A fuel rod vibration contact identification method based on dual-threshold frequency domain analysis

Abstract

The application discloses a fuel rod vibration contact identification method based on double-threshold frequency domain analysis and relates to the technical field of nuclear reactor flow-induced vibration. In view of the problems of the traditional contact identification method, such as difficult installation, weak anti-interference and easy misjudgment, the method is characterized in that an acceleration sensor is arranged in the axial direction of a fuel rod to collect signals, and after pretreatment, the signals are combined with double-threshold time domain determination, frequency domain power spectrum density verification and multi-point time difference positioning to realize accurate identification and position determination of a contact event. The method does not require additional sensors, reduces hardware dependence and arrangement complexity, can adjust parameters according to different working conditions, is suitable for various test scenes, is not only suitable for nuclear reactor fuel assembly vibration test, but also can be popularized to flow-induced vibration contact analysis of long and thin structures such as ocean engineering and chemical equipment, and has reliability and versatility.

Description

Technical Field

[0001] This invention relates to the field of nuclear reactor flow-induced vibration technology, and in particular to a fuel rod vibration contact identification method based on dual-threshold frequency domain analysis. Background Technology

[0002] Flow-induced vibration testing is an important tool for studying the dynamic response characteristics of nuclear fuel assemblies under coolant flow conditions. Its purpose is to simulate the vibration behavior and nonlinear contact mechanisms of fuel rods under reactor operating conditions. Under the scouring action of high-velocity coolant, fuel rods with extremely high aspect ratios will exhibit vibration behavior. When instantaneous contact collisions occur between fuel rods or between fuel rods and the supporting grid, sudden dynamic impact loads and abrupt changes in local acceleration will be induced in the structure. These dynamic contact behaviors are highly random, extremely short-lived (typically on the order of milliseconds), and accompanied by nonlinear characteristics and modal coupling effects, making it very difficult to directly measure the contact using conventional sensing methods.

[0003] In vibration testing, traditional contact behavior identification relies heavily on mechanical contact sensors or acoustic emission sensors. However, these methods are difficult to install within fuel rod bundle channels under high-temperature, high-flow-rate conditions and are susceptible to structural vibrations and flow field noise, resulting in poor signal stability and low recognition rates. Practical tests show that directly measuring contact force or contact location is often impractical. Therefore, indirect methods that analyze structural response signals have become mainstream. Among these, using accelerometers to acquire the acceleration time-history signals of fuel rods and identifying contact events through signal processing has become an important technical approach in flow-induced vibration testing due to its non-invasiveness and ease of installation. However, traditional methods relying solely on acceleration peak values ​​or single time-domain thresholds are prone to misjudgments or omissions under strong background noise conditions. Therefore, in the field of flow-induced vibration testing technology, there is an urgent need for a contact behavior identification method that combines dual-threshold time-domain determination of acceleration time-history signals with frequency-domain power spectrum analysis to stably and accurately identify fuel rod contact collision behavior under complex operating conditions and strong noise backgrounds. Summary of the Invention

[0004] The purpose of this invention is to provide a fuel rod vibration contact identification method based on dual-threshold frequency domain analysis. Addressing the lack of current methods for identifying contact collision behavior in fuel rod flow-induced vibration tests, this invention can quantitatively identify and calculate the occurrence and end times, duration, peak acceleration, velocity change, and contact location of contact events. This reduces testing and analysis costs, improves identification efficiency and result reliability, and provides technical support for the study of fuel rod flow-induced vibration mechanisms, safety assessments, and structural design optimization.

