A base frequency automatic identification method of a marine mooring cable tension monitoring signal

Abstract

The application discloses a kind of marine mooring cable tension monitoring signal fundamental frequency automatic identification method, comprising: the cable vibration signal of acquisition is carried out FFT transformation, obtains spectrogram, and is carried out smoothing, removes noise point processing;According to certain rule, all spectral peak values are selected and marked;Adjacent peak frequency difference is calculated in sequence respectively and is recorded as Δf1, Δf1…Δfn-1, the above frequency difference is statistically analyzed, and the mean value is calculated after selecting all effective frequencies;The ratio of each order frequency peak value and mean value is counted, and the effectiveness of mean value is determined;Effective mean value is used as the initial value of fundamental frequency, wavelet processing signal is applied, and the spectrogram around the value is intercepted in spectrogram, if there is a certain amplitude peak value in spectrogram, it is the fundamental frequency value, if it does not exist, the initial value is the fundamental frequency value.The method can accurately obtain the fundamental frequency of cable vibration.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of ship mooring, in particular to a base frequency automatic identification method of a ship mooring cable tension monitoring signal. BACKGROUND

[0002] The mooring cable is a key component of the ship mooring system, and the tension of the mooring cable changes constantly due to factors such as flow rate, tide, swell, wind speed, and operation. For example, when the ship is at the wave crest or is pushed to the shore by the wave, the mooring cable tension increases, and when the ship height increases during the high tide, the mooring cable tension increases. Especially in severe weather such as typhoon, under the combined action of tide, swell, wind speed, etc., the cable tension will inevitably increase sharply. And the increase of the tension of part of the cable may cause uneven force of other cables, causing unstable berthing of the ship, and if not timely warned and adjusted, it may cause the cable to break. The breakage of the cable will inevitably lead to instability of the mooring, especially in severe weather such as typhoon, which may eventually cause the ship to lose control and cause serious safety accidents and heavy economic losses. In addition, due to factors such as cable material, service time, friction at the guide cable, and improper mooring operation, cable breakage has become one of the common mooring safety accidents.

[0003] At present, the commonly used cable tension test methods include numerical simulation calculation, direct measurement and indirect measurement of pressure sensor, and the indirect measurement method mainly includes vibration frequency method and torque balance method. The vibration frequency method is widely used in the tension monitoring of ship mooring cables due to its simple measuring device, low running and maintenance cost, and continuous monitoring. The principle of the frequency method for testing cable tension is to use an acceleration sensor fixed on the cable to collect the vibration acceleration signal of the cable under artificial excitation or environmental excitation, and after filtering, amplifying and spectrum analysis, the base frequency of the cable is determined from the obtained frequency spectrum diagram, and then the cable tension is determined according to the relationship between the base frequency of the cable and the cable tension. Therefore, when the frequency method is used to test the cable tension, only the accurate base frequency needs to be obtained, and the cable tension can be obtained through the corresponding relationship between the tension and the base frequency. This method can achieve very high precision. This method has been widely used because of its economy, practicality, simple operation, reusable equipment and high testing precision. Therefore, the accurate automatic identification and extraction of the base frequency of the cable vibration determines the accuracy of the cable tension calculation and the reliability of the cable tension monitoring system.

[0004] The main automatic fundamental frequency identification algorithms are the power spectrum frequency difference method, the fundamental frequency method and the new fundamental frequency method. The power spectrum frequency difference method is based on the theory of string vibration. When the cable tension is fixed, the higher order frequencies become integer multiples of the fundamental frequency, which will appear as a series of equally spaced peaks in the power spectrum, where the spacing of the peaks represents the fundamental frequency. By capturing these consecutive peaks and calculating the spacing of adjacent peaks, and then averaging these spacings, the desired fundamental frequency can be obtained. The fundamental frequency method: first, select a resonance peak with the largest amplitude from the frequency spectrum and record its frequency fn, which should be the nth order natural frequency. Assuming it is the peak caused by the n1th order resonance frequency of the cable, the assumed fundamental frequency F1 can be calculated. According to the theory of string vibration, other resonance peaks should be integer multiples of F1. Therefore, if the ratio of all resonance peaks to F1 is very close to an integer (with a deviation of no more than ±0.05), F1 can be considered as the true fundamental frequency F. Otherwise, the value of n1 can be adjusted appropriately and the process continues until the fundamental frequency F is determined. The new fundamental frequency method analyzes the part of the signal that responds better to the frequency difference method to obtain an estimate of the fundamental frequency fw1. Then, use this estimate as the initial value of the fundamental frequency to find the maximum frequency fn in the frequency spectrum. Next, calculate the value of n, i.e. n = fn / fw1. Finally, by the fundamental frequency method, fw can be calculated.

