Pipe blockage quantification and identification method and system based on three-dimensional morphology and dynamic response

Abstract

This invention discloses a method and system for quantitative identification of pipeline blockage based on three-dimensional morphology and dynamic response, belonging to the field of intelligent pipeline detection technology. This invention simultaneously acquires three-dimensional point cloud data of pipeline blockages and micro-vibration dynamic force response signal sequences. It calculates the three-dimensional occupancy gradient (3DOG), characterizing the radial spatial distribution density change of the blockage, and the dynamic response entropy (DRE), quantifying the complexity of the material's response energy distribution. A pre-set nonlinear fusion model yields the Pipeline Blockage Comprehensive Index (PCCI). Based on the PCCI threshold range, the blockage severity is graded, and a clearing strategy is automatically matched. This invention is the first to achieve simultaneous quantification of the spatial morphology and material properties of blockages in two dimensions, establishing a complete intelligent evaluation chain. It effectively solves the problems of traditional detection relying on manual experience, having a single evaluation dimension, and being unable to quantify, significantly improving the intelligence level of blockage detection in high-speed rail tunnels and various pipelines, and enhancing the scientific nature of clearing decisions.

Description

Technical Field

[0001] This invention relates to the field of intelligent pipeline detection technology, specifically to a method and system for quantitative identification of pipeline blockage based on three-dimensional morphology and dynamic response. Background Technology

[0002] The drainage system of high-speed railway tunnels is the "lifeline" for ensuring train operation safety. Once blocked, the accumulated water will directly threaten track stability and structural safety. However, the clearing operation of high-speed railway tunnels has extremely different constraints from those of conventional municipal pipelines: the maintenance window is less than 4 hours, and it must be successful in one go with no room for trial and error; the tunnel environment is closed and sensitive, and personnel cannot enter directly, so the clearing operation must be precise and non-destructive; the composition of the blockage material is extremely complex, including high-strength concrete rebound material and dense chemical crystals, with alternating soft and hard textures and diverse qualities; the spatial shape of the blockage material is varied, and it may be a crystalline ring attached to the wall, a long longitudinal accumulation, or a local boulder, which directly affects the choice of clearing strategy.

[0003] Current pipeline blockage detection and assessment technologies have significant shortcomings: the commonly used manual inspection relies on subjective experience and judgment, which is inefficient and prone to decision-making errors; flow and pressure monitoring can only provide qualitative alarms and cannot locate the blockage point or learn the characteristics of the blockage; sonar or laser cross-sectional scanning can only measure the water flow area and obtain only two-dimensional cross-sectional information, which cannot identify the longitudinal distribution pattern, let alone perceive the material hardness, a key parameter that determines the success or failure of clearing the blockage.

[0004] In summary, existing technologies are all general tools for conventional municipal environmental design and cannot meet the extreme requirements of "zero safety risk, minute-level decision-making, and successful blockage clearing in one go" for high-speed rail tunnels. They lack effective means to simultaneously quantify the "spatial distribution pattern" and "material response characteristics" of blockages, resulting in high-speed rail tunnel drainage and blockage clearing decisions remaining at the level of experience and guesswork for a long time, with extremely low levels of intelligence. Summary of the Invention

[0005] This invention aims to solve the fundamental problems of pipeline blockage assessment being based on a single dimension, lack of quantification, and reliance on experience for decision-making. It proposes a pipeline blockage quantitative identification method and system based on three-dimensional morphology and dynamic response, which realizes the synchronous quantitative perception of the spatial morphology and material properties of the blockage, and completes the scientific assessment of the difficulty of clearing the blockage and intelligent decision output.

[0006] To achieve the above objectives, this invention provides a method for quantitative identification of pipe blockage based on three-dimensional morphology and dynamic response, comprising the following steps performed sequentially: S1. Dual-modal data synchronous acquisition: The structured light 3D scanner integrated into the detection equipment acquires 3D point cloud data of the blockage area in the pipeline. At the same time, the piezoelectric force sensor on the detection equipment simultaneously acquires the dynamic force response signal sequence generated by the contact with the blockage when the equipment performs a preset standard micro-amplitude vibration excitation on the blockage. The 3D point cloud data is the basic data for subsequent spatial distribution feature quantification, and the dynamic force response signal sequence is the basic data for subsequent material response feature quantification. S2. Spatial distribution feature quantification: Using the three-dimensional point cloud data obtained in step S1 as input, the three-dimensional occupancy gradient index 3DOG is calculated to quantify the degree of drastic change in the density of the blockage from the center to the pipe wall on the cross-section of the pipe. The obtained 3DOG index is the input parameter for the subsequent comprehensive difficulty index calculation. S3. Material response characteristic quantification: Using the dynamic force response signal sequence collected in step S1 as input, calculate the dynamic response entropy index DRE to quantify the disorder or complexity of the energy distribution of the response signal of the overall material structure of the blockage under micro-amplitude vibration excitation. The obtained DRE index is another input parameter for the subsequent calculation of the comprehensive difficulty index. S4. Calculation of Comprehensive Difficulty Index for Clearing Blockages: The 3DOG index obtained in step S2 and the DRE index obtained in step S3 are used as dual inputs. The calculation is performed through a preset nonlinear fusion function, and the output is the scalar form of the comprehensive pipeline blockage index PCCI. PCCI is a quantitative representation of the comprehensive difficulty of clearing blockages, and serves as the core basis for subsequent quantitative evaluation and strategy output. S5. Quantitative assessment and congestion clearing strategy output: Based on the preset threshold range of the PCCI value obtained in step S4, automatically match and output the corresponding qualitative description of the congestion level, and recommend a suitable congestion clearing operation strategy according to the congestion level.

