Leakage detection method of trenchless underground water supply and drainage network based on multi-data fusion

Abstract

This invention relates to the field of intelligent detection technology, and more specifically, to a method for detecting leakage in trenchless underground water supply and drainage pipe networks based on multi-data fusion. The method includes: Step 1: Deploying acoustic sensors at equal intervals along the pipe network; each acoustic sensor collects the acoustic time-domain signal generated by pipe leakage, and then calculating the spectral intensity of the acoustic time-domain signal at the pipe's characteristic frequency; Step 2: Calculating the cross-correlation function between two different acoustic sensors based on the spectral intensity; determining the acoustic delay of the acoustic time-domain signal between different acoustic sensors by finding the peak position of the cross-correlation function; Step 3: Calculating the location of the leakage point based on the acoustic delay and the principle of sound wave propagation; and assessing the leakage volume based on the location of the leakage point and combined with pipe usage data. This method, through multi-data fusion and intelligent analysis, effectively combines trenchless operation, high sensitivity, precise positioning, and scientific quantitative assessment.

Description

Technical Field

[0001] This invention belongs to the field of intelligent detection technology, specifically relating to a method for detecting leakage in trenchless underground water supply and drainage networks based on multi-data fusion. Background Technology

[0002] Underground water supply and drainage networks are a crucial component of urban infrastructure, undertaking key functions such as water supply, drainage, and sewage treatment. However, due to factors such as long-term underground burial, extended service life, and complex environments, these networks commonly face leakage problems. Statistics show that the average leakage rate of urban water supply systems worldwide reaches 30% to 40%, with some older urban areas even exceeding 50%. Network leakage not only wastes water resources and increases energy consumption but can also lead to serious consequences such as ground subsidence and groundwater pollution, posing a significant threat to urban safety and the ecological environment. Therefore, accurately and efficiently detecting and locating leakage points in underground pipe networks is of great importance for ensuring the safe operation of urban infrastructure and improving resource utilization efficiency.

[0003] Traditional methods for detecting leaks in underground pipelines mainly include direct excavation inspection, pressure testing, tracer methods, and acoustic detection. Direct excavation inspection is simple and intuitive, but it is highly destructive, costly, inefficient, and causes serious disruption to urban traffic and residents' lives. Pressure testing determines the presence of leaks by monitoring pressure changes within the pipe section, offering some detection sensitivity, but it struggles to accurately pinpoint leak locations and is ineffective at detecting small leaks. Tracer methods involve injecting specific tracer substances (such as fluorescent dyes or radioactive isotopes) into the pipeline and then monitoring for leaks at the surface. While offering high detection sensitivity, it is complex, costly, and poses environmental pollution risks. Acoustic detection is one of the most widely used trenchless leak detection technologies, primarily based on acoustic signals generated at the leak site for detection and location. This method offers advantages such as no excavation required, fast detection speed, and minimal environmental interference. Existing acoustic detection techniques mainly include listening leaks, correlation methods, and noise recording methods. The sound-listening method uses specialized equipment (such as electronic leak detectors) to listen for leaks on the ground or at pipe fittings. It is simple to operate, but its accuracy depends heavily on the operator's experience and is severely affected by ambient noise. The noise recording method deploys multiple noise recorders in the pipe network system and analyzes changes in the recorded acoustic signal intensity to determine the leak location. It has the advantage of a high degree of automation, but its location accuracy is limited. Summary of the Invention

[0004] The main objective of this invention is to provide a trenchless underground water supply and drainage network leakage detection method based on multi-data fusion. This method achieves an effective combination of trenchless operation, high sensitivity, precise positioning, and scientific quantitative assessment through multi-data fusion and intelligent analysis, providing a more efficient, accurate, economical, and comprehensive solution for underground pipeline network leakage management.

[0005] To solve the above problems, the technical solution of the present invention is implemented as follows:

[0006] A method for detecting leakage in trenchless underground water supply and drainage networks based on multi-data fusion, the method comprising:

[0007] Step 1: Deploy acoustic sensors at equal intervals along the pipeline network. Each acoustic sensor collects the acoustic time-domain signal generated by pipeline leakage, and then calculates the spectral intensity of the acoustic time-domain signal at the characteristic frequency of the pipeline.

[0008] Step 2: Calculate the cross-correlation function between two different acoustic sensors based on the spectral intensity. By finding the peak position of the cross-correlation function, determine the acoustic delay of the acoustic time-domain signal between different acoustic sensors.

[0009] Step 3: Calculate the location of the leak point based on acoustic delay and the principle of sound wave propagation; assess the leakage volume based on the location of the leak point and the pipeline usage data.

[0010] Furthermore, pipeline usage data includes: pipeline service time (t). service The unit is year; the pipe wall thickness T is in meters; the pipe elastic modulus E is in Pa; the pipe diameter D is... pipe The unit is meters; the soil cover depth H soil The unit is m; the density of the soil cover ρ soil The unit is kg / m³ 3 Poisson's ratio v of the pipe material; speed of sound propagation c in the pipe material. pipe The unit is m / s; the internal pressure of the pipeline, P. int The unit is Pa.

[0011] Furthermore, the characteristic frequency f of the pipeline is:

[0012]

[0013] Where, ρ water This is the density of water, expressed in kg / m³. 3 .

[0014] Furthermore, the Poisson's ratio (v) ranges from 0.21 to 0.26 for cast iron pipes; from 0.27 to 0.30 for steel pipes; from 0.31 to 0.34 for copper pipes; from 0.35 to 0.38 for PVC pipes; from 0.39 to 0.42 for PE pipes; from 0.43 to 0.45 for PP pipes; from 0.46 to 0.48 for ABS pipes; from 0.15 to 0.20 for concrete pipes; and from 0.10 to 0.14 for fiberglass pipes.

[0015] Furthermore, the acoustic time-domain signal s acquired by the i-th acoustic sensor i (t) Spectral intensity S at the characteristic frequency f of the pipe i (f) is:

[0016]

[0017] Where t0 is the start time of acquiring acoustic time-domain signals, in seconds; t1 is the stop time of acquiring acoustic time-domain signals, in seconds; t is the time variable, in seconds; z is the imaginary number sign; C soil The speed of sound wave propagation in the soil cover is expressed in m / s.

