CN112627023A - Intelligent bridge detection method and system and intelligent bridge detection robot - Google Patents
Intelligent bridge detection method and system and intelligent bridge detection robot Download PDFInfo
- Publication number
- CN112627023A CN112627023A CN202011322256.6A CN202011322256A CN112627023A CN 112627023 A CN112627023 A CN 112627023A CN 202011322256 A CN202011322256 A CN 202011322256A CN 112627023 A CN112627023 A CN 112627023A
- Authority
- CN
- China
- Prior art keywords
- bridge
- detection
- information
- detected
- area
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- E—FIXED CONSTRUCTIONS
- E01—CONSTRUCTION OF ROADS, RAILWAYS, OR BRIDGES
- E01D—CONSTRUCTION OF BRIDGES, ELEVATED ROADWAYS OR VIADUCTS; ASSEMBLY OF BRIDGES
- E01D19/00—Structural or constructional details of bridges
- E01D19/10—Railings; Protectors against smoke or gases, e.g. of locomotives; Maintenance travellers; Fastening of pipes or cables to bridges
- E01D19/106—Movable inspection or maintenance platforms, e.g. travelling scaffolding or vehicles specially designed to provide access to the undersides of bridges
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/017—Detecting movement of traffic to be counted or controlled identifying vehicles
- G08G1/0175—Detecting movement of traffic to be counted or controlled identifying vehicles by photographing vehicles, e.g. when violating traffic rules
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/052—Detecting movement of traffic to be counted or controlled with provision for determining speed or overspeed
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/065—Traffic control systems for road vehicles by counting the vehicles in a section of the road or in a parking area, i.e. comparing incoming count with outgoing count
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N5/00—Details of television systems
- H04N5/76—Television signal recording
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Architecture (AREA)
- Civil Engineering (AREA)
- Structural Engineering (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Manipulator (AREA)
Abstract
The invention discloses an intelligent bridge detection method, which comprises the following specific steps: collecting basic information related to the detected bridge; analyzing the basic situation of the bridge to be detected by using an image fusion detection method according to the acquired basic information, and generating a maintenance suggestion of the bridge to be detected; the basic information of the detected bridge comprises bridge environment information, traffic information and bridge basic information. The invention also discloses a detection system for realizing the bridge detection method according to the intelligent bridge detection method, and discloses a functional bridge detection robot, and the bridge detection method and the detection system are used for realizing bridge detection. The method can realize the accurate positioning of bridge diseases, comprehensively detect the bridge by using ultrasonic waves and an infrared thermal imaging mode, and make a bridge maintenance strategy for the reference of maintenance personnel according to the detection and analysis result of the bridge.
Description
Technical Field
The invention belongs to the technical field of intelligent traffic facilities, and particularly relates to an intelligent bridge detection method and system and an intelligent bridge detection robot.
Background
The highway bridge is an important component of highway traffic, is related to highway traffic and transportation, and is further related to the development of national economic construction. In order to fully exert the function of the highway bridge and prolong the service life of the highway bridge, various measures are required to perform necessary maintenance on the highway bridge in the process of operating the highway bridge.
The maintenance work of the highway bridge is directly related to the driving safety and smoothness of the highway, and the deterioration of the highway bridge is effectively avoided by maintaining the highway bridge. The highway bridge bears the abrasion of vehicles and the damage and the erosion of natural factors for a long time within the design service life, so that the bridge decays; with the rapid development of highway traffic industry, the traffic capacity and the bearing capacity of a highway bridge are continuously increased, so that the disease development of the highway bridge is increased, if the bridge is not subjected to regular and standard maintenance, the function of the bridge is reduced in long-time high-load use, even the function of the bridge is completely lost, the vehicle driving safety is seriously influenced, even accidents occur, and casualties are caused. How to timely and accurately master the health state of an existing highway bridge becomes a key for effective maintenance management of the highway bridge, so bridge detection is becoming more and more important.
