Heating pipeline damage detection method and system based on ultrasonic technology

Abstract

The invention relates to the technical field of intelligent sensors, in particular to a heating pipeline damage detection method and system based on the ultrasonic technology. Determining a time-frequency defect index by combining the energy contribution degree of the main characteristic components of the composite ultrasonic signal in the frequency domain to the composite ultrasonic signal; determining a defect characteristic value by analyzing the difference between the time-frequency defect index of the composite ultrasonic signal and the clustering center of the normal clustering cluster and combining the abnormal probability; and on the basis of the defect characteristic value, adjusting the capability of the intelligent sensor array for capturing defect echoes at the deep layer of the heating pipeline. The problem of insufficient penetrating power of high-frequency ultrasonic signals caused by multiple materials and false alarm of defects of the heating pipeline caused by difference of pipeline installation conditions are solved, and the accuracy and reliability of damage detection of the heating pipeline are improved.

Description

Technical Field

[0001] The present application relates to the field of intelligent sensor technology, and in particular to a method and system for detecting damage in heating pipes based on ultrasonic technology. Background Art

[0002] With the acceleration of urbanization, the scale of underground heating pipelines continues to expand. Problems such as pipeline aging, corrosion, and third-party construction damage have led to frequent leakage accidents. Traditional detection methods rely on pressure testing, which has defects such as low efficiency, poor positioning accuracy, and the need to stop operations for inspection. Against this background, non-invasive ultrasonic detection technology has come into being. By analyzing the changes in the sound wave characteristics when ultrasonic waves propagate in the pipeline, it can accurately identify defects such as pipeline cracks, holes, and wall thinning.

[0003] In the detection of heating pipe damage, ultrasonic technology can stably propagate and accurately feedback defect information in metal pipes due to the homogeneity of the material and high acoustic conductivity. However, when facing non-metallic or composite pipes such as plastic, fiberglass, and PE, the difference in acoustic impedance causes ultrasonic energy to attenuate, and the high-frequency signal has insufficient penetration, making it difficult to detect deep and tiny defects, reducing the accuracy and reliability of heating pipe damage detection. In addition, the multi-layer complex structure wrapping causes multiple reflections, refractions, and scattering, resulting in disordered sound wave paths and energy dispersion, which causes false detection or missed detection of heating pipe damage, reducing the accuracy and reliability of heating pipe damage detection. Summary of the Invention

[0004] In order to solve the above technical problems, the purpose of this application is to provide a heating pipe damage detection method and system based on ultrasonic technology. The technical solutions adopted are as follows:

[0005] In a first aspect, an embodiment of the present application provides a method for detecting damage to a heating pipe based on ultrasonic technology, the method comprising the following steps:

[0006] In the heating pipes, an intelligent sensor array is used to obtain composite ultrasonic signals within a preset time period before each moment;

[0007] Decomposing the composite ultrasonic signal into components at multiple scales, analyzing the energy distribution of the components at each scale, and determining the energy distribution value of the composite ultrasonic signal at each moment; determining the energy contribution value of the composite ultrasonic signal at each moment based on the energy contribution of the main characteristic components of the composite ultrasonic signal in the frequency domain to the composite ultrasonic signal, and determining the time-frequency defect index of the composite ultrasonic signal at each moment in combination with the energy distribution value;

[0008] Analyzing the degree of dispersion of all data in the time domain and the extreme distribution of all peaks in the frequency domain of the composite ultrasonic signal at each moment, and combining the time-frequency defect index, to evaluate the abnormal probability of the composite ultrasonic signal at each moment; clustering the time-frequency defect indices at the current moment and all moments within a preset time period before the current moment to obtain normal clusters; determining the defect characteristic value of the composite ultrasonic signal at the current moment by analyzing the difference between the cluster center of the normal cluster and the time-frequency defect index of the composite ultrasonic signal at the current moment, and combining the abnormal probability;

[0009] Based on the defect characteristic value, the ability of the intelligent sensor array to capture defect echoes deep in the heating pipe is adjusted.

[0010] Preferably, the method for acquiring the composite ultrasonic signal is:

[0011] An intelligent sensor array is used to collect various signals of the heating pipes at each moment and within a preset time period before that, and the signal-to-noise ratio of various signals is obtained. The product of the normalized signal intensity value of the frequency domain signals of various signals at each frequency and the signal-to-noise ratio is calculated. The frequency domain signals of all kinds of signals at the same frequency are added together to obtain the signal intensity at the corresponding frequency of the frequency domain signal of the composite ultrasonic signal. The frequency domain signal of the composite ultrasonic signal is converted into the time domain to obtain a composite ultrasonic signal, wherein the various signals include: low-frequency ultrasonic echoes of non-metallic pipes, high-frequency ultrasonic echoes of metal pipes, environmental noise signals and pipeline mechanical vibration signals.