[0007] Preferably, the specific steps for calculating the 3D occupancy gradient index 3DOG in step S2 are as follows, and the steps are sequentially connected: S2.1 Radial Region Division: Based on the theoretical central axis of the pipeline, the cross-section of the pipeline to be analyzed is uniformly divided into K continuous concentric annular zones along the radial direction, where K≥3; S2.2 Point Cloud Density Calculation: Based on the 3D point cloud data from step S1, count the number of point clouds Ni falling into the i-th annular zone region, and calculate the point cloud spatial density based on the volume Vi of that region. = / , where i=1,2,...,K, i=1 is the innermost central region, and i=K is the outermost region close to the pipe wall; S2.3 Gradient Calculation and Weighting: Based on the spatial density ρi of the point cloud in each annular zone obtained in step S2.2, calculate the point cloud density gradient value of the adjacent outer ring region relative to the inner ring region. = - , where i = 1, 2, ..., K-1; represents each gradient value. Assign weight coefficients The weighting rule is that the gradient closer to the pipe wall corresponds to... The larger; S2.4 Normalization Output: Summate all weighted absolute gradient values ​​obtained in step S2.3, and then normalize them to obtain the final 3DOG value. The normalization process eliminates the influence of the absolute value of the overall point cloud density, so that 3DOG only represents the relative distribution pattern of the blockage.

[0008] Preferably, the specific steps for calculating the dynamic response entropy index (DRE) in step S3 are as follows, and the steps are sequentially connected: S3.1 Signal Time-Frequency Decomposition: The dynamic force response time-domain signal sequence {x(t)} of length L acquired in step S1 is decomposed using J-level wavelet packet transform to obtain 2 signals distributed in different frequency subbands. J A sequence of wavelet packet coefficients { }, where j=1,2,...,2 J For sub-band index; S3.2 Subband Energy Calculation: Based on the wavelet packet coefficient sequence obtained in step S3.1 { } Calculate the signal energy of the j-th sub-band. ; S3.3 Energy Probability Distribution: Based on the sub-band signal energy Ej obtained in step S3.2, calculate the proportion of each sub-band energy to the total energy. ; S3.4 Shannon Entropy Calculation: Based on the energy probability distribution obtained in step S3.3 p j Calculate Shannon entropy: The Shannon entropy value is used as the final DRE value.

[0009] Preferably, the expression for the preset nonlinear fusion function in step S4 is: ; Wherein, α and β are both positive weighting coefficients, determined by regression analysis of historical case data with known congestion clearing difficulty. α is used to adjust the influence of 3DOG on the basic difficulty of congestion clearing, β is used to adjust the nonlinear amplification of congestion clearing difficulty by DRE, and e is a natural constant.

[0010] Preferably, the preset threshold range in step S5 includes three gradient ranges, and each range corresponds to a unique degree of congestion and a congestion clearing strategy, specifically: Two calibration thresholds, θ1 and θ2, are set to satisfy 0 < θ1 < θ2. The thresholds are determined by grouped experiments based on pipe material, pipe diameter, and common blockage types. When PCCI≤ At this time, the output blockage level is "mild sludge", and the recommended unblocking strategy is high-pressure water jet flushing; when <PCCI≤ When the output blockage level is "moderate caking", the recommended unblocking strategy is mechanical scraping or reaming, or a combination of medium and high pressure water jet and mechanical scraping / reaming. When PCCI> When the output blockage level is "severely solidified / hard blockage", the recommended unblocking strategy is a combination of cutting or a combination of chemical swelling treatment and mechanical removal.

[0011] Preferably, in step S1, the preset standard micro-amplitude vibration excitation is a composite sweep frequency waveform containing multiple frequency components from 2 to 100 Hz, and the vibration amplitude is 1-5 mm; the sampling frequency of the piezoelectric force sensor is not less than 5 times the highest frequency of the excitation, and the collected dynamic force response signal is a triaxial force signal, which can be subsequently synthesized into a total force signal or the axial force signal of the main excitation direction can be selected for analysis.

[0012] A pipeline blockage quantitative identification system, the system having a modular integrated structure and a connected data transmission link, comprising: The three-dimensional morphology acquisition unit is used to perform the three-dimensional point cloud data acquisition operation in step S1 and output high-precision three-dimensional point cloud data of the pipe blockage area to the central processing and control unit. The dynamic response sensing unit is used to perform the dynamic force response signal acquisition operation in step S1 and output the dynamic force response signal sequence generated by contact with the blockage to the central processing and control unit. The central processing and control unit receives the output data from the three-dimensional morphology acquisition unit and the dynamic response sensing unit, and sequentially executes the 3DOG calculation in step S2, the DRE calculation in step S3, the PCCI calculation in step S4, and then executes the PCCI threshold comparison and evaluation strategy matching in step S5, and transmits the quantitative indicators and matching results to the evaluation result output unit. The evaluation result output unit is connected to the central processing and control unit to receive quantitative indicators and matching results, perform visualization display, and realize remote data transmission.

