Cooling tower resonance detection method

Abstract

The invention provides a cooling tower resonance detection method, which comprises the following steps of S1, arranging a vibration sensor at a preset position of a cooling tower, and collecting a first vibration data set of a cooling tower body; s2, a vibration meter is installed on a fan transmission system, a fan is started, and the operation frequency of a motor is adjusted in a stepping mode; s3, synchronously collecting a second vibration data set of the fan at the corresponding frequency at the adjusting point of each frequency, and recording the second vibration data set as a sequence [b, b,..., bn]; and S4, analyzing the sequence of the first vibration data set and the sequence of the second vibration data set through spectrum comparison, and identifying the characteristic frequency of the fan causing the resonance of the cooling tower from [b, b,..., bn] according to the vibration frequency coincidence degree and the waveform similarity. According to the invention, the resonant frequency is accurately locked by synchronously collecting the response of the tower body structure and the vibration data of the fan excitation source and adopting the stepping frequency scanning and multi-dimensional frequency spectrum comparative analysis technology.

Description

Technical Field

[0001] The present invention relates to the technical field of industrial equipment vibration monitoring and fault diagnosis, and more specifically, to a method for detecting cooling tower resonance. The method is particularly suitable for identifying abnormal tower vibration problems caused by "rotational excitation of the fan drive system" in large mechanical ventilation cooling towers. Background Art

[0002] Cooling towers are critical heat dissipation equipment in industries such as power, chemical engineering, and metallurgy. Their long-term, stable operation is crucial to the energy efficiency of industrial systems. The fan drive system (including the motor, reducer, and blades), the core moving component of a cooling tower, generates periodic rotational excitation during operation. When the excitation frequency approaches the natural frequency of the cooling tower structure, it can induce "harmful structural resonance," leading to serious failures such as cracking of tower welds, collapse of filler materials, loosening of connectors, and even safety accidents.

[0003] At present, the detection of cooling tower resonance problems in the industry mainly has the following technical bottlenecks:

[0004] 1. Limitations of offline testing: Traditional methods rely on knock tests or modal analysis after shutdown to obtain the structural natural frequency. These methods cannot reflect the impact of dynamic loads and fluid coupling on frequency characteristics under actual operating conditions, resulting in a disconnect between test results and actual resonance risks.

[0005] 2. Broadband monitoring blind spots: Existing technologies often independently monitor wind turbine shaft vibration or tower vibration, lacking a mechanism for synchronously collecting and comparing data from both at the same frequency and phase. While conventional online vibration monitoring systems can capture the tower vibration spectrum, they can only identify the resonant peak frequency and cannot directly correlate it with the excitation source.

[0006] Furthermore, existing methods lack frequency scanning accuracy and data relevance. Specifically, wind turbine frequency regulation typically uses fixed speed or coarse step-by-step adjustments, which can easily miss narrowband resonance points. Vibration sensors are often deployed at single points on the tower, ignoring spatial vibration mode differences (the effect of X / Y / Z axial vibration coupling), resulting in incomplete resonance mode identification. Furthermore, spectrum comparison often relies on amplitude thresholds and fails to incorporate time-domain waveform envelope similarity criteria, making it susceptible to noise interference and misjudgment.

[0007] Therefore, in order to solve the problem of resonance between the cooling tower fan and the cooling tower at different operating frequencies, an accurate and efficient cooling tower resonance detection method is invented to solve the resonance phenomenon between the cooling tower fan and the cooling tower structure during the frequency conversion adjustment process. Summary of the Invention

[0008] The present application provides a method for detecting resonance of a cooling tower, so as to solve the problem in the prior art that resonance occurs between a cooling tower fan and the cooling tower structure during frequency conversion regulation.

[0009] The present application adopts the following technical solution: a method for detecting resonance of a cooling tower, comprising a vibration sensor for receiving vibration and a vibration meter electrically connected thereto, the vibration meter being used to perform spectrum analysis on the vibration received by the vibration sensor and output vibration data externally, comprising the following steps:

[0010] S1: placing a vibration sensor at a preset position of the cooling tower to collect a first vibration data set of the cooling tower body;

[0011] S2: Install the vibration meter on the fan drive system, start the fan and adjust the motor operating frequency in steps;

[0012] S3: At each frequency adjustment point, synchronously collect the second vibration data set of the fan at the corresponding frequency, recorded as a sequence [b1, b2, ..., bn];

[0013] S4: By comparing and analyzing the first vibration data set and the second vibration data set sequence through spectrum, the characteristic frequency of the fan that causes the cooling tower resonance is identified from [b1, b2, …, bn] according to the vibration frequency overlap and waveform similarity.

[0014] Preferably, the first vibration data set is the natural frequency of the cooling tower, and the second vibration data set is the excitation frequency of the fan; the first vibration data set and the second vibration data set [b1, b2, ..., bn] both include frequency domain amplitude spectra and time domain waveforms.