The traditional bridge detection method is manual surface visual inspection, the working environment is severe and dangerous, the difficulty is high, the efficiency is low, the experience of professional detection personnel is taken as an important basis, the subjectivity is strong, and the influence of environmental factors is easy to realize. Along with the continuous exploration of bridge detection at home and abroad, the bridge detection is gradually transited to nondestructive detection technologies such as precision instrument crack detection by manual detection, and automatic equipment such as a vehicle-mounted bridge detection vehicle, a robot or an unmanned aerial vehicle replaces detection personnel to complete bridge detection.
However, the inventors found in the course of research that the following problems exist in the existing bridge inspection apparatus.
Firstly, the vehicle-mounted bridge inspection vehicle has the problems of high cost and large volume. The price of the bridge inspection vehicle is generally one or two million, the daily maintenance cost is very high, a plurality of medium and small bridge maintenance units are often unable to bear the bridge inspection vehicle, and the rental price is still very expensive; secondly, the efficiency is high, the bridge inspection vehicle is limited by the length of the truss, when the bottom of the bridge is constructed, the bridge which is too wide can be constructed only on one side, and the construction efficiency and quality are reduced; and the bridge inspection vehicle is large in size, and often needs to perform certain traffic control during operation, occupies lanes and blocks traffic. Even if the existing bridge inspection vehicle carries detection equipment to detect a bridge, detection personnel still need to drive the inspection vehicle to move the position.
Secondly, the bridge detection robot is still in a laboratory research stage at present, detection equipment with large weight cannot be carried under the influence of self load, the coverage area of single scanning is small during robot detection, and the detection efficiency is not improved; the robot which is adsorbed on the surface of the bridge by adopting the magnetic force principle for detection can only be used for detecting the bridge containing iron materials, cannot be universally used for detecting all highway bridges, and does not improve the detection efficiency due to slow movement speed; the robot which is adsorbed on the surface of the bridge by adopting the negative pressure principle to detect cannot safely realize the movement of the bridge structure change surface from the bridge deck to the bridge piers and the like, cannot detect all areas to be detected of the bridge, and has limited detection range.
Thirdly, the guardrail robot bridge inspection device has the problem that the existing expressway bridge inspection device cannot be universally checked. The form of high-speed bridge guardrail changes variously, mainly divides four big types: the fence type bridge inspection device comprises a fence type bridge inspection device body, a lattice type bridge inspection device body and a lattice type bridge inspection device body, the fence type bridge inspection device body is applicable to the fence type bridge inspection device body, the lattice type bridge inspection device body is applicable to the fence type bridge inspection device body, the fence type bridge inspection device body is adaptive to a rack, guide rails need to be installed or.
Fourthly, the unmanned aerial vehicle bridge inspection device has the problem that the automation degree is low and the stability and the reliability cannot be guaranteed. Bridge detection of the unmanned aerial vehicle is only used as an auxiliary mode of manual detection at present, detection personnel are required to control the unmanned aerial vehicle on site, but when the unmanned aerial vehicle is positioned below a bridge, the unmanned aerial vehicle is out of control due to the fact that the bridge blocks signals such as a GPS (global positioning system) and the like, and bridge detection cannot be carried out; in addition, unmanned aerial vehicle easily receives a great deal of environmental factor such as wind and rain to disturb and leads to out of control, can't carry out the bridge and detect, appears unmanned aerial vehicle crash even, causes loss of property.
Disclosure of Invention
Aiming at the defects of the prior art, the invention aims to provide an intelligent bridge detection method, an intelligent bridge detection system and an intelligent bridge detection robot, which are convenient for bridge detection and have high detection precision.
In order to achieve the purpose, the invention adopts the following technical scheme:
an intelligent bridge detection method is characterized by comprising the following specific steps:
s1: acquiring basic information of the bridge according to the bridge environment information, the traffic information around the bridge and the bridge;
s2: acquiring a detected bridge detection image and information on site, and analyzing by using an image fusion detection method according to the acquired detection image and information to obtain a bridge detection result;
s3: and analyzing and generating a maintenance suggestion of the bridge to be detected according to the bridge environment information, the traffic information around the bridge, the basic bridge information and the bridge detection result.
