Key structure dynamic characteristic long-term tracking identification method for wind driven generator

Abstract

The invention discloses a key structure dynamic characteristic long-term tracking identification method for a wind driven generator, which comprises the following steps: acquiring acceleration vibration data and generator shaft rotating speed data when a fan is in long-term service, and performing grouping processing on the acquired acceleration vibration data and generator shaft rotating speed data; judging the operation state of the fan corresponding to each group of data according to the rotating speed signal of the generator shaft; and according to a time history, independently drawing respective related frequency and modal damping time history diagrams for different key components of the wind driven generator, and realizing long-term tracking identification of the key structure dynamic characteristics of the wind driven generator. According to the invention, modal parameter identification analysis can be regularly carried out on the key structure of the wind driven generator on line, so that long-term tracking monitoring of the dynamic characteristics of the key structure is realized.

Description

Technical Field

[0001] The invention relates to the technical field of structure monitoring, and in particular to a method for long-term tracking and identifying key structural dynamic characteristics of a wind turbine. Background Art

[0002] During the operation of wind turbines, due to the long-term influence of complex and changeable environmental factors, their key structures such as towers and blades may suffer from cumulative faults such as wear, fatigue, and cracks. If these problems are not discovered and handled in a timely manner, they may affect the normal operation of the equipment at the least, or cause serious accidents at the worst. Therefore, it is of great significance to carry out research on structural health monitoring technology of wind turbines. The dynamic characteristics of the structure are important indicators for evaluating the vibration performance and health status of the structure. Among them, the change of structural modal frequency over time can reflect the health status of the structure; the change of structural modal damping can reflect the form of structural vibration. In order to better understand the health status and performance evolution of the key structures of wind turbines during service, the abnormal vibration of the key structures is monitored and timely intervention can be carried out. It is very important to track and study the long-term dynamic characteristics of the key structures of wind turbines, including modal frequency and modal damping information, and analyze their characteristic laws.

[0003] As wind turbine structural health monitoring has gradually gained attention, many wind turbines have deployed wind turbine supervisory control and data acquisition systems, which can realize real-time monitoring of wind turbine structural vibration response and environmental data. Massive monitoring databases can provide a basis for studying the long-term dynamic characteristics of key structures of wind turbines. Therefore, there is an urgent need for a method for tracking and identifying dynamic characteristics of long-term monitoring signals extracted from wind turbines. Summary of the invention

[0004] In view of this, the present invention provides a method for long-term tracking and identification of key structural dynamic characteristics of a wind turbine.

[0005] The present invention discloses a method for long-term tracking and identifying key structural dynamic characteristics of a wind turbine, which comprises:

[0006] Step 1: Acquire acceleration vibration data and generator shaft speed data of the wind turbine during long-term service, and group the acquired acceleration vibration data and generator shaft speed data;

[0007] Step 2: Determine the operating state of the fan to which each set of data corresponds based on the generator shaft speed signal;

[0008] Step 3: According to the time history, independently draw the frequency and modal damping time history diagrams related to different key components of the wind turbine to achieve long-term tracking and identification of the key structural dynamic characteristics of the wind turbine.

[0009] Furthermore, the step 2 comprises:

[0010] Step 21: When the wind turbine is in operation, the vibration acceleration signal is synchronously processed using synchronous information processing technology; the interference information related to the operation of the shaft is removed from each group of data obtained through synchronous processing in combination with a bandpass filter; the original sampling clock is used to resample each group of data after the interference is removed to restore the signal, and the modal frequency information of each component in the signal spectrum diagram of each group of data is obtained using the peak picking method.

[0011] Furthermore, the step 21 comprises:

[0012] Step 211: Processing the vibration acceleration signal using a synchronous information processing method according to the motor shaft speed data to obtain an order spectrum of a synchronous order signal in a periodic domain;

[0013] Step 212: filtering the synchronous order signal in the periodic domain with respect to the order spectrum signal of the vibration acceleration signal, eliminating the order response related to the rotating shaft, and obtaining a synchronous signal with operation response interference eliminated;

[0014] Step 213: using resampling in synchronous information processing technology, reversely processing the acceleration vibration according to the timing sampling sequence of the original vibration acceleration signal, so as to restore the time domain signal of the original vibration acceleration signal with the operation response interference removed;

[0015] Step 214: Perform spectrum analysis on the time domain signal, extract information in the spectrum by using a peak picking method, and read the coordinates corresponding to the peak points to obtain the modal frequencies of each key structure.

