Intelligent Detection Device and Method for Voiding Damage in Tall Composite Structures Based on Acoustic Signals

Abstract

This invention discloses an intelligent detection device for void damage on tall composite structures based on acoustic signals. The device includes a wall-climbing robot capable of walking on the structure; the robot is equipped with an electromagnetic hammer to generate high-definition acoustic signals; and a signal receiver to collect these signals. A control system receives the acoustic signals from the receiver, processes and analyzes them, and constructs a three-class convolutional CNN neural network model based on void, healthy, and invalid data. The wall-climbing robot, combined with the three-class convolutional CNN model and a computing card, enables real-time identification of void damage during the inspection process and marks the void damage site. This invention enables automated detection and identification of void damage on the surface of tall composite structures, significantly reducing the workload and safety risks for personnel working on structures near high edges.

Description

Technical Field

[0001] This invention relates to the field of intelligent detection of void damage in tall composite structures. More specifically, this invention relates to an intelligent detection device for void damage in tall composite structures based on acoustic signals. Background Technology

[0002] With the construction of ultra-long span bridges, tall composite bridge towers are widely used due to their advantages such as high load-bearing capacity, lightweight structure, and rapid construction. However, due to factors such as construction technology and harsh service conditions, void defects are prone to occur in the structure, and internal void defects may cause structural damage and failure, thereby affecting the load-bearing capacity of the structure. Therefore, void damage detection should be carried out in a timely manner after the installation and construction of the composite structure.

[0003] Currently, the detection of void damage mainly relies on manual tapping combined with signal acquisition from coupled sensors or human ear listening to sound signals for analysis and judgment. However, for the detection of dangerous operations such as ultra-high cable towers, personnel need to tap and judge multiple points on each side of the high position, which is difficult and extremely dangerous. Therefore, there is an urgent need for an intelligent detection device and identification method for void damage of high-rise composite structures. Summary of the Invention

[0004] To achieve these objectives and other advantages according to the present invention, a preferred embodiment of the present invention provides an intelligent detection device for void damage in tall composite structures based on acoustic signals, comprising:

[0005] A wall-climbing robot that can walk on tall, modular structures; the wall-climbing robot is equipped with an electromagnetic hammer that generates high-definition sound signals, and the wall-climbing robot is also equipped with a signal receiver to collect the sound signals;

[0006] The control system receives the acoustic signals collected by the signal receiver, performs data processing and analysis on the collected acoustic signals, and constructs a three-class convolutional CNN neural network model based on empty, healthy, and invalid data.

[0007] The wall-climbing robot combines a three-class convolutional CNN neural network model and a computing card to achieve real-time identification of the detachment damage status during the walking inspection process and to mark the detachment damage site.

[0008] Preferably, the wall-climbing robot is a magnetic wall-climbing robot, and its walking mechanism uses magnetic tracks; and the wall-climbing robot can autonomously plan its path and walk on the towering composite structure.

[0009] Preferably, the wall-climbing robot is equipped with a binocular camera and a lidar, which can detect whether there are obstacles in the surrounding area during the walking process, so as to intelligently avoid them during the walking process.

[0010] Preferably, the front end of the magnetic wall-climbing robot is equipped with an electromagnetic hammer, an array microphone, and a defect marking device. The signal receiver is an array microphone, and the defect marking device is used to mark the site of the detachment damage.

[0011] Preferably, the electromagnetic hammer is triggered by the control system to perform instantaneous vibrations at equal intervals during the wall-climbing robot's movement, and the magnitude of the vibration force can be adjusted by the trigger current; the array microphone collects short-term sound signals while performing instantaneous vibrations.

[0012] Preferably, the collected acoustic signals are processed and analyzed, specifically including the following steps:

[0013] The collected acoustic signals are converted into digital signals, filtered, and then subjected to time-frequency analysis to obtain wavelet and Mel-time spectrograms. The wavelet and Mel-time spectrograms are then fed into a deep learning network model for analysis and recognition.

[0014] Preferably, the wall-climbing robot is equipped with a computing card, which is pre-written with signal analog-to-digital conversion, time-frequency analysis, and deep learning CNN neural network model algorithms; the collected acoustic signals are transmitted to the computing card for signal conversion, time-frequency analysis, and model recognition to obtain the real-time signal classification and recognition results of the delamination damage state.

[0015] Preferably, when the computing power card analyzes and finds that the detection area is a void damage or invalid data, it sends a signal to the control system. The control system sends an instruction to the defect marking device to mark the defect. The defect marking device has a color block at the bottom, which is excited by current to probe downward to the structural surface to form a mark.