[0005] To achieve the above objectives, this invention provides a fuel rod vibration contact identification method based on dual-threshold frequency domain analysis, comprising the following steps: S1. Sensor arrangement and wave velocity calibration: At least two acceleration sensors are arranged along the axial direction of the fuel rod. The axial coordinates of each sensor are recorded. The structural vibration wave velocity is obtained by impact test and cross-correlation calculation. S2. Acceleration signal acquisition: Under steady-state conditions, acquire the transient acceleration time history signals from each sensor. a i ( t Single-segment acquisition duration ≥ 60s; S3. Signal preprocessing: The original acceleration signal is sequentially processed by DC offset elimination, bandpass filtering, and notch filtering of power frequency and its harmonics signals. S4. Dual threshold setting: Based on the statistics of vibration signals in the collision-free zone, calculate 3 times the standard deviation as the dynamic baseline threshold, and then set the collision threshold through the sensitivity coefficient. S5. Initial identification of time-domain events: By comparing the signal amplitude with the dynamic baseline threshold and the collision threshold, the start point, peak time, peak amplitude and end point of the contact event are determined. S6. Amplitude-duration joint criterion screening: Based on the duration of the contact event and the quantification conditions of the peak amplitude and the dynamic baseline threshold, real contact impact signals are screened. S7. Frequency domain power spectral density verification: Select the contact event time window and the adjacent vibration time window, calculate the power spectral density of the two windows, and determine the authenticity of the event by the high-frequency energy ratio; S8. Contact Position Location: For collision events verified by S7, the axial position of the collision event is calculated based on the cross-correlation time delay of multi-sensor signals, combined with sensor coordinates and vibration wave velocity.

[0006] Preferably, in S1, the sensor arrangement is based on the lower end of the fuel rod as the zero point. The impact test is used to verify the synchronicity of the response channel and the clarity of the peak shape. The vibration wave velocity is determined by the formula... Calculation, where The distance between the two sensors, This is the time difference between the arrival of the tap response.

[0007] Preferably, S3 specifically includes: S3.1 Calculate the mean value of the signal for each channel, and subtract the mean value from the original signal to eliminate DC offset; S3.2. Combining the modal characteristic frequencies of the fuel rod structure and the frequency response range of the sensor, a bandpass filter is used to limit the signal frequency range; S3.3. Use a narrowband notch filter to filter out 50Hz, 100Hz, 150Hz power frequency signals and their harmonics.

[0008] Preferably, in S4, the collision-free interval refers to the steady-state vibration period during which no contact collision occurs, which is determined by collision-free working condition data or signal characteristics in the early stage of the test; The formula for calculating the dynamic baseline threshold is as follows:

[0009] Where σ bg The standard deviation of the vibration signal in the collision-free zone; The formula for calculating the collision threshold is as follows:

[0010] The typical value of the sensitivity coefficient k is 3.

[0011] Preferably, S5 is executed according to the following sub-steps: S5.1, using the pre-processed acceleration signal For the analysis object, when the signal amplitude first exceeds the collision threshold When, mark as the event start point. :

[0012] in, This is the preprocessed acceleration signal; S5.2, in to During the duration of the collision, the peak moment of the collision signal is recorded. and amplitude The expression is:

[0013]

[0014] S5.3 When the signal amplitude falls back to the dynamic baseline threshold and When the event ends, mark it as the event end point. The expression is:

[0015] Avoid misinterpreting normal vibrations as a continuation of a collision.

[0016] Preferably, in S6, the joint criterion is specifically: (1) Duration criterion: Calculate the duration of the collision impact. ,like If the event is determined to be a normal amplitude fluctuation of vibration rather than a collision event, the event is removed. (2) Amplitude ratio criterion: Define peak amplitude ratio ,like If the amplitude fluctuation is determined to be normal, the event is excluded. Only when and If so, retain the event as a candidate collision event.

[0017] Preferably, S7 is executed according to the following sub-steps: S7.1 Select the collision event time window and adjacent collision-free vibration time windows , Typical values ​​are 50-100ms, ensuring the vibration window. It does not contain any collision behavior; S7.2 Calculate the collision event time window using the Welch method. and collision-free vibration time window The power spectral density is used to divide the time window into overlapping segments, and the Hanning window function is applied. w ( n ) Window the spectrum, then perform an FFT and average the results to obtain the PSD; set the frequency resolution and window length to ensure the stability and resolution of the spectrum estimation; S7.3. Divide the spectrum into modal frequency bands based on the structural characteristics of the fuel rods. and contact collision frequency band Modal frequency band Used to cover the main natural frequencies of the fuel rods, contact collision frequency band. Used to cover the high-frequency energy generated by collision impact; S7.4 Calculate the frequency domain energy integral within the modal band and the contact / collision band, expressed as follows:

[0018]

[0019] S7.5: Calculate the high-frequency energy ratio of collision events R , represented as:

[0020] S7.6: Set the high-frequency energy ratio threshold ,like If a collision is identified as a genuine collision, it is retained for localization; otherwise, it is identified as a non-collision event and discarded.