[0012] Preferably, the method for determining the energy distribution value of the composite ultrasonic signal at each moment is:

[0013] The composite ultrasonic signal at each moment is used as the input of the wavelet transform, and the components at a preset number of scales are output. The energy mean of the components at each scale is calculated, and the cumulative sum of the energy mean of the components at the preset number of scales is used as the energy distribution value of the composite ultrasonic signal at each moment.

[0014] Preferably, the method for determining the energy contribution value of the composite ultrasonic signal at each moment is:

[0015] The composite ultrasonic signal at each moment is used as the input of principal component analysis, the variance contribution rates of all principal components are output, and the sum of the first preset number of variance contribution rates in the results of descending order of variance contribution rates is used as the energy contribution value of the composite ultrasonic signal at each moment.

[0016] Preferably, the time-frequency defect index of the composite ultrasonic signal at each moment is a result of forward fusion of the energy distribution value and the energy contribution value of the composite ultrasonic signal at each moment.

[0017] Preferably, the evaluating the abnormal probability of the composite ultrasonic signal at each moment includes:

[0018] The variance of all data of the composite ultrasonic signal in the time domain at each moment, the maximum peak value in the frequency domain, and the time-frequency defect index at each moment are used as inputs of the logistic regression algorithm to output the abnormal probability of the composite ultrasonic signal at each moment.

[0019] Preferably, the normal cluster is: a cluster with the smallest mean value of the time-frequency defect index among all clusters obtained by clustering the time-frequency defect indexes of the current moment and all moments within a preset period before the current moment.

[0020] Preferably, the expression of the defect characteristic value of the composite ultrasonic signal at the current moment is: Where D represents the difference between the time-frequency defect index of the composite ultrasonic signal at the current moment and the cluster center of the normal cluster; P n represents the abnormal probability of the composite ultrasonic signal at time n within the preset time period before the current time; N represents the length of the preset time period; norm() represents the normalization function.

[0021] Preferably, the adjusting the ability of the smart sensor array to capture defect echoes deep in the heating pipe comprises:

[0022] If the defect characteristic value of the composite ultrasonic signal at the current moment is greater than a preset threshold, the intelligent sensor array automatically adjusts to a low-frequency mode to capture the echo of a deep defect in the heating pipe. Otherwise, the intelligent sensor array does not adjust. In a second aspect, an embodiment of the present application also provides a heating pipe damage detection system based on ultrasonic technology, comprising a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements any of the steps of the aforementioned method for detecting heating pipe damage based on ultrasonic technology.

[0023] This application has at least the following beneficial effects:

[0024] This application addresses the problems of large differences in acoustic impedance of non-metallic pipes, rapid ultrasonic attenuation, and sound wave scattering caused by multi-layer wrapping structures. By analyzing the degree of energy contribution of the main characteristic components of the composite ultrasonic signal in the frequency domain to the composite ultrasonic signal, and combining the energy distribution value, a time-frequency defect index is constructed, which reflects the defect-related time-domain transient energy intensity and frequency-domain abnormal energy distribution, solves the positioning ambiguity problem caused by the echo of tiny defects in deep thick-walled non-metallic pipes being masked by noise and the disorder of the sound wave path, and improves the accuracy and reliability of heating pipe damage detection; further, this application uses a clustering algorithm to classify the time-frequency defect index to address misjudgments caused by temperature fluctuations and differences in pipe materials. , and combined the time domain variance and frequency domain peak to supervise the training of the logistic regression model, and obtained the defect characteristic value, which reflects the global deviation degree of the current signal from the historical normal mode and the defect accumulation risk, reduces the false alarm of heating pipe defects caused by changes in ambient temperature and differences in pipeline installation conditions, and improves the accuracy and reliability of heating pipe damage detection; further, the application optimizes the working mode of the intelligent sensor array in real time based on the defect characteristic value, enhances the detection depth of ultrasound in non-metallic pipes, solves the problem of insufficient penetration of high-frequency ultrasonic signals caused by multiple materials, reduces the probability of misjudgment or missed detection of heating pipe defects, and improves the accuracy and reliability of heating pipe damage detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present application or the prior art, the following is a brief introduction to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0026] Figure 1 A flowchart of a method for detecting damage to a heating pipe based on ultrasonic technology according to an embodiment of the present application;

[0027] Figure 2 A schematic diagram of the abnormality probability extraction process provided by one embodiment of the present application. DETAILED DESCRIPTION

[0028] To further illustrate the technical means and effectiveness of this application's implementation of the intended invention, the following, in conjunction with the accompanying drawings and preferred embodiments, describes in detail the ultrasonic-based heating pipe damage detection method and system proposed in this application, including its specific implementation, structure, features, and effectiveness. In the following description, references to different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.