[0013] Preferably, the three-dimensional morphology acquisition unit includes a structured light projection module, a binocular camera, and an equipment calibration module. The structured light projection module and the binocular camera work together to achieve non-contact three-dimensional scanning. The equipment calibration module is used for coordinate accuracy calibration before scanning. The acquired three-dimensional point cloud data includes the three-dimensional coordinates (X, Y, Z) of each point, and the point cloud spacing is less than 1% of the pipe diameter.

[0014] Preferably, the dynamic response sensing unit includes a piezoelectric force sensor, a signal conditioning circuit, and a micro-excitation execution module; the micro-excitation execution module is a retractable probe or a flexible contact plate, used to apply a preset standard micro-amplitude vibration excitation to the blockage; the piezoelectric force sensor is a high dynamic range three-dimensional force sensor, used to collect triaxial dynamic force response signals; the signal conditioning circuit is used to filter and amplify the collected raw signals before transmitting them to the central processing and control unit.

[0015] Preferably, the central processing and control unit includes an embedded processor, a memory, and an algorithm execution module; the memory stores 3DOG calculation algorithms, DRE calculation algorithms, nonlinear fusion function models, PCCI graded thresholds, and a blockage clearing strategy mapping library; the algorithm execution module, driven by the embedded processor, retrieves the algorithms and data from the memory to complete the calculation of quantitative indicators and the matching of evaluation strategies; The quantitative indicators output by the evaluation result output unit include 3DOG value, DRE value and PCCI value. The output matching results include a qualitative description of the degree of congestion, a recommendation of clearing operation strategy, and support for the output of refined parameter suggestions for the strategy. The advantages of this invention compared to the prior art are: (1) This invention achieves the first simultaneous quantitative perception of the spatial morphology and material properties of blockages in high-speed railway tunnels, filling a gap in dual-dimensional detection technology. The invention accurately depicts the density distribution trend of blockages from the center to the tunnel wall using a three-dimensional occupancy gradient index, clearly distinguishing different spatial morphologies such as uniform siltation, wall-attached crystalline rings, and localized boulders. Furthermore, it quantifies the energy distribution complexity of the material's response to micro-vibrations using a dynamic response entropy index, effectively identifying the characteristics of different materials such as high-strength concrete, dense crystalline structures, and soft-hard composites. The introduction of these two indicators provides a quantifiable and comparable scientific description system for the "black box" of blockages in high-speed railway tunnels for the first time.

[0016] (2) A unique comprehensive pipeline blockage index model has been developed, enabling a leap from experience-based judgment to quantitative assessment of blockage clearance difficulty. This invention organically combines spatial distribution difficulty with material property difficulty through a nonlinear fusion function, and the output comprehensive index directly corresponds to the level of blockage clearance difficulty. This mechanism completely changes the traditional extensive mode of relying on manual experience to "guess the difficulty by watching videos," allowing maintenance personnel to accurately predict the required tools, time, and processes before entering the tunnel, providing a scientific basis for successful blockage clearance during the maintenance window.

[0017] (3) Establish an intelligent closed loop of detection and decision-making to accurately match the extreme operation requirements of high-speed railway tunnels. This invention automatically outputs graded evaluation results and corresponding unblocking strategies based on a comprehensive index, from high-pressure water jet flushing to composite cutting processes, to achieve intelligent recommendation of tool selection and operation plan. This function directly responds to the extreme constraints of "minute-level decision-making and zero trial-and-error space" during the high-speed railway maintenance window, and greatly improves the success rate and efficiency of unblocking operations.

[0018] (4) The algorithm is lightweight and the sensors are mature, enabling engineering implementation and self-evolution. The structured light scanning and piezoelectric sensing technologies assisted by this invention are both mature and reliable, and the algorithm has a moderate computational load, which can be integrated into existing tunnel inspection robot platforms. At the same time, the model parameters and grading thresholds can be continuously optimized through the historical case database, so that the evaluation accuracy can be continuously improved with the accumulation of data, forming a self-evolving intelligent system that becomes more accurate with use.

[0019] In summary, this invention establishes for the first time a complete quantitative evaluation chain from raw data collection to recommended blockage clearing strategies, fundamentally solving the industry pain points of "unclear visibility, inaccurate measurement, and difficult decision-making" in high-speed railway tunnel drainage systems, and significantly improving the intelligence level of tunnel maintenance and the ability to ensure traffic safety. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 This is a flowchart of the method of the present invention; Figure 2 This is a schematic diagram of the pipe blockage point cloud and partitioning in S2 of the present invention; In the diagram: 1-pipe wall; 2-pipe blockage; 3-first quantization partition; 4-i-th quantization partition. Detailed Implementation

[0022] The embodiments of this application are described in detail below. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain this application, and should not be construed as limiting this application.

[0023] In the description of this application, it should be noted that, unless otherwise expressly specified and limited, the terms "installation," "connection," and "joining" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components. Those skilled in the art can understand the specific meaning of the above terms in this application according to the specific circumstances.

[0024] The present invention will now be described in further detail with reference to the accompanying drawings.

[0025] This invention aims to solve the fundamental problems of existing technologies in assessing pipeline blockage, such as a single dimension, inability to quantify, and reliance on experience for decision-making. Specifically, it includes: (1) How to transform the three-dimensional spatial distribution pattern of the blockage (not just the two-dimensional cross-section) into a calculable quantitative indicator with clear physical meaning.

[0026] (2) How to quantify the overall material properties of the blockage (such as loose / dense, elastic / rigid) without destructive or contact (or micro-contact) means.

