Search

CN-121977166-A - Intelligent flaw detection and leakage positioning system for building pipeline

CN121977166ACN 121977166 ACN121977166 ACN 121977166ACN-121977166-A

Abstract

The invention discloses an intelligent flaw detection and leakage positioning system for a building pipeline, which relates to the technical field of building pipeline detection and comprises a data acquisition module, a multi-modal fusion module, a defect identification module, a leakage positioning module, a digital twin module, a health prediction module and a decision control module, wherein the data acquisition module is used for deploying various sensors along the building pipeline, acquiring acoustic emission signals, infrared thermal imaging images, pressure fluctuation data and fluxgate detection signals under the running state of the pipeline and outputting original sensing data, the multi-modal fusion module is connected with the data acquisition module and used for respectively carrying out feature extraction on the original sensing data to construct multi-modal feature vectors, dynamically adjusting each modal weight based on environmental parameters and outputting weighted fusion feature vectors, and the defect identification module is connected with the multi-modal fusion module and used for inputting the weighted fusion feature vectors into a deep neural network model.

Inventors

  • CONG XIAOGUANG
  • CHEN WEIHAO
  • ZHANG HONGDOU

Assignees

  • 中建八局第二建设有限公司

Dates

Publication Date
20260505
Application Date
20251217

Claims (10)