[0029] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.

[0030] The specific scheme of the heating pipe damage detection method and system based on ultrasonic technology provided by this application is described in detail below with reference to the accompanying drawings.

[0031] See also Figure 1 , which shows a flowchart of a method for detecting damage to a heating pipe based on ultrasonic technology according to an embodiment of the present application, the method comprising the following steps:

[0032] Step S1: In the heating pipe, a smart sensor array is used to obtain a composite ultrasonic signal within a preset time period before each moment.

[0033] The heating pipe damage detection system based on ultrasonic technology includes: ultrasonic transmitting / receiving array module, multi-channel signal acquisition module, embedded processing unit module, and dynamic parameter optimization module. The functions of each module are as follows:

[0034] Ultrasonic transmitting / receiving array module: This module uses an array composed of wide-band intelligent sensors, supporting multi-band coverage and adaptive signal transmission. It can not only improve the defect resolution of metal pipes at high frequencies, but also enhance the ultrasonic penetration ability of non-metallic pipes at low frequencies.

[0035] Multi-channel signal acquisition module: The multi-channel signal acquisition module integrates a 24-bit high-precision ADC and a 1MHz sampling rate design. It can capture ultrasonic echo signals in real time and simultaneously record data such as environmental noise and pipeline vibration, providing the raw data basis for subsequent analysis.

[0036] Embedded processing unit module: Based on FPGA chip, it realizes real-time processing at the edge and executes two sub-modules: time-frequency domain feature extraction module and joint risk analysis module.

[0037] Dynamic parameter optimization module: Dynamically optimizes ultrasonic parameters based on the analysis results of the two sub-modules: time-frequency domain feature extraction module and joint risk analysis module.

[0038] An ultrasonic transmitter / receiver array composed of wideband smart sensors achieves multi-band coverage, utilizing high frequencies to improve defect resolution in metallic pipes while enhancing ultrasonic penetration in non-metallic pipes through low frequencies. The smart sensors integrate a 24-bit high-precision ADC with a 1MHz sampling rate, capturing ultrasonic echo signals in real time while simultaneously recording ambient noise and pipe vibration data, providing the raw data foundation for subsequent analysis. An embedded processing unit, based on an FPGA chip, enables real-time edge processing, extracting time- and frequency-domain features and conducting joint risk analysis, dynamically optimizing ultrasonic parameters to suit different pipe materials and operating conditions.

[0039] Wideband intelligent sensor arrays and multi-channel signal acquisition modules are installed on the outer wall of the pipeline and at key nodes along the pipeline to collect low-frequency ultrasonic echo signals from non-metallic pipelines and high-frequency ultrasonic echo signals from metallic pipelines, which are used to enhance the deep penetration of non-metallic materials and the resolution of metal surface defects. Intelligent environmental noise sensors and intelligent pipeline vibration sensors are deployed simultaneously. The noise spectrum and vibration waveform are analyzed in real time through the embedded processing unit. The environmental noise spectrum and pipeline mechanical vibration waveform data are captured in real time along the distributed fixed network of the pipeline for the construction of a background interference feature library and abnormal vibration correlation analysis.

[0040] An intelligent sensor array is used to collect various signals of the heating pipes at each moment and within a preset time period before each moment, and the signal-to-noise ratio of various signals is obtained. The product of the normalized signal intensity value of the frequency domain signals of various signals at each frequency and the signal-to-noise ratio is calculated. The frequency domain signals of all kinds of signals at the same frequency are added together to obtain the signal intensity at the corresponding frequency of the frequency domain signal of the composite ultrasonic signal. The frequency domain signal of the composite ultrasonic signal is converted into the time domain to obtain a composite ultrasonic signal, wherein the various signals include: low-frequency ultrasonic echoes of non-metallic pipes, high-frequency ultrasonic echoes of metal pipes, environmental noise signals and mechanical vibration signals of pipes. In this embodiment, the data acquisition frequency is set to f.

[0041] It should be noted that the values ​​of the preset duration and data acquisition frequency f are both manually set. In this embodiment, the value of the preset duration is 1ms, and the value of the data acquisition frequency f is 1MHz. In actual application, as other implementation methods, the implementer can also set them according to the specific situation. This embodiment does not impose any special restrictions.

[0042] In particular, if the time before the current moment is less than the preset time length, signal collection is not performed for the current moment.

