should be noted that the environmental data includes wind speed, lightning, air temperature, rain and snow, etc. The wind turbine digital twin can simulate the simulated vibration data generated by the leaf under the current environmental data based on various data. The simulated vibration data is the vibration data of the leaf when the leaf does not appear abnormal, which is beneficial to improve the accuracy of determining the normal vibration data of the leaf, and further improve the accuracy of subsequent judgment of whether the leaf is abnormal.

[0149] The deviation comparison step: calculating the deviation coefficient of the vibration data of the leaf and the simulated vibration data through the leaf monitoring model, and comparing the deviation coefficient with the preset deviation interval;

[0150] In this step, the monitoring system calculates the deviation coefficient of the vibration data of the leaf and the simulated vibration data through the leaf monitoring model after obtaining the simulated vibration data of the leaf, and compares the deviation coefficient with the preset deviation interval. The preset deviation interval is set in the leaf monitoring model in advance. The formula for calculating the deviation coefficient is:

[0151]

[0152] Wherein, A is the deviation coefficient, N is the number of sampling points which is set in the leaf monitoring model according to experience, x n is the frequency value corresponding to the nth sampling point in the vibration data of the leaf, x 1 n is the frequency value corresponding to the nth sampling point in the simulated vibration data of the leaf. The monitoring system samples N data in the vibration data of the leaf and the simulated vibration data of the leaf according to the preset number of sampling points N, and calculates the deviation coefficient A of the vibration data of the leaf and the simulated vibration data of the leaf according to the above formula. Taking N sampling points in the vibration data of the leaf and the simulated vibration data of the leaf respectively, without obtaining all the data in the vibration data and the simulated vibration data for calculation, the calculation efficiency is improved while the accuracy is ensured.

[0153] The time domain feature set extraction step: if the deviation coefficient is not in the preset deviation interval, the time domain feature set of the vibration data of the blade is extracted.

[0154] In this step, if the monitoring system determines that the deviation coefficient is not in the preset deviation interval, the vibration data of the blade is subjected to time-frequency analysis to obtain the time domain feature set of the vibration data of the blade. For example, the monitoring system obtains the time domain feature set of the vibration data of the blade by the formula:

[0155]

[0156] The time-frequency analysis is performed on the vibration data of the blade to calculate the time-frequency curve C x (t,f) of the vibration data of the blade, where t is the unit time corresponding to the vibration data, f is the frequency of the vibration data, a is a coefficient, a is generally 0.5, j is an imaginary unit, and i is a preset time shift parameter. After obtaining the time-frequency curve, the monitoring system performs frequency energy conversion on the unit time signal of the time-frequency curve to obtain the frequency mode and the frequency peak value of the vibration data as the time domain feature set.

[0157] The blade monitoring step: inputting the time domain feature set into the digital twin of the wind turbine to determine the fault position of the blade and determining the fault severity according to the time domain feature set, the sound data and the temperature data of the blade to obtain the blade monitoring result.

[0158] In this step, after obtaining the time domain feature set, the monitoring system inputs the time domain feature set into the digital twin of the wind turbine. The digital twin of the wind turbine simulates work according to the frequency mode and the frequency peak value in the time domain feature set to determine the fault position of the blade. After determining the fault position of the blade, the monitoring system determines the fault severity according to the time domain feature set, the sound data and the temperature data of the blade. For example, the monitoring system performs dimensionless processing on the time domain feature set, the sound data and the temperature data of the transmission chain, calculates the fault severity coefficient according to the dimensionless processed time domain feature set, the sound data and the temperature data of the transmission chain, and the formula for calculating the fault severity coefficient is as follows:

[0159]

[0160] where k is the fault severity coefficient, S i is the dimensionless processed sound data at the i th time point, W iFor the temperature data of the i-th time point after dimensionless processing, f is the time domain feature set after dimensionless processing, a, b and c are corresponding weight coefficients respectively, and t is the time point of the sound data and the temperature data. Since the fault severity of the blade is related to various data, the fault severity is calculated by integrating the time domain feature set, the sound data of the transmission chain and the temperature data, thereby improving the accuracy of the fault severity.

[0161] After the monitoring system calculates the fault severity coefficient, the monitoring system compares the fault severity coefficient with the preset fault severity coefficient table to determine the severity of the fault, and then obtains the blade monitoring result according to the fault position and the fault severity, and displays the fault position and the fault severity to the relevant personnel through the digital twin of the wind turbine, and simultaneously performs alarm processing when the fault severity is high.