[0016] Furthermore, the step 211 includes:

[0017] The instantaneous phase of the motor shaft speed data is extracted, and a synchronous acquisition clock with a full period is generated by inverse function conversion. The vibration acceleration signal is resampled using the synchronous acquisition clock to obtain a synchronous order signal, and the synchronous order signal is Fourier transformed to obtain an order spectrum.

[0018] Further, the step 212 includes:

[0019] Step 2121: for the order spectrum signal of the vibration acceleration signal, use a bandpass filter to filter the synchronous order signal in the periodic domain to obtain a plurality of synchronous full-order signals related to the rotating shaft; remove the order response related to the rotating shaft from the order spectrum signal, and perform bandpass filtering on the order spectrum signal after removing the order response;

[0020] Step 2122: Eliminate the bandpass filtered signal from the original vibration acceleration signal, that is, obtain a synchronous signal with the operation response interference eliminated.

[0021] Furthermore, the step 2 comprises:

[0022] Step 22: When the wind turbine is in shutdown state, band-pass filtering is performed on the frequency interval where the multi-order modal energy is concentrated; the random reduction method is used to convert the response acceleration signal of the multi-order mode induced by the random excitation of the environment into a free decay signal, and the damping information of the different order modal components of the structure of the wind turbine in each group of data is obtained by fitting the exponential function model.

[0023] Further, the step 22 comprises:

[0024] Step 221: performing bandpass filtering on the vibration acceleration signal to obtain single-mode acceleration data of the key structure of the wind turbine;

[0025] Step 222: converting the single-mode acceleration data into a free decay signal by random reduction method;

[0026] Step 223: Perform a Hibbert transform on the free decay signal, take the logarithm of the transformed amplitude and obtain the modal damping by fitting.

[0027] Furthermore, after step 2 and before step 3, the method further includes:

[0028] Repeat step 2 to obtain the multi-order frequencies and modal damping of the wind turbine generator set at each time number corresponding to the grouped data.

[0029] Furthermore, the step 1 comprises:

[0030] The acceleration vibration data and generator shaft speed data of the fan during long-term service are obtained from the fan supervisory control and data acquisition system; at the same time, the obtained acceleration vibration data and generator shaft speed data are grouped and processed according to time.

[0031] Due to the adoption of the above technical solution, the present invention has the following advantages:

[0032] 1. The present invention is suitable for long-term tracking of key structural dynamic characteristics of wind turbines. The present invention can effectively consider the impact of different operating conditions data, and adopts different analysis methods for processing and analysis of different operating conditions data. The operating conditions of wind turbine data are judged according to the shaft speed signal; for data under operating conditions, the vibration acceleration signal and the speed signal are processed using synchronous information processing technology, and then the modal frequency of each key structure is obtained by combining spectrum analysis with peak picking method. For vibration signals under shutdown conditions, a bandpass filter is used to obtain the single modal component signal of the structure caused by environmental excitation, and then the random reduction method is used to extract the free decay signal from the single modal component for fitting damping.

[0033] 2. The present invention obtains acceleration signals from the supervisory control and data acquisition system of the wind turbine and identifies multiple order modal frequencies and damping of key structures such as the tower and blades of the wind turbine. The analysis of cases shows that the accuracy of modal parameter identification and the degree of automation are significantly reliable and practical.

[0034] 3. The method of the present invention can be used to track and identify the long-term dynamic characteristics of the key structures of wind turbines, which can provide technical support for studying the evolution of the key structural frequency and damping of wind turbines with respect to time and wind speed. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the embodiments of the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0036] Figure 1 A flow chart of a method for long-term tracking and identification of key structural dynamic characteristics of a wind turbine according to the present invention;

[0037] Figure 2 This is a flow chart of the synchronous information processing technology involved in the present invention;

[0038] Figure 3 To set the order spectrum and bandpass filter width of the synchronization signal;

[0039] Figure 4 The modal frequencies of each key part are obtained by peak extraction method;

[0040] Figure 5 To extract the multi-order modal damping corresponding to different key structures. DETAILED DESCRIPTION

[0041] The present invention is further described in conjunction with the accompanying drawings and embodiments, and the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by ordinary technicians in this field should fall within the scope of protection of the embodiments of the present invention.