[0016] Preferably, the control system includes remote real-time image monitoring based on a camera, local storage of detected sound signal data, and cloud transmission functions.

[0017] On the other hand, another technical solution of the present invention provides a detection method for a smart detection device for void damage in tall composite structures, comprising the following steps:

[0018] S1: Place the wall-climbing robot at the location of the towering composite structure wall. The wall-climbing robot performs path planning and automatically avoids obstacles during the climbing process through binocular cameras and LiDAR.

[0019] S2: The wall-climbing robot moves along the wall by driving the tracks with a motor. During the movement, the control system triggers an electromagnetic hammer to perform a striking operation based on the given walking distance.

[0020] S3: At the same time as the hammering action is generated, the front array microphone of the wall-climbing robot is triggered to collect the sound signal generated by the hammering vibration. The control system converts the collected sound signal into a digital signal and performs interception, filtering and time-frequency analysis processing on the digital signal data.

[0021] S4: Construct a CNN neural network model for detachment damage based on acoustic signals that have been trained under the same conditions. Input the signal features extracted and analyzed in S3 into the network model for analysis and identification to obtain the corresponding three classification results: detachment, healthy data, and invalid data.

[0022] S5: The computing card transmits the classification results to the control system. When the result is empty or invalid data, a current is applied to momentarily trigger the marking device to mark the damage. When the result is healthy, the wall-climbing robot continues to move forward to the next point.

[0023] S6. Repeat S1-S5 until the wall-climbing robot has completed the inspection of all the tower walls of the towering composite structure.

[0024] The present invention has at least the following beneficial effects: The present invention provides an intelligent detection device for void damage in tall composite structures, which can realize the automated detection and identification of void damage on the surface of tall composite structures, replacing manual point-by-point tapping detection on each surface, greatly reducing the labor intensity of personnel on tall adjacent structures and reducing safety risks. At the same time, the algorithm model is used to accurately evaluate whether void damage has occurred inside the steel shell cable tower, reducing the interference of human factors on the accuracy of the result evaluation, thereby improving the detection efficiency and accuracy of void damage on the surface of composite structures.

[0025] Other advantages, objectives and features of the present invention will become apparent in part from the following description, and in part from those skilled in the art through study and practice of the invention. Attached Figure Description

[0026] Figure 1 This is a schematic diagram of the wall-climbing robot in this invention. Detailed Implementation

[0027] The present invention will now be described in further detail with reference to the accompanying drawings, so that those skilled in the art can implement it based on the description.

[0028] The following description is intended to disclose the present invention and enable those skilled in the art to implement it. The preferred embodiments described below are merely examples, and other obvious modifications will occur to those skilled in the art. The basic principles of the invention defined in the following description can be applied to other embodiments, modifications, improvements, equivalents, and other technical solutions that do not depart from the spirit and scope of the invention.

[0029] Those skilled in the art should understand that, in the disclosure of this invention, the terms "longitudinal," "lateral," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, the above terms should not be construed as limiting this invention.

[0030] It is understood that the term "a" should be understood as "at least one" or "one or more", that is, in one embodiment, the number of an element can be one, while in another embodiment, the number of the element can be multiple, and the term "a" should not be understood as a limitation on the number.

[0031] like Figure 1 As shown, a preferred embodiment of the present invention provides an intelligent detection device for void damage in tall composite structures based on acoustic signals, comprising:

[0032] A wall-climbing robot 1 can walk on a tall, modular structure; the wall-climbing robot 1 is equipped with an electromagnetic hammer 2, which generates a high-definition sound signal; the wall-climbing robot 1 is also equipped with a signal receiver 3 to collect the sound signal.

[0033] The control system receives the acoustic signals collected by the signal receiver, performs data processing and analysis on the collected acoustic signals, and constructs a three-class convolutional CNN neural network model based on empty, healthy, and invalid data.

[0034] The wall-climbing robot combines a three-class convolutional CNN neural network model and a computing card to achieve real-time identification of the detachment damage status during the walking inspection process and to mark the detachment damage site.

[0035] The above technical solution can realize the automated detection and identification of void damage on the surface of tall composite structures, replacing manual point-by-point tapping inspection, greatly reducing the labor intensity of personnel on tall edge structures and reducing safety risks. At the same time, the algorithm model is used to accurately evaluate whether void damage has occurred inside the steel shell cable tower, reducing the interference of human factors on the accuracy of the evaluation results, thereby improving the detection efficiency and accuracy of void damage on the surface of composite structures.