[0027] (3) How to scientifically integrate the above two different quantitative indicators to form a comprehensive index that can directly and reliably correspond to the actual difficulty of clearing blockages.

[0028] (4) How to establish an objective and automated evaluation and decision-making output process based on this comprehensive index to replace subjective experience judgment.

[0029] This invention provides a quantitative identification method for pipeline blockage based on three-dimensional morphology and dynamic response. Through a unique algorithm framework, it transforms sensor information acquired from different physical dimensions into quantitative indicators with clear engineering significance, ultimately achieving a scientific assessment of the difficulty of clearing blockages. The specific solution is implemented according to the following steps; see the flowchart below. Figure 1 : S1. Simultaneous acquisition of dual-modal sensing data This step aims to obtain the two most crucial physical characteristics of the blockage: its three-dimensional spatial geometry and the material's dynamic response properties. Specifically, it includes: 3D point cloud data acquisition: A non-contact scanning method is used to scan the target blockage area inside the pipeline using a structured light 3D scanner or LiDAR mounted on the front end of the detection equipment. Equipment calibration is required before scanning to ensure coordinate accuracy. The acquired raw point cloud data must contain the 3D coordinates (X, Y, Z) of each point, and its density and accuracy should meet the requirements of subsequent morphological analysis (e.g., the point cloud spacing should be less than 1% of the pipeline diameter).

[0030] Dynamic response signal acquisition: During the same period of 3D scanning or in the immediately following "probe mode", initiate the dynamic excitation and acquisition process: Standard micro-excitation application: The contact actuator (such as a retractable probe or flexible contact plate) at the front end of the control detection equipment is subjected to micro-amplitude vibration in a preset, standardized pattern. This vibration pattern is usually designed as a composite waveform containing multiple frequency components (such as a 2-100Hz sweep frequency signal), with the amplitude controlled at the millimeter level (such as 1-5mm) to ensure sufficient excitation to the blockage while avoiding damage or large displacement.

[0031] Synchronous recording of response signals: A high-sensitivity piezoelectric force sensor integrated on the contact actuator is used to synchronously acquire the time-domain force signal sequence {F} fed back by the blockage during the aforementioned micro-excitation process. x(t) ,F y(t) ,F z(t) The signal sampling frequency must be at least five times the highest excitation frequency (e.g., ≥500Hz) to fully capture dynamic response details.

[0032] S2. Spatial Distribution Morphology Quantification – 3D Occupancy Gradient (3DOG) Calculation, such as Figure 2 As shown in the figure, the center of the circle is the first quantization partition 3; This step transforms the 3D point cloud data into a single index that can quantify the "eccentricity" and "gradient steepness" of the blockage's radial distribution in the pipe. The core focus is analyzing the trend of point cloud density variation from the pipe center to the pipe wall. Specific sub-steps are as follows: S2.1 Coordinate System Establishment and Radial Partitioning: A cylindrical coordinate system is established with the theoretical central axis of the pipeline as the Z-axis. At the cross-sectional location to be analyzed (usually the densest point cloud or the most congested section identified from the image), the cross-section is uniformly divided radially into K consecutive concentric annular zones (K≥3, usually 6-10). Let the inner radius of the pipeline be R, then the inner and outer radii of the i-th annular zone (the i-th quantization partition 4) are respectively... and ,in , .

[0033] S2.2 Partition Point Cloud Density Calculation: For the i-th quantized partition, count the number N of all 3D point clouds falling within that region. i Calculate the approximate volume of the annulus. Where Δz is the effective thickness of the analysis section (which can be determined based on the distribution of the point cloud along the axial direction). Then the spatial density ρ of the point cloud in this ring zone... i =N i / V i .

[0034] S2.3 Radial Density Gradient Calculation and Weighting: Calculate the density gradient g between adjacent rings from the inside out. i =ρ i+1 -ρ i Where i = 1, 2, ..., K-1. To reflect the difference in the impact of gradients at different locations on the "difficulty of clearing congestion", a gradient g is defined for each gradient. i Assign a weight coefficient ω i The weighting principle is as follows: the gradient between the annular zones closer to the pipe wall is more important for determining whether the blockage has "attached to the wall and caked up," and therefore should be given a larger weight. A linear or exponentially increasing weighting strategy can be used, such as ω... i =i / sum(1:K-1) or ω i =base i-1 (base>1).

[0035] S2.4 Gradient Index Normalization Output: The weighted absolute values ​​of the gradients are summed and divided by a normalization factor to obtain the final 3DOG value. The calculation formula is as follows: Formula Explanation: The numerator is the weighted gradient sum, reflecting the overall drastic degree of density change. In the denominator, max(ρ) is used to eliminate biases caused by differences in the absolute number of points in the overall point cloud (related to scanning distance and reflectivity), ensuring that 3DOG focuses on the relative distribution morphology. The weighted sum in the denominator ensures the result falls within a reasonable numerical range. 3DOG≈0 indicates a uniform density distribution; a larger 3DOG value indicates that blockages are more significantly concentrated and accumulated near the pipe wall, forming difficult-to-treat ring-shaped or lateral caking.