The method comprises the following steps of collecting detected images and information of a detected bridge on site, analyzing the detected images and information by using an image fusion detection method according to the collected detected images and information, and obtaining a bridge detection result, wherein the method comprises the following specific steps:
s21: controlling an ultrasonic generator to send ultrasonic waves to a detection area for excitation; acquiring infrared temperature data of a detected area of the bridge by using a thermosensitive sensor to form an infrared image, and qualitatively judging bridge diseases to obtain a bridge detection qualitative result;
the bridge detected area comprises all bottom surface structures, bottom structures and auxiliary structures;
s22: collecting video image data of a bridge detected area, and carrying out quantitative judgment on a bridge disease area through a neural network algorithm to obtain a bridge detection quantitative result;
s23: and performing row matrix coupling calculation on the bridge detection qualitative result and the bridge detection quantitative result to further obtain a final bridge detection result.
The bridge detection qualitative result is a bridge disease area and a bridge health area;
and the bridge detection quantitative result is the type, size and specific position of the bridge diseases.
The specific steps of step S21 are:
s211: the ultrasonic generator sends an excitation signal to the detection area, and a controller on the detection robot records the time for sending the excitation signal;
s212: a thermosensitive sensor on the detection robot acquires a feedback signal of the detected surface of the bridge after receiving an excitation signal of the ultrasonic generator, acquires infrared temperature data of a detected area of the bridge to form an infrared image, and a controller on the detection robot records the time of receiving the feedback signal;
s213: and judging the crack depth and the crack size of the bridge according to the time from sending the excitation signal to collecting the feedback signal and the infrared image, so as to identify the cracks of the bridge and obtain a bridge detection qualitative result.
The method comprises the following steps of acquiring infrared temperature data of the surface of a bridge to be detected by using a thermosensitive sensor to form an infrared image, and specifically comprises the following steps:
s2121: the thermal sensor captures the surface temperature of the bridge measured area before and after receiving the excitation signal and acquires infrared imaging data;
s2122: and generating an infrared image of the detected area of the bridge according to the acquired imaging data.
The calculation formula of the crack depth of the bridge in the bridge detection qualitative result is as follows:
wherein alpha is thermal diffusion efficiency, rho is the density of the material in the measured area of the bridge, c is specific heat, d is the crack depth of the bridge, tmIs the optimum detection time at tmMeasured maximum temperature difference DeltaTmQ is pulsed excitation;
and judging the bridge disease area and the healthy bridge area by using the infrared image.
The specific steps of step S22 are:
s221: acquiring the existing healthy bridge image and various bridge disease images, and marking the images with labels;
s222: the collected image marked with the label is used as a training sample, various bridge disease images marked with the label are used as positive training samples, and the healthy bridge image marked with the label is used as a negative training sample;
s223: manufacturing a neural network model, and training the neural network model by adopting a positive training sample and a negative training sample;
and S224, inputting the visual image data of the disease area collected by the camera into the trained neural network model to obtain a bridge detection quantitative result.
An intelligent bridge detection system, comprising:
a collection end: the system is used for acquiring detection information, environmental information and traffic information of a bridge to be detected and acquiring basic information of the bridge;
a background analysis server: the bridge maintenance system is used for analyzing bridge disease information according to the relevant information collected by the collection end, determining the position of a bridge disease to obtain a bridge detection result, formulating a bridge renovation and maintenance plan according to the analysis result of the data analysis module on the detection information, generating a bridge maintenance suggestion, and sending the bridge detection result and the bridge maintenance suggestion to the client;
a client: and receiving and displaying the bridge detection result and the bridge maintenance suggestion.
The background analysis server comprises:
a data receiving module: receiving detected bridge detection information, environment information and traffic information acquired by an acquisition end, and acquiring basic bridge information;
bridge detection analysis module: the system comprises a collecting terminal, a data acquisition terminal and a data processing terminal, wherein the collecting terminal is used for collecting image information and detection data of a bridge to be detected, so as to determine a disease area and a health area of the bridge to be detected and determine the disease type of the disease area;
the environment information analysis module: analyzing environmental information acquired by a sensor to generate an environmental analysis result;
a traffic information analysis module: recording the traffic information of the bridge to be detected by using a camera device on the bridge;
the bridge maintenance suggestion generation module: generating bridge renovation and maintenance results according to the analysis results of the data analysis module, the environmental information analysis module and the traffic information module;
a data sending module: and sending the bridge detection result and the bridge maintenance suggestion to the client.
