A tip timing arrival time calculation method based on multi-adjacent shaft key method

Abstract

The application discloses a kind of based on the tip timing arrival time calculation method of multi-adjacent shaft key method, comprising the following steps: first: for the blade and shafting of turbomachinery, blade position sensor device, shaft key position sensor device and shaft key structure are designed and arranged, and the arrival signal of blade position sensor and shaft key position sensor is measured by blade position sensor and shaft key position sensor respectively;Second step: each group of data obtained by sensor is carried out data noise reduction;Third step: using multi-adjacent shaft key method, the sensor data after noise reduction processing is handled, and the tip arrival time difference value is obtained, and then the vibration displacement of blade is obtained.The present application can realize more accurate calculation of tip timing arrival time, effectively improve the accuracy of vibration data measured by tip timing.

Description

Technical Field

[0001] This invention belongs to the field of non-contact vibration monitoring of turbine blades, specifically involving a method for calculating the timing arrival time of the blade tip based on the multi-adjacent axis key method. Background Technology

[0002] Turbine blades are the core components of turbine machinery, and their safety and reliability are crucial factors for the safe and stable operation of such machinery. Turbine blades operate in extremely complex environments, enduring vibration stresses and are highly susceptible to high-cycle fatigue, leading to blade failure. Therefore, real-time monitoring of the vibration status of turbine blades is essential and of significant engineering value.

[0003] Traditional vibration measurement typically employs contact methods, requiring the installation of displacement sensors on the blade surface. This approach is complex to install, inefficient, and can negatively impact blade structural performance. Tip-timed vibration measurement, a non-contact method, is simple in structure, easy to install, and has minimal impact on the blade. Therefore, tip-timed vibration measurement is increasingly widely used in vibration monitoring of turbine blades.

[0004] The principle of the blade tip timing method is as follows: sensors are installed around the working shell of the turbine blade and near the shaft to measure the arrival time of the blade tip's rotation. The difference between this value and the theoretical arrival time is used to obtain the blade displacement value, and thus the vibration parameters. Therefore, the key to the blade tip timing method is the accurate extraction of the blade tip arrival time.

[0005] Currently, traditional blade tip timing methods mainly include the keyless method and the single-key method. The keyless method lacks a reference time from the key, resulting in significant measurement errors; while the single-key method suffers from a substantial increase in vibration error for blades far from the key, making it impossible to uniformly measure the entire blade circumference. These traditional methods generally suffer from low computational accuracy and large relative errors, making it impossible to accurately extract the blade tip arrival time. Therefore, the key to blade tip timing technology is the stability of the sensor signal. However, under actual background noise, the sensor signal will have significant errors, making it difficult to extract the blade tip arrival time data. Summary of the Invention

[0006] The purpose of this invention is to provide a method for calculating the arrival time of blade tips based on the multi-adjacent key method. This method uses sensors arranged at the key and the blade tip to obtain arrival time measurement signals of the blade tip and adjacent multi-key points. After noise reduction processing of the signals, the vibration displacement of each blade at its arrival time is calculated using the multi-adjacent key method. This invention enables more accurate calculation of the arrival time of blade tips, effectively improving the accuracy of vibration data measured at blade tips.

[0007] The present invention is achieved using the following technical solution:

[0008] A method for calculating the tip timing arrival time based on the multi-adjacent-axis key method includes the following steps:

[0009] Step 1: Design and arrange blade position sensor devices, keyway position sensor devices, and keyway structures for the blades and shaft system of the turbine machinery. Measure the arrival signals of the blade position sensor and keyway position sensor respectively.

[0010] Step 2: Perform noise reduction on the data obtained by the sensor.

[0011] Step 3: Using the multiple adjacent axis key method, the sensor data after noise reduction is processed to obtain the time difference of the blade tip arrival, and then the vibration displacement of the blade is obtained.