[0021] Preferably, S8 is executed according to the following sub-steps: S8.1 For the time window of a real collision event, take the signal from each pair of adjacent sensors. and Calculate the cross-correlation function:

[0022] in, For time delay; S8.2 Find the time delay corresponding to the maximum cross-correlation value:

[0023] S8.3, Based on the axial coordinates of adjacent sensors , The collision location is calculated using the time difference between the arrival of the vibration wave velocity c specified in S1.

[0024] S8.4 If more than two acceleration sensors are arranged on the fuel rod, the arithmetic mean of the positioning results calculated by multiple pairs of adjacent sensors shall be taken as the axial position of the final collision event.

[0025] According to specific embodiments provided by the present invention, the present invention discloses the following technical effects: 1. This method only requires the acquisition of conventional acceleration sensor signals and is applicable to most existing flow-induced vibration testing and data acquisition systems. It does not require the addition of additional contact sensors, which significantly reduces hardware dependence and layout complexity.

[0026] 2. This method uses a dual threshold determination method (amplitude threshold + shortest duration threshold) to quickly capture contact events, and combines power spectral density feature verification to achieve an identification accuracy of over 95%, effectively reducing false positives and false negatives caused by background noise.

[0027] 3. This method uses the time difference of arrival method combined with the vibration wave transmission effect to locate the contact event. The spatial accuracy can be controlled within ±10mm, providing spatial distribution information support for experimental research.

[0028] 4. This method can adjust the threshold and positioning parameters according to different operating states, boundary conditions and fuel rod constraints to achieve reliable identification under various test conditions and maintain the consistency and comparability of results.

[0029] 5. The fuel rod contact behavior identification method proposed in this invention is not only applicable to the vibration test of fuel assemblies in nuclear reactors such as pressurized water reactors and lead-bismuth fast reactors, but can also be extended to other fields such as marine engineering and chemical equipment flow-induced vibration contact analysis, and has good versatility and promotion value.

[0030] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0031] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0032] Figure 1 This is a schematic flowchart of an embodiment of a fuel rod vibration contact identification method based on dual-threshold frequency domain analysis according to the present invention. Figure 2 shows the signal diagrams of two acceleration sensors according to an embodiment of the present invention; (a) is the waveform of the original acceleration time history signal collected by sensor 1; (b) is the waveform of the original acceleration time history signal collected by sensor 2; Figure 3 This is a schematic diagram of the sensor arrangement and positioning results according to an embodiment of the present invention; Figure 4 is a comprehensive diagram of detection statistics and feature distribution in an embodiment of the present invention; (a) is a preliminary confirmation diagram of dual acceleration sensor signals and collision signals; (b) is a diagram of the number of collisions counted by different detection processes; (c) is a diagram of the verification results of frequency domain power spectral density; (d) is a scatter plot of collision event feature distribution; (e) is a diagram of the probability distribution of each collision position. Figure 5 shows a multi-dimensional feature analysis diagram of a typical collision event #1 according to an embodiment of the present invention; (a) is an acceleration time history curve; (b) is a power spectral density curve; (c) is a cross-correlation function curve. Figure 6 shows a multi-dimensional feature analysis diagram of a typical collision event #2 according to an embodiment of the present invention; (a) is an acceleration time history curve; (b) is a power spectral density curve; (c) is a cross-correlation function curve. Figure 7 is a multi-dimensional feature analysis diagram of a typical collision event #3 according to an embodiment of the present invention. (a) is an acceleration time history curve; (b) is a power spectral density curve; and (c) is a cross-correlation function curve. Detailed Implementation

[0033] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0034] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0035] Example Take the vibration contact identification process of a fuel rod as an example.

[0036] I. Test Subjects and Test Parameters The total length of the fuel rod is 1500mm, with the lower end of the fuel rod as the zero point of the axial coordinate. For example... Figure 3 As shown in Figure 2, two accelerometers are positioned at 375mm (sensor 1) and 1125mm (sensor 2) along the fuel rod axis to record the axial coordinates of the sensors. The distance between the two sensors is Δx = 1125mm - 375mm = 750mm. The sampling frequency is set to 1000Hz, and the acquisition time for a single acceleration signal is 60s, meeting the requirement of a signal acquisition time ≥ 60s under steady-state conditions. As shown in Figure 2, the original acceleration signal exhibits significant sudden impact characteristics (including positive and negative impacts), and there is a fixed time difference between the impact signals from sensor 1 and sensor 2.