[0162] The monitoring system of the embodiment improves the accuracy of the obtained simulation vibration data by simulating in combination with the digital twin of the wind turbine, thereby improving the accuracy of judging whether the blade has a fault, and determines the fault severity of the blade in combination with the sound data, the temperature data and the time domain feature set, thereby improving the accuracy of the alarm.

[0163] Reference Figure 5 , Figure 5 The flowchart of the fourth embodiment of the wind turbine monitoring method of the present application is shown in the figure, and the tower monitoring sub-step includes:

[0164] Deformation data calculation sub-step: input the environmental data, the working condition data set, the tower sway data and the inclination data of the tower into the tower monitoring model, and calculate the deformation data of the tower through the tower monitoring model;

[0165] Tower monitoring sub-step: obtain the time domain feature set of the deformation data through the tower monitoring model, input the time domain feature set of the deformation data into the digital twin of the wind turbine, determine the fault position and the fault severity of the tower, and obtain the tower monitoring result.

[0166] In the deformation data calculation sub-step and the tower monitoring sub-step, the monitoring system inputs the environmental data, the working condition data set, the tower sway data and the inclination data of the tower into the tower monitoring model, and calculates the deformation data of the tower through the tower monitoring model according to the above data; for example, the tower monitoring model calculates the deformation data of the tower according to the following formula:

[0167] x(t)=(m+nt)h r +ch s cosωt

[0168] Wherein, ω is the vibration frequency of the tower drum, which is calculated by the tower drum monitoring system according to the swing data and the inclination data of the tower drum, c is a dynamic deformation coefficient, s is a dynamic deformation index, m is a static deformation coefficient, n is a quasi-static deformation coefficient, c, s, m and n can be calculated by the tower drum monitoring system according to the environmental data and the working condition data of the tower drum, r is a constant, and h is the height of the tower drum. The monitoring system calculates the deformation data of the tower drum through the tower drum monitoring model, and then obtains the deformation curve of the tower drum. The time domain feature set of the deformation curve is obtained through the tower drum monitoring model, and the time domain feature set is input into the digital twin of the wind turbine. The fault position and the fault severity of the tower drum are determined by simulating the time domain feature set through the digital twin of the wind turbine, so as to obtain the tower drum monitoring result. Then, the fault position and the fault severity are displayed to the relevant personnel through the digital twin of the wind turbine and an alarm is given.

[0169] The monitoring system of the embodiment calculates the deformation data of the tower drum through the tower drum monitoring model combined with the environmental data, the working condition data set of the tower drum, the swing data and the inclination data of the tower drum, and then simulates according to the time domain features of the deformation data through the digital twin of the wind turbine, thereby improving the accuracy of determining whether the tower drum has a fault.

[0170] The application also provides a wind turbine monitoring device. The wind turbine monitoring device comprises:

[0171] The classification module is configured to, when the start instruction is detected, collect target data through the pre-installed sensor and data collector, and classify the target data to obtain a classification result.

[0172] The input module is configured to input the target data into a corresponding monitoring model according to the classification result, and obtain a monitoring result by combining the monitoring model with a pre-created digital twin of the wind turbine.

[0173] The display module is configured to display the monitoring result based on the digital twin of the wind turbine.

[0174] Further, the classification module is further configured to:

[0175] The sound, vibration and temperature integrated sensor is configured to collect sound data, vibration data and temperature data of the transmission chain and the blade.

[0176] The swing sensor is configured to collect swing data of the tower drum, and the inclination sensor is configured to collect inclination data of the tower drum.

[0177] The data collector is configured to collect a working condition data set and environmental data of the wind turbine, and obtain target data by combining the sound data, the vibration data, the temperature data, the swing data and the inclination data.

[0178] Further, the input module is further used for:

[0179] According to the classification result, input the environment data, the transmission chain working condition data in the working condition data set, and the sound data, vibration data and temperature data of the transmission chain into the transmission chain monitoring model, and obtain transmission chain monitoring result by combining the wind turbine digital twin created in advance through the transmission chain monitoring model;

[0180] Input the environment data, the blade working condition data in the working condition data set, and the sound data, vibration data and temperature data of the blade into the blade monitoring model, and obtain blade monitoring result by combining the wind turbine digital twin created in advance through the blade monitoring model;

[0181] Input the environment data, the working condition data set, and the sway data and the inclination data of the tower into the tower monitoring model, and obtain tower monitoring result by combining the wind turbine digital twin created in advance through the tower monitoring model.