[0012] A further improvement of the present invention is that, in the first step, in order to ensure the measurement accuracy of the blades and reduce the influence of the rotor's eccentric vibration, the shaft keys are arranged in a uniform manner, that is, the central angles between the shaft keys are all equal.

[0013] A further improvement of the present invention is that, in the first step, the ratio N / n of the number of blades to the number of shaft keys is taken as an integer or the closest integer, and the number of shaft keys arranged is greater than or equal to 4, so as to ensure that there are adjacent multi-shaft keys.

[0014] A further improvement of the present invention is that, in the first step, the cross-sectional shape of the shaft key is a rectangle, a regular trapezoid, an inverted trapezoid, or an irregular quadrilateral.

[0015] A further improvement of this invention is that, in the first step, a total of K sets of blade tip position sensor signal data are obtained, namely {W1, W2, W3... W... K A total of L sets of shaft key position sensor signal data were obtained, namely {X1, X2, X3……X}. L}

[0016] A further improvement of this invention lies in the fact that, in the second step, a Fourier transform noise reduction method is used to reduce the noise of the signals obtained from the sensors. Noise reduction processing is performed sequentially on the K groups of blade tip position sensor signal data and the L groups of shaft key position sensor signal data to obtain the processed signals {W′1, W′2, W′3……W K {X1′} and {X2′, X3′……X} L ′}.

[0017] A further improvement of this invention is that, in the second step, noise reduction is performed on the signal obtained from the sensor. For the first set of key position signals X1, the specific steps are as follows:

[0018] (201) Transform the signal data from the time domain to the frequency domain, that is:

[0019] (202) According to the frequency domain signal X1 T Based on the characteristics of the signal, noise reduction processing is performed to remove the signal with noise frequency and retain the signal with peak frequency.

[0020] (203) Convert the processed frequency domain signal into a time domain signal, i.e.

[0021] (204) Perform noise reduction processing steps (201) to (203) sequentially on the K group of blade tip position sensor signal data and the L group of shaft key position sensor signal data to obtain the processed signals {W1′, W2′, W3′……W K {X1′} and {X2′, X3′……X} L ′}.

[0022] A further improvement of this invention is that, in the third step, the specific implementation steps are as follows:

[0023] (301) The noise-reduced signal is processed using the multi-nearest axis key method;

[0024] (302) Assume the reference position of the blade root. Assume that there are m blades between every two adjacent shaft keys. Then, based on the phase distance between the blade root and the shaft key, insert m blade root reference position points between adjacent shaft keys to ensure that in the whole circle of blades, each blade or blade root has a relative position reference point between two shaft keys.

[0025] (303) Define the arrival times of adjacent key, blade root position, and blade tip position. Select three timing times from the key arrival time signal X1′, and use t n-1 t n With t n+1 In other words, t n With t n+1 When two adjacent shaft keys are timed, m signals are obtained from the tip arrival time signal W1′, and used... The expression represents the presence of m blades between two adjacent axial keys; i∈[1,m], where t n With t n+1 Between the arrival times of the two shaft keys, based on the phase and rotational speed distributions, insert m blade root reference points, using {t' n,1 ,t' n,2 ...t' n,i ...t' n,m} represents the expression where: m indicates that there are m leaf root reference points between two adjacent axial keys; i∈[1,m];

[0026] (304) Calculate the time difference and vibration displacement of the blade arrival times by using the arrival times of adjacent shaft keys, blade root positions, and blade tip positions. For m blades between the nth shaft key and the (n+1)th shaft key, calculate the blade arrival times. Arrival time of leaf root reference position {t' n,1 ,t' n,2 ...t' n,i ...t' n,m The difference between} is {Δt} n,1 ,Δt n,2 ……Δt n,i ……Δt n,m}, then determine the time difference of arrival times of these m blades; general term Δt n,i This represents the time difference between the actual arrival time of the i-th blade at the sensor and the theoretical value. Through time difference Δt n,i The vibration displacement y of the i-th blade at the time of arrival after the key n is obtained. n,i ,