[0018] Furthermore, the cross-correlation function R between the i-th acoustic sensor and the j-th acoustic sensor ij (τ) is:

[0019]

[0020] Where τ is the time delay variable, in seconds; f max f is the upper limit of the characteristic frequency; min This is the lower limit of the characteristic frequency. The acoustic time-domain signal s acquired by the j-th acoustic sensor i (t) Spectral intensity S at the characteristic frequency f of the pipe j (f) conjugate.

[0021] Furthermore, the acoustic delay Δt is:

[0022]

[0023] Furthermore, the location of the leakage point L leak for:

[0024]

[0025] Among them, the location of the leakage point L leak This represents the distance measured along the pipe axis from a reference starting point, which is the position of the first acoustic sensor; g is the acceleration due to gravity, measured in m / s². 2 ;v soil For the soil cover, Poisson's ratio; L total This represents the distance between the acoustic sensor and other adjacent acoustic sensors.

[0026] Furthermore, the leakage rate Q leak for:

[0027]

[0028] Among them, L i Let t be the position of the i-th acoustic sensor along the pipe axis; ref Reference service life for pipelines; pH is the pH value of the cover soil; Cl - The chloride ion concentration in the cover soil is expressed in mg / kg; pH ref For reference pH value; For reference chloride ion concentration.

[0029] The trenchless underground water supply and drainage network leakage detection method based on multi-data fusion of this invention has the following beneficial effects: First, this method breaks through the limitations of traditional acoustic detection technology. By calculating the spectral intensity of the acoustic time-domain signal at the characteristic frequency of the pipeline, it significantly improves the signal-to-noise ratio and the accuracy of feature extraction. Especially in complex underground environments and under different pipe material conditions, this method can effectively identify and extract leakage signal features, making the detection results more reliable and stable. Second, this method innovatively introduces the concept of pipeline characteristic frequency and establishes a complete calculation model. By considering parameters such as the sound wave propagation speed, diameter, wall thickness, elastic modulus, and Poisson's ratio of the pipeline material, the characteristic frequencies of different pipelines are accurately calculated, laying the foundation for subsequent spectral analysis and signal processing. In particular, it provides detailed Poisson's ratio value ranges for different pipe materials, enabling the detection system to adapt to various pipeline network environments. Third, this method significantly improves the accuracy of leakage location by constructing a cross-correlation function that considers environmental factors. Compared to traditional methods, this cross-correlation function incorporates environmental parameters such as cover depth, cover density, and sound wave propagation velocity, effectively compensating for changes in sound wave propagation characteristics in complex underground environments and reducing location errors. Fourth, this method not only achieves precise leakage location but also accurately assesses leakage volume, providing a quantitative basis for pipeline maintenance. The leakage assessment model comprehensively considers pipeline physical parameters, acoustic signal characteristics, service time, and environmental factors. In particular, it innovatively incorporates the influence of cover pH and chloride ion concentration on pipeline aging, making the assessment results more comprehensive and accurate. Finally, this method achieves trenchless detection throughout the entire process, minimizing interference with urban operations and residents' lives. The detection process is efficient and rapid, significantly reducing detection costs and time, and improving pipeline maintenance efficiency. Furthermore, this method is applicable to various pipe materials and burial conditions, showing broad application prospects. Attached Figure Description

[0030] Figure 1 This is a schematic diagram of the method flow for a trenchless underground water supply and drainage network leakage detection method based on multi-data fusion, provided in an embodiment of the present invention. Detailed Implementation

[0031] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0032] Example 1, Reference Figure 1A method for detecting leakage in trenchless underground water supply and drainage networks based on multi-data fusion, the method comprising:

[0033] Step 1: Deploy acoustic sensors at equal intervals along the pipeline network. Each acoustic sensor collects the acoustic time-domain signal generated by pipeline leakage, and then calculates the spectral intensity of the acoustic time-domain signal at the characteristic frequency of the pipeline.

[0034] Step 2: Calculate the cross-correlation function between two different acoustic sensors based on the spectral intensity. By finding the peak position of the cross-correlation function, determine the acoustic delay of the acoustic time-domain signal between different acoustic sensors.

[0035] Step 3: Calculate the location of the leak point based on acoustic delay and the principle of sound wave propagation; assess the leakage volume based on the location of the leak point and the pipeline usage data.

[0036] Underground water supply and drainage pipe networks are a crucial component of urban infrastructure, and their safe and stable operation is vital for ensuring residents' lives and industrial production. However, due to various factors such as pipe aging, corrosion, and external damage, pipe network leakage occurs frequently. Traditional leakage detection methods, such as listening rods and correlators, often rely on manual experience, resulting in low efficiency and poor performance for deeply buried pipes or those in noisy environments. Excavation-based detection is not only costly and time-consuming but also significantly impacts traffic and the environment. Therefore, developing an efficient, accurate, and trenchless intelligent method for detecting leakage in underground pipe networks has significant practical implications and application value. A trenchless underground water supply and drainage pipe network leakage detection method based on multi-data fusion is a technical solution proposed to address this challenge. It utilizes modern sensing technology, signal processing technology, and data fusion concepts to achieve rapid location and quantitative assessment of underground pipe leakage.

[0037] The core of this method lies in utilizing acoustic principles. When pressurized fluid (usually water) inside a pipe is ejected from a leak point, it interacts in complex ways with the pipe wall and surrounding medium (such as soil), generating vibrations and sound waves. These sound waves propagate along multiple paths, including the pipe wall, the internal fluid, and the external soil. The first step of this method is to uniformly deploy a series of highly sensitive acoustic sensors at predetermined, equal intervals along the pipeline network to be monitored. These sensors, like a doctor's stethoscope, are placed close to or near the pipe (usually installed at contact points such as valves and fire hydrants, or drilled to approach the pipe body), specifically designed to capture the weak acoustic signals that may be generated by pipe leaks. The purpose of uniform deployment is to form an effective monitoring network, ensuring that signals can be received by multiple sensors, providing the necessary data redundancy and spatial information for subsequent location calculations. Each sensor operates independently, acquiring acoustic vibration signals at its location in real time and converting these continuously changing vibrations into a time-domain signal data stream. These raw time-domain signals contain a variety of information, including target leakage signals, as well as environmental noise (such as vehicle movement and ground construction) and the operating noise of the pipeline system itself (such as water flow noise and pump station vibration). Therefore, directly analyzing the raw time-domain signals often makes it difficult to effectively identify leakage.