[0043] It should be noted that there are many methods for performing time-frequency conversion on signals. In this embodiment, short-time fast Fourier transform is used to perform time-frequency conversion on signals. In actual application, as other implementation methods, implementers may also use other time-frequency conversion algorithms such as fast Fourier transform. This embodiment does not impose any special restrictions on the selection of time-frequency conversion algorithms.

[0044] Among them, the process of obtaining the signal-to-noise ratio and the short-time fast Fourier transform algorithm are well-known technologies, and the specific process of obtaining the signal-to-noise ratio and converting the time domain signal into the frequency domain using the short-time fast Fourier transform algorithm will not be repeated.

[0045] Step S2: Decompose the composite ultrasonic signal into components at multiple scales, analyze the energy distribution of the components at each scale, and determine the energy distribution value of the composite ultrasonic signal at each moment; determine the energy contribution value of the composite ultrasonic signal at each moment based on the energy contribution degree of the main characteristic components of the composite ultrasonic signal in the frequency domain to the composite ultrasonic signal, and determine the time-frequency defect index of the composite ultrasonic signal at each moment in combination with the energy distribution value.

[0046] Since the ultrasonic impedance characteristics of non-metallic pipes are significantly different from those of metals, the ultrasonic energy attenuates sharply during propagation, especially in thick-walled or large-diameter pipes, where high-frequency signals lack penetration and the echoes of tiny defects are easily masked by background noise. At the same time, the insulation layer, anti-corrosion layer or buried structure covering the pipeline causes multiple reflections and scattering of sound waves, resulting in a disordered sound wave path and blurred defect positioning.

[0047] Therefore, based on the above analysis, the composite ultrasonic signal is decomposed into components at multiple scales, and the energy distribution of the components at each scale is analyzed to determine the energy distribution value of the composite ultrasonic signal at each moment. Based on the energy contribution of the main characteristic components of the composite ultrasonic signal in the frequency domain to the composite ultrasonic signal, the energy contribution value of the composite ultrasonic signal at each moment is determined. In combination with the energy distribution value, the time-frequency defect index of the composite ultrasonic signal at each moment is determined. The specific process is as follows:

[0048] (1) In this embodiment, the composite ultrasonic signal is decomposed into components at multiple scales, and the energy distribution of the components at each scale is analyzed to determine the energy distribution value of the composite ultrasonic signal at each moment. Specifically:

[0049] As an implementation method, in this embodiment, the composite ultrasonic signal at each moment is used as the input of the wavelet transform, wherein the wavelet basis is selected as "db4", the decomposition scale is set to a preset number, and the components at the preset number of layer scales are output;

[0050] It should be noted that the value of the preset number is set manually. In this embodiment, the value of the preset number is 5. The implementer can also set it by himself according to the specific situation. This embodiment does not impose any special restrictions.

[0051] Among them, wavelet transform is a well-known technology, and the specific process of using it to perform multi-scale decomposition of signals will not be described in detail.

[0052] Furthermore, the energy mean of the components at each scale level is calculated, and the cumulative sum of the energy mean of the components at a preset number of scale levels is used as the energy distribution value of the composite ultrasonic signal at each moment, which is used to characterize the local characteristics of the composite ultrasonic signal in different frequency bands and times, and to distinguish defect signals from noise interference. If the energy distribution value is larger, it means that the defect-related energy in the composite ultrasonic signal is stronger, the possibility that the composite ultrasonic signal is a defect signal is greater, and the possibility that the heating pipe is damaged is greater.

[0053] (2) Furthermore, this embodiment determines the energy contribution value of the composite ultrasonic signal at each moment based on the degree of energy contribution of the main characteristic components of the composite ultrasonic signal in the frequency domain to the composite ultrasonic signal, specifically:

[0054] As an implementation method, this embodiment uses the composite ultrasonic signal at each moment as the input of principal component analysis, outputs the variance contribution rate of all principal components, and uses the sum of the first preset number of variance contribution rates in the results of descending order of the variance contribution rates as the energy contribution value of the composite ultrasonic signal at each moment, reflecting the degree of abnormality in the frequency domain energy distribution of the composite ultrasonic signal. By reducing the dimension, the key features of the frequency domain are extracted, the redundant noise is suppressed, and the frequency components related to the defects are highlighted. The larger the value, the more likely it is that the frequency segment where the frequency domain energy is concentrated corresponds to local defects such as cracks or delamination of the heating pipe.

[0055] It should be noted that the value of the preset number is set manually. In this embodiment, the value of the preset number is 3. In actual application, the implementer can also set it by himself according to the specific situation. This embodiment does not impose any special restrictions.