[0036] S3. Quantification of Material Dynamic Response Characteristics – Calculation of Dynamic Response Entropy (DRE) This step transforms the time-domain force response signal into an index that can quantify the "complexity" or "disorder" of the blockage material structure. The basic principle is that different material structures (such as homogeneous solids, particle packing, viscoelastic gels, and fiber entanglements) exhibit significantly different energy distribution patterns in the frequency domain in response to the same micro-excitation. Specific sub-steps are as follows: S3.1 Response Signal Preprocessing and Selection: The acquired triaxial force signals are synthesized, typically by calculating the resultant force magnitude sequence. Alternatively, an axial force signal consistent with the main excitation direction can be selected. The sequence is then preprocessed, including detrending and bandpass filtering (passband covering the excitation frequency range).

[0037] S3.2 Signal Time-Frequency Decomposition: The preprocessed signal F(t) undergoes a J-level wavelet packet transform (WPT). Wavelet packet transform provides finer frequency band division than traditional Fourier transform or wavelet transform. The decomposition yields 2... J Each wavelet packet node corresponds to a coefficient sequence for a specific frequency band. }

[0038] S3.3 Sub-band Energy Calculation: Calculate the signal energy of the j-th sub-band (node). .

[0039] S3.4 Energy Distribution Entropy Calculation: Calculate the energy distribution entropy of each sub-band. This is considered as the source of a probability distribution. First, calculate the proportion of energy in each frequency band relative to the total energy: Then, the Shannon entropy of this energy distribution is calculated, which yields the DRE value: Physical interpretation: If the response energy is highly concentrated in a few frequency bands (such as the response of a rigid body), then the energy distribution is concentrated. The energy distribution is uneven, resulting in a lower calculated entropy value (DRE). If the response energy is widely dispersed across many frequency bands (e.g., in porous, layered, or internally frictional materials), the energy distribution is uniform. When the values ​​are nearly equal, the calculated entropy value (DRE) is higher. Therefore, a low DRE value usually corresponds to rigid, homogeneous blockages; a high DRE value usually corresponds to complex, viscoelastic, or loosely structured blockages.

[0040] S4. Comprehensive Calculation of Blockage Clearing Difficulty Index – Pipeline Congestion Index (PCCI) The core innovation of this step lies in fusing two indicators, representing "spatial distribution difficulty" and "material property difficulty," into a unified difficulty index through a nonlinear model based on physical mechanisms. The specific calculation is as follows: Fusion model: Employs a nonlinear function combining product and exponentiation. Model parameters: α and β are positive weighting coefficients, whose values ​​were determined through regression analysis of a large amount of historical case data with known congestion clearing difficulty. α mainly adjusts the intensity of the impact of 3DOG on the basic difficulty, while β mainly adjusts the degree of nonlinear amplification of difficulty by DRE.

[0041] Model Mechanism Explanation: The design of this model is based on the following engineering insights: 1. 3DOG as a multiplicative basis: The "eccentricity" (3DOG) of spatial distribution directly affects the accessibility and operation mode of the work tool, and is the basic linear factor in the composition of difficulty.

[0042] 2. DRE exhibits nonlinear amplification through an exponential function: The impact of material complexity (DRE) on difficulty is often nonlinear. For example, when a material changes from "simple rigidity" to "complex viscoelasticity," the strategies, energy consumption, and time required for unclogging may increase dramatically, rather than increasing linearly. The exponential function exp(β) DRE can effectively simulate this "complexity threshold effect".

[0043] 3. Synergistic effect of both: The final difficulty is the result of the combined effect of spatial distribution and material properties, so a product form is adopted.

[0044] S5. Quantitative Grading Assessment and Congestion Clearance Strategy Mapping Output The calculated PCCI values ​​are mapped to preset task difficulty levels, and corresponding strategy suggestions are output.

[0045] Tiered threshold setting: Based on a large amount of practical data, two thresholds are set. and (0< < These thresholds can be grouped and calibrated according to pipe material, pipe diameter, and common blockage types.

[0046] Hierarchical mapping rules: If PCCI≤ The condition is classified as "mild siltation." Characteristics: evenly distributed or centrally accumulated; simple material. Recommended strategy: high-pressure water jet flushing.

[0047] like <PCCI≤ The condition is classified as "moderate compaction." Characteristics: There is some wall-side buildup or the material is becoming increasingly complex. Recommended strategy: Mechanical scraping, reaming, or a combination of medium- and high-pressure water jetting and mechanical methods.

[0048] If PCCI> The condition is classified as "severely solidified / hardened blockage." Characteristics: Significant wall-attached crusting and / or highly complex and hard material. Recommended strategy: Powerful composite processes, such as "high-frequency impact crushing + mechanical gripping," "high-pressure water jet (ultra-high pressure) cutting," and "chemical solvent pretreatment + mechanical removal."

[0049] To more clearly illustrate the specific embodiments of the present invention, an example is provided below: Step S1: Simultaneous acquisition of dual-modal data (1) The robot stopped when it was about 0.5 meters away from the suspected blockage.

[0050] (2) 3D morphology acquisition: The structured light 3D scanner (0.5mm resolution) integrated in the robot's head is activated to scan the blockage section about 1.5 meters long in front, and obtain 3D point cloud data containing hundreds of thousands of points. The point cloud accurately depicts the 3D shape of the exposed surface of the blockage.

[0051] (3) Dynamic response acquisition: The robot's head extends out of the contact plate and gently touches the surface of the blockage. The control unit issues a command to drive the contact plate to perform a set of standardized micro-amplitude composite vibrations: the amplitude is 3 mm, the frequency varies between 2 Hz and 100 Hz in a specific pattern, and the duration is 8 seconds. At the same time, the piezoelectric three-dimensional force sensor (range ±500 N, sampling frequency 2000 Hz) integrated on the contact plate synchronously acquires force signals in three directions and synthesizes the total force signal sequence {F(t)}, obtaining a total of 16,000 data points.