[0038] To extract useful leakage information from complex mixed signals, signal processing is required. The unique aspect of this method is that it doesn't analyze the entire frequency band, but instead focuses on a specific frequency—the pipe's characteristic frequency. This characteristic frequency is not arbitrarily chosen but closely related to the pipe's physical properties, such as the pipe's material (which determines the speed of sound propagation within the pipe wall and the material's elastic modulus and Poisson's ratio), the pipe's diameter, wall thickness, and the density of the medium (water) inside the pipe. The pipe structure itself can be considered as a filter or resonator; the energy generated by leakage may be more concentrated or prominent at certain structure-related frequencies. Determining this characteristic frequency through calculation or experience allows the focus of analysis to be concentrated on the frequency band most likely to carry leakage information. Calculating the spectral intensity at a specific frequency essentially utilizes signal processing techniques such as Fourier transform to analyze the energy level of the original time-domain signal at that characteristic frequency point. This step not only effectively extracts the signal components related to the pipe structure's response but also initially suppresses noise in other frequency bands. More importantly, this method introduces the concept of multi-data fusion when calculating spectral intensity. It uses factors such as pipe geometry (diameter, wall thickness), material properties (elastic modulus), operating parameters (internal pressure), and external environmental information (cover depth, density, sound wave propagation speed in soil) as adjustment factors or weights to correct or standardize the calculated spectral intensity. This means that the obtained spectral intensity value not only reflects the acoustic signal energy but also indirectly reflects the relative significance of that energy level under current pipe, pressure, and soil conditions. This makes spectral intensities measured at different locations and under different conditions more comparable, laying the foundation for subsequent analysis.

[0039] After obtaining the spectral intensity information at each sensor location and performing preliminary processing and fusion, the next step is to determine the time difference between the arrival of the leakage signal at different sensors, i.e., the acoustic delay. This is a crucial step in achieving leakage localization. This method employs cross-correlation function analysis. Cross-correlation is a classic method for measuring the similarity and time lag relationship between two signals. Specifically, it involves selecting spectral intensity signals (which can be understood as a narrowband signal or its representative value near a characteristic frequency point) calculated from two different sensors (usually adjacent or a pair of sensors separated by a certain distance) and calculating their cross-correlation function. The value of the cross-correlation function changes with the time shift (delay) of one signal relative to the other. The cross-correlation function reaches its peak when the similar components in the two signals (mainly those from the same leakage source) are time-aligned. The horizontal axis (time delay) corresponding to this peak precisely represents the difference in time required for the leakage sound wave to propagate to the two sensors. For example, if the cross-correlation peak occurs at a positive delay time, it indicates that the signal arrives at the first sensor first; if it occurs at a negative delay time, it indicates that the signal arrives at the second sensor first. The magnitude of the peak value also reflects the correlation and strength of the signal. When calculating the cross-correlation function, the concept of data fusion can also be introduced. For example, parameters related to the pipeline, soil, and pressure can be used to weight or correct the cross-correlation calculation process to improve the robustness and accuracy of the calculation. By systematically calculating the cross-correlation function between multiple sensor pairs on the pipeline network and finding the peak value, a series of acoustic delay data can be obtained.

[0040] After obtaining the precise acoustic delay, the third step is to use this time difference information to estimate the exact location of the leak. Its basic principle is similar to satellite navigation systems or earthquake source location, both belonging to location methods based on the time difference of signal arrival. Assuming the leak point is located in the pipe between two sensors, given the distance between the two sensors, the time difference of sound wave propagation from the leak point to the two sensors (i.e., the acoustic delay), and the effective propagation speed of the sound wave in the propagation medium, the precise location of the leak point relative to the two sensors can be calculated using simple geometric relationships or algebraic equations. However, in the actual situation of underground pipelines, the propagation path and speed of sound waves are complex. Sound waves may primarily propagate along the pipe wall, or through the water inside the pipe, or through the surrounding soil, or even a combination of these paths. The propagation speed of each path is different and is affected by various factors such as pipe material, size, internal pressure, water properties, and soil type, density, and water content. Therefore, to achieve accurate location, in-depth multi-data fusion is necessary again. The method requires comprehensive consideration of the propagation speed of sound waves in pipe materials, water, and soil, as well as the properties of these media (such as the pipe's elastic modulus, Poisson's ratio, soil density, Poisson's ratio, and cover depth). This method utilizes the principles of sound wave propagation, combined with previously calculated acoustic delays and various physical parameters provided in the pipe usage data, to establish a physical model or calculation formula that more closely reflects actual propagation conditions, thereby determining the location of leakage points. This calculation process demonstrates a high degree of integration between acoustic measurement data and the physical model of the pipe system, as well as environmental parameters.

[0041] Simply locating the leak point is insufficient; assessing the magnitude of the leak is equally important for deciding whether and how to repair it immediately. The final part of this method involves combining the located leak point with pipe usage data to assess the leakage volume. The principle behind leak volume assessment is generally based on the assumption that the severity of the leak (i.e., the size of the leak) is correlated with the intensity of the acoustic signal generated by the leak. Generally, larger leaks produce stronger acoustic signals at the same pressure. However, sound waves attenuate during propagation, and the signal strength received by sensors farther from the leak point is weaker. Therefore, the leakage volume cannot be simply determined directly from the received signal strength. This method takes this into account, utilizing previously calculated spectral intensity values ​​from each sensor (especially those closest to the leak point) and combining them with the determined leak point location L. leak and the position L of each sensor i Calculate the signal propagation distance |L i -L leak Then, based on the pipe's material and dimensions (diameter D)... pipeA sound wave attenuation model is established based on factors such as wall thickness (T) and elastic modulus (E) to compensate for the measured spectral intensity and estimate the sound intensity at the leakage source. Furthermore, the leakage rate is also related to the internal pressure (P) of the pipe. int Directly related: the higher the pressure, the greater the leakage from the same leak point. Pipeline service life t service and the chemical properties of the surrounding soil environment (such as pH value, chloride ion concentration Cl). - This also affects the corrosion status of the pipeline and the formation and development of leaks, thus influencing the relationship between leakage volume and acoustic signal intensity. This method creatively incorporates these factors into a comprehensive model, integrating acoustic measurement data (spectral intensity, location information), pipeline physical data (size, material, pressure), pipeline historical data (service time), and environmental chemical data (soil pH, chloride ion concentration) to ultimately provide a correlation between leakage volume Q and the acoustic signal intensity. leak This quantitative assessment, achieved through the fusion of multi-dimensional data, allows for a more comprehensive understanding of the physical processes and influencing factors of the entire leakage event, rather than solely relying on sound intensity. This enhances the accuracy and reliability of the assessment.