[0015] Preferably, the comparison result indicates that the difference between the excitation frequency of the second vibration data set and the natural frequency of the first vibration data set is within ±5% of the natural frequency, and a significant amplitude integral (much higher than the background vibration level) is observed on the spectrum graph. In this case, the current second vibration data bn is determined to be a characteristic frequency.

[0016] Preferably, before comparing the first vibration data set and the second vibration data set, the method further includes preprocessing the first vibration data set and the second vibration data set, wherein the preprocessing includes signal amplification, filtering, and digital-to-analog conversion.

[0017] Preferably, the frequency regulation in S2 adopts a step size of 0.1 Hz, continuously increasing the motor operating frequency within the range of 25 Hz-50 Hz, and collecting data after each frequency point runs stably for ≥30 seconds.

[0018] Preferably, the step size can be adaptively adjusted to 0.05 Hz-0.15 Hz.

[0019] Preferably, the preset position in S1 is the three-dimensional orthogonal directions of the X-axis, Y-axis and Z-axis of the cooling tower, and the first vibration data set includes independent spectrum analysis results in the three axes.

[0020] Preferably, the second vibration data set bn of each frequency point of the fan is obtained by taking the average value of three repeated measurements, and the duration of a single measurement is ≥1 min.

[0021] Preferably, the vibration sensor is a three-axis acceleration sensor, and the vibration meter is integrated with an FFT analysis module.

[0022] The present invention synchronously collects the vibration data of the tower structure response and the fan excitation source, adopts stepped frequency scanning and multi-dimensional spectrum comparison and analysis technology, and realizes the precise locking of the resonant frequency and the determination of the excitation source.

[0023] Traditional methods can only detect excessive tower vibration but cannot distinguish between resonance and ordinary forced vibration, leading to incorrect maintenance. This invention uses dual-channel synchronous acquisition and dual frequency and time domain judgment to eliminate irrelevant frequency interference such as bearing failures. Simultaneously, time domain envelope similarity verifies phase synchronization, avoiding false resonances such as fluid excitation.

[0024] This technology overcomes the bottleneck of narrowband weak resonance detection. Existing technologies, due to fixed step-size scanning and single-point sensor placement, result in a high rate of missed detection of narrowband weak resonances. This invention utilizes innovative adaptive step-size adjustment, dynamically switching between 0.05Hz and 0.15Hz. Furthermore, the vibration sensor utilizes a three-axis orthogonal sensor layout, enabling simultaneous capture of multimodal resonances in bending, torsion, and tension. This improves the reliability of resonance data. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings of the embodiments will be briefly introduced below. Obviously, the drawings described below only relate to some embodiments of the present application, and are not limitations to the present application.

[0026] Figure 1 It is a logic diagram of the cooling tower resonance detection method of the present invention. DETAILED DESCRIPTION

[0027] To make the purpose, technical solutions, and advantages of the embodiments of the present application more clear, the present application will be further described in detail below with reference to the accompanying drawings. Generally, the components of the embodiments of the present application described and shown in the drawings herein can be arranged and designed in various different configurations. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0028] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, it does not need to be further defined or explained in subsequent drawings.

[0029] Unless otherwise defined, technical or scientific terms used in this patent document shall have the ordinary meaning as understood by persons of ordinary skill in the art to which this application belongs. The terms "first," "second," and similar expressions used in this patent specification and claims do not denote any order, quantity, or importance, but are merely used to distinguish one component from another. Similarly, terms such as "a," "an," or "the" do not denote a limitation of quantity, but rather denote the presence of at least one. Terms such as "include" or "comprising" mean that the elements or objects preceding the phrase "include" or "comprising" include the elements or objects listed after the phrase and their equivalents, and do not exclude other elements or objects. Terms such as "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer" are used solely to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly. These terms are used solely to facilitate the description of this application and to simplify the description. They are not intended to indicate or imply that the device or component referred to must have a specific orientation, be constructed, or operate in a specific orientation, and are not to be construed as limitations on this application.

[0030] In the description of this application, it should be noted that, unless otherwise specified or limited, the terms "mounted," "connected," and "connected" should be understood broadly. For example, they may refer to fixed, removable, or integral connections; mechanical or electrical connections; direct or indirect connections through an intermediary; and internal communication between two components. Those skilled in the art will understand the specific meanings of these terms in this application based on their specific circumstances.

[0031] The following describes some embodiments of the present application in detail with reference to the accompanying drawings. In the absence of conflict, the features of the following embodiments can be combined with each other.