[0027] The present invention has at least the following beneficial technical effects:

[0028] Regarding blade tip timing methods, compared to existing conventional methods such as the keyless method, single-axis key method, and multi-axis key method, this invention proposes a blade tip timing arrival time calculation method based on the multi-adjacent axis key method, which can effectively improve measurement accuracy. Currently, the main factor affecting blade vibration measurement is the rotational speed fluctuation of the blade shaft. The keyless method lacks a reference axis key, and its measurement results are greatly affected by the rotational speed fluctuation, making it unsuitable for blade tip timing measurement. The single-axis key method, due to its limited number of axes keys, experiences measurement errors caused by rotational speed fluctuations that accumulate with the increase in the distance between the blade and the axis key, increasing almost linearly. While the multi-axis key method has a larger number of axes keys, the vibration measurement of a single blade still references a single axis key, thus still exhibiting a relatively linear increase in measurement error locally. The proposed adjacent multi-axis key method ensures that each blade has two reference axes keys, thereby significantly reducing the accumulation of local rotational speed fluctuation errors and improving the accuracy of blade tip timing measurement.

[0029] The relative measurement error of the keyless method reaches a maximum of about 130%, which is obviously no longer suitable for measuring blade vibration. The relative measurement error of the single-axis key method increases almost linearly with the increase of the blade's distance from the key, exceeding 40% at its maximum, which is no longer suitable for blades far from the key. When calculating the vibration displacement of blades far from the first mark, the multi-axis key method reduces the cumulative error by shortening the time difference term due to multiple keys as references, with a maximum vibration displacement error of about 15%. The adjacent multi-axis key method further reduces the maximum relative measurement error to about 7%, effectively improving the measurement accuracy.

[0030] Furthermore, in response to the impact of background noise on measurement accuracy, the second step of the blade tip timing arrival time calculation method based on the multi-adjacent axis key method proposed in this invention, namely, using time-frequency noise reduction to reduce noise in the measurement data, can effectively reduce the impact of noise and improve the accuracy of vibration measurement. Attached Figure Description

[0031] Figure 1 A geometric diagram showing the sensor layout.

[0032] Figure 2 This is a schematic diagram of the noise reduction process and a schematic diagram of the leaf tip timing signal partitioning method. Figure 2 (a) is the time-domain curve of the initial data with noise. Figure 2 (b) shows the frequency domain curve of the initial data with noise. Figure 2 (c) shows the frequency domain curve after noise reduction. Figure 2 (d) is the time-domain curve after noise reduction.

[0033] Figure 3 This is a schematic diagram illustrating the implementation of the multi-adjacent shaft key method under variable speed operating conditions.

[0034] Figure 4 This is a flowchart of a method for calculating the tip timing arrival time based on the multi-adjacent axis key method according to the present invention.

[0035] Explanation of reference numerals in the attached figures:

[0036] 1 is the blade position sensor; 2 is the shaft key position sensor; 3 is the shaft key. Detailed Implementation

[0037] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the disclosure to those skilled in the art. It should be noted that, unless otherwise specified, the embodiments and features described herein can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0038] like Figure 4 As shown, the present invention provides a method for calculating the tip timing arrival time based on the multi-adjacent-axis key method, which includes the following steps:

[0039] The first step involves designing and arranging blade position sensor devices, keyway position sensor devices, and keyway structures for the turbine's blades and shaft system. The blade position sensors and keyway position sensors obtain data on the arrival time of each blade tip and the arrival time of each keyway, respectively. A geometric schematic diagram of the sensor arrangement in this first step is shown below. Figure 1 As shown. Figure 1 In the diagram, 1 is the blade position sensor; 2 is the shaft key position sensor; and 3 is the shaft key. The specific process of the first step is as follows: (1) Design and arrange the shaft keys according to the number of blades and shaft structure of the object to be measured. Assuming there are N blades in the whole circle and n shaft keys are evenly installed on the rotor, the arrangement of the shaft keys should meet the following conditions:

[0040] 1. In order to ensure the measurement accuracy of the blades and reduce the influence of rotor eccentric vibration, the shaft keys should be arranged uniformly, that is, the central angles between the shaft keys should be equal.