[0056] The principal component analysis is a well-known technology, and the process of analyzing the energy contribution of the principal components in the composite ultrasonic signal using the principal component analysis is not described in detail here.

[0057] (3) Furthermore, this embodiment determines the time-frequency defect index of the composite ultrasonic signal at each moment based on the energy contribution value of the composite ultrasonic signal at each moment and in combination with the energy distribution value, specifically:

[0058] In this embodiment, the result of the forward fusion of the energy distribution value and the energy contribution value of the composite ultrasonic signal at each moment is used as the time-frequency defect index of the composite ultrasonic signal at each moment, which is used to characterize the global possibility of monitoring pipeline damage at each moment. If the time-frequency defect index at the current moment is larger, it indicates that there are significant time-domain instantaneous energy and frequency-domain concentrated energy in the composite ultrasonic signal, that is, the larger the energy distribution value and the energy contribution value, the greater the possibility of heating pipeline damage; conversely, if the time-frequency defect index at the current moment is smaller, it indicates that there is a smaller possibility of significant time-domain instantaneous energy and frequency-domain concentrated energy in the composite ultrasonic signal, that is, the smaller the energy distribution value and the energy contribution value, the smaller the possibility of heating pipeline damage.

[0059] It should be understood that forward fusion refers to combining two or more indicators through addition or multiplication to obtain a comprehensive indicator, thereby more comprehensively and accurately evaluating a phenomenon or problem. This fusion method is not limited to simple arithmetic operations and can also include more complex statistical models and analysis methods. Implementers can choose according to their specific circumstances and this embodiment does not impose any special restrictions.

[0060] Preferably, in this embodiment, the sum of the energy distribution value and the energy contribution value of the composite ultrasonic signal at each moment is used as the time-frequency defect index of the composite ultrasonic signal at each moment; in actual application, as another implementation method, the product of the energy distribution value and the energy contribution value of the composite ultrasonic signal at each moment is used as the time-frequency defect index of the composite ultrasonic signal at each moment.

[0061] At this point, the composite ultrasonic signal is decomposed into multiple scale components by using wavelet transform, and the energy mean of the components at each scale is calculated and accumulated to obtain the energy distribution value. Furthermore, the main characteristic components of the signal frequency domain and their variance contribution rates are extracted through principal component analysis, and the variance contribution rates are accumulated to obtain the energy contribution value, which reflects the degree of abnormality of the energy distribution in the frequency domain. Finally, the energy distribution value and the energy contribution value are forward fused to obtain the time-frequency defect index, which comprehensively evaluates the possibility of pipeline damage. It effectively integrates time domain and frequency domain information, suppresses noise interference, highlights defect characteristics, and improves the accuracy of non-metallic pipeline damage detection through multi-scale decomposition and feature extraction.

[0062] Step S3: Analyze the discrete degree of all data of the composite ultrasonic signal in the time domain and the extreme distribution of all peaks in the frequency domain at each moment, and combine the time-frequency defect index to evaluate the abnormal probability of the composite ultrasonic signal at each moment; cluster the time-frequency defect indices of the current moment and all moments in the preset time period before it to obtain normal clustering clusters; determine the defect characteristic value of the composite ultrasonic signal at the current moment by analyzing the difference between the time-frequency defect index of the composite ultrasonic signal at the current moment and the cluster center of the normal clustering cluster, and combining the abnormal probability.

[0063] The time-frequency defect index at a single moment is insufficiently adaptable to complex working conditions such as temperature fluctuations and differences in pipeline materials, leading to an increased risk of misjudgment. For example, the time-frequency defect index values ​​of normal vibration and minor defects overlap, or different pipe sections present similar time-frequency defect index values ​​due to differences in installation conditions but have very different actual conditions.

[0064] Therefore, based on the above analysis, the degree of dispersion of all data in the time domain and the extreme distribution of all peaks in the frequency domain of the composite ultrasonic signal at each moment are analyzed, and combined with the time-frequency defect index, the abnormal probability of the composite ultrasonic signal at each moment is evaluated; by analyzing the difference between the time-frequency defect index of the composite ultrasonic signal at the current moment and the cluster center of the normal cluster, and combining the abnormal probability, the defect characteristic value of the composite ultrasonic signal at the current moment is determined. The specific process is as follows:

[0065] (1) In this embodiment, the discrete degree of all data in the time domain and the extreme distribution of all peaks in the frequency domain of the composite ultrasonic signal at each moment are analyzed respectively, and combined with the time-frequency defect index to evaluate the abnormal probability of the composite ultrasonic signal at each moment. Specifically:

[0066] In this embodiment, the variance of all data of the composite ultrasonic signal in the time domain at each moment, the maximum peak in the frequency domain, and the time-frequency defect index at each moment are used as inputs of the logistic regression algorithm. In this embodiment, the regularization coefficient C=1.0 is set, and the Sigmoid function is used for mapping. The time-frequency defect index of all moments in the preset period before the current moment is randomly divided into a training set and a test set in a ratio of 7:3. The training set is used to fit the logistic regression model, and the test set is used to verify the generalization performance of the model, and the abnormal probability of the composite ultrasonic signal at each moment is output.