[0042] Furthermore, pipeline usage data includes: pipeline service time (t). service The unit is year; the pipe wall thickness T is in meters; the pipe elastic modulus E is in Pa; the pipe diameter D is... pipe The unit is meters; the soil cover depth H soil The unit is m; the density of the soil cover ρ soil The unit is kg / m³ 3 Poisson's ratio v of the pipe material; speed of sound propagation c in the pipe material. pipe The unit is m / s; the internal pressure of the pipeline, P. int The unit is Pa.

[0043] Pipeline service time t service This parameter, obtained by reviewing pipeline laying records, reflects the aging degree of the pipeline and directly affects the fatigue state of the pipeline material and the likelihood of leakage. As service time increases, pipeline materials deteriorate to varying degrees, leading to a decrease in pipe wall strength and thus increasing the risk of leakage. In trenchless inspection, this parameter serves as a crucial input to the leakage assessment model, used to correct for calculating leakage. Pipeline wall thickness T and diameter D... pipeThese parameters can usually be obtained from pipeline design drawings or measured at exposed points on the pipeline using an ultrasonic thickness gauge. For pipe sections completely buried underground, non-destructive testing using ground-penetrating radar technology can be used. These two parameters are closely related to the structural strength of the pipeline, affecting the propagation characteristics of sound waves in the pipeline and the spectral characteristics of leakage signals. The pipeline's elastic modulus E and Poisson's ratio v are important parameters describing the mechanical properties of the pipeline material, usually obtained from standard material handbooks depending on the pipeline material type. For pipelines with long service lives, these parameters may change and need to be corrected through sampling inspection or on-site dynamic response testing. In leakage detection, these parameters affect the propagation speed and modal characteristics of sound waves, thus affecting the accuracy of leakage location. Soil cover depth H soil and soil cover density ρ soil These parameters can be obtained through geological exploration or estimated using pipeline construction records and geological data. They affect the intensity of the leakage signal received by surface acoustic sensors; the thicker and denser the overburden layer, the more significant the signal attenuation. In trenchless detection methods, these parameters are used to compensate for signal attenuation and ensure accurate location of the leakage. The speed of sound propagation in the pipeline material, c... pipe This is the core parameter for leak location calculation, which can be obtained through theoretical calculation or field calibration. Theoretical calculation is based on the elastic modulus and density of the pipe material, while field calibration is determined by measuring the sound wave propagation time through a tapping test on the pipe at a known location. This parameter directly determines the conversion relationship between acoustic delay and leak location, and is a key factor in the accuracy of leak location. The internal pressure P of the pipe... int This is a real-time parameter reflecting the operating status of the pipeline network, typically obtained through real-time monitoring by pressure sensors installed at key nodes in the network. Fluctuations in internal pressure affect the acoustic signal characteristics generated by leaks; higher internal pressures result in increased frequency and intensity of the leak's acoustic signal. In leak detection systems, this parameter is used to adjust the leak identification threshold and estimate the leakage volume.

[0044] Furthermore, the characteristic frequency f of the pipeline is:

[0045]

[0046] Where, ρ water This is the density of water, expressed in kg / m³. 3 .

[0047] According to the theory of thin-walled cylinders, the deformation of a cylinder during radial vibration is related to the change in internal pressure. The governing equation for the radial vibration of a cylinder can be derived using the Lagrange equation. When establishing the Lagrange equation for pipe vibration, the kinetic and potential energy of the pipe needs to be considered. The kinetic energy of the pipe is mainly determined by the mass of the pipe wall and the vibration velocity, while the potential energy consists of the elastic potential energy of the pipe wall. For the radial vibration of the pipe wall, its kinetic energy can be expressed as… Where ρ pipe It is the density of the pipe material, v r Where is the radial vibration velocity, and L is the pipe length. The elastic potential energy of the pipe wall can be expressed as... Where u r It is radial displacement.

[0048] When considering fluid-structure coupling, the internal fluid exerts pressure on the pipe wall, which causes radial vibration of the pipe wall. According to fluid dynamics theory, the pressure fluctuation of the internal fluid can be expressed as... Where ρ water It is the density of water, c water Let be the speed of sound propagation in water, and u be the fluid displacement vector. Combining fluid pressure with the elastic deformation of the pipe wall, a fluid-structure coupling equation is established. In the simplified one-dimensional model, the wave equation for the radial vibration of the pipe can be obtained:

[0049] A variational analysis is performed on this wave equation, assuming the displacement u r The form is u r =Asin(kx-ωt), where k is the wave number and ω is the angular frequency. Substituting this expression into the wave equation, we obtain the characteristic equation: From the characteristic equation, we can obtain the relationship between the angular frequency ω and the wave number k: Considering the relationship between wavenumber k and wavelength λ and the speed of sound wave propagation c in the pipe pipe Relationship with wavelength and frequency c pipe =λf, from which the expression for frequency f can be further derived. The wavenumber k can be expressed as... Substituting into the expression for angular frequency ω (ω = 2πf), we get: Simplifying the above expression, we get:

[0050] In pipe vibration analysis, the mass effect of the pipe wall is usually much smaller than the mass effect of the fluid, i.e., ρ pipe <<ρ water Therefore, it can be approximated as ρ pipe +ρ water ≈ρ water .

