CN-122016161-A - Intelligent detection system and method for leakage of large-thickness bottom plate of existing building based on multi-mode sensor array

CN122016161ACN 122016161 ACN122016161 ACN 122016161ACN-122016161-A

Abstract

The invention belongs to the technical field of leakage detection, and particularly relates to an intelligent detection system and method for leakage of a large-thickness bottom plate of an existing building based on a multi-mode sensor array. The intelligent detection system for the leakage of the large-thickness bottom plate of the existing building based on the multi-mode sensor array comprises a cloud end analysis server and a field movement detection platform, wherein the cloud end server performs deep learning reasoning, big data analysis and digital twin modeling, realizes data interaction and collaborative work with the field movement detection platform through 5G or WiFi wireless communication, and integrates an intelligent mobile trolley, an active excitation subsystem, a multi-mode sensor array subsystem and an edge calculation unit as an operation execution unit. The invention provides an intelligent detection system and method for leakage of a large-thickness bottom plate of an existing building based on a multi-mode sensor array, which realize the transition from surface layer to deep layer full-thickness detection coverage, from qualitative judgment to quantitative evaluation, and from manual operation to intelligent automation.

Inventors

GAO CHANGLING
YANG YUAN
WANG JIAOLONG
LI WEI
CHEN XIAORUN
LIU TAO
WANG XUDONG
HE QINGQUAN
XIAO ZHENYANG
ZHAO XINGKUN
LI YUBING

Assignees

中国五冶集团有限公司

Dates

Publication Date: 20260512
Application Date: 20260108

Claims (10)