[0067] It should be noted that the value of the preset time period length is set manually. In this embodiment, the length of the preset time period is 0.1s. In actual application, as other implementation methods, the implementer can also set it by himself based on the specific situation. This embodiment does not impose any special restrictions.

[0068] Among them, the Sigmoid function and the logistic regression algorithm are both well-known technologies, and the concept of the Sigmoid function and the specific process of using the logistic regression algorithm to evaluate the abnormal probability of the composite ultrasound signal will not be repeated here.

[0069] Preferably, the abnormal probability extraction process diagram provided in this embodiment is as follows Figure 2 shown.

[0070] (2) Furthermore, this embodiment clusters the time-frequency defect indices at the current moment and all moments within a preset period before the current moment to obtain normal clusters. Specifically:

[0071] In this embodiment, the time-frequency defect index of all moments in a preset period before the current moment is used as the input of the clustering algorithm, wherein the absolute value of the difference between the time-frequency defect indices is used as the metric distance of the clustering algorithm, the number of clusters is set to q, q clusters are output, the mean time-frequency defect index of each cluster is calculated, and the cluster with the smallest mean video defect index is used as the normal cluster.

[0072] It should be noted that there are many commonly used clustering algorithms. In this embodiment, the k-means clustering algorithm is used to cluster the time-frequency defect index. In actual application, as other implementation methods, implementers can also use other clustering methods such as the DPC density peak clustering algorithm based on specific circumstances. Regarding the selection of clustering algorithms, this embodiment does not impose any special restrictions.

[0073] The k-means clustering algorithm is a well-known technology, and the specific process of clustering the time-frequency defect index using it will not be described in detail.

[0074] It is supplemented that the value of the number of clusters q is artificially set. In this embodiment, the value of the number of clusters q is 2. The implementer can also set it by himself according to the specific situation. This embodiment does not impose any special restrictions.

[0075] (3) Furthermore, this embodiment determines the defect characteristic value of the composite ultrasonic signal at the current moment by analyzing the difference between the time-frequency defect index of the composite ultrasonic signal at the current moment and the cluster center of the normal cluster, and combining the abnormal probability, specifically:

[0076] As an implementation manner, in this embodiment, the expression of the defect characteristic value of the composite ultrasonic signal at the current moment is: Where D represents the difference between the time-frequency defect index of the composite ultrasonic signal at the current moment and the cluster center of the normal cluster; P n represents the abnormal probability of the composite ultrasonic signal at time n within the preset time period before the current time; N represents the length of the preset time period; norm() represents the normalization function.

[0077] It should be noted that there are many methods for measuring the differences between data. In this embodiment, the absolute value of the difference between the time-frequency defect index of the composite ultrasonic signal at the current moment and the cluster center of the normal cluster cluster is used as the difference between the time-frequency defect index of the composite ultrasonic signal at the current moment and the cluster center of the normal cluster cluster. In actual application, as other implementation methods, the implementer may also adopt other methods for measuring the differences between data, such as the square or ratio of the difference, based on the specific circumstances. This embodiment does not impose any special restrictions on the selection of methods for measuring the differences between data.

[0078] According to the defect characteristic value of the composite ultrasonic signal at the current moment, it can be understood that the defect characteristic value reflects the possibility of defects in the heating pipe. If the difference between the time-frequency defect index of the composite ultrasonic signal at the current moment and the cluster center of the normal cluster is greater, it means that the energy distribution of the current composite ultrasonic signal is more different from the historical normal state. At this time, the heating pipe is more likely to be damaged. The larger the time-frequency defect index, the more similar the signal characteristics corresponding to the composite ultrasonic signal and the historical defect samples are, indicating that the heating pipe is more likely to have defects at the current moment, and the larger the defect characteristic value finally obtained;

[0079] On the contrary, if the difference between the time-frequency defect index of the composite ultrasonic signal at the current moment and the cluster center of the normal cluster is smaller, it means that the energy distribution of the current composite ultrasonic signal is smaller than the historical normal state. At this time, the possibility of damage to the heating pipe is smaller, and the smaller the time-frequency defect index is, the greater the difference between the signal characteristics corresponding to the composite ultrasonic signal and the historical defect samples is, indicating that the possibility of defects in the heating pipe at the current moment is smaller, and the final defect characteristic value is smaller.