The utility model provides an intelligent bridge inspection robot, its concrete structure is: the detection robot comprises walking wheels, a steel cable detection device, a camera, a walking speed control device and a protection device, wherein the walking wheels are installed at the bottom end of the detection robot, the steel cable detection device is arranged between the walking wheels, the camera is installed on the detection robot, the walking speed control device is arranged obliquely above the camera, and the protection device is arranged on the outer side of the detection robot; protection device and inspection robot's connected mode is for bonding, and protection device is the rubber material, inspection robot shanghai is provided with temperature and humidity sensor, baroceptor and air quality sensor.
The invention has the beneficial effects that: the bridge is detected by using the detection method provided by the invention, the accurate positioning of the bridge diseases can be realized, the qualitative bridge result is obtained by using ultrasonic and infrared thermal imaging, the image of the detected bridge is collected by using the camera, the qualitative bridge result is fused with the quantitative bridge result obtained by image recognition through a neural network algorithm, the detection result of the bridge is finally obtained, the bridge is comprehensively detected, the detection efficiency and the accuracy are improved, and meanwhile, the bridge maintenance strategy is formulated according to the detection and analysis result of the bridge for the reference of maintenance personnel.
Drawings
FIG. 1 is a flow chart of a bridge inspection method of the present invention;
FIG. 2 is a system block diagram of a bridge detection method;
FIG. 3 is a schematic view of a maintenance decision model;
FIG. 4 is a rear view structural diagram of the inspection robot of the present invention;
FIG. 5 is a left side view structural diagram of the inspection robot of the present invention;
FIG. 6 is a schematic structural diagram of a fourth embodiment of the present invention;
FIG. 7 is a schematic diagram of the overall top view of the present invention;
FIG. 8 is an enlarged view of the structure at A of FIG. 7 in accordance with the present invention;
FIG. 9 is a schematic structural diagram of a fifth embodiment of the present invention;
FIG. 10 is a schematic top-down sectional view of a fifth embodiment of a bridge pier connected to a second fastener according to the present invention;
fig. 11 is a side sectional view of a second fixing member according to a fifth embodiment of the present invention.
Wherein, 1, a travelling wheel; 2. a wire rope detection device; 3. a camera; 4. a traveling speed control device; 5. a guard, 6 detecting the robot; 7. a bridge body; 8. an abutment; 9. a first fixing member; 10. a first cable body; 11. a bridge pier; 12. a limiting plate; 13, a groove; 14. a first limit groove; 15. a first locking lever, 16 a second fastener, 17 a second cable body; 18. guide rail, 19, card slot; 20. and a second limit groove 21 is formed in the second locking rod.
Detailed Description
The invention will be further described with reference to the accompanying drawings and the detailed description below:
example one
As shown in fig. 1, an intelligent bridge detection method specifically includes the steps of:
s1: acquiring basic information of the bridge according to the bridge environment information, the traffic information around the bridge and the bridge;
s2: acquiring a detected bridge detection image and information on site, and analyzing by using an image fusion detection method according to the acquired detection image and information to obtain a bridge detection result;
s3: and analyzing and generating a maintenance suggestion of the bridge to be detected according to the bridge environment information, the traffic information around the bridge, the basic bridge information and the bridge detection result.
The bridge environment information comprises temperature, humidity, air pressure and air quality; the traffic information around the bridge comprises traffic flow, vehicle type and vehicle running speed, and the traffic information around the bridge is acquired through a camera device arranged on the bridge; the basic bridge information is obtained from the existing data, including bridge type, bridge drawing, construction unit, etc.