[0042] The present invention extracts different working condition data of the wind turbine generator when it is in long-term service obtained by the wind turbine supervisory control and data acquisition system; pre-processes the vibration acceleration monitoring data and calculates the time-frequency spectrum of each signal segment in segments to analyze and judge the working condition of the wind turbine. When the judgment result is the operating condition, the vibration acceleration signal is synchronously processed by using synchronous information processing technology, and the interference information related to the operation of the rotating shaft is eliminated by combining with a bandpass filter; then the signal spectrum diagram after eliminating the interference is further analyzed, and the modal frequency of each component in the signal is obtained by using the peak picking method. When the wind turbine is judged to be in a shutdown condition, the multi-order modal energy concentration frequency interval that needs to be paid attention to is band-pass filtered; the response acceleration signal induced by the random excitation of the multi-order mode that needs to be paid attention to by the random reduction method is converted into a free decay signal, and the damping information of the different order modal components of the structure that needs to be paid attention to in the signal is obtained by fitting the exponential function model; according to the multi-order modal parameters of the key structure of the wind turbine, a frequency and modal damping time history diagram is drawn according to the time history, so as to realize the long-term tracking and identification of the key structural dynamic characteristics of the wind turbine.

[0043] Specifically, see Figures 1 to 5 The present invention provides an embodiment of a method for long-term tracking and identifying key structural dynamic characteristics of a wind turbine, which comprises the following steps:

[0044] Step S1: Obtain different operating condition data of the fan during long-term service from the fan supervisory control and data acquisition system; at the same time, group the acquired data by hour, including acceleration data Y o And the motor shaft speed data V o , where o represents the hourly number of the data.

[0045] Step S2: Determine the operating state of the fan corresponding to each set of data according to the motor shaft speed signal, and execute step S3 when the determination result is an operating condition; and execute step S4 when the determination result is a shutdown condition.

[0046] Step S3: extracting the multi-order modal frequencies corresponding to different key structures, including the following steps:

[0047] S31: According to the motor shaft speed data V o Signal, using synchronous information processing technology to analyze the vibration acceleration signal Y o Processing is performed to obtain the order spectrum of the synchronous order signal in the periodic domain. Synchronous information processing technology is used to process the motor shaft speed data Vo The instantaneous phase of the signal is extracted, and a synchronous acquisition clock with a full cycle is generated by inverse function conversion; and the synchronous acquisition clock is used to collect the vibration acceleration signal Y o After resampling, the synchronous order signal is obtained, and the order spectrum is obtained by Fourier transforming the synchronous order signal.

[0048] S32: Filter the periodic domain signal using a bandpass filter for the order spectrum signal of the vibration acceleration signal, and then obtain multiple synchronous full-order signals Y related to the rotating shaft o (s), s = 1, 2, ..., n; where n is the number of order responses related to the shaft that need to be eliminated. The filtering frequency band of the bandpass filter is [f star ,f cut ]; filter starting frequency f star =0.95*n, the filter cutoff frequency is f cut =1.05*n.

[0049] S33: Remove the filtered signal from the original signal, that is, obtain the signal synchronization signal YT without the operation response interference o =Y o -Y o (s).

[0050] S34: resample the signal after removing interference using the original sampling clock, and sample the synchronization signal YT according to the timing sampling sequence of the original signal. O The signal is processed inversely and then restored to the original time domain signal with the running response interference removed.

[0051] S35: For the original signal Perform spectrum analysis, extract information from the spectrum by peak picking method, and read the horizontal coordinate corresponding to the peak point to obtain the modal frequency of each key structure.

[0052] Step S4: extracting the multi-order modal damping corresponding to different key structures. When extracting a certain order modal damping of a single structure separately, the following steps are included:

[0053] S41: First, the original signal Y o Bandpass filtering is performed to obtain the single-mode acceleration data of the structure of interest.

[0054] S42: Acceleration data by random reduction method Convert to free decay signal Among them, the acceleration signal of a single group of free decay can be expressed as

[0055] S43: Perform Hibbert transformation on a single group of free decay signals, i.e. construct an analytical equation H(t) = U(t)e iθ(t) ; where U(t) represents the amplitude, which can be further expressed as

[0056] S44: Take the logarithm of the amplitude U(t) and obtain the modal damping by fitting.

[0057] Step S5: Repeat steps S2 and S4 to obtain the multi-order frequencies and modal damping of the wind turbine generator set at each time number corresponding to the grouped data. The calculation results of each key structure are plotted into a frequency and modal damping time history diagram according to the time history, so as to realize the long-term tracking and identification of the dynamic characteristics in the key structure of the wind turbine. It is possible to further study the evolution law of the modal frequency and damping in the key structure of the wind turbine over time, and summarize the evolution characteristics of the key structural health status and dynamic characteristics of the wind turbine from long-term tracking.