[0005] The power spectrum peak method is theoretically an effective method for determining the tension of a cable. However, in practical applications, this simple relationship can be affected by various factors, resulting in errors. First, the cable in reality often has a certain sag, which makes its tension distribution non-uniform throughout its length. Second, the self-weight of the cable also affects its tension, especially in the case of long cables. Due to these nonlinear factors, the measured frequencies of each order and the multiple of the fundamental frequency can only be approximate. Simply measuring these high-order frequencies and then trying to determine the fundamental frequency through them can result in significant errors. Therefore, the power spectrum peak method is mainly suitable for preliminary estimation of the range of the fundamental frequency, providing a starting point or reference for more accurate measurement methods. For applications that require accurate measurement of the fundamental frequency, other more complex and accurate methods may be needed. The fundamental frequency method for determining the tension of a cable is undoubtedly a more accurate method, whose principle is mainly based on the relationship between the vibration frequency of the cable and the tension. However, in practical applications, this method is not always directly or easily executed. Random environmental excitations, such as wind, rain, or sea waves, can affect the cable, resulting in multiple resonance peaks on the frequency spectrum. In this case, it becomes very difficult to identify which resonance peak is related to the fundamental frequency or an integer multiple of the cable. Theoretically, the resonance peak with the largest amplitude on the frequency spectrum can be selected and its frequency fnrecorded, then it is assumed to be the nth natural frequency. However, the problem is that it is difficult to determine this n order without knowing the approximate range of the fundamental frequency. To determine this n value, multiple iterations of calculations and analyses may be needed. This not only consumes a lot of time, but also requires relevant professional knowledge and experience. The new fundamental frequency method When the signal response is not obvious enough or is masked by noise due to external factors, frequency estimation becomes difficult. Noise can come from various sources, including the equipment itself, environmental factors, or other unknown interference sources. In such cases, obtaining an accurate fundamental frequency estimate becomes more difficult; simply selecting the largest frequency fnfrom the frequency spectrum may not be the best choice in many cases. Especially when there are multiple significant peaks on the frequency spectrum or interference caused by other signal sources, simply selecting the largest peak can lead to misleading results; this method requires multiple iterations and calculations, which can increase the time and complexity of the measurement process; this method is largely dependent on the results of the preliminary analysis. If the preliminary analysis is not accurate enough, the subsequent calculations can also be error-prone. SUMMARY

[0006] Based on the problems of the above-mentioned automatic fundamental frequency identification method, the present application provides a method for automatically identifying the fundamental frequency of a mooring cable tension monitoring signal for marine use, which can accurately obtain the fundamental frequency of cable vibration.

[0007] The technical scheme adopted by the present application is as follows: a base frequency automatic identification method for a marine mooring cable tension monitoring signal, comprising:

[0008] Step one: denoising the cable vibration signal, and performing fast Fourier transform on the denoised signal to obtain a frequency spectrum graph;

[0009] Step two: performing smoothing processing on the frequency spectrum graph to remove noise points, obtaining all resonance peak points by a peak value extraction method, and sorting all resonance peaks according to frequency size and marking as f1, f2, f3…fn;

[0010] Step three: calculating adjacent peak frequency difference values and marking as Δf1, Δf1…Δfn-1, performing statistical analysis on the above frequency difference values, calculating the mean value after selecting all effective frequencies, and determining the effectiveness of the mean value by calculating the ratio of each order frequency peak value to the mean value;

[0011] Step four: taking the effective mean value as the base frequency initial value, performing filtering processing on the cable vibration signal, and intercepting the spectrum graph near the value in the frequency spectrum graph, if there is a peak value with an amplitude of more than 5% of the highest amplitude in the spectrum graph, the frequency corresponding to the peak value is the base frequency value, and if there is no such peak value, the initial value is the base frequency value.