[0036] During actual testing, the wall-climbing robot is placed at the location of the tower wall of the tall composite structure. The wall-climbing robot can autonomously plan its path on the tall composite structure. The overall plan follows an S-shaped back-and-forth motion from the bottom to the top, which can ensure all-round coverage of the walking path. The wall-climbing robot strikes the tower wall with an electromagnetic hammer 2. The high-definition sound signal emitted is reflected by the tower wall and then received by a signal receiver.

[0037] Moreover, the wall-climbing robot can be a magnetic wall-climbing robot, with its walking mechanism using magnetic tracks 6, and the towering combined structure made of steel, which can be attracted by magnets. Therefore, the use of magnetic tracks can combine walking and adsorption functions, realizing multi-angle vertical adsorption and climbing.

[0038] In another technical solution, the wall-climbing robot is equipped with a binocular camera 4 and a lidar 5, which can detect whether there are obstacles around it during the walking process, so as to intelligently avoid them during the walking process.

[0039] When detecting whether there are obstacles in the surroundings during the walking process, if there are obstacles in front of the walking path, the control system will control the wall-climbing robot to change the path to avoid the obstacles.

[0040] In another technical solution, the front end of the magnetic wall-climbing robot is equipped with an electromagnetic hammer, an array microphone, and a defect marking device. The signal receiver is an array microphone, and the defect marking device 7 is used to mark the site of the detachment damage.

[0041] In another technical solution, the electromagnetic hammer is triggered by the control system to perform instantaneous vibrations at equal intervals during the wall-climbing robot's movement. The magnitude of the vibration force can be adjusted by the trigger current. The array microphone collects short-term sound signals at the same time as the instantaneous vibration. This is because the effective time for the sound signal to be generated and lasts is only 0.02 seconds. Therefore, in order to capture the effective sound signal, signal acquisition must start at the same time as the excitation; otherwise, the effective sound signal cannot be collected.

[0042] In another technical solution, the collected acoustic signals are processed and analyzed, specifically including the following steps:

[0043] The collected acoustic signals are converted into digital signals, filtered, and then subjected to time-frequency analysis to obtain wavelet and Mel-time spectrograms. The wavelet and Mel-time spectrograms are then fed into a deep learning network model for analysis and recognition.

[0044] In another technical solution, the wall-climbing robot is equipped with a computing card, which is pre-written with signal analog-to-digital conversion, time-frequency analysis, and deep learning CNN neural network model algorithms. The collected acoustic signals are transmitted to the computing card for signal conversion, time-frequency analysis, and model recognition to obtain the real-time signal classification and recognition results of the delamination damage state.

[0045] In another technical solution, when the computing power card analyzes and finds that the detection area is a void damage or invalid data, it sends a signal to the control system. The control system sends an instruction to the defect marking device to mark the defect. The defect marking device has a color block at the bottom, which is excited by current to probe downward to the structural surface to form a mark.

[0046] In another technical solution, the control system includes remote real-time image monitoring based on a camera, local storage of detected sound signal data, and cloud transmission functions.

[0047] On the other hand, another technical solution of the present invention provides a detection method for a smart detection device for void damage in tall composite structures, comprising the following steps:

[0048] S1: Place the wall-climbing robot at the location of the towering composite structure wall. The wall-climbing robot performs path planning and automatically avoids obstacles during the climbing process through binocular cameras and LiDAR.

[0049] S2: The wall-climbing robot moves along the wall by driving the tracks with a motor. During the movement, the control system triggers an electromagnetic hammer to perform a striking operation based on the given walking distance.

[0050] S3: At the same time as the hammering action is generated, the front array microphone of the wall-climbing robot is triggered to collect the sound signal generated by the hammering vibration. The control system converts the collected sound signal into a digital signal and performs interception, filtering and time-frequency analysis processing on the digital signal data.

[0051] S4: Construct a CNN neural network model for detachment damage based on acoustic signals that have been trained under the same conditions. Input the signal features extracted and analyzed in S3 into the network model for analysis and identification to obtain the corresponding three classification results: detachment, healthy data, and invalid data.

[0052] S5: The computing card transmits the classification results to the control system. When the result is empty or invalid data, a current is applied to momentarily trigger the marking device to mark the damage. When the result is healthy, the wall-climbing robot continues to move forward to the next point.

[0053] S6. Repeat S1-S5 until the wall-climbing robot has completed the inspection of all the tower walls of the towering composite structure.

[0054] The above technical solution can realize the automated detection and identification of void damage on the surface of tall composite structures, replacing manual point-by-point tapping inspection, greatly reducing the labor intensity of personnel on tall edge structures and reducing safety risks. At the same time, the algorithm model is used to accurately evaluate whether void damage has occurred inside the steel shell cable tower, reducing the interference of human factors on the accuracy of the evaluation results, thereby improving the detection efficiency and accuracy of void damage on the surface of composite structures.