[0058] The present invention can regularly perform online modal parameter identification and analysis on the key structures of the wind turbine, thereby realizing long-term tracking and monitoring of the dynamic characteristics of the key structures.

[0059] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A method for long-term tracking and identification of key structural dynamic characteristics of a wind turbine, characterized in that: include: Step 1: Acquire acceleration vibration data and generator shaft speed data of the wind turbine during long-term service, and group the acquired acceleration vibration data and generator shaft speed data; Step 2: Determine the operating state of the fan to which each set of data corresponds based on the generator shaft speed signal; Step 3: According to the time history, independently draw the frequency and modal damping time history diagrams related to different key components of the wind turbine to achieve long-term tracking and identification of the key structural dynamic characteristics of the wind turbine.

2. The method according to claim 1, characterized in that The step 2 comprises: Step 21: When the wind turbine is in operation, the vibration acceleration signal is synchronously processed using synchronous information processing technology; the interference information related to the operation of the shaft is removed from each group of data obtained through synchronous processing in combination with a bandpass filter; the original sampling clock is used to resample each group of data after the interference is removed to restore the signal, and the modal frequency information of each component in the signal spectrum diagram of each group of data is obtained using the peak picking method.

3. The method according to claim 2, characterized in that The step 21 comprises: Step 211: Processing the vibration acceleration signal using a synchronous information processing method according to the motor shaft speed data to obtain an order spectrum of a synchronous order signal in a periodic domain; Step 212: filtering the synchronous order signal in the periodic domain with respect to the order spectrum signal of the vibration acceleration signal, eliminating the order response related to the rotating shaft, and obtaining a synchronous signal with operation response interference eliminated; Step 213: using resampling in synchronous information processing technology, reversely processing the acceleration vibration according to the timing sampling sequence of the original vibration acceleration signal, so as to restore the time domain signal of the original vibration acceleration signal with the operation response interference removed; Step 214: Perform spectrum analysis on the time domain signal, extract information in the spectrum by using a peak picking method, and read the coordinates corresponding to the peak points to obtain the modal frequencies of each key structure.

4. The method according to claim 3, characterized in that The step 211 comprises: The instantaneous phase of the motor shaft speed data is extracted, and a synchronous acquisition clock with a full period is generated by inverse function conversion. The vibration acceleration signal is resampled using the synchronous acquisition clock to obtain a synchronous order signal, and the synchronous order signal is Fourier transformed to obtain an order spectrum.

5. The method according to claim 3, characterized in that: The step 212 includes: Step 2121: for the order spectrum signal of the vibration acceleration signal, use a bandpass filter to filter the synchronous order signal in the periodic domain to obtain a plurality of synchronous full-order signals related to the rotating shaft; remove the order response related to the rotating shaft from the order spectrum signal, and perform bandpass filtering on the order spectrum signal after removing the order response; Step 2122: Eliminate the bandpass filtered signal from the original vibration acceleration signal, that is, obtain a synchronous signal with the operation response interference eliminated.

6. The method according to claim 1, characterized in that The step 2 comprises: Step 22: When the wind turbine is in shutdown state, band-pass filtering is performed on the frequency interval where the multi-order modal energy is concentrated; the random reduction method is used to convert the response acceleration signal of the multi-order mode induced by the random excitation of the environment into a free decay signal, and the damping information of the different order modal components of the structure of the wind turbine in each group of data is obtained by fitting the exponential function model.

7. The method according to claim 6, characterized in that The step 22 comprises: Step 221: performing bandpass filtering on the vibration acceleration signal to obtain single-mode acceleration data of the key structure of the wind turbine; Step 222: converting the single-mode acceleration data into a free decay signal by random reduction method; Step 223: Perform a Hibbert transform on the free decay signal, take the logarithm of the transformed amplitude and obtain the modal damping by fitting.

8. The method according to claim 1, characterized in that After step 2 and before step 3, the method further includes: Repeat step 2 to obtain the multi-order frequencies and modal damping of the wind turbine generator set at each time number corresponding to the grouped data.

9. The method according to claim 1, characterized in that: The step 1 comprises: The acceleration vibration data and generator shaft speed data of the fan during long-term service are obtained from the fan supervisory control and data acquisition system; at the same time, the obtained acceleration vibration data and generator shaft speed data are grouped and processed according to time.

Images