[0032] Core principle of the invention

[0033] To address the difficulty in tracing cooling tower resonance sources, this invention innovatively establishes a synchronous comparison mechanism between the fan excitation frequency and the tower response frequency. This mechanism generates a series of excitation frequencies by stepping the fan speed (step size ≤ 0.1 Hz). The tower vibration response and fan vibration data are simultaneously collected. The resonance point is then pinpointed using a dual criterion: frequency-domain amplitude spectrum coincidence and time-domain envelope waveform similarity (threshold > 85%). This overcomes the excitation-response separation limitations of traditional single-point monitoring and enables precise location of the resonance source.

[0034] Example 1: Basic resonance detection process

[0035] Step S1: Tower vibration data collection

[0036] A triaxial ICP accelerometer (PCB 356A16, measuring range ±50g) was selected as the vibration sensor and installed on the side wall of the lower pool of the cooling tower (3m above the tower base) (it should be noted that during the actual resonance detection operation, the installation height of the sensor should be determined according to the specific height of the tower). The XYZ axes were aligned with the radial, tangential, and axial directions of the tower body, respectively.

[0037] The first vibration data set is obtained by collecting the original vibration signal through a vibrometer with an integrated FFT module at a sampling frequency of 2 kHz and continuously collecting data for 10 minutes, including:

[0038] Frequency domain amplitude spectrum (FFT resolution 0.5Hz)

[0039] Time domain waveform (0-5s time period)

[0040] Step S2: Fan excitation frequency sweep

[0041] A Wilcoxon 786M vibration probe was installed on the fan. The motor frequency was increased continuously from 25 Hz to 50 Hz in 0.1 Hz increments. The data collection began after 30 seconds of stable operation at each frequency point.

[0042] Step S3: Synchronous collection of fan vibration

[0043] The fan vibration signal is synchronously collected at each frequency point (e.g., 30.1 Hz). Each collection lasts 1 minute and is repeated three times to take the average value to generate the second vibration data set sequence [b1, b2, …, bn] (n = 250 frequency points).

[0044] The data set includes: fan vibration frequency domain amplitude spectrum and time domain waveform envelope.

[0045] Step S4: Spectrum comparison and resonance source identification

[0046] Compare the first tower vibration data set with the fan sequence [b1,…,bn] point by point;

[0047] Preferably, if the frequency domain overlap is greater than 90% and the time domain similarity is greater than 85%, the fan frequency is determined to be a characteristic frequency.

[0048] Preferably, if the comparison result indicates that the difference between the excitation frequency of the second vibration data set and the natural frequency of the first vibration data set is within the range of ±5% of the natural frequency, and a significant amplitude integral (much higher than the background vibration level) is observed on the spectrum graph, then the current second vibration data bn is determined to be a characteristic frequency.

[0049] It should be noted that in this technical solution, the calculation of frequency domain overlap and time domain similarity can be directly obtained through the vibration meter. The calculation formulas of frequency domain overlap and time domain similarity and the principles of spectrum comparison and resonance source identification are not repeated in this technical solution and also fall within the scope of protection of this application.

[0050] Comparative Experiment 1: Traditional Broadband Monitoring Method

[0051] Detection method Identify resonance points False positive rate Positioning accuracy Traditional broadband monitoring 2 / 5 38% 45% Method of the present invention 5 / 5 5% 98%

[0052] Traditional methods fail to establish an excitation-response relationship and are unable to distinguish between the tower resonance peaks caused by fan excitation and bearing fault frequency (the two have similar vibration frequencies), leading to misjudgment. The present invention reduces the misjudgment rate from 38% to 5% through synchronous scanning and dual-criteria filtering.

[0053] Example 2: Three-axis vibration separation analysis

[0054] On the basis of the previous embodiment, the improvements in this technical solution are as follows:

[0055] In step S1, three uniaxial sensors (or one triaxial sensor) are placed in three-dimensional orthogonal directions of the cooling tower:

[0056] X-axis: parallel to the air inlet surface (detecting lateral bending vibration)

[0057] Y-axis: perpendicular to the air inlet surface (detecting torsional vibration)

[0058] Z axis: vertical direction (detecting axial tensile vibration)

[0059] The spectrum analysis of the three-axis data was performed independently, and the first vibration data set was expanded to {F 1χ , F1ᵧ, F1ᵦ}.

[0060] Resonance determination logic optimization:

[0061] When any axial fan-tower data pair (such as b2, F 1χ ) satisfies the dual criteria (frequency domain overlap > 90% + time domain similarity > 85%), which means that resonance occurs.

[0062] Comparative Experiment 2: Single-Axis vs. Tri-Axis Layout

[0063] Sensor layout Identify the number of resonance modes Missed detection rate Single axis (Z direction only) 1 (axial tension) 67% Three-axis layout 3 (bending + twisting + stretching 0%

[0064] In this technical solution, cooling tower resonance is directionally sensitive (for example, bending modes only appear on the X-axis), and three-axis separate acquisition can avoid mode omissions caused by single-point layout (reduced from 67% to 0%).