[0041] 2. The ratio N / n of the number of blades to the number of shaft keys should be an integer or the closest integer, and the number of shaft keys arranged should be greater than or equal to 4 to ensure that there are adjacent multi-shaft keys.

[0042] 3. The cross-sectional shape of the shaft key is generally rectangular, but it can also be a regular trapezoid, an inverted trapezoid, an irregular quadrilateral, etc.

[0043] (2) Install blade position sensors and keyway position sensors. First, install the blade position sensors, which are installed at a small distance from the blade tip. Install a total of K blade position sensors, numbered 1-K, as blade tip timing sensors. Next, install the keyway position sensors, which are installed at a small distance from the keyway. Install a total of L keyway position sensors, numbered 1-L, as keyway timing sensors. Eddy current sensors, etc., can be used for both blade and keyway position sensors.

[0044] (3) Debug the blade position sensor and the shaft key position sensor. After the sensor is installed, test the blade tip timing signal measurement and the shaft key timing signal measurement respectively. Connect the sensor to the signal acquisition device and measure the sensor data acquisition effect. If the signal is poor or weak, adjust and rearrange the sensor position until the signal is clear.

[0045] (4) Measure the blade tip timing signal and shaft key timing signal data. After the blade position sensor and shaft key position sensor are installed and properly adjusted, measure the signals from the blade position sensor and shaft key position sensor. A total of K sets of blade tip position sensor signal data are obtained, namely {W1, W2, W3... W... K A total of L sets of shaft key position sensor signal data were obtained, namely {X1, X2, X3...X...}. L}

[0046] The second step is to perform noise reduction on the data acquired by the sensor. (Refer to...) Figure 2 As shown. Among them, Figure 2 (a) is the time-domain curve of the initial data with noise. Figure 2 (b) shows the frequency domain curve of the initial data with noise. Figure 2 (c) shows the frequency domain curve after noise reduction. Figure 2 (d) shows the time-domain curve after noise reduction. This invention uses the Fourier transform noise reduction method to reduce the noise of the signal obtained from the sensor. Taking the first set of key position signals X1 as an example, the detailed steps are as follows.

[0047] (1) Transform the signal data from the time domain to the frequency domain, that is:

[0048]

[0049] After this operation is performed, the signal data will be... Figure 2 (a) time-domain data converted to Figure 2 (b) Frequency domain data.

[0050] (2) According to the frequency domain signal X1 T Based on the characteristics of the signal, noise reduction processing is performed to remove noise frequencies from the signal, leaving only the peak frequency signal. After this operation, the signal data will be... Figure 2 (b) frequency domain curve changed Figure 2 (c) The frequency domain curve after noise reduction.

[0051] (3) Convert the processed frequency domain signal into a time domain signal, that is:

[0052]

[0053] After this operation, the signal data will be further processed by... Figure 2 (c) The frequency domain curve after noise reduction becomes Figure 2 (d) is the time-domain curve after noise reduction.

[0054] (4) Perform noise reduction processing steps (1) to (3) sequentially on the K group of blade tip position sensor signal data and the L group of shaft key position sensor signal data to obtain the processed signals {W1′, W2′, W3′……W K {X1′} and {X2′, X3′……X} L ′}, for use in the next step.

[0055] The third step is to use the multiple adjacent axis key method to process the sensor data after noise reduction to obtain the time difference of the blade tip arrival, and then obtain the vibration displacement of the blade. The detailed operation steps are as follows.