[0012] Further, the method for denoising the cable vibration signal in step one adopts wavelet denoising.

[0013] Further, the smoothing processing on the frequency spectrum graph in step two adopts a three-point smoothing method, and the peak value extraction algorithm adopts a quadratic polynomial fitting method.

[0014] Further, the effective value elimination method in step three adopts a ratio method or a probability statistical method to eliminate abnormal values.

[0015] Further, the wavelet analysis method or low-pass filtering is adopted for filtering the cable vibration signal.

[0016] The base frequency automatic identification method provided by the present application first classifies according to the frequency spectrum signal characteristics, performs statistical analysis on the characteristic values, obtains the base frequency initial value, then performs wavelet processing on the original vibration signal according to the base frequency initial value, obtains the spectrum graph near the base frequency initial value, and finally determines the base frequency value according to the peak value amplitude, so that the base frequency of the cable vibration can be accurately obtained. BRIEF DESCRIPTION OF DRAWINGS

[0017] Figure 1 is a processing flowchart of the base frequency automatic identification method of the present application.

[0018] Figure 2 is an original cable vibration signal graph in the embodiment of the present application.

[0019] Figure 3This is a signal image after wavelet denoising in an embodiment of the present invention.

[0020] Figure 4 This is the signal spectrum diagram after wavelet denoising in an embodiment of the present invention.

[0021] Figure 5 This is a spectrum diagram after three-point smoothing in an embodiment of the present invention.

[0022] Figure 6 This is the spectrum extracted by wavelet analysis in an embodiment of the present invention.

[0023] Figure 7 This is a spectrum diagram near the effective value in an embodiment of the present invention. Detailed Implementation

[0024] The technical solutions in the embodiments of the present invention will be clearly and completely described below. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the 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.

[0025] like Figure 1 As shown, the present invention provides an automatic fundamental frequency identification method for marine mooring cable tension monitoring signals. The method involves performing an FFT transform on the collected cable vibration signal to obtain a spectrum, followed by smoothing and noise removal. All spectral peaks are selected and marked according to certain rules. The frequency differences between adjacent peaks are calculated sequentially and denoted as Δf1, Δf1…Δfn-1. Statistical analysis is performed on these frequency differences, and the mean is calculated after selecting all effective frequencies. The ratio of each frequency peak to the mean is statistically analyzed to determine the validity of the mean. The effective mean is used as the initial fundamental frequency value. Wavelet processing is applied to the signal, and a spectrum near this value is extracted from the spectrum. If a peak with an amplitude greater than 5% of the highest amplitude exists in the spectrum, the frequency corresponding to that peak is the fundamental frequency value; otherwise, the initial value is the fundamental frequency value.

[0026] The specific implementation process is as follows:

[0027] Step 1: Analyze the cable vibration signal (e.g.) Figure 2 The signal after wavelet denoising (as shown) is obtained by wavelet denoising (wavelet basis selected as db02). Figure 3 Perform a Fast Fourier Transform to obtain the spectrum (as shown). Figure 4 (as shown)

[0028] Step 2: Smooth the spectrum to remove noise (using a three-point averaging method, where the current value after smoothing is the average of the previous, current, and next values ​​before smoothing). Figure 5), all resonance peak points are obtained by peak extraction method (the peak extraction algorithm is to sequentially fit each group of data in the data points by using a quadratic polynomial, the number of data points used in the fitting is specified by the width, for each wave peak, the quadratic fitting can be compared with the threshold, and the wave peak below the threshold is ignored, wherein the width is 5, and the threshold is 500), and all resonance peaks are sorted by frequency size and marked as f1=2.637Hz, f2=5.2246Hz, f3=7.5684Hz, f4=10.0586Hz, f5=10.2539Hz, f6=10.8887Hz, f7=12.7441Hz, f8=14.9902Hz, f9=15.5273Hz;