The method comprises the following steps of collecting detected images and information of a detected bridge on site, analyzing the detected images and information by using an image fusion detection method according to the collected detected images and information, and obtaining a bridge detection result, wherein the method comprises the following specific steps:
s21: controlling an ultrasonic generator to send ultrasonic waves to a detection area for excitation; acquiring infrared temperature data of a detected area of the bridge by using a thermosensitive sensor to form an infrared image, and qualitatively judging bridge diseases to obtain a bridge detection qualitative result;
the bridge detected area comprises all bottom surface structures, bottom structures and auxiliary structures;
s22: collecting video image data of a bridge detected area, and carrying out quantitative judgment on a bridge disease area through a neural network algorithm to obtain a bridge detection quantitative result;
s23: and performing row matrix coupling calculation on the bridge detection qualitative result and the bridge detection quantitative result to further obtain a final bridge detection result.
The bridge detection qualitative result is a bridge disease area and a bridge health area;
and the bridge detection quantitative result is the type, size and specific position of the bridge diseases.
Controlling an ultrasonic generator to send ultrasonic waves to a detection area, and exciting a local part of the bridge by using the ultrasonic waves, wherein the method comprises the following specific steps:
s211: the ultrasonic generator sends an excitation signal to the detection area, and a controller on the detection robot records the time for sending the excitation signal;
s212: a thermosensitive sensor on the detection robot acquires a feedback signal of the detected surface of the bridge after receiving an excitation signal of the ultrasonic generator, acquires infrared temperature data of a detected area of the bridge to form an infrared image, and a controller on the detection robot records the time of receiving the feedback signal;
s213: and judging the crack depth and the crack size of the bridge according to the time from sending the excitation signal to collecting the feedback signal and the infrared image, so as to identify the cracks of the bridge and obtain a bridge detection qualitative result.
The method comprises the following steps of acquiring infrared temperature data of the surface of a bridge to be detected by using a thermosensitive sensor to form an infrared image, and specifically comprises the following steps:
s2121: the thermal sensor captures the surface temperature of the bridge measured area before and after receiving the excitation signal and acquires infrared imaging data;
s2122: and generating an infrared image of the detected area of the bridge according to the acquired imaging data.
Utilizing a thermosensitive sensor on the surface of a detected area of the bridge to generate detection data according to the heat conductivity difference of materials at different parts of the bridge, then forming an image and judging a fault position; the infrared imaging data acquired by the thermal sensor are wirelessly transmitted to a bridge detection and analysis server, the surface temperature of the detected area of the bridge is captured to generate a temperature distribution map of the surface of the detected area of the bridge, abnormal signals in the temperature distribution map are analyzed to obtain the position and the size of cracks of the detected area of the bridge, and the detected area of the bridge comprises a damaged area of the bridge and/or a healthy area of the bridge;
the temperature difference occurs due to different heat conduction between the damaged bridge surface and the healthy bridge surface, and the heat conduction equation is as follows:
wherein alpha is thermal diffusion efficiency, k is the thermal conductivity of the material of the measured area of the bridge, rho is the density of the material of the measured area of the bridge, c is specific heat, T (x, T) is the measured area x of the bridge, the temperature of the surface at time T, and T is time;
the partial permeability of the material of the detected area of the bridge;
d is the crack thickness of the measured area of the bridge, and when the crack thickness is excited by energy q, the temperature change is T, namely:
q＝q0δ(x)δ(t)
the boundary conditions are as follows:
T(x,t)＝T0
t is bridge quiltMeasuring the temperature of the surface of the area, q is the pulse excitation of the ultrasonic generator, d is the crack thickness of the measured area of the bridge, q0Is the boundary value of the pulse excitation energy of the ultrasonic generator; e is a natural constant equal to about 2.7; the temperature of the damaged bridge surface and the healthy bridge surface changes along with the time, and the temperature difference delta T is expressed by the following equation:
obtaining the maximum temperature difference by solving an extreme value;
at the optimum detection time tmMeasured maximum temperature difference DeltaTmThe equation is:
the temperature change of the surface of the bridge detected area can occur after the bridge detected area is excited, under the condition of the same time, the temperature T of the bridge disease surface is inversely proportional to the temperature T of the healthy bridge surface, and the temperature of the bridge disease surface is lower than that of the healthy bridge surface.