1. The intelligent detection system for the leakage of the large-thickness bottom plate of the existing building based on the multi-mode sensor array is characterized by comprising a cloud analysis server and a field movement detection platform, wherein the cloud server performs deep learning reasoning, big data analysis and digital twin modeling, realizes data interaction and collaborative work with the field movement detection platform through 5G or WiFi wireless communication, and integrates an intelligent mobile trolley, an active excitation subsystem, a multi-mode sensor array subsystem and an edge calculation unit as a work execution unit.
2. The intelligent detection system for leakage of the large-thickness bottom plate of the existing building based on the multi-mode sensor array is characterized in that the multi-mode sensor array subsystem comprises a thermal infrared imager, a ground penetrating radar, an ultrasonic phased array, a microwave sensor and a vibration sensor, the thermal infrared imager is used for surface layer detection, the ground penetrating radar achieves deep penetration, the ultrasonic phased array provides high-resolution focusing imaging, the microwave sensor measures water content, and the vibration sensor obtains structural dynamic response.
3. The intelligent detection system for leakage of the large-thickness bottom plate of the existing building based on the multi-mode sensor array according to claim 1, wherein the active excitation subsystem comprises a heating device, an ultrasonic excitation source and a water pressure simulation device, and the active excitation subsystem improves the identification sensitivity of micro-leakage by applying energy disturbance.
4. The intelligent detection system for leakage of the large-thickness bottom plate of the existing building based on the multi-mode sensor array, which is disclosed in claim 1, is characterized in that the edge calculation unit is responsible for real-time processing and preliminary diagnosis of field data, converts original sensor signals into structural characteristic data, and realizes quick response of an edge side and reasonable distribution of cloud computing load.
5. An intelligent detection method for leakage of an existing building large-thickness bottom plate based on a multi-mode sensor array, which is characterized by comprising the following steps of: s1, infrared detection, passive multi-time phase, active thermal excitation, heating for 15-30min and cooling for 30-60min, and characteristic extraction, wherein the passive multi-time phase comprises 5-6 am, 14-15 pm and 18-19 pm, and the active thermal excitation comprises temperature gradient T, time derivative T/ T, thermal time constant τ; S2, GPR detection, multi-frequency scanning, B-scan section, C-scan plane, 3D reconstruction, water content inversion, dielectric constant epsilon r and volume water content theta, wherein the multi-frequency scanning comprises 800MHz coarse scanning, 1.5GHz fine scanning and 2.5GHz fine scanning; s3, ultrasonic detection, phased array focusing, electronic scanning, echo analysis, SAFT imaging, namely, improving resolution, wherein the depth is 100-600 mm; s4, microwave humidity measurement, array measurement, 5X 5 sensor penetration of 100-200mm, gradient analysis, spatial gradient Theta and high value area identification; S5, vibration analysis, namely passive environment vibration, multipoint synchronous acquisition, 1-500Hz, active ultrasonic excitation, 40-100kHz sweep frequency, abnormal identification, namely damping alpha change and transfer function H (omega); S6, multi-mode data fusion, namely spatial registration, wherein the precision is less than 10mm, feature extraction, namely 41-dimensional vector, infrared 12-dimensional, GPR 8-dimensional, ultrasonic 10-dimensional, microwave 5-dimensional and vibration 6-dimensional, deep learning fusion, namely multi-head attention, a transducer encoder and multi-task output, and physical constraint correction, namely a thermal diffusion equation, a seepage mechanics and a wave equation; S7, three-dimensional reconstruction, namely generating point cloud, namely generating a voxel of 50mm 3 , and mapping colors, and carrying out path tracking, namely carrying out A-scale algorithm, gradient tracking, BIM integration and AR display; S8, quantitatively evaluating the comprehensive index ISI: =Σwi.Ii, leakage classification, namely, 0-25 of grade I, 25-50 of grade II, 50-75 of grade III and 75-100 of grade IV, and estimating leakage quantity, namely, Q=k.i.A. and accuracy +/-15%; And S9, intelligent diagnosis, namely, cause diagnosis, namely, bayesian network and decision tree, and maintenance scheme, namely, technical selection, cost estimation and construction period planning.
6. The intelligent detection method for leakage of the large-thickness floor of the existing building based on the multi-mode sensor array according to claim 5, wherein in step S1: The passive detection parameter setting comprises the steps of collecting 5-6 early morning points, 14-15 afternoon points and 18-19 evening points, collecting distance of 1-2 meters, single-point collecting frame number of more than or equal to 10 frames, temperature difference requirement of more than 5 ℃, and moving step of 1 meter/step; the active excitation parameter configuration comprises target temperature rise +10-15 ℃, heating time 15-30 minutes, acquisition interval 30 seconds/time, cooling observation time 30-60 minutes, acquisition interval 60 seconds/time and heating power 500-2000W.
7. The intelligent detection method for leakage of the large-thickness bottom plate of the existing building based on the multi-mode sensor array according to claim 5, wherein in step S2, a multi-frequency adaptive scanning strategy is adopted: Coarse scanning, frequency 800MHz, detection depth >500mm, resolution of 150mm, grid spacing of 0.5m, scanning area/hour of 200-300m 2 , global rapid general investigation of application targets and large-scale anomaly identification; Fine scanning, namely, the frequency is 1.5GHz, the detection depth is 300-400mm, the resolution is 80mm, the grid spacing is 0.2m, the scanning area per hour is 100-150m 2 , and the application target is the middle layer of an abnormal area and accurately positions defects; Fine scanning, namely, the frequency is 2.5GHz, the detection depth is less than 300mm, the resolution is 50mm, the grid spacing is 0.1m, the scanning area per hour is 50-80m 2 , and the application target is the surface layer of a key area and high-resolution imaging.
8. The intelligent detection method for leakage of the large-thickness bottom plate of the existing building based on the multi-mode sensor array according to claim 5, wherein in step S3, phased array focusing parameters are configured: 64 array elements are configured, wherein the array element distance is 2mm, the center frequency is 100kHz, the pulse width is 2-5 cycles, the focusing depth is 100-600mm, the focal length is 50mm, the angle scanning is +/-30 degrees, and the sampling rate is more than or equal to 10MHz; 128 array elements are arranged, wherein the array element distance is 1mm, the center frequency is 150kHz, the pulse width is 2-3 cycles, the focusing depth is 100-600mm, the focal length is 50mm, the angle scanning is +/-45 degrees, and the sampling rate is more than or equal to 20MHz.
9. The intelligent detection method for leakage of the large-thickness floor of the existing building based on the multi-mode sensor array according to claim 5, wherein in step S4, the microwave sensor array is configured by: Standard configuration, namely 5×5 array scale, 0.5m sensor spacing, 2 seconds of measurement time/point, 50 seconds of total measurement time, 4m 2 coverage area, 0.5m spatial resolution and application scene, namely conventional detection; High-precision configuration, namely, array scale 7 multiplied by 7, sensor spacing 0.3m, measurement time/point 3 seconds, total measurement time 147 seconds, coverage area 3.6m 2 , spatial resolution 0.3m, and application scene key area; The rapid scanning configuration comprises 3X 3 array scale, 0.8m sensor interval, 1 second measuring time/point, 9 seconds total measuring time, 3.2m 2 coverage area, 0.8m spatial resolution and preliminary screening of applicable scenes.
10. The intelligent detection method for leakage of the large-thickness floor of the existing building based on the multi-mode sensor array according to claim 5, wherein in step S5, vibration test configuration parameters are as follows: Passive environment vibration, excitation source, sensor type, MEMS triaxial accelerometer, grid-shaped arrangement mode, interval 0.5-1m, sampling rate 2kHz, acquisition time length 5-10 min, sensitivity >1000mV/g, dynamic range >60dB, measuring frequency band 1-500Hz, application target, low frequency integral response, damping identification; Active ultrasonic excitation, wherein an excitation source is an ultrasonic excitation source, 40-100kHz sweeps, a sensor type is an MEMS triaxial accelerometer, the arrangement mode is grid-shaped, the distance is 0.5-1m, the sampling rate is 500kHz, the acquisition time length is 0.5 seconds per frequency point, the sensitivity is more than 1000mV/g, the dynamic range is more than 60dB, the measuring frequency range is 40-100kHz, and the application target is high-frequency local defects and modal analysis.