[0051] Further simplifying the above expression, we get: Solve for the frequency f: Considering the cyclic boundary conditions and the geometric constraints of the pipe, the wavelength λ is related to the pipe diameter D. pipe There exists a relationship: λ = 2D pipe This means c pipe =λf=2D pipe f, therefore Combining the two expressions for frequency f above, we can obtain: Rearranging the above equations, we get: Solve for c pipe ,get: Substitution The final characteristic frequency expression of the pipeline is obtained as follows: After further mathematical transformations and simplifications, and taking into account empirical correction coefficients in practical applications, the final expression for the pipe characteristic frequency f is:

[0052] Furthermore, the Poisson's ratio ν ranges from 0.21 to 0.26 for cast iron pipes; from 0.27 to 0.30 for steel pipes; from 0.31 to 0.34 for copper pipes; from 0.35 to 0.38 for PVC pipes; from 0.39 to 0.42 for PE pipes; from 0.43 to 0.45 for PP pipes; from 0.46 to 0.48 for ABS pipes; from 0.15 to 0.20 for concrete pipes; and from 0.10 to 0.14 for fiberglass pipes.

[0053] Poisson's ratio (v) is a dimensionless parameter describing the relationship between lateral and axial deformation of a material. Different pipe materials exhibit different Poisson's ratio ranges due to differences in molecular structure and internal composition. Metal pipes, such as cast iron pipes, have a Poisson's ratio between 0.21 and 0.26; steel pipes have a slightly higher ratio, between 0.27 and 0.30; and copper pipes have a ratio between 0.31 and 0.34. The relatively low Poisson's ratio of these metal materials indicates that they exhibit smaller lateral deformation under axial force, which is related to the crystal structure and the strength of internal molecular bonds. Plastic pipes, due to their polymer structure, typically have higher Poisson's ratios. PVC pipes have a Poisson's ratio between 0.35 and 0.38, PE pipes between 0.39 and 0.42, PP pipes between 0.43 and 0.45, and ABS pipes between 0.46 and 0.48. These plastic pipes have a high Poisson's ratio, meaning they exhibit greater lateral deformation under axial stress. This is related to the molecular chain structure and weaker intermolecular forces of the plastic material. Inorganic materials such as concrete pipes and fiberglass pipes have lower Poisson's ratios. Concrete pipes have a Poisson's ratio between 0.15 and 0.20, while fiberglass pipes have an even lower ratio, only 0.10 to 0.14. The lower Poisson's ratio of these materials indicates that they have very small lateral deformation under axial stress, which is closely related to their brittle properties and internal microstructure. In the intelligent detection of pipeline leaks, accurate Poisson's ratio parameters are crucial for calculating the pipe's characteristic frequency, sound wave propagation velocity, and leak location. After the acoustic sensor collects the leak signal, the system needs to select the appropriate Poisson's ratio range according to the pipe material type to ensure the accuracy of subsequent cross-correlation function calculations and leak location estimations. Different Poisson's ratio values ​​significantly affect the propagation mode and velocity of sound waves in the pipe, thus affecting the final leak location accuracy.

[0054] Furthermore, the acoustic time-domain signal s acquired by the i-th acoustic sensor i (t) Spectral intensity S at the characteristic frequency f of the pipe i (f) is:

[0055]

[0056] Where t0 is the start time of acquiring acoustic time-domain signals, in seconds; t1 is the stop time of acquiring acoustic time-domain signals, in seconds; t is the time variable, in seconds; z is the imaginary number sign; C soil The speed of sound wave propagation in the soil cover is expressed in m / s.

[0057] First, for the time-domain signal s acquired by the i-th acoustic sensor i (t) needs to be transformed from the time domain to the frequency domain using a Fourier transform. The Fourier transform is a fundamental mathematical tool for analyzing the frequency characteristics of time-varying signals, and its standard form is: Where j is the imaginary unit and ω is the angular frequency. In practical applications, signals can only be acquired within a finite time window [t0, t1], therefore the Fourier transform is modified to... Considering the angular frequency ω = 2πf, and to maintain sign consistency, the imaginary unit in the formula is represented by z, resulting in the acoustic signal s. i (t) Preliminary spectral expression at frequency f This expression describes the energy distribution of a signal at a specific frequency *f*, but it does not yet consider the influence of the pipeline system and the surrounding environment on sound wave propagation. In a pipeline leakage scenario, the sound wave generated at the leak point propagates to the acoustic sensor through the pipe wall and surrounding soil, and its energy attenuation is affected by various factors along the way. To accurately reflect the spectral intensity of the actual leakage signal, the preliminary spectrum needs to be corrected by introducing the influence factor of the transmission path. According to sound wave propagation theory, when the sound wave propagates from the leak point to the sensor, its energy attenuation is affected by the pipe geometry, material properties, and the properties of the surrounding environment.

[0058] First, consider the pipe size. According to the theory of sound wave propagation in circular pipes, the sound wave energy is proportional to the cube of the pipe diameter, expressed as: This is because the larger the pipe diameter, the smaller the divergence effect of sound waves during propagation, and the slower the energy attenuation. Secondly, the internal pressure P of the pipe... int The pressure affects the initial energy of the sound waves generated by the leak; the higher the pressure, the greater the sound wave energy at the leak point. Therefore, the spectral intensity is directly proportional to the internal pressure. When considering the pipe material properties, the elastic modulus E and wall thickness T are two key parameters. According to the theory of elasticity, the greater the elastic modulus of the material, the smaller the energy loss of the sound wave during propagation; while the greater the wall thickness, the more significant the energy attenuation of the sound wave when penetrating the pipe wall. Combining these two factors, the spectral intensity is directly proportional to the internal pressure. Proportional. Finally, the influence of the soil cover environment on sound wave propagation needs to be considered. Soil cover depth H soil The larger the density ρ, the longer the sound wave travels and the more significant the energy attenuation; soil The larger the impedance of the sound wave in the soil, the greater the energy loss; while the speed of sound propagation C in the soil cover... soil This reflects the acoustic properties of the soil and affects the energy attenuation of sound waves. According to the theory of sound wave propagation in non-homogeneous media, the spectral intensity and... Proportional. Taking all the above factors into account, the complete expression for the spectral intensity is obtained: This formula can be further rigorously derived using wave equation theory. Consider the wave equation in a pipeline system. Where p is the sound pressure level and c is the speed of sound. This is the Laplace operator. Applying the wave equation to a pipe in cylindrical coordinates, and considering the boundary conditions—the elastic boundary of the pipe wall and the impedance boundary of the soil—we can obtain an expression for the energy change of the sound wave during propagation. The wave equation is solved analytically, and the result is substituted into the energy relation E=∫p 2 The relationship between acoustic wave energy and pipe and environmental parameters can be derived from dt. Combining the energy conservation property of Fourier transform, the expression for spectral intensity is finally obtained, consistent with the formula obtained through physical analysis above. This rigorous derivation from the wave equation verifies the theoretical rationality of the spectral intensity formula. In practical applications, this formula enables the system to calculate the standardized spectral intensity based on the acquired raw acoustic signal and known pipe parameters, thus providing accurate frequency domain data for subsequent cross-correlation analysis and leak location. This method eliminates the interference of pipe parameters and environmental factors on spectral analysis, improving the accuracy and reliability of leak detection.