[0055] Although embodiments of the present invention have been disclosed above, they are not limited to the applications listed in the specification and embodiments. They can be applied to various fields suitable for the present invention. For those skilled in the art, other modifications can be easily made. Therefore, without departing from the general concept defined by the claims and their equivalents, the present invention is not limited to the specific details and illustrations shown and described herein.

Claims

1. A kind of high-rise combined structure void damage intelligent detection device based on acoustic signal, it is characterized in that, include: A wall-climbing robot capable of walking on towering composite structures; the wall-climbing robot is equipped with an electromagnetic hammer that generates high-definition sound signals, and also equipped with a signal receiver to collect the sound signals; the wall-climbing robot is a magnetic wall-climbing robot, and its walking mechanism uses magnetic tracks; and the wall-climbing robot can autonomously plan its path and walk on towering composite structures. The control system receives the acoustic signals collected by the signal receiver, performs data processing and analysis on the collected acoustic signals, and constructs a three-class convolutional CNN neural network model based on empty, healthy, and invalid data. The wall-climbing robot combines a three-class convolutional CNN neural network model and a computing card to achieve real-time identification of the detachment damage status during the walking inspection process and to mark the detachment damage site. The front end of the magnetic wall-climbing robot is equipped with an electromagnetic hammer, an array microphone, and a defect marking device. The signal receiver is an array microphone, and the defect marking device is used to mark the site of the detachment damage. The electromagnetic hammers are triggered by the control system to perform instantaneous vibrations at equal intervals during the wall-climbing robot's movement, and the magnitude of the vibration force can be adjusted by the trigger current; the array microphones collect short-term sound signals while performing instantaneous vibrations. The collected acoustic signals undergo data processing and analysis, specifically including the following steps: The collected acoustic signals are converted into digital signals. After filtering the digital signals, time-frequency analysis is performed to obtain wavelet and Mel-time spectrograms. The wavelet and Mel-time spectrograms are then fed into a deep learning network model for analysis and recognition. The wall-climbing robot is equipped with a computing card, which is pre-written with signal analog-to-digital conversion, time-frequency analysis, and deep learning CNN neural network model algorithms. The collected acoustic signals are transmitted to the computing card for signal conversion, time-frequency analysis, and model recognition to obtain the real-time signal classification and recognition results of the delamination damage status. When the computing power card analyzes and finds that the detection area is a void damage or invalid data, it sends a signal to the control system. The control system sends an instruction to the defect marking device to mark the defect. The defect marking device has a color block at the bottom, which is excited by current to probe downward to the structural surface to form a mark.

2. The acoustic signal based intelligent detection device for out-of-plumb damage of high-rise combined structure according to claim 1, characterized in that, The wall-climbing robot is equipped with a binocular camera and a lidar, which can detect whether there are obstacles in the surrounding area during the walking process, so as to intelligently avoid them.

3. The acoustic signal based intelligent detection device for out-of-plumb damage of high-rise combined structure according to claim 1, characterized in that, The control system includes remote real-time image monitoring based on a camera, local storage of sound signal data, and cloud transmission functions.

4. A detection method for a smart detection device for void damage in tall composite structures based on any one of claims 1-3, characterized in that, Includes the following steps: S1: Place the wall-climbing robot at the location of the towering composite structure wall. The wall-climbing robot performs path planning and automatically avoids obstacles during the climbing process through binocular cameras and LiDAR. S2: The wall-climbing robot moves along the wall by driving the tracks with a motor. During the movement, the control system triggers an electromagnetic hammer to perform a striking operation based on the given walking distance. S3: At the same time as the hammering action is generated, the front array microphone of the wall-climbing robot is triggered to collect the sound signal generated by the hammering vibration. The control system converts the collected sound signal into a digital signal and performs interception, filtering and time-frequency analysis processing on the digital signal data. S4: Construct a CNN neural network model for detachment damage based on acoustic signals that have been trained under the same conditions. Input the signal features extracted and analyzed in S3 into the network model for analysis and identification to obtain the corresponding three classification results: detachment, healthy data, and invalid data. S5: The computing card transmits the classification results to the control system. When the result is empty or invalid data, the current is applied to momentarily trigger the marking device to mark the damage. If the result is healthy, the wall-climbing robot continues to move forward to the next point; S6. Repeat S1-S5 until the wall-climbing robot has completed the inspection of all the tower walls of the towering composite structure.

Images