[0056] (1) The noise-reduced signal is processed using the multiple nearest neighbor key method. Taking the first set of noise-reduced blade tip arrival time signal W1′ and key arrival time signal X1′ as an example, the key arrival time signal X1′ will contain multiple key timing times {t1, t2, t3……t}. n The leaf tip arrival time signal W1′ will contain multiple leaf tip timing moments.

[0057] (2) Assumptions are made regarding the blade root reference positions. Since the shaft keys are evenly distributed, there will be a fixed number of blades between every two adjacent shaft keys, that is, there will be a fixed number of blade roots between the shaft keys. Assuming that there are m blades between every two adjacent shaft keys, based on the phase distance between the blade roots and the shaft keys, m blade root reference position points are inserted between adjacent shaft keys to ensure that in the entire circle of blades, each blade (blade root) has a relative position reference point between two shaft keys.

[0058] (3) Define the arrival times of adjacent axis keys, leaf root positions and leaf tip positions. Figure 3 An implementation example of the multi-adjacent-shaft key method under variable speed conditions is shown. (Refer to...) Figure 3 As shown, three timing points are selected from the key arrival time signal X1′, and t is used as the time interval. n-1 t n With t n+1 These represent the timing times of three adjacent keys (key n-1, key n, and key n+1) as measured by the key position sensor.

[0059] With t n With t n+1 Taking the timing of two adjacent shaft keys as an example, m signals are obtained from the tip arrival time signal W1′, and then... This indicates that there are m blades between two adjacent axial keys; i ∈ [1, m]. General term This indicates the moment detected by the tip timing sensor when the i-th blade, following the key n, passes the sensor probe.

[0060] In t n With t n+1 Between the arrival times of the two shaft keys, based on the phase and rotational speed distributions, insert m blade root reference points, using {t' n,1 ,t' n,2 ...t' n,i ...t' n,m} represents, where: m indicates that there are m leaf root reference points between two adjacent axial keys; i∈[1,m]. General term t' n,i This indicates the time of the reference point at the root of the i-th blade inserted after the key n according to the number of blades.

[0061] It is worth noting that, because Figure 3 The case in the example is a variable speed operation, so the interval between the m blade root reference positions is not uniform. If it is changed to a uniform speed operation, the interval between the blade root reference positions will be uniform.

[0062] (4) Calculate the time difference and vibration displacement of the blade arrival times by using the arrival times of adjacent shaft keys, blade root positions, and blade tip positions. Taking m blades between the nth and (n+1)th shaft keys as an example, calculate the blade arrival times... Arrival time of leaf root reference position {t' n,1 ,t' n,2 ...t' n,i ...t' n,m The difference between} is {Δt} n,1 ,Δt n,2 ……Δt n,i ……Δt n,m Then, the arrival time difference of these m blades can be determined. General term Δt n,i This represents the time difference between the actual arrival time of the i-th blade at the sensor and the theoretical value.

[0063]

[0064] Where: m indicates that there are m blades between two adjacent axes; i∈[1,m].

[0065] Furthermore, through the time difference Δt n,i We can obtain the vibration displacement y at the moment when the i-th blade arrives after the n-th shaft key. n,i The formula is:

[0066]

[0067] In the formula: V—the velocity of the blade tip i between the theoretical and actual moments when it passes through the sensor / mm·s -1V n ' ,i —Theoretical velocity of blade tip i passing through the sensor after the nth key / mm·s -1 ;y err —Systematic error.

[0068] By taking values ​​from 1 to m for i, the vibration displacements of the m blades between key n and key n+1 can be obtained, denoted by {y}. n,1 ,y n,2 ...y n,m}express.

[0069] In the third step, the systematic error y of the blade tip timing system in equation (4) is addressed. d An analysis was conducted. It should be noted that between the actual and theoretical time when the blade tip passes the sensor, the rotating blade may still be operating under a variable speed condition, but due to Δt... n,i The value is very small, and the change in velocity is very insignificant, affecting the velocity V′. n,i The effect is almost negligible, so to simplify the calculation, the blade can be considered to be moving at a constant speed during this time. Therefore, the systematic error y d The impact is negligible.