[0052] Step S2: Calculate the 3D occupancy gradient (3DOG) (1) Data preprocessing: coordinate correction and noise filtering are performed on the acquired 3D point cloud.

[0053] (2) Radial partitioning (corresponding to S2.1): Set the theoretical inner diameter of the pipe to 800mm. With the central axis of the pipe as the Z-axis, at the cross-section of interest (in this example, the cross-section at the point of most severe blockage), divide the cross-section radially into K=6 concentric rings. The radius of the innermost ring (i=1) ranges from 0 to 133mm, and the radius of the outermost ring (i=6, attached to the wall) ranges from 667 to 800mm.

[0054] (3) Calculate density and gradient (corresponding to S2.2-S2.3): Count the number of point clouds within each annulus and calculate the volume density. Assume the following: ρ1 = 50 pts / L, ρ2 = 120 pts / L, ρ3 = 300 pts / L, ρ4 = 450 pts / L, ρ5 = 600 pts / L, ρ6 = 580 pts / L. (Unit: points per liter, for illustrative purposes only) Calculate the adjacent gradients: g1=70, g2=180, g3=150, g4=150, g5=-20.

[0055] Set a weight vector to reflect that "the gradient near the pipe wall is more important": ω=[0.1,0.15,0.2,0.25,0.3] (the larger i is, the larger ω is).

[0056] (4) Normalization calculation (corresponding to S2.4): Sum of the absolute values ​​of the weighted gradients: Sum=0.1 70+0.15 180 + 0.2 150 + 0.25 150 + 0.3 20 = 7 + 27 + 30 + 37.5 + 6 = 107.5; Maximum density ρ max =600, weights and Sum ω =1.0.

[0057] Calculate 3DOG = 107.5 / (600) 1.0)≈0.179.

[0058] Interpretation of results: The 3DOG value is approximately 0.18, which is moderately high. Combined with the original density data, the density reaches its peak at i=4 and 5 rings (i.e., the outer part of the pipe), and then decreases slightly at the wall-adhering area (i=6). This indicates that the blockage mainly accumulates in a "thick ring" shape in the middle section of the pipe, and is not completely adhered to the wall, nor is it accumulated in the center.

[0059] Step S3: Calculate the dynamic response entropy (DRE) (1) Signal preprocessing: Detrending and bandpass filtering (2-150Hz) are performed on the total force signal {F(t)}.

[0060] (2) Wavelet packet decomposition (corresponding to S3.1): Using the “db4” wavelet basis, perform J=5 level wavelet packet decomposition to obtain 2 5 =32 frequency sub-bands.

[0061] (3) Calculate energy and entropy (corresponding to S3.2-S3.4): Calculate the energy E of each subband j .

[0062] Assume that the calculated energy is mainly distributed in about 15 subbands, with the energy in other subbands being very weak.

[0063] Calculate probability p j Using the Shannon entropy, we get DRE≈3.8 (bits).

[0064] Interpretation of results: A DRE value of 3.8 is considered moderate. This indicates that the blockage's response energy to micro-vibrations is somewhat dispersed, rather than a purely rigid response (rigid response DRE is typically <2.5), suggesting that the material may have some viscoelasticity or an inhomogeneous internal structure, such as moist deposits mixed with some solid waste.

[0065] Step S4: Calculate the Pipeline Congestion Index (PCCI) (1) Call the model parameters: Call the pre-trained model parameters of the current pipe type (concrete drainage pipe) from the system database: α=1.2, β=0.25.

[0066] (2) Calculate PCCI: Step S5: Quantitative Assessment and Strategy Output (1) Call difficulty threshold: The difficulty threshold for calling the current pipeline type from the system: =0.4, =1.0.

[0067] (2) Comparison and output: The calculated PCCI value is 0.556.

[0068] judge: (0.4)<0.556≤ (1.0).

[0069] Automatically generate evaluation report: Quantitative indicators: 3DOG=0.179, DRE=3.8, PCCI=0.556.

[0070] Blockage level: Moderate caking.

[0071] Characteristics: The material is stacked in a ring shape in space and has viscoelastic properties.

[0072] Recommended unblocking strategy: The preferred approach is a combined process of "high-pressure water jet (pressure recommended ≥15MPa) and mechanical reamer (low speed, high torque)". The high-pressure water jet can impact and soften the sticky parts, while the mechanical reamer can break up the solids encased within and remove the ring-shaped deposits.

[0073] Through the above-described clear and quantifiable steps, this invention successfully transforms a vague "pipeline blockage" problem into a precise engineering task defined by three specific values ​​(3DOG, DRE, PCCI) and a clear action recommendation (unblocking strategy), significantly improving the intelligence and scientific level of pipeline maintenance.

[0074] It should be noted that the "effective thickness Δz of the analysis section" mentioned in step S2.2 is not the total physical length of the blockage along the pipe axis, but rather refers to the axial integral thickness of the unit annular analysis voxel selected for statistical point cloud density calculation during three-dimensional occupancy gradient (3DOG) calculation. Its definition and value determination method are as follows: Definition: The effective thickness Δz is a virtual axial slice thickness used to discretize the 3D point cloud space into a series of radially partitioned thin annular cylinders (i.e., analysis voxels) to calculate the point cloud density per unit volume. The selection principle of Δz is: to fully reflect the radial density distribution characteristics of the blockage on the pipe cross-section while ensuring the spatial representativeness and noise resistance of the statistical results, and at the same time to avoid the density characteristics of different radial regions being blurred by axial averaging due to excessive thickness.