[0037] II. Key Parameter Settings 1. Vibration wave velocity calibration: Verify the synchronicity of the response channel and the clarity of the peak shape through a tapping test (e.g., Figure 3 As shown), the time difference between the arrival of the impact response at the two sensors is measured. According to the formula Calculate the structural vibration wave velocity and obtain The calibration has been completed during the experiment and will be used for subsequent positioning calculations.

[0038] 2. Dual threshold settings: Dynamic baseline threshold: The standard deviation of the vibration signal in the collision-free zone is calculated based on the statistics of the vibration signal in the collision-free zone. According to the formula Calculation, final .

[0039] Collision threshold: Take the sensitivity coefficient k=3, and use the formula... Calculation yields .

[0040] 3. Frequency domain analysis parameter settings: Time window setting: Collision event time window Time window of collision-free vibration with neighboring All values ​​are set to 100ms to ensure that the collision-free window does not contain any collision behavior.

[0041] Characteristic frequency band division: setting contact and collision frequency bands Used to cover the high-frequency energy generated by collisions and impacts; modal frequency band Covers the main natural frequency of the fuel rod.

[0042] High-frequency energy ratio threshold After experimental verification and setting an appropriate threshold, the high-frequency energy ratio of the collision event was finally verified. .

[0043] 4. Positioning calculation parameters: The TDOA (Time Difference of Arrival) calculation window is set to 0.1s to extract the cross-correlation time delay of sensor signals.

[0044] III. Experimental Procedure and Results (a) Signal processing flow, such as Figure 1 As shown.

[0045] 1. Signal Preprocessing: The acquired acceleration signals were sequentially processed by DC offset cancellation (subtracting the signal mean), bandpass filtering (matching the fuel rod modal characteristic frequency with the sensor frequency response range), and notch filtering (filtering out 50Hz, 100Hz, 150Hz power frequency signals and their harmonics). After processing, the impact characteristics (including positive and negative) in Figures 5, 6, and 7 are more prominent, and noise interference is significantly reduced.

[0046] 2. Initial identification of time-domain events: By comparing the preprocessed signal with... , In contrast, mark the starting point of the contact event. Peak time and peak amplitude End point .

[0047] As shown in Figure 5, events #1: S1 = 2.37g, S2 = 1.57g, both exceeding... ,mark , , ; As shown in Figure 6, event #2: S1=1.55g, S2=2.86g, both exceeding... ,mark , , ; As shown in Figure 7, in event #3: the absolute value of the amplitude of S1 is 1.22g and the absolute value of the amplitude of S2 is 1.51g, both exceeding... ,mark , , .

[0048] As shown in Figure 4(a), a total of 12 suspected events were initially identified, all of which met the judgment rules.

[0049] 3. Joint criterion screening: Candidate collision events are screened based on a joint criterion of amplitude and duration. Duration criterion: Calculate the duration of the collision impact. Three typical events All values ​​are 5ms ≤ 10ms, which meets the requirements. Amplitude ratio criterion: Define peak amplitude ratio , Regarding event #1, ; Regarding event #2, ; Regarding event #3, ; Ultimately, 6 conditions were satisfied. and The candidate events (including the above 3) are selected, and the results are shown in Figure 4(b).

[0050] 4. Frequency Domain Power Spectral Density Verification: The Welch method was used to calculate the power spectral density of the candidate event window and the collision-free window, as shown in Figure 5(b), Figure 6(b), and Figure 7(c). Stable spectrum estimates were obtained through Hanning windowing, FFT transformation, and averaging. The R-values ​​for event #1 were 3.5, for event #2 4.2, and for event #3 3, respectively. It was determined to be a real collision.

[0051] All 6 candidate events passed the verification, with a pass rate of 100%, as shown in Figure 4(c).

[0052] 5. Contact position positioning Calculate the cross-correlation function of adjacent sensor signals Obtain the time delay corresponding to the maximum cross-correlation value. The τ for event #1 is 0.23ms, the τ for event #2 is 0.26ms, and the τ for event #3 is 0.24ms. According to the formula Calculate the axial position of the collision, where , These are the axial coordinates of sensors 1 and 2, respectively; for event #1, x = (0.23 × 3000 + 375 + 1125) / 2 = 68 mm; for event #2, x = 72 mm; for event #3, x = 69 mm. The location results of the six events were concentrated in the 65-75mm range, with an average of 0mm and a positioning accuracy of ±5mm, as shown in Figure 4(e).