[0080] At this point, the abnormality probability is evaluated using a logistic regression model by comprehensively considering the time domain discreteness, frequency domain peak extreme distribution and time-frequency defect index of the composite ultrasonic signal at each moment; further, the time-frequency defect index at the current moment and the previous preset time period is clustered to determine the normal clustering cluster. By calculating the difference between the time-frequency defect index at the current moment and the center of the normal clustering cluster and combining it with the abnormality probability, the defect characteristic value is obtained, which effectively overcomes the problem of insufficient adaptability of the time-frequency defect index at a single moment to complex working conditions. By introducing multi-dimensional feature analysis in the time domain and frequency domain and cluster comparison of historical data, the accuracy and reliability of heating pipe damage detection are improved.

[0081] Step S4: Based on the defect characteristic value of the composite ultrasonic signal at the current moment, the ability of the intelligent sensor array to capture the defect echo in the deep layer of the heating pipe is adjusted.

[0082] Due to the significant difference in acoustic impedance characteristics between non-metallic pipe materials and metals, as well as the multiple scattering effect of sound waves caused by the multi-layer wrapping structure of heating pipes, traditional ultrasonic monitoring methods have problems in application. Due to the insufficient penetration of high-frequency ultrasonic waves in non-metallic pipes, the echo signal intensity of defects such as tiny cracks and holes in the deep areas of thick-walled or large-diameter pipes drops sharply, or is even completely masked by background noise, which will cause serious missed detection of heating pipe damage.

[0083] Therefore, this embodiment is based on the defect characteristic value B, and dynamically adjusts the hardware parameters and signal processing strategies of the detection system by real-time calculation of the numerical changes of the defect characteristic value B, specifically including: in this embodiment, the preset threshold is set to 0.75. When the defect characteristic value B exceeds the preset threshold, the intelligent sensor array automatically switches to the low-frequency mode and adjusts the gain to achieve dynamic optimization of penetration depth and signal sensitivity. Specifically, the heating pipe damage detection system based on ultrasonic technology automatically triggers the intelligent sensor parameter adaptation mechanism, and the intelligent sensor array automatically switches to the low frequency band, such as: 20kHz-50kHz, to capture the echo of deep defects in the heating pipe, and at the same time increase the gain value of the multi-channel signal acquisition module to enhance the penetration depth of ultrasonic waves in non-metallic pipes, compensate for energy attenuation, and ensure the effective capture of deep defect echoes; conversely, when the defect characteristic value B is less than or equal to the preset threshold of 0.75, the intelligent sensor array is not adjusted.

[0084] The value of the preset threshold is set manually, and the implementer can also set it according to the specific situation. This embodiment does not impose any special restrictions.

[0085] Thus, this embodiment acquires multi-source signals through an intelligent sensor array, extracts time-domain and frequency-domain features using wavelet transform and principal component analysis, and evaluates damage probability by calculating energy distribution and energy contribution values ​​in combination with a time-frequency defect index. Furthermore, a normal state model is established through cluster analysis, defect characteristic values ​​at the current moment are calculated, and intelligent sensor array parameters are dynamically adjusted to optimize detection performance. This effectively integrates multi-dimensional information and improves the accuracy and reliability of non-metallic pipeline damage detection.

[0086] Based on the same inventive concept as the above method, an embodiment of the present application also provides a heating pipe damage detection system based on ultrasonic technology, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, the steps of any one of the above-mentioned heating pipe damage detection methods based on ultrasonic technology are implemented.

[0087] It should be noted that the order in which the embodiments of the present application are presented is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. Furthermore, the foregoing descriptions of specific embodiments of this specification are provided. Furthermore, the processes depicted in the accompanying drawings do not necessarily require the specific order or sequential sequence shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0088] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments.

[0089] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the principles of the present application shall be included in the scope of protection of the present application.