[0075] The specific method for determining the effective thickness is as follows: An adaptive method based on local axial resolution of point clouds: Based on the average spacing δ (i.e., the nominal distance between two adjacent points) of the point cloud acquired by the 3D scanner, Δz is set to an integer multiple of δ: Δz=m×δ Where m is a positive integer greater than or equal to 5 (usually taken as 5 to 10). This method is computationally simple and can ensure that each analysis voxel contains at least 5 to 10 layers of point cloud along the axis, thereby effectively suppressing the influence of random noise on density statistics. For example, if the point cloud spacing δ = 1 mm, then Δz = 5 mm to 10 mm can be taken.

[0076] Based on the above definition and value method, "effective thickness" has a clear physical meaning and a repeatable calculation basis, which can be directly implemented by those skilled in the art without creative effort.

[0077] Finally, any aspects not fully described in this invention utilize existing mature products and technologies.

[0078] The present invention and its embodiments have been described above. This description is not restrictive, and the accompanying drawings are only one embodiment of the present invention; the actual structure is not limited thereto. In conclusion, if those skilled in the art are inspired by this description and design similar structures and embodiments without departing from the spirit of the invention, such designs should fall within the protection scope of the present invention.

Claims

1. A method for quantitative identification of pipe blockage based on three-dimensional morphology and dynamic response, characterized in that, The following steps are performed sequentially: S1. Dual-modal data synchronous acquisition: The structured light 3D scanner integrated into the detection equipment acquires 3D point cloud data of the blockage area in the pipeline. At the same time, the piezoelectric force sensor on the detection equipment simultaneously acquires the dynamic force response signal sequence generated by contact with the blockage when the equipment performs a preset standard micro-amplitude vibration excitation on the blockage. The 3D point cloud data is the basic data for subsequent spatial distribution feature quantification, and the dynamic force response signal sequence is the basic data for subsequent material response feature quantification. S2. Spatial distribution feature quantification: Using the three-dimensional point cloud data obtained in step S1 as input, the three-dimensional occupancy gradient index 3DOG is calculated to quantify the degree of drastic change in the density of the blockage from the center to the pipe wall on the cross-section of the pipe. The obtained 3DOG index is the input parameter for the subsequent comprehensive difficulty index calculation. S3. Material response characteristic quantification: Using the dynamic force response signal sequence collected in step S1 as input, calculate the dynamic response entropy index DRE to quantify the disorder or complexity of the energy distribution of the response signal of the overall material structure of the blockage under micro-amplitude vibration excitation. The obtained DRE index is another input parameter for the subsequent calculation of the comprehensive difficulty index. S4. Calculation of Comprehensive Difficulty Index for Clearing Blockages: The 3DOG index obtained in step S2 and the DRE index obtained in step S3 are used as dual inputs. The calculation is performed through a preset nonlinear fusion function to output the scalar form of the Comprehensive Pipeline Blockage Index (PCCI). The PCCI is a quantitative representation of the comprehensive difficulty of clearing blockages and serves as the core basis for subsequent quantitative evaluation and strategy output. S5. Quantitative assessment and blockage clearing strategy output: Based on the preset threshold range of the PCCI value obtained in step S4, automatically match and output the corresponding qualitative description of the blockage degree, and recommend a suitable blockage clearing operation strategy according to the blockage degree.

2. The pipeline blockage quantitative identification method based on three-dimensional morphology and dynamic response according to claim 1, characterized in that: The specific steps for calculating the 3D occupancy gradient index (3DOG) in step S2 are as follows, and each step is sequentially connected: S2.1 Radial Region Division: Based on the theoretical central axis of the pipeline, the cross-section of the pipeline to be analyzed is uniformly divided into K continuous concentric annular zones along the radial direction, where K≥3; S2.2 Point Cloud Density Calculation: Based on the 3D point cloud data from step S1, count the number of point clouds Ni falling into the i-th annular zone region, and calculate the point cloud spatial density based on the volume Vi of that region. = / , where i=1,2,...,K, i=1 is the innermost central region, and i=K is the outermost region close to the pipe wall; S2.3 Gradient Calculation and Weighting: Based on the spatial density ρi of the point cloud in each annular zone obtained in step S2.2, calculate the point cloud density gradient value of the adjacent outer ring region relative to the inner ring region. = - , where i = 1, 2, ..., K-1; represents each gradient value. Assign weight coefficients The weighting rule is that the gradient closer to the pipe wall corresponds to... The larger; S2.4 Normalization Output: Summate all weighted absolute gradient values ​​obtained in step S2.3, and then normalize them to obtain the final 3DOG value. The normalization process eliminates the influence of the absolute value of the overall point cloud density, so that 3DOG only represents the relative distribution pattern of the blockage.