[0182] Further, the input module further includes a transmission chain monitoring module, and the transmission chain monitoring module is used for:

[0183] Input the environment data, the transmission chain working condition data, and the sound data, vibration data and temperature data of the transmission chain into the transmission chain monitoring model;

[0184] According to the vibration data of the transmission chain, generate a corresponding time domain waveform graph through the transmission chain monitoring model, and calculate a time domain characteristic value corresponding to the vibration data of the transmission chain according to the time domain waveform graph;

[0185] According to the environment data, the transmission chain working condition data, and the sound data and temperature data of the transmission chain, generate an analog time domain characteristic value of the transmission chain by combining the wind turbine digital twin through the transmission chain monitoring model;

[0186] If the time domain characteristic value is greater than the analog time domain characteristic value, generate a corresponding time domain waveform graph according to the time domain waveform graph;

[0187] Obtain transmission chain monitoring result according to the time domain waveform graph and the wind turbine digital twin through the transmission chain monitoring model.

[0188] Further, the transmission chain monitoring module is further used for:

[0189] Perform wavelet multi-resolution analysis on the time domain waveform graph through the transmission chain monitoring model to generate a frequency band signal set;

[0190] signal reconstruction is performed on each frequency band signal in the set of frequency band signals, and a refined spectrum analysis and an envelope spectrum analysis are performed on each frequency band signal after the signal reconstruction to obtain a fault characteristic frequency value;

[0191] The fault characteristic frequency value is input into the digital twin of the wind turbine for simulation to determine a fault location and a fault severity of the transmission chain to obtain a transmission chain monitoring result.

[0192] Further, the input module further comprises a blade monitoring module, and the blade monitoring module is configured to:

[0193] According to a time sequence, the environment data, blade working condition data in the set of working condition data, and sound data, vibration data and temperature data of the blade are aligned respectively and input into the blade monitoring model;

[0194] The blade monitoring model generates simulated vibration data according to the aligned environment data, blade working condition data, and sound data and temperature data of the blade in combination with the digital twin of the wind turbine;

[0195] The blade monitoring model calculates a deviation coefficient of the vibration data of the blade and the simulated vibration data, and compares the deviation coefficient with a preset deviation interval;

[0196] If the deviation coefficient is not within the preset deviation interval, a time domain feature set of the vibration data of the blade is extracted, and the time domain feature set is input into the digital twin of the wind turbine to determine a fault location and a fault severity of the blade to obtain a blade monitoring result.

[0197] Further, the input module further comprises a tower monitoring module, and the tower monitoring module is configured to:

[0198] The environment data, the set of working condition data, the sway data and the inclination data of the tower are input into the tower monitoring model, and the tower monitoring model calculates deformation data of the tower;

[0199] The tower monitoring model obtains a time domain feature set of the deformation data, and the time domain feature set of the deformation data is input into the digital twin of the wind turbine to determine a fault location and a fault severity of the tower to obtain a tower monitoring result.

[0200] The application further provides a wind turbine monitoring system.

[0201] The wind turbine monitoring system comprises a memory, a processor, and a wind turbine monitoring program stored in the memory and executable on the processor, and the wind turbine monitoring program, when executed by the processor, implements the steps of the wind turbine monitoring method.

[0202] The method implemented when the wind turbine monitoring program executable on the processor is executed can refer to the embodiments of the wind turbine monitoring method of the present application, and will not be described here.

[0203] The present application also provides a computer readable storage medium.

[0204] The computer readable storage medium stores the wind turbine monitoring program, and the wind turbine monitoring program, when executed by the processor, implements the steps of the wind turbine monitoring method.

[0205] The method implemented when the wind turbine monitoring program executable on the processor is executed can refer to the embodiments of the wind turbine monitoring method of the present application, and will not be described here.

[0206] It should be noted that in this document, the terms "comprising", "including", or any other variant thereof are intended to cover non-exclusive inclusions, so that processes, methods, articles, or systems that include a series of elements not only include those elements, but also include other elements not explicitly listed, or inherent to such processes, methods, articles, or systems. Without more limitations, the element defined by the statement "comprising a" does not exclude the presence of other identical elements in the process, method, article, or system that includes the element.