The step S22 is to collect video image data of the detected area of the bridge, and quantitatively judge the bridge disease area through a neural network algorithm to obtain a bridge detection quantitative result, and the specific steps are as follows:
s221: acquiring the existing healthy bridge image and various bridge disease images, and marking the images with labels;
collecting an image to be inspected, wherein the image comprises an image of cracks, bursts, peels and exposed steel bars, marking the collected image according to the damage type, and the peels refer to the peeling of the surface of the concrete, but the steel bars are not exposed, and the damaged part is still covered by the concrete;
the method has the advantages that the resolution ratio of the image is firstly ensured when the image is collected, the image sample used by the method is 4000 multiplied by 3000 pixels, the resolution ratio of the image is high, and the precision of a trained neural network model can be ensured;
s222: the collected image marked with the label is used as a training sample, various bridge disease images marked with the label are used as positive training samples, and the healthy bridge image marked with the label is used as a negative training sample;
using the collected image marked with the label as a training sample; in order to generate training samples that cover all the scope of bridge damage, each training image needs to be processed as follows: (a) randomly scaling and cropping the image with a probability of 1/2; (b) flipping the image horizontally or vertically with a probability of 1/3; (c) randomly manipulating the image with a probability of 1/4 to apply motion blur, change brightness, or add noise;
s223: manufacturing a neural network model, and training the neural network model by adopting a positive training sample and a negative training sample;
making a neural network model, and applying the network model by using training and training; the raw images are used for validation and testing during training.
And S224, inputting the visual image data of the disease area collected by the camera into the trained neural network model to obtain a bridge detection quantitative result.
And inputting the visual image data of the fault position acquired by the camera into the trained neural network model, and finally determining the bridge diseases.
The neural network employs yolov3 network.
In practical application, due to the change of the viewpoint and the damage distance of the camera, the shot bridge damage has great change in proportion and aspect ratio, and in order to accurately analyze the shot bridge image by using a neural network model, the shot images with various damage sizes and shapes need to be zoomed and cut.
The bridge detection is carried out by using the method, and the detection speed is greatly improved compared with the traditional bridge detection speed.
In the quantitative analysis, the result of the qualitative analysis and the result of the preliminary quantitative analysis of the collected image data by the neural network algorithm are coupled and calculated to obtain the final quantitative analysis result, the classification of the detected area of the bridge is carried out according to the result of the qualitative analysis of the bridge detection, and the result of the qualitative analysis of the bridge after classification and the result of the preliminary quantitative analysis are subjected to matrix fusion to obtain the final quantitative analysis result.
According to bridge detection information obtained by the detection robot through field detection, analyzing a bridge detection result obtained by adopting an image fusion detection method, wherein the bridge detection result is a final quantitative analysis result;
the bridge environment information, the traffic information around the bridge and the basic bridge information are used for generating a bridge maintenance decision result;
establishing a maintenance decision model, wherein the maintenance decision model generates a bridge maintenance decision according to a bridge detection result, bridge environment information, traffic information around the bridge and basic bridge information;
as shown in fig. 3, the maintenance decision model includes four parallel coupling networks, a first neural network, a second neural network, a third neural network and a fourth neural network, the four neural networks respectively have independent input layers, convolutional layers, pooling layers and full-connection layers, the four neural networks are coupled by using one coupling layer, and outputs of the four neural networks are all input into the coupling layer;
the bridge detection result is an input to the first neural network,
the bridge environment information is input to the second neural network,
the bridge traffic information is input to the third neural network,
the basic information of the bridge is the input of the fourth neural network,
the convolution kernels with different sizes are utilized to respectively extract the characteristics of the input data,
the FM algorithm is adopted at the coupling layer to generate a bridge maintenance decision,
the neural network employs a cnn network,
constructing a vector V by an input layer, wherein the convolutional layer comprises n neurons, carrying out convolution operation on the vector V of the input layer to generate new characteristics, and each neuron i in the convolutional layer uses a filter K in a convolution window with the size of ti，RiThe convolution formula is as follows:
Ri＝f(Ki*V+bi)
bias bias term, convolution operator, f () function of ReLU, specifically adopting slope function, effectively avoiding the problems of gradient explosion and disappearance in network operation by said convolution operation, and biThe random gradient descent convergence speed is increased, and the training speed is increased.