[0059] Furthermore, the cross-correlation function R between the i-th acoustic sensor and the j-th acoustic sensor ij (τ) is:

[0060]

[0061] Where τ is the time delay variable, in seconds; f max f is the upper limit of the characteristic frequency; min This is the lower limit of the characteristic frequency. The acoustic time-domain signal s acquired by the j-th acoustic sensor i (t) Spectral intensity S at the characteristic frequency f of the pipe j (f) conjugate.

[0062] In digital signal processing, the cross-correlation function of two discrete-time signals x[n] and y[n] is defined as... Where n represents the time delay. Extending this definition to continuous time-domain signals, the cross-correlation function of two continuous signals x(t) and y(t) can be expressed as: Where τ is the time delay variable. In the frequency domain, according to the correlation theorem, the cross-correlation function can be calculated using the Fourier transform: Where X(f) and Y(f) are the Fourier transforms of x(t) and y(t) respectively, and Y... * (f) denotes the conjugate of Y(f), This represents the inverse Fourier transform. The definition of the inverse Fourier transform is... Therefore, the cross-correlation function can be expressed as In pipeline leakage detection applications, the focus is on the acoustic signals s collected by the i-th and j-th acoustic sensors.i (t) and s j (t) Cross-correlation within a specific frequency range.

[0063] Since the leakage signal is mainly concentrated in the characteristic frequency range of the pipeline [f min ,f max Within this range, the integration range is limited to this interval, resulting in... Where S i (f) and S j (f) are s i (t) and s j (t) Spectral intensity at frequency f It is S j (f) is the conjugate of z, where z is the imaginary sign. This formula represents the basic cross-correlation calculation, but in actual leakage detection, the influence of environmental and pipeline conditions on the cross-correlation function also needs to be considered. When the leakage sound wave propagates from the leakage point to sensors at two different locations, the signal will be affected by attenuation and phase change along the propagation path. To compensate for these effects and obtain a more accurate time delay estimate, a correction factor needs to be introduced. First, the propagation speed C of the sound wave in the overburden... soil Directly affecting the propagation time of sound waves, the cross-correlation function needs to be related to C. soil Proportional to allow for standardized comparisons under different soil conditions. Secondly, the internal pressure P of the pipeline. int This will affect the initial energy of the leakage signal; the cross-correlation function needs to be related to P. int It is inversely proportional to eliminate the influence of pressure fluctuations on the cross-correlation calculation.

[0064] Finally, pipe diameter D pipe The wall thickness T also affects the propagation characteristics of sound waves in the pipe; the geometric mean of these two parameters... It is inversely proportional to the cross-correlation function. Taking all these factors into account, we obtain the modified expression for the cross-correlation function: The physical meaning of this formula can be further understood through the solution of the wave equation. Consider the equation for sound wave propagation in a pipe. Where p is sound pressure level and c is sound speed. It is the Laplace operator. When a sound wave propagates from the leak point to the sensor, its propagation characteristics are affected by the pipe and soil parameters. By solving the wave equation, the relationship between the sound pressure p and the propagation distance r and time t can be obtained: p Where A is the amplitude, ω is the angular frequency, and k is the wave number. The signals received by the two sensors can be expressed as: and Where r i and r j This is the distance from the leak point to the two sensors. The cross-correlation function of these two signals is... Proportional. Considering Distance difference (r) i -r j The relationship between (r) and time delay τ is (r i -r j The phase term in the cross-correlation function can be written as e = cτ. j2πfτ Amplitude term Influenced by pipeline parameters and soil conditions, the correction factor derived above can be used. Let R be the expression. Thus, through the analytical solution of the wave equation, we can obtain the same expression for the cross-correlation function as described above. In practical applications, the cross-correlation function R... ij The calculation of (τ) requires first obtaining the spectral intensities S of the two sensors. i (f) and S j (f), and then in the characteristic frequency range [f] min ,f max The calculation method involves frequency domain multiplication and integration, followed by the application of a correction factor. This method effectively eliminates interference from environmental and pipeline conditions, improves the accuracy and stability of cross-correlation analysis, and lays the foundation for subsequent acoustic delay determination and leak location.

[0065] Furthermore, the acoustic delay Δt is:

[0066]

[0067] Furthermore, the location of the leakage point L leak for:

[0068]

[0069] Among them, the location of the leakage point L leak This represents the distance measured along the pipe axis from a reference starting point, which is the position of the first acoustic sensor; g is the acceleration due to gravity, measured in m / s². 2 ;ν soil For the soil cover, Poisson's ratio; L total This represents the distance between the acoustic sensor and other adjacent acoustic sensors.

[0070] The acoustic delay Δt is determined based on cross-correlation function analysis. The cross-correlation function Rt ij (τ) describes the relationship between the similarity between the signals acquired by the i-th and j-th sensors as a function of time delay τ. From signal processing theory, when two signals originate from the same sound source but propagate along different paths, the cross-correlation function reaches its maximum value at a specific time delay. This time delay corresponds to the time difference between the sound wave's propagation from the sound source to the two sensors. Therefore, the acoustic delay Δt can be calculated by solving the cross-correlation function R. ijThe time delay corresponding to the maximum value of (τ) is used to determine the value, i.e., Δt = arg max. τ R ij (τ). Here, the arg max operation represents finding the value of the independent variable when the function reaches its maximum value. Mathematically, this can be obtained by solving the condition that the derivative of the cross-correlation function with respect to τ is equal to zero, i.e. and Due to the cross-correlation function It is a complex integral expression, and in practical applications, numerical methods are usually used to calculate its maximum point, such as applying interpolation algorithms after discrete sampling or using fast Fourier transform to improve computational efficiency.