3. The pipeline blockage quantitative identification method based on three-dimensional morphology and dynamic response according to claim 1, characterized in that: The specific steps for calculating the dynamic response entropy (DRE) index in step S3 are as follows, and each step is sequentially connected: S3.1 Signal Time-Frequency Decomposition: The dynamic force response time-domain signal sequence {x(t)} of length L acquired in step S1 is decomposed using J-level wavelet packet transform to obtain 2 signals distributed in different frequency subbands. J A sequence of wavelet packet coefficients { }, where j=1,2,...,2 J For sub-band index; S3.2 Subband Energy Calculation: Based on the wavelet packet coefficient sequence obtained in step S3.1 { } Calculate the signal energy of the j-th sub-band. ; S3.3 Energy Probability Distribution: Based on the sub-band signal energy Ej obtained in step S3.2, calculate the proportion of each sub-band energy to the total energy. ; S3.4 Shannon Entropy Calculation: Based on the energy probability distribution obtained in step S3.3 p j Calculate Shannon entropy: The Shannon entropy value is used as the final DRE value.

4. The pipeline blockage quantitative identification method based on three-dimensional morphology and dynamic response according to claim 1, characterized in that: The expression for the preset nonlinear fusion function mentioned in step S4 is: ； Wherein, α and β are both positive weighting coefficients, determined by regression analysis of historical case data with known congestion clearing difficulty. α is used to adjust the influence of 3DOG on the basic difficulty of congestion clearing, β is used to adjust the nonlinear amplification of congestion clearing difficulty by DRE, and e is a natural constant.

5. The pipeline blockage quantitative identification method based on three-dimensional morphology and dynamic response according to claim 1, characterized in that: The preset threshold range mentioned in step S5 includes three gradient ranges, and each range corresponds to a unique degree of congestion and a congestion clearing strategy, specifically: Two calibration thresholds θ1 and θ2 are set to satisfy 0 < θ1 < θ2. The thresholds are calibrated by grouping experiments based on pipe material, pipe diameter and common blockage types. When PCCI≤ At this time, the output blockage level is "mild sludge", and the recommended unblocking strategy is high-pressure water jet flushing; when <PCCI≤ When the output blockage level is "moderate caking", the recommended unblocking strategy is mechanical scraping or reaming, or a combination of medium and high pressure water jet and mechanical scraping / reaming. When PCCI> When the output blockage level is "severely solidified / hard blockage", the recommended unblocking strategy is a combination of cutting or a combination of chemical swelling treatment and mechanical removal.

6. The pipeline blockage quantitative identification method based on three-dimensional morphology and dynamic response according to claim 1, characterized in that: The preset standard micro-amplitude vibration excitation mentioned in step S1 is a composite sweep frequency waveform containing multiple frequency components from 2 to 100 Hz, and the vibration amplitude is 1-5 mm; the sampling frequency of the piezoelectric force sensor is not less than 5 times the highest excitation frequency, and the collected dynamic force response signal is a triaxial force signal, which is convenient for subsequent synthesis of total force signal or selection of axial force signal in the main excitation direction for analysis.

7. A pipeline blockage quantitative identification system implementing the method of any one of claims 1-6, characterized in that: The system has a modular integrated structure and a connected data transmission link, including: The three-dimensional morphology acquisition unit is used to perform the three-dimensional point cloud data acquisition operation in step S1 and output high-precision three-dimensional point cloud data of the pipe blockage area to the central processing and control unit. The dynamic response sensing unit is used to perform the dynamic force response signal acquisition operation in step S1 and output the dynamic force response signal sequence generated by contact with the blockage to the central processing and control unit. The central processing and control unit receives the output data from the three-dimensional morphology acquisition unit and the dynamic response sensing unit, and sequentially executes the 3DOG calculation in step S2, the DRE calculation in step S3, the PCCI calculation in step S4, and then executes the PCCI threshold comparison and evaluation strategy matching in step S5, and transmits the quantitative indicators and matching results to the evaluation result output unit. The evaluation result output unit is connected to the central processing and control unit, receives the quantitative indicators and matching results, displays them visually, and enables remote data transmission.

8. The pipeline blockage quantitative identification system according to claim 7, characterized in that: The three-dimensional morphology acquisition unit includes a structured light projection module, a binocular camera, and an equipment calibration module. The structured light projection module and the binocular camera work together to achieve non-contact three-dimensional scanning. The equipment calibration module is used for coordinate accuracy calibration before scanning. The acquired three-dimensional point cloud data includes the three-dimensional coordinates (X, Y, Z) of each point, and the point cloud spacing is less than 1% of the pipe diameter.

9. The pipeline blockage quantitative identification system according to claim 7, characterized in that: The dynamic response sensing unit includes a piezoelectric force sensor, a signal conditioning circuit, and a micro-excitation execution module; the micro-excitation execution module is a retractable probe or a flexible contact plate, used to apply a preset standard micro-amplitude vibration excitation to the blockage. The piezoelectric force sensor is a high dynamic range three-dimensional force sensor used to acquire triaxial dynamic force response signals; the signal conditioning circuit is used to filter and amplify the acquired raw signals before transmitting them to the central processing and control unit.

10. The pipeline blockage quantitative identification system according to claim 7, characterized in that: The central processing and control unit includes an embedded processor, a memory, and an algorithm execution module; the memory stores 3DOG calculation algorithm, DRE calculation algorithm, nonlinear fusion function model, PCCI graded threshold, and congestion clearing strategy mapping library; Driven by the embedded processor, the algorithm execution module retrieves the algorithm and data from the memory to complete the calculation of quantitative indicators and the matching of evaluation strategies. The quantitative indicators output by the evaluation result output unit include 3DOG value, DRE value and PCCI value. The output matching results include a qualitative description of the degree of congestion, a recommendation of congestion clearing operation strategy, and support the output of refined parameter suggestions for the strategy.

Images