[0053] (II) Test Results 1. Verification of adaptability to multiple scenarios: The positive impacts in Figures 5 and 6 and the negative impacts in Figure 7 were accurately identified, and the feature parameters all met the criteria requirements, proving that the method is not affected by the impact direction and has strong adaptability.

[0054] 2. Collision position distribution: Through positioning calculation, the axial positions of the six collision events are all concentrated around 70mm, and the spatial positioning accuracy is controlled within ±10mm, which meets the test accuracy requirements.

[0055] 3. Key Feature Validation: The duration of all real collision events was ≤10ms, and the peak amplitude was ≤10ms. High frequency energy ratio This meets the criteria requirements; Power spectral density analysis showed that the collision event had significant high-frequency energy concentration in the 200-500Hz frequency band, which was clearly distinguishable from the non-collision vibration signal. The cross-correlation analysis of sensor signals accurately captures the time delay of the collision wave reaching the two sensors, providing reliable data support for positioning calculations.

[0056] In summary, this embodiment successfully achieved accurate identification and localization of fuel rod vibration contact events through the above parameter settings and process execution, verifying the effectiveness of the dual-threshold frequency domain analysis method: no additional contact sensors are required, and detection can be completed using only conventional accelerometers, reducing the complexity of hardware layout; the combination of time-domain dual-threshold judgment and frequency-domain power spectral density verification effectively avoids misjudgment and missed judgment caused by background noise; the positioning accuracy reaches within ±10mm, providing accurate spatial distribution information for the study of fuel rod flow-induced vibration mechanism; the experimental parameters can be flexibly adjusted according to different working conditions, demonstrating good adaptability and versatility.

[0057] The remaining technical features in the above embodiments can be flexibly selected by those skilled in the art to meet different specific practical needs according to actual circumstances. Modifications and variations made by those skilled in the art that do not depart from the spirit and scope of the present invention should be within the protection scope of the appended claims. In the above description, numerous specific details have been set forth to provide a thorough understanding of the present invention. However, it will be apparent to those skilled in the art that these specific details are not necessary to implement the present invention. In other instances, to avoid obscuring the present invention, well-known techniques, such as specific construction details, operating conditions, and other technical conditions, have not been specifically described.

[0058] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. Furthermore, those skilled in the art will recognize that, based on the ideas of the present invention, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A method for identifying the vibration contact of fuel rods based on dual-threshold frequency domain analysis, characterized in that, Includes the following steps: S1. Sensor arrangement and wave velocity calibration: At least two acceleration sensors are arranged along the axial direction of the fuel rod. The axial coordinates of each sensor are recorded. The structural vibration wave velocity is obtained by impact test and cross-correlation calculation. S2. Acceleration signal acquisition: Under steady-state conditions, acquire the transient acceleration time history signals from each sensor. a i ( t Single-segment acquisition duration ≥ 60s; S3. Signal preprocessing: The original acceleration signal is sequentially processed by DC offset elimination, bandpass filtering, and notch filtering of power frequency and its harmonics signals. S4. Dual threshold setting: Based on the statistics of vibration signals in the collision-free zone, calculate 3 times the standard deviation as the dynamic baseline threshold, and then set the collision threshold through the sensitivity coefficient. S5. Initial identification of time-domain events: By comparing the signal amplitude with the dynamic baseline threshold and the collision threshold, the start point, peak time, peak amplitude and end point of the contact event are determined. S6. Amplitude-duration joint criterion screening: Based on the duration of the contact event and the quantification conditions of the peak amplitude and the dynamic baseline threshold, real contact impact signals are screened. S7. Frequency domain power spectral density verification: Select the contact event time window and the adjacent vibration time window, calculate the power spectral density of the two windows, and determine the authenticity of the event by the high-frequency energy ratio; S8. Contact Position Location: For collision events verified by S7, the axial position of the collision event is calculated based on the cross-correlation time delay of multi-sensor signals, combined with sensor coordinates and vibration wave velocity.

2. The fuel rod vibration contact identification method based on dual-threshold frequency domain analysis according to claim 1, characterized in that: In S1, the sensor arrangement is based on the lower end of the fuel rod as the zero point. The impact test is used to verify the synchronization of the response channels and the clarity of the peak shape. The vibration wave velocity is calculated using the formula... Calculation, where The distance between the two sensors, This is the time difference between the arrival of the tap response.