[0071] Once the acoustic delay Δt is determined, the next step is to derive the location L of the leakage point. leak The calculation formula is based on the geometric relationships of sound wave propagation and wave theory. Let two adjacent acoustic sensors be located on the axis of the pipe, L... i and L j Position, the distance between the sensors is L total =|L j -L i |. When the pipe is at position L leak When a leak occurs, the sound waves generated at the leak point will propagate to two sensors, with propagation times of t and t respectively. i and t j The acoustic delay Δt is the difference between these two propagation times, Δt = |t| i -t j According to the principle of sound wave propagation, propagation time is directly proportional to propagation distance, that is... and Where v eff This is the effective propagation speed of sound waves in a pipeline system. Considering the complex propagation path of sound waves in underground pipeline systems, involving the influence of pipeline materials, internal fluids, and surrounding soil, the effective propagation speed v... eff Several factors need to be considered. According to wave dynamics theory, wave velocity in multiphase media is affected by the elastic properties and density of the medium. For buried pipeline systems, the propagation path of sound waves from the leak point to the sensor mainly includes two parts: propagation through the pipe wall and propagation through the soil. The propagation velocity through the pipe wall is mainly determined by the elastic modulus and density of the pipe material, while the propagation velocity through the soil is affected by the cover depth, soil density, and soil elastic properties. According to elastic wave theory, the sound wave propagation velocity C in the soil... soil Related to the elastic properties and density of the soil, it can be expressed as: Where E soil It is the elastic modulus of the soil, ρ soil It is soil density, vsoil It is the Poisson's ratio of the soil. For underground pipeline systems, due to the soil cover depth H... soil The presence of soil cover also means that the propagation of sound waves in the soil is affected by soil static pressure. Static pressure increases with soil depth, causing changes in the sound wave propagation speed. Considering the complexity of underground pipeline systems, the effective propagation speed v... eff It can be represented as Where g is the acceleration due to gravity, this expression reflects the combined influence of soil properties, pipe properties, and internal pressure on sound wave propagation. Substituting the effective propagation velocity into the relationship between acoustic delay and propagation distance, we can obtain... Further transformation yields This indicates that the effective propagation speed can be determined by the sensor spacing and the measured acoustic delay. For the leakage point location L... leak Determining the location requires considering the specific positional relationship between the leakage point and the two sensors. Let the reference starting point be the position L of the first acoustic sensor. i Leakage point location L leak This represents the distance measured along the pipe axis from the reference starting point. The relative location of the leak point can be deduced based on the time difference of the acoustic signals acquired by the two sensors. When the leak point is located between the two sensors, |L leak -L i |-|L leak -L j |=v eff ·Δt. Considering L j =L i +L total And let L i ≤L leak ≤L j Then there is (L) leak -L i )-(L j -L leak ) = v eff ·Δt, i.e. 2L leak -L i -L j =v eff ·Δt. Substitute into L j =L i +L total 2L leak -2L i -L total =v eff ·Δt, i.e. Because of L i It is the reference starting point, which can be set to zero. The effective propagation speed v effSubstituting the expression, we get This is the formula for calculating the location of leaks. It takes into account the geometric characteristics of the pipeline system, soil properties, and sound wave propagation characteristics, and can accurately locate the leak point based on the measured acoustic delay. In practical applications, it is necessary to pay attention to correcting for installation deviations of the acoustic sensors and the complexity of the sound wave propagation path to improve positioning accuracy.

[0072] Furthermore, the leakage rate Q leak for:

[0073]

[0074] Among them, L i Let t be the position of the i-th acoustic sensor along the pipe axis; ref Reference service life for pipelines; pH is the pH value of the cover soil; Cl - The chloride ion concentration in the cover soil is expressed in mg / kg; pH ref For reference pH value; For reference chloride ion concentration.

[0075] First, based on the fundamental principles of fluid mechanics, the flow rate at a pipe leakage point can be expressed by the orifice flow rate formula, i.e. Where Q is the flow rate, and C is the flow rate. d Here, A is the flow coefficient, P is the leak area, and ρ is the pressure difference. In underground pipe network leakage scenarios, the pressure difference is mainly determined by the internal pipe pressure P. int Therefore, the leakage volume is proportional to the square root of the internal pressure of the pipe. Considering the difficulty in directly measuring the leak area in practical engineering, an indirect method is needed for estimation. The intensity distribution of the acoustic signal is an important basis for estimating the leak area. When a pipe leaks, the sound waves generated at the leak point propagate through the pipe and are collected by acoustic sensors deployed along the pipeline. According to acoustic theory, the sound wave energy is related to the intensity of flow disturbance in the medium, which in turn is related to the leak area and flow rate. Therefore, the scale of the leak can be inferred by analyzing the spectral characteristics of the acoustic signal. In frequency domain analysis, the spectral intensity S... i (f) reflects the energy distribution of the acoustic signal acquired by the i-th sensor at a specific frequency f. For acoustic signals caused by leakage, the distribution characteristics of their spectral intensity are directly related to the scale of leakage. In particular, the maximum spectral intensity max measured among all sensors... i |S i (f)|and minimum spectral intensity min i |S iThe ratio of (f)| can reflect the relative scale of leakage. This is because the larger the leakage scale, the stronger the generated sound wave energy, and the more significant the difference in spectral intensity between sensors. According to the theory of sound wave energy attenuation, sound wave energy attenuates exponentially with propagation distance, and the attenuation coefficient is related to the medium properties.