[0207] The above-mentioned embodiment numbers of the present application are only for description, and do not represent the advantages and disadvantages of the embodiments.

[0208] From the above description of the embodiments, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software and the necessary general hardware platform, of course, they can also be implemented by hardware, but in many cases the former is a better embodiment. Based on this understanding, the technical solutions of the present application can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes a plurality of instructions for making a terminal device (which can be a mobile phone, computer, server, or network device) execute the methods described in the embodiments of the present application.

[0209] The above merely describes the preferred embodiments of the present application, and is not intended to limit the patent scope of the present application, and any equivalent structure or equivalent process conversion, or direct or indirect application in other related technical fields, which are made by using the content of the present application specification and drawings, are also included in the patent protection scope of the present application.

Claims

1. A wind turbine monitoring method, characterized in that: The wind turbine monitoring method comprises the following steps: A classification step: when a start command is detected, target data is collected by pre-installed sensors and data collectors, and the target data is classified to obtain a classification result, wherein the sensors include: an integrated sound, vibration and temperature sensor installed in a transmission chain of the wind turbine, an integrated sound, vibration and temperature sensor installed at a blade root of the wind turbine, a sway sensor and an inclination sensor installed at a top end of a tower of the wind turbine, and an inclination sensor installed at a bottom end of the tower; Monitoring step: according to the classification result, inputting the target data into a corresponding monitoring model, and obtaining a monitoring result by combining the monitoring model with a pre-created digital twin of the wind turbine, wherein the monitoring model includes: a transmission chain monitoring model, a blade monitoring model, and a tower monitoring model; Presentation step: presenting the monitoring results based on the wind turbine digital twin; The step of inputting the target data into a corresponding monitoring model according to the classification result, and obtaining a monitoring result by combining the monitoring model with a pre-created digital twin of the wind turbine comprises: inputting the environmental data collected by the data collector and the transmission chain operating condition data in the operating condition data set, as well as the sound data, vibration data, and temperature data of the transmission chain collected by the integrated sound, vibration, and temperature sensor into the transmission chain monitoring model according to the classification result; Generate a corresponding time domain waveform diagram based on the vibration data of the transmission chain using the transmission chain monitoring model, and calculate a time domain eigenvalue corresponding to the vibration data of the transmission chain based on the time domain waveform diagram; By combining the transmission chain monitoring model with the wind turbine digital twin, a simulated time domain characteristic value of the transmission chain is generated according to the environmental data and the transmission chain operating condition data, and the time domain characteristic value is compared with the simulated time domain characteristic value; If the time domain eigenvalue is greater than the simulated time domain eigenvalue, the transmission chain monitoring result is obtained by combining the transmission chain monitoring model with the wind turbine digital twin based on the sound data and temperature data of the transmission chain and the time domain waveform diagram.

2. The wind turbine monitoring method according to claim 1, wherein: The classification step comprises: Transmission chain and blade data collection sub-step: collecting sound data, vibration data and temperature data of the transmission chain and the blades through the sound, vibration and temperature integrated sensor; Tower data collection sub-step: collecting the tower shaking data by the shaking sensor, and collecting the tower tilt data by the tilt sensor; Determine the target data sub-step: collect the operating data set and environmental data of the wind turbine through a data collector, and combine the sound data, the vibration data, the temperature data, the shaking data and the tilt data to obtain the target data.

3. The wind turbine monitoring method according to claim 2, characterized in that: The step of inputting the target data into a corresponding monitoring model according to the classification result, and obtaining a monitoring result by combining the monitoring model with a pre-created digital twin of the wind turbine further includes: a blade monitoring sub-step: inputting the environmental data, the blade operating condition data in the operating condition data set, and the sound data, vibration data, and temperature data of the blade into the blade monitoring model, and combining the blade monitoring model with a pre-created wind turbine digital twin to obtain a blade monitoring result; Tower monitoring sub-step: input the environmental data, the operating condition data set, and the tower shaking data and the tilt data into the tower monitoring model, and combine the tower monitoring model with the pre-created wind turbine digital twin to obtain the tower monitoring results.