[0029] Step three: the frequency difference of adjacent peaks is calculated and marked as Δf1=2.5879, Δf2=2.3437, Δf3=2.4902, Δf4=0.1953, Δf5=0.6348, Δf6=1.8555, Δf7=2.2461, Δf8=0.5371, statistical analysis is performed on the above frequency differences, the mean value Δf=1.6113Hz is calculated after selecting all effective frequencies, the ratio of each order frequency peak value to the mean value (1.6, 1.5, 1.5, 0.1, 0.4, 1.2, 1.4, 0.3) is calculated, and the effectiveness of the mean value is determined (it is determined that the ratio of each order frequency peak value to the mean value is greater than 1.5 or less than 0.5 is invalid, and thus Δf1, Δf4, Δf5, and Δf8 are removed, and the mean value is recalculated as 2.2339Hz after removing the invalid values);

[0030] Step four: the effective mean value 2.2339Hz is used as an initial value of the fundamental frequency, wavelet processing is applied to the cable vibration signal, wavelet denoising (the wavelet base is selected as db02, the frequency spectrum diagram obtained by performing 1 decomposition on the original signal is as shown in Figure 6 ), and the spectrum diagram near the value is intercepted in the frequency spectrum diagram (as shown in Figure 7 ), and if there is a peak value with an amplitude greater than 1000 in the spectrum diagram, the peak value is the fundamental frequency value 2.636Hz.

[0031] The fundamental frequency calculated by the power spectrum frequency difference method is 2.5300Hz, the fundamental frequency method is 2.5635, the new fundamental frequency method is 2.5635Hz, and the actual fundamental frequency is 2.636Hz. It can be seen that the fundamental frequency recognition algorithm proposed in the application can accurately obtain the fundamental frequency of the cable vibration.

[0032] The above content is only an example and description of the structure of the application, and those skilled in the art can make various modifications or supplements or use similar ways to replace the described specific embodiments, as long as the modifications or supplements do not deviate from the structure of the application or exceed the scope defined by the claims, and the modifications or supplements should belong to the protection scope of the application.

Claims

1. A method for automatic identification of the fundamental frequency of marine mooring cable tension monitoring signals, characterized in that, include: Step 1: Denoise the cable vibration signal and perform a Fast Fourier Transform on the denoised signal to obtain the spectrum. Step 2: Smooth the spectrum to remove noise, obtain all resonant peaks using the peak extraction method, and sort all resonant peaks by frequency magnitude, labeling them as f1, f2, f3…fn; Step 3: Calculate the frequency difference between adjacent peaks and record them as Δf1, Δf2, ..., Δfn-1 respectively. Perform statistical analysis on the frequency differences, select all frequency differences to calculate the mean, and calculate the ratio of each frequency difference to the mean to determine the validity of the mean. Specifically, set the ratio of each frequency difference to the mean to be greater than 1.5 or less than 0.5 as invalid, remove invalid values, and recalculate the mean. Step 4: Use the effective mean as the initial value of the fundamental frequency, filter the cable vibration signal, and extract the spectrum near this value from the spectrum graph. If there is a peak in the spectrum with an amplitude greater than 5% of the highest amplitude, the frequency corresponding to the peak is the fundamental frequency value. If there is no peak, the initial value is the fundamental frequency value.

2. The method for automatic identification of the fundamental frequency of marine mooring cable tension monitoring signal as described in claim 1, characterized in that: The method for denoising the cable vibration signal in step one is wavelet denoising.

3. The method for automatic identification of the fundamental frequency of marine mooring cable tension monitoring signal as described in claim 1, characterized in that: In step two, the smoothing of the spectrum is performed using the three-point smoothing method, and the peak extraction algorithm uses the quadratic polynomial fitting method.

4. The method for automatic identification of the fundamental frequency of marine mooring cable tension monitoring signal as described in claim 1, characterized in that: Step 3 involves removing outliers using either a ratio method or a probability statistical method.

5. The method for automatic identification of the fundamental frequency of marine mooring cable tension monitoring signal as described in claim 1, characterized in that: Wavelet analysis or low-pass filtering can be used to filter the cable vibration signal.

Images