[0065] Example 3: Adaptive step size adjustment

[0066] On the basis of the above two embodiments, the improvements in this technical solution are as follows:

[0067] In step S2, set the step size adaptive adjustment strategy:

[0068] 1. Initially scan at 0.1 Hz step size;

[0069] 2. When the frequency domain overlap is >70% but the time domain similarity is <85%, the system switches to a 0.05Hz microstep size and rescans within the current frequency ±0.1Hz range;

[0070] 3. When there is no overlap among five consecutive frequency points, switch to 0.15Hz step size accelerated scanning.

[0071] Example 4: Data preprocessing optimization

[0072] Add a preprocessing module before S4 spectrum comparison:

[0073] 1. Signal amplification: Apply 100 times gain to the sensor signal (for weak vibrations <5mV);

[0074] 2. Bandpass filter: cut-off frequency 20-1000Hz (filters out motor electromagnetic interference and structural low-frequency noise);

[0075] 3. Digital-to-analog conversion: 24-bit ADC quantization (dynamic range 144dB), sampling rate synchronously locked to 2kHz.

[0076] In this technical solution, the unprocessed signal in the initial state may cause the time domain envelope similarity calculation value to be artificially high due to the 50Hz power frequency interference, thereby causing misjudgment; when the signal is band-pass filtered and the interference is eliminated, the similarity error can be reduced.

[0077] The cooling tower resonance detection implementation process in this application is as follows:

[0078] 1. Install a three-axis sensor on the tower (position: intersection of the X / Y / Z axial support beams);

[0079] 2. Scan from 25 Hz to 50 Hz (250 points in total) with a step size of 0.1 Hz, collecting data for 1 minute three times at each point;

[0080] 3. After preprocessing, compare the data sets and compare the frequency domain overlap and time domain envelope similarity to ultimately determine the resonant frequencies of the fan and cooling tower.

[0081] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in the present application should be included in the scope of protection of the present application. Therefore, the scope of protection of the present application should be based on the scope of protection of the claims.

Claims

1. A method for detecting resonance of a cooling tower, comprising a vibration sensor for receiving vibrations and a vibration meter electrically connected thereto, wherein the vibration meter is configured to perform spectrum analysis on the vibrations received by the vibration sensor and output vibration data, wherein: S1: placing a vibration sensor at a preset position of the cooling tower to collect a first vibration data set of the cooling tower body; S2: Install the vibration meter on the fan drive system, start the fan and adjust the motor operating frequency in steps; S3: At each frequency adjustment point, synchronously collect the second vibration data set of the fan at the corresponding frequency, recorded as a sequence [b1, b2, ..., bn]; S4: By comparing and analyzing the first vibration data set and the second vibration data set sequence through spectrum, the characteristic frequency of the fan that causes the cooling tower resonance is identified from [b1, b2, ..., bn] according to the vibration frequency overlap and waveform similarity.

2. The method for detecting cooling tower resonance according to claim 1, wherein: The first vibration data set is the natural frequency of the cooling tower, and the second vibration data set is the excitation frequency of the fan; the first vibration data set and the second vibration data set [b1, b2, ..., bn] both include a frequency domain amplitude spectrum and a time domain waveform diagram.

3. The method for detecting cooling tower resonance according to any one of claim 2, wherein: The comparison result indicates that the difference between the excitation frequency of the second vibration data set and the natural frequency of the first vibration data set is within ±5% of the natural frequency, and the current second vibration data bn is determined to be a characteristic frequency.

4. The method for detecting cooling tower resonance according to claim 3, wherein: Before comparing the first vibration data set and the second vibration data set, the method further includes preprocessing the first vibration data set and the second vibration data set, wherein the preprocessing includes signal amplification, filtering, and digital-to-analog conversion.

5. The method for detecting cooling tower resonance according to claim 1, wherein: The frequency regulation in S2 adopts a step size of 0.1 Hz, and the motor operating frequency is continuously increased in the range of 25 Hz-50 Hz. Data is collected after each frequency point runs stably for ≥30 s.

6. The method for detecting cooling tower resonance according to claim 5, characterized in that: The step size can be adaptively adjusted to 0.05 Hz-0.15 Hz.

7. The method for detecting cooling tower resonance according to claim 6, characterized in that: The preset positions in S1 are the three-dimensional orthogonal directions of the X-axis, Y-axis and Z-axis of the cooling tower, and the first vibration data set includes the results of independent spectrum analysis in the three axes.

8. The method for detecting cooling tower resonance according to claim 1, wherein: The second vibration data set bn of each frequency point of the fan is obtained by taking the average value of three repeated measurements, and the duration of a single measurement is ≥1 min.

9. The method for detecting cooling tower resonance according to claim 1, wherein: The vibration sensor is a three-axis acceleration sensor, and the vibration meter is integrated with an FFT analysis module.

Images