[0076] In actual pipe networks, the spectral intensity ratio It can be used as an indicator of leakage scale, as it is directly proportional to the area of ​​the leak, and therefore also directly proportional to the leakage volume. From the perspective of pipe physical characteristics, the pipe diameter D... pipe This has a significant impact on leakage volume. According to the pipe flow rate formula in fluid mechanics, under the same pressure conditions, the flow rate is proportional to the 2.5th power of the pipe diameter. This is because larger diameter pipes have a greater flow capacity, and a proportional increase in leakage will result in a larger leakage volume. Therefore, the leakage volume calculation formula incorporates [the factor that affects the leakage rate]. The structural strength of a pipeline also affects the development of leakage and the amount of water loss. The compressive strength of a pipeline is determined by both the elastic modulus E of the material and the wall thickness T. A higher elastic modulus means a stiffer material, resulting in less deformation under the same pressure; a thicker wall means a higher overall strength of the pipeline. Therefore, the amount of water loss is related to... The ratio is directly proportional, meaning that the lower the pipe strength, the greater the leakage under the same conditions. In practical applications, the distance between the acoustic sensor and the leak point also affects the accuracy of the measurement results. According to the theory of sound wave propagation, sound wave energy decreases exponentially with propagation distance, and the attenuation rate is determined by the medium properties. For underground pipe networks, the energy attenuation of sound waves during propagation can be approximately expressed as... Where α is the attenuation coefficient, with an empirical value of approximately 0.5, |L i -L leak | represents the distance from the i-th sensor to the leak point. This attenuation factor needs to be applied to leak estimation to compensate for measurement bias caused by distance. The aging condition of the pipeline has a significant impact on leak development and leak volume.

[0077] Pipeline service time t service The longer the length of the cover soil, the greater the degree of material aging and the faster the leakage develops. Simultaneously, environmental factors such as the pH value and chloride ion concentration of the cover soil also contribute to this. - This also accelerates the corrosion and aging of pipeline materials. According to the material corrosion kinetics model, the corrosion rate of materials is closely related to the pH and chloride ion concentration of the environment. Acidic environments (low pH) and environments with high chloride ion concentrations accelerate the corrosion of metal pipelines and the degradation of non-metallic pipelines. Integrating these factors into the aging influencing factors can be expressed as follows: Where t ref This refers to the reference service time, pH. ref It is a reference pH value. This refers to the chloride ion concentration. This expression describes the increasing trend of pipeline leakage as service time increases and environmental corrosivity intensifies. When service time is close to zero or environmental corrosivity is very low, the aging effect factor is close to zero, indicating almost no aging effect; when service time is very long or environmental corrosivity is very high, the aging effect factor is close to 1, indicating that the aging effect reaches its maximum. Taking all the above factors into account, the leakage rate Q... leak The calculation formula can be expressed as: This formula theoretically considers the impact of acoustic characteristics, pipe physical properties, location, and aging on leakage, achieving an accurate estimation of leakage. In practical applications, experimental data is needed to calibrate and verify the parameters in the formula to ensure the accuracy and reliability of the estimation results. This multi-data fusion method allows for a comprehensive assessment of underground pipe network leakage without excavation, providing a scientific basis for pipe network maintenance and repair.

[0078] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that the technical solutions described in the above embodiments can still be modified, or some of the technical features thereof can be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for detecting leakage in trenchless underground water supply and drainage networks based on multi-data fusion, characterized in that, The method includes: Step 1: Deploy acoustic sensors at equal intervals along the pipeline network. Each acoustic sensor collects the acoustic time-domain signal generated by pipeline leakage, and then calculates the spectral intensity of the acoustic time-domain signal at the characteristic frequency of the pipeline. Step 2: Calculate the cross-correlation function between two different acoustic sensors based on the spectral intensity. By finding the peak position of the cross-correlation function, determine the acoustic delay of the acoustic time-domain signal between different acoustic sensors. Step 3: Calculate the location of the leak point based on acoustic delay and the principle of sound wave propagation; assess the leakage volume based on the location of the leak point and pipeline usage data. The pipeline usage data includes the pipeline's service life. The unit is year; pipe wall thickness The unit is meters; the elastic modulus of the pipe. The unit is Pa; pipe diameter The unit is meters; soil cover depth The unit is m; soil cover density The unit is kg / m³ 3 Poisson's ratio of pipe materials The speed of sound wave propagation in pipe materials The unit is m / s; internal pressure of the pipeline. The unit is Pa; the characteristic frequency of the pipeline. for: in, This is the density of water, expressed in kg / m³. 3 ; No. Acoustic time-domain signals acquired by an acoustic sensor At the characteristic frequency of the pipeline Spectral intensity at for: in, The start time for acquiring acoustic time-domain signals, measured in seconds; The stopping time for acquiring acoustic time-domain signals, measured in seconds; The variable is time, and the unit is seconds (s). It is the symbol for imaginary numbers; The speed of sound waves in the soil cover is expressed in m / s. No. The acoustic sensor and the first Cross-correlation function between acoustic sensors for: in, This is a time delay variable, measured in seconds (s). The upper limit of the characteristic frequency; This is the lower limit of the characteristic frequency. For the first Acoustic time-domain signals acquired by an acoustic sensor At the characteristic frequency of the pipeline Spectral intensity at The conjugate; Acoustic delay for: ; Leakage point location for: Among them, the location of the leakage point This represents the distance measured along the pipe axis from a reference starting point, which is the position of the first acoustic sensor. This is the acceleration due to gravity, measured in m / s². 2 ; Poisson's ratio for soil covering; The distance between the acoustic sensor and other adjacent acoustic sensors; Leakage for: in, For the first along the pipeline axis direction The location of each acoustic sensor; This is a reference service life for the pipeline. The pH value of the cover soil; The concentration of chloride ions in the cover soil is expressed in mg / kg. For reference pH value; For reference chloride ion concentration.

2. The method for detecting leakage in trenchless underground water supply and drainage networks based on multi-data fusion as described in claim 1, characterized in that, If the pipe is a cast iron pipe, Poisson's ratio The value range is from 0.21 to 0.26; if the pipeline is a steel pipe, Poisson's ratio... The value ranges from 0.27 to 0.30; if the pipe is copper, Poisson's ratio... The value range is from 0.31 to 0.34; if the pipe is a PVC pipe, Poisson's ratio is... The value range is from 0.35 to 0.38; if the pipe is a PE pipe, Poisson's ratio is... The value range is from 0.39 to 0.42; if the pipe is PP pipe, Poisson's ratio is... The value range is from 0.43 to 0.45; if the pipe is ABS, Poisson's ratio is... The value range is from 0.46 to 0.48; if the pipe is a concrete pipe, Poisson's ratio is... The value range is from 0.15 to 0.20; if the pipe is a fiberglass pipe, Poisson's ratio... The value range is from 0.10 to 0.14.

Images