  1. 1. An intelligent flaw detection and leakage positioning system for a building pipeline is characterized by comprising the following components: The system comprises a data acquisition module, a multi-mode fusion module, a defect identification module, a leakage positioning module, a digital twin module, a health prediction module and a decision control module; The data acquisition module is used for deploying various sensors along the line of the building pipeline, acquiring acoustic emission signals, infrared thermal imaging images, pressure fluctuation data and fluxgate detection signals under the running state of the pipeline and outputting original sensing data; The multi-modal fusion module is connected with the data acquisition module and is used for respectively carrying out feature extraction on the original sensing data, constructing multi-modal feature vectors, dynamically adjusting the modal weights based on environmental parameters and outputting weighted fusion feature vectors; the defect identification module is connected with the multi-mode fusion module and is used for inputting the weighted fusion feature vector into a deep neural network model, identifying whether a pipeline has a crack, corrosion or perforation defect, and outputting a defect type label and confidence information; the leakage positioning module is connected with the data acquisition module and is used for calculating signal arrival time differences among the plurality of sensors by adopting a cross-correlation function method after the leakage type defect is identified, and estimating the space coordinates of the leakage points by combining a nonlinear propagation model and a weighted least square optimization method; the digital twin module is connected with the data acquisition module and is used for establishing a three-dimensional virtual model according to the physical structure and the material property of the pipeline, receiving real-time monitoring data and synchronously updating the model state to form a digital twin body consistent with the actual pipeline; The health prediction module is connected with the digital twin module and is used for training a prediction model based on historical flaw detection data and real-time state data and outputting a pipeline health index and a residual service life prediction value; The decision control module is connected with the defect identification module, the leakage positioning module and the health prediction module and is used for synthesizing defect types, leakage positions and health state prediction results and generating maintenance suggestions and maintenance instructions.
  2. 2. The intelligent flaw detection and leakage positioning system for building pipelines according to claim 1, wherein the process of extracting features of the original sensing data in the multi-mode fusion module comprises the following steps: performing frequency band energy decomposition on the acoustic emission signal by adopting wavelet packet transformation, and extracting the energy entropy value of each frequency band; carrying out texture feature extraction on the infrared thermal imaging image by adopting a gray level co-occurrence matrix algorithm to obtain contrast, correlation and energy parameters; Performing frequency domain analysis on the pressure fluctuation data by adopting Fourier transformation, and extracting main frequency amplitude and bandwidth information; Carrying out fluctuation intensity quantification on the fluxgate detection signal by adopting a sliding window root mean square algorithm to obtain a metal wall thickness variation trend index; And constructing a multidimensional feature vector based on the features, and taking the multidimensional feature vector as a basic input of subsequent weighted fusion.
  3. 3. The intelligent flaw detection and leakage positioning system for building pipelines according to claim 2, wherein the process of dynamically adjusting the modal weights comprises the following steps: acquiring the current temperature, humidity and fluid medium type by adopting an environment sensing unit; inputting the environmental parameters into a preset weight distribution function, wherein the function expression is as follows: Wherein w i represents a weight coefficient corresponding to the i-th type sensor, E represents a current environment parameter set, f i (E) represents a sensitivity function of the i-th type sensor in the current environment, and beta i is an empirical adjustment factor; And carrying out weighted fusion on various feature vectors according to the calculation result to generate weighted fusion feature vectors.
  4. 4. The intelligent flaw detection and leakage positioning system for building pipelines according to claim 3, wherein the deep neural network model in the flaw identification module adopts the following structure and flow: The input layer receives the weighted fusion feature vector output by the multi-mode fusion module; the hidden layer sequentially comprises a convolution layer, a pooling layer and a long-period memory network layer and is used for extracting local time sequence characteristics and modeling long-period dependency relationship; The output layer adopts a Softmax function to map the final characteristics to four classification label spaces, and classification labels respectively comprise normal, crack, corrosion and perforation; The classification result is determined by the following formula: wherein C represents the type of defect ultimately determined, z k represents the output score of the kth class; if the output label is a crack, corrosion or perforation, judging that the leakage type defect exists, and triggering a leakage positioning process.
  5. 5. The intelligent flaw detection and leakage positioning system for building pipelines according to claim 4, wherein the implementation process of estimating the spatial coordinates of the leakage points in the leakage positioning module comprises the following steps: after the leakage type defect is identified, collecting sound wave signals received by a plurality of sensors; and calculating an arrival time difference of signals between any two sensors by adopting a cross-correlation function method, wherein the cross-correlation function is defined as: S i (t),s j (t) respectively represents signals acquired by the ith sensor and the jth sensor, and tau represents a time delay variable; taking tau ij corresponding to the maximum value of the cross-correlation function as the signal arrival time difference between the two sensors; constructing a time delay matrix tau= [ tau ij ] among all the sensors; Combining the known sensor coordinates (x i ,y i ) and the set propagation velocity v, a system of nonlinear positioning equations is built: ||(x l ,y l )-(x i ,y i )||＝v·t i ,i=1,2,...,N; Wherein, (x l ,y l ) represents the coordinates of the leak point to be solved, and t i represents the signal arrival time at the i-th sensor; And a weighted least square optimization method is introduced to solve the optimal leakage point coordinates, and the objective function is as follows: Wherein omega i is sensor confidence weight, and is obtained according to historical error statistics; and finally outputting the space coordinates (x l ,y l ) of the leakage point.
  6. 6. The intelligent flaw detection and leakage positioning system for building pipelines according to claim 5, wherein the implementation of synchronously updating the model state in the digital twin module comprises the following steps: establishing a three-dimensional topological model according to the geometric dimension, the material property and the connection mode of an actual pipeline; receiving real-time monitoring data from a data acquisition module, wherein the real-time monitoring data comprises temperature, pressure, flow and defect position information; Simulating the internal stress distribution and damage expansion trend of the pipeline by using a finite element simulation method; in each update period, the following state update formula is adopted: S t ＝F(S t-1 ,D t ); Wherein S t represents the state of the digital twin body at the t-th moment, S t-1 represents the state at the last moment, D t represents the actual data input at the t-th moment, and F ()' represents a state transfer function, and the physical model and the data drive are jointly modeled; The updated digital twin state is used for data input by the health prediction module.
  7. 7. The intelligent flaw detection and leakage positioning system for building pipelines according to claim 6, wherein the process of training the prediction model in the health prediction module comprises the following steps: collecting historical flaw detection records and operation data to form a training sample set; Extracting an input feature vector X of each sample, wherein the input feature vector X comprises a current health index, accumulated running time, historical defect types and leakage times; Defining a target output Y as a health index and a residual service life predicted value of k steps in the future; modeling the sequence data by using a long and short term memory network LSTM, wherein a hidden state update formula is as follows: h t ＝LSTM(h t-1 ,X t ); Y t ＝W o ·h t +b o ; Wherein h t represents a hidden state at the t moment, and W o ,b o is a weight matrix and a bias term of the output layer respectively; After model training is completed, state data updated by the digital twin module is received periodically, and a health index H i and a residual service life predicted value are output.
  8. 8. The intelligent inspection and leakage localization system for building pipelines of claim 7, wherein the process of generating maintenance advice and maintenance instructions by the decision control module comprises: the defect type output by the defect identification module, the leakage coordinate output by the leakage positioning module, the health index output by the health prediction module and the residual service life prediction value are synthesized; setting a three-level early warning mechanism, comprising: Generating a "continuously monitoring" instruction when the health index is below a threshold T 1 and no leakage occurs; generating a "scheduled patrol" instruction when the health index is below a threshold T 2 or a slight leak is identified; when the health index is lower than a threshold T 3 or serious leakage is identified, generating an 'immediate maintenance' instruction and closing a valve in a designated area in parallel; and the maintenance instruction is sent to the remote control terminal through the communication interface, and meanwhile, the operation log is saved for tracing.
  9. 9. A computer device comprises a memory and a processor, wherein the memory stores a computer program, and the computer program is characterized in that the processor realizes the steps of the intelligent flaw detection and leakage positioning system for the building pipeline according to any one of claims 1-8 when executing the computer program.
  10. 10. A computer readable storage medium having a computer program stored thereon, wherein the computer program when executed by a processor performs the steps of the intelligent flaw detection and leakage localization system for a building pipe according to any one of claims 1 to 8.