Description

Intelligent detection system and method for leakage of large-thickness bottom plate of existing building based on multi-mode sensor array Technical Field The invention belongs to the technical field of leakage detection, and particularly relates to an intelligent detection system and method for leakage of a large-thickness bottom plate of an existing building based on a multi-mode sensor array. Background The leakage detection of the large-thickness bottom plate of the existing building faces a series of technical problems, and the problems severely restrict the accuracy and efficiency of leakage diagnosis. The detection depth of the traditional infrared detection technology is seriously insufficient, and the temperature abnormality of the surface layer within 50 mm can be detected generally. However, the thickness of the base plate of the existing buildings such as basements, subway stations and civil air defense projects is generally 300-800 mm, leakage often occurs in the deep layer or the bottom of the base plate, and infrared signals on the surface layer are weak or even cannot be detected. This limitation of the detection depth results in a large number of deep leaks being missed, which often results in serious damage when the water seepage penetrates the surface, missing the best maintenance opportunity. Insufficient positioning accuracy is another prominent problem. While the prior art is able to approximate the leak area, there are significant drawbacks in accurately locating the source of the leak. The plane positioning error is usually more than 100 mm, and the depth direction lacks an effective positioning means, so that whether leakage occurs at the upper part, the middle part or the bottom of the bottom plate cannot be determined. The uncertainty of the positioning brings great trouble to maintenance construction, constructors can only excavate blindly or treat in a large scale by experience, materials and working hours are wasted, and normal structures can be damaged. The lack of quantitative assessment capability severely affects leak management decisions. Most of traditional detection methods can only give qualitative descriptions of [ slight leakage ], [ moderate leakage ], [ severe leakage ], and the like, and the leakage degree and leakage amount cannot be quantified. Property managers and maintenance units have difficulty in formulating reasonable maintenance schemes and budgets based on the reasonable maintenance schemes and budgets, and insurance claims are often disputed due to lack of quantitative data. Poor environmental suitability is an important factor limiting the reliability of detection. The surface of the existing building bottom plate is often covered by decoration materials such as ceramic tiles, floor paint and the like, and the materials can change the surface thermal characteristics and reflection characteristics and interfere detection signals. The change of the ambient temperature and the humidity can also obviously influence the detection result, and especially the passive infrared detection is highly dependent on the temperature difference between day and night and almost cannot work in overcast and rainy days. The low detection efficiency increases the detection cost and the time cost. The traditional method needs multiple on-site detection, and after each detection, a professional is required to conduct complex data interpretation and analysis, and the whole detection period often needs days or even weeks. For continuously operated facilities such as subway stations, the detection time window is very limited, and low efficiency means that large-area detection cannot be completed. The lack of leak path tracking capability makes the problem of addressing the symptoms without addressing the root cause ubiquitous. The prior art only can find out the signs of leakage on the surface, and cannot trace back where the water seepage is from the water seepage, and what path to reach the detection position. This results in maintenance often being only a management of surface symptoms, while the actual source and path of the leak is not found or treated, and the leak will quickly reappear elsewhere, creating a "patch-not-patch" situation. Insufficient intelligentization results in a high degree of reliance on professionals. The existing detection method needs the participation of expertise with abundant experience in each link from data acquisition, analysis to diagnosis and scheme formulation, has strong subjectivity in manual interpretation and poor result consistency, and is easy to misjudge due to fatigue or negligence. The shortage and high expense of professionals also limit the popularity of detection services. The difficulty in identifying early micro-leaks makes preventative maintenance difficult to implement. The passive detection method is very weak in signal for early micro-leakage with low water content and small leakage amount, and is often submerged in backgroun