3. The fuel rod vibration contact identification method based on dual-threshold frequency domain analysis according to claim 2, characterized in that: S3 specifically includes: S3.1 Calculate the mean value of the signal for each channel, and subtract the mean value from the original signal to eliminate DC offset; S3.

2. Combining the modal characteristic frequencies of the fuel rod structure and the frequency response range of the sensor, a bandpass filter is used to limit the signal frequency range; S3.

3. Use a narrowband notch filter to filter out 50Hz, 100Hz, 150Hz power frequency signals and their harmonics.

4. The fuel rod vibration contact identification method based on dual-threshold frequency domain analysis according to claim 3, characterized in that: In S4, the collision-free interval refers to the steady-state vibration period during which no contact collision occurs, which is determined by collision-free working condition data or signal characteristics in the early stage of the test. The formula for calculating the dynamic baseline threshold is as follows: The formula for calculating the collision threshold is as follows: Where, σ bg The standard deviation of the vibration signal in the collision-free zone is given, and the typical value of the sensitivity coefficient k is 3.

5. The fuel rod vibration contact identification method based on dual-threshold frequency domain analysis according to claim 4, characterized in that: S5 is executed according to the following sub-steps: S5.1, using the pre-processed acceleration signal For the analysis object, when the signal amplitude first exceeds the collision threshold When, mark as the event start point. : in, This is the preprocessed acceleration signal; S5.2, in to During the duration of the collision, the peak moment of the collision signal is recorded. and amplitude The expression is: S5.3 When the signal amplitude falls back to the dynamic baseline threshold and When the event ends, mark it as the event end point. The expression is: Avoid misinterpreting normal vibrations as a continuation of a collision.

6. The fuel rod vibration contact identification method based on dual-threshold frequency domain analysis according to claim 5, characterized in that: In S6, the joint criterion is specifically as follows: (1) Duration criterion: Calculate the duration of the collision impact. ,like If the event is determined to be a normal amplitude fluctuation of vibration rather than a collision event, the event is removed. (2) Amplitude ratio criterion: Define peak amplitude ratio ,like If the amplitude fluctuation is determined to be normal, the event is excluded. Only when and If so, retain the event as a candidate collision event.

7. The fuel rod vibration contact identification method based on dual-threshold frequency domain analysis according to claim 6, characterized in that: S7 is executed according to the following sub-steps: S7.1 Select the collision event time window and adjacent collision-free vibration time windows , Typical values ​​are 50-100ms, ensuring the vibration window. It does not contain any collision behavior; S7.2 Calculate the collision event time window using the Welch method. and collision-free vibration time window The power spectral density is used to divide the time window into overlapping segments, and the Hanning window function is applied. w ( n ) Add a window, then perform an FFT and take the average to obtain the PSD; Setting the frequency resolution and window length ensures the stability and resolution of the spectrum estimation; S7.

3. Divide the spectrum into modal frequency bands based on the structural characteristics of the fuel rods. and contact collision frequency band Modal frequency band Used to cover the main natural frequencies of the fuel rods, contact collision frequency band. Used to cover the high-frequency energy generated by collision impact; S7.4 Calculate the frequency domain energy integral within the modal band and the contact / collision band, expressed as follows: S7.5: Calculate the high-frequency energy ratio of collision events R , is represented as: S7.6: Set the high-frequency energy ratio threshold ,like If a collision is identified as a genuine collision, it is retained for localization; otherwise, it is identified as a non-collision event and discarded.

8. The fuel rod vibration contact identification method based on dual-threshold frequency domain analysis according to claim 7, characterized in that: S8 is executed according to the following sub-steps: S8.1 For the time window of a real collision event, take the signal from each pair of adjacent sensors. and Calculate the cross-correlation function: in, For time delay; S8.2 Find the time delay corresponding to the maximum cross-correlation value: S8.3, Based on the axial coordinates of adjacent sensors , The collision location is calculated using the time difference between the arrival of the vibration wave velocity c specified in S1. S8.4 If more than two acceleration sensors are arranged on the fuel rod, the arithmetic mean of the positioning results calculated by multiple pairs of adjacent sensors shall be taken as the axial position of the final collision event.

Images