Claims

1. A method for detecting damage to heating pipes based on ultrasonic technology, characterized in that: The method comprises the following steps: In the heating pipes, an intelligent sensor array is used to obtain composite ultrasonic signals within a preset time period before each moment; Decomposing the composite ultrasonic signal into components at multiple scales, analyzing the energy distribution of the components at each scale, and determining the energy distribution value of the composite ultrasonic signal at each moment; determining the energy contribution value of the composite ultrasonic signal at each moment based on the energy contribution of the main characteristic components of the composite ultrasonic signal in the frequency domain to the composite ultrasonic signal, and determining the time-frequency defect index of the composite ultrasonic signal at each moment in combination with the energy distribution value; Analyzing the degree of dispersion of all data in the time domain and the extreme distribution of all peaks in the frequency domain of the composite ultrasonic signal at each moment, and combining the time-frequency defect index, to evaluate the abnormal probability of the composite ultrasonic signal at each moment; clustering the time-frequency defect indices at the current moment and all moments within a preset time period before the current moment to obtain normal clusters; determining the defect characteristic value of the composite ultrasonic signal at the current moment by analyzing the difference between the cluster center of the normal cluster and the time-frequency defect index of the composite ultrasonic signal at the current moment, and combining the abnormal probability; Based on the defect characteristic value, the ability of the intelligent sensor array to capture defect echoes deep in the heating pipe is adjusted.

2. The method for detecting damage to heating pipes based on ultrasonic technology according to claim 1, characterized in that: The method for obtaining the composite ultrasonic signal is as follows: An intelligent sensor array is used to collect various signals of the heating pipes at each moment and within a preset time period before that, and the signal-to-noise ratio of various signals is obtained. The product of the normalized signal intensity value of the frequency domain signals of various signals at each frequency and the signal-to-noise ratio is calculated. The frequency domain signals of all kinds of signals at the same frequency are added together to obtain the signal intensity at the corresponding frequency of the frequency domain signal of the composite ultrasonic signal. The frequency domain signal of the composite ultrasonic signal is converted into the time domain to obtain a composite ultrasonic signal, wherein the various signals include: low-frequency ultrasonic echoes of non-metallic pipes, high-frequency ultrasonic echoes of metal pipes, environmental noise signals and pipeline mechanical vibration signals.

3. The method for detecting damage to heating pipes based on ultrasonic technology according to claim 1, characterized in that: The method for determining the energy distribution value of the composite ultrasonic signal at each moment is: The composite ultrasonic signal at each moment is used as the input of the wavelet transform, and the components at a preset number of scales are output. The energy mean of the components at each scale is calculated, and the cumulative sum of the energy mean of the components at the preset number of scales is used as the energy distribution value of the composite ultrasonic signal at each moment.

4. The method for detecting damage to heating pipes based on ultrasonic technology according to claim 1, wherein: The method for determining the energy contribution value of the composite ultrasonic signal at each moment is: The composite ultrasonic signal at each moment is used as the input of principal component analysis, the variance contribution rates of all principal components are output, and the sum of the first preset number of variance contribution rates in the results of descending order of variance contribution rates is used as the energy contribution value of the composite ultrasonic signal at each moment.

5. The method for detecting damage to heating pipes based on ultrasonic technology according to claim 1, characterized in that: The time-frequency defect index of the composite ultrasonic signal at each moment is the result of forward fusion of the energy distribution value and the energy contribution value of the composite ultrasonic signal at each moment.

6. The method for detecting damage to heating pipes based on ultrasonic technology according to claim 1, characterized in that: The evaluating of the abnormal probability of the composite ultrasonic signal at each moment includes: The variance of all data of the composite ultrasonic signal in the time domain at each moment, the maximum peak value in the frequency domain, and the time-frequency defect index at each moment are used as inputs of the logistic regression algorithm to output the abnormal probability of the composite ultrasonic signal at each moment.

7. The method for detecting damage to heating pipes based on ultrasonic technology according to claim 1, wherein: The normal cluster is: a cluster with the smallest mean value of the time-frequency defect index among all clusters obtained by clustering the time-frequency defect indexes at the current moment and all moments within a preset period before the current moment.

8. The method for detecting damage to heating pipes based on ultrasonic technology according to claim 1, characterized in that: The expression of the defect characteristic value of the composite ultrasonic signal at the current moment is: Where D represents the difference between the time-frequency defect index of the composite ultrasonic signal at the current moment and the cluster center of the normal cluster; P n represents the abnormal probability of the composite ultrasonic signal at time n within the preset time period before the current time; N represents the length of the preset time period; norm() represents the normalization function.

9. The method for detecting damage to heating pipes based on ultrasonic technology according to claim 1, wherein: The ability of adjusting the intelligent sensor array to capture defect echoes deep in the heating pipe includes: If the defect characteristic value of the composite ultrasonic signal at the current moment is greater than the preset threshold, the intelligent sensor array automatically adjusts to the low-frequency mode to capture the echo of the deep defect of the heating pipe. Otherwise, the intelligent sensor array does not adjust.

10. A heating pipe damage detection system based on ultrasonic technology, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the steps of the heating pipe damage detection method based on ultrasonic technology as described in any one of claims 1 to 9 are implemented.

Images