4. The wind turbine monitoring method according to claim 3, characterized in that: The steps of obtaining the transmission chain monitoring result include: The frequency band signal set production step includes: performing wavelet multi-resolution analysis on the time domain waveform through the transmission chain monitoring model to generate a frequency band signal set; Obtaining the fault characteristic frequency value: performing a refined spectrum analysis and an envelope spectrum analysis on each frequency band signal in the frequency band signal set to obtain the fault characteristic frequency value; The transmission chain monitoring result is determined in the following steps: the fault characteristic frequency value is input into the digital twin of the wind turbine for simulation, the fault location of the transmission chain is determined, and the fault severity is determined based on the fault characteristic frequency value, the sound data and the temperature data of the transmission chain to obtain the transmission chain monitoring result.

5. The wind turbine monitoring method according to claim 3, characterized in that: The blade monitoring sub-step includes: Blade monitoring model input sub-step: aligning the environmental data, the blade operating condition data in the operating condition data set, and the sound data, vibration data, and temperature data of the blade according to the time series, and inputting them into the blade monitoring model; The simulated vibration data generation step includes: generating simulated vibration data by using the blade monitoring model according to the aligned environmental data and the blade operating condition data, combined with the wind turbine digital twin; Deviation comparison step: calculating a deviation coefficient between the vibration data of the blade and the simulated vibration data using the blade monitoring model, and comparing the deviation coefficient with a preset deviation range; The time domain feature set extraction step: if the deviation coefficient is not within the preset deviation interval, extracting the time domain feature set of the vibration data of the blade; The blade monitoring sub-step is as follows: inputting the time domain feature set into the wind turbine digital twin to determine the fault location of the blade, and determining the severity of the fault based on the time domain feature set, the sound data and the temperature data of the blade to obtain the blade monitoring result.

6. The wind turbine monitoring method according to claim 3, characterized in that: The tower monitoring sub-step includes: The deformation data calculation step includes inputting the environmental data, the operating condition data set, the sway data and the tilt data of the tower into the tower monitoring model, and calculating the deformation data of the tower through the tower monitoring model; The tower monitoring step includes: obtaining the time domain feature set of the deformation data through the tower monitoring model, inputting the time domain feature set of the deformation data into the wind turbine digital twin, determining the fault location and fault severity of the tower, and obtaining the tower monitoring result.

7. A wind turbine monitoring device, characterized in that: The wind turbine monitoring device comprises: a classification module, configured to, upon detecting a start command, collect target data through pre-installed sensors and data collectors, and classify the target data to obtain a classification result, wherein the sensors include: an integrated sound, vibration, and temperature sensor installed in a transmission chain of the wind turbine, an integrated sound, vibration, and temperature sensor installed at a blade root of the wind turbine, a sway sensor and an inclination sensor installed at a top end of a tower of the wind turbine, and an inclination sensor installed at a bottom end of the tower; An input module is configured to input the target data into a corresponding monitoring model based on the classification result, and obtain a monitoring result by combining the monitoring model with a pre-created digital twin of the wind turbine, wherein the monitoring model includes a transmission chain monitoring model, a blade monitoring model, and a tower monitoring model; A display module, configured to display the monitoring results based on the wind turbine digital twin; The input module is further used in the transmission chain monitoring sub-step: according to the classification result, the environmental data collected by the data collector and the transmission chain operating condition data in the operating condition data set, as well as the sound data, vibration data and temperature data of the transmission chain collected by the sound, vibration and temperature integrated sensor are input into the transmission chain monitoring model, and the transmission chain monitoring model is combined with the pre-created wind turbine digital twin; Generate a corresponding time domain waveform diagram based on the vibration data of the transmission chain using the transmission chain monitoring model, and calculate a time domain eigenvalue corresponding to the vibration data of the transmission chain based on the time domain waveform diagram; By combining the transmission chain monitoring model with the wind turbine digital twin, a simulated time domain characteristic value of the transmission chain is generated according to the environmental data and the transmission chain operating condition data, and the time domain characteristic value is compared with the simulated time domain characteristic value; If the time domain eigenvalue is greater than the simulated time domain eigenvalue, the transmission chain monitoring result is obtained by combining the transmission chain monitoring model with the wind turbine digital twin based on the sound data and temperature data of the transmission chain and the time domain waveform diagram.

8. A wind turbine monitoring system, characterized in that: The wind turbine monitoring system includes: a memory, a processor, and a wind turbine monitoring program stored in the memory and executable on the processor. When the wind turbine monitoring program is executed by the processor, the steps of the wind turbine monitoring method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a wind turbine monitoring program, which, when executed by a processor, implements the steps of the wind turbine monitoring method according to any one of claims 1 to 6.

Images