Description

Intelligent flaw detection and leakage positioning system for building pipeline Technical Field The invention relates to the technical field of building pipeline detection, in particular to an intelligent flaw detection and leakage positioning system for a building pipeline. Background Building pipe detection technology refers to the generic term for a variety of technical means for detecting and assessing the operational status and structural integrity of various pipes within a building. Therefore, how to improve the intelligent level and safety of the detection of the building pipeline by using advanced technical means is one of the problems to be solved in the current urgent need. In the field of building pipeline detection, the existing leakage positioning method is greatly interfered by noise, and has the defects of high positioning error and large sensor performance fluctuation under different working conditions caused by irregular propagation paths in a complex pipe network, so that a detection result is unstable, and meanwhile, the traditional method focuses on post detection, potential faults cannot be predicted in advance, and maintenance is delayed. Disclosure of Invention The present invention has been made in view of the above-described problems occurring in the prior art. Therefore, the invention provides an intelligent flaw detection and leakage positioning system for a building pipeline, which solves the problems that the existing leakage positioning method is greatly interfered by noise, the propagation path in a complex pipe network is irregular, the positioning error is high, the performance fluctuation of a sensor is large under different working conditions, the detection result is unstable, and meanwhile, the traditional method focuses on the post detection, the potential faults cannot be predicted in advance, and the maintenance is delayed. In order to solve the technical problems, the invention provides the following technical scheme: In a first aspect, the present invention provides an intelligent inspection and leak location system for a building pipe, comprising: The system comprises a data acquisition module, a multi-mode fusion module, a defect identification module, a leakage positioning module, a digital twin module, a health prediction module and a decision control module; The data acquisition module is used for deploying various sensors along the line of the building pipeline, acquiring acoustic emission signals, infrared thermal imaging images, pressure fluctuation data and fluxgate detection signals under the running state of the pipeline and outputting original sensing data; The multi-modal fusion module is connected with the data acquisition module and is used for respectively carrying out feature extraction on the original sensing data, constructing multi-modal feature vectors, dynamically adjusting the modal weights based on environmental parameters and outputting weighted fusion feature vectors; the defect identification module is connected with the multi-mode fusion module and is used for inputting the weighted fusion feature vector into a deep neural network model, identifying whether a pipeline has a crack, corrosion or perforation defect, and outputting a defect type label and confidence information; the leakage positioning module is connected with the data acquisition module and is used for calculating signal arrival time differences among the plurality of sensors by adopting a cross-correlation function method after the leakage type defect is identified, and estimating the space coordinates of the leakage points by combining a nonlinear propagation model and a weighted least square optimization method; the digital twin module is connected with the data acquisition module and is used for establishing a three-dimensional virtual model according to the physical structure and the material property of the pipeline, receiving real-time monitoring data and synchronously updating the model state to form a digital twin body consistent with the actual pipeline; The health prediction module is connected with the digital twin module and is used for training a prediction model based on historical flaw detection data and real-time state data and outputting a pipeline health index and a residual service life prediction value; The decision control module is connected with the defect identification module, the leakage positioning module and the health prediction module and is used for synthesizing defect types, leakage positions and health state prediction results and generating maintenance suggestions and maintenance instructions. As a preferable scheme of the intelligent flaw detection and leakage positioning system for the building pipeline, the invention comprises the following steps of respectively extracting characteristics of original sensing data in the multi-mode fusion module: performing frequency band energy decomposition on the acoustic emission signal by adopting wav