[0070] (5) Repeat steps (1) to (4) above for all shaft keys and blades to obtain the vibration displacement of all blades at the corresponding arrival time of the whole circle.

[0071] Although the present invention has been described in detail above with general descriptions and specific embodiments, modifications or improvements can be made to it, which will be obvious to those skilled in the art. Therefore, all such modifications or improvements made without departing from the spirit of the present invention fall within the scope of protection claimed by the present invention.

Claims

1. A method for calculating the tip timing arrival time based on the multi-adjacent-axis key method, characterized in that, Includes the following steps: Step 1: Design and arrange blade position sensor devices, keyway position sensor devices, and keyway structures for the blades and shaft system of the turbine machinery. Measure the arrival signals of the blade position sensor and keyway position sensor respectively; obtain a total of K sets of blade tip position sensor signal data, i.e. { , , ... A total of L sets of shaft key position sensor signal data were obtained, namely { , , ... The key arrangement is uniform, meaning the central angles between keys are all equal; number of blades. N Number of shaft keys n ratio N / n Take the integer or the closest integer; the number of axial keys arranged is greater than or equal to 4 to ensure that there are adjacent multi-axial keys; Step 2: Use Fourier transform noise reduction to denoise each set of data obtained by the sensor; for the first set of key position signals... The specific steps are as follows: (201) Transform the signal data from the time domain signal to the frequency domain signal, that is: ; (202) According to the frequency domain signal Based on the characteristics of the signal, noise reduction processing is performed to remove the signal with noise frequency and retain only the signal with peak frequency. (203) Convert the processed frequency domain signal into a time domain signal, i.e. ; (204) Perform noise reduction processing steps (201) to (203) sequentially on the K group blade tip position sensor signal data and the L group shaft key position sensor signal data to obtain the processed signal { , , ... }and{ , , ... }; Step 3: (301) Process the noise-reduced signal using the multi-nearest-axis key method; (302) Assume the reference position of the leaf root, assuming that the number of blades between every two adjacent axes is m If one, then based on the phase distance between the leaf root and the shaft key, insert between adjacent shaft keys. m Each blade root has a reference position point to ensure that every blade or blade root in the entire circumference has a relative position reference point between two shaft keys, which is useful under variable speed conditions. m The intervals between the aforementioned leaf root reference points are uneven, under uniform speed conditions. m The intervals between the leaf root reference positions are uniform; (303) Define the arrival times of adjacent key, blade root position, and blade tip position, starting from the key arrival time signal. Select three time points and use , and It means that, with and When two adjacent shaft keys are timed, the arrival time signal from the blade tip is... Obtain m A signal, using { , ... ... } indicates that: m This indicates that there exists between two adjacent axial keys. m One leaf; ,exist and Based on the phase and rotational speed distributions, insert the two key shafts at their arrival times. m A leaf root reference point, using { , ... ... } indicates that: m This indicates that there exists between two adjacent axial keys. m One leaf root reference point; ; (304) Calculate the time difference and vibration displacement of the blade arrival time using the arrival times of adjacent shaft keys, blade root positions, and blade tip positions. For the first... n The first key and the first n +1 axis key m There are several blades, and their arrival times are calculated. , ... ... } and leaf root reference position arrival time { , ... ... The difference between} , ... ... }, then determine this m The time difference of arrival time of each blade; general term Indicates the first i The time difference between the actual arrival time of each blade at the sensor and the theoretical value; = Through time difference Obtain the shaft key n The next i Vibration displacement at the arrival time of each blade , .

2. The method for calculating the tip timing arrival time based on the multi-adjacent-axis key method according to claim 1, characterized in that, In the first step, the cross-sectional shape of the shaft key can be rectangular, trapezoidal, inverted trapezoidal, or irregular quadrilateral.

Images