Example two
As shown in fig. 2, the system for implementing the intelligent bridge detection method includes:
a collection end: the system is used for acquiring detection information, environmental information and traffic information of a bridge to be detected and acquiring basic information of the bridge;
a background analysis server: the bridge maintenance system is used for analyzing bridge disease information according to the relevant information collected by the collection end, determining the position of a bridge disease to obtain a bridge detection result, formulating a bridge renovation and maintenance plan according to the analysis result of the data analysis module on the detection information, generating a bridge maintenance suggestion, and sending the bridge detection result and the bridge maintenance suggestion to the client;
a client: and receiving and displaying the bridge detection result and the bridge maintenance suggestion.
Further, the background analysis server comprises:
a data receiving module: receiving detected bridge detection information, environment information and traffic information acquired by an acquisition end, and acquiring basic bridge information;
bridge detection analysis module: the system comprises a collecting terminal, a data acquisition terminal and a data processing terminal, wherein the collecting terminal is used for collecting image information and detection data of a bridge to be detected, so as to determine a disease area and a health area of the bridge to be detected and determine the disease type of the disease area;
the method comprises the steps of obtaining a bridge image, analyzing and processing the obtained bridge image to determine whether a bridge has a disease position and determining the bridge disease position; before analysis of the collected image data, data fusion processing needs to be carried out on the collected image data to obtain analysis data with higher credibility;
the method comprises the steps that an image acquisition module and an ultrasonic detection device move in the same direction, the ultrasonic detection device firstly passes through a detection area of the bridge, when the ultrasonic detection device detects that the detection area of the bridge has cracks, the image acquisition device shoots m images of the detection area, when the ultrasonic detection device detects that the detection area of the bridge has no cracks, the image acquisition device shoots n images of the detection area, wherein m is larger than n, and m and n are positive integers;
the environment information analysis module: analyzing environmental information acquired by a sensor to generate an environmental analysis result;
a traffic information analysis module: recording the traffic information of the bridge to be detected by using a camera device on the bridge;
the bridge maintenance suggestion generation module: generating bridge renovation and maintenance results according to the analysis results of the data analysis module, the environmental information analysis module and the traffic information module;
a data sending module: and sending the bridge detection result and the bridge maintenance suggestion to the client.
EXAMPLE III
As shown in fig. 4 and 5, the intelligent bridge inspection robot using the inspection method and system has the following specific structure:
including walking wheel 1, steel cable detection device 2, camera 3, walking speed controlling means 4 and protector 5, and walking wheel 1 installs in inspection robot's bottom, and be provided with steel cable detection device 2 between the walking wheel 1, camera 3 installs on inspection robot, and camera 3's oblique top is provided with walking speed controlling means 4, inspection robot's the outside is provided with protector 5, protector 5 and inspection robot's connected mode is for bonding, and protector is the rubber material, inspection robot top is provided with temperature and humidity sensor, baroceptor and air quality sensor.
The robot can be used for detecting bridges and piers, and is different from a matching device between the robot and the bridges or piers only and used for detecting the bridges or piers.
Example four
As shown in fig. 6, 7 and 8, the matching device required to be matched with the robot for bridge detection comprises a bridge body including a bridge body 7, an abutment 8, a first fixing member 9, a first steel cable body 10, a pier 11, a limiting plate 12, a groove 13, a limiting groove 14 and a locking rod 15, the abutment 8 is arranged at the bottom of the bridge body 7, and the outer side of the abutment 8 is provided with a first fixing piece 9, a first steel cable body 10 is erected between the first fixing pieces 9, the detection robot 5 is arranged on the first steel cable body 10, the bridge pier 11 is fixedly connected with the bottom of the abutment 8, the left side and the right side of the first fixing piece 9 are both provided with a limiting plate 12, and the limiting plate 12 and the first fixing piece 9 are both connected with the abutment 8 through a groove 13, and recess 13 sets up on abutment 8, and first spacing groove 14 has all been seted up to the inside of limiting plate 12 and abutment 8, and the internal connection of first spacing groove 14 has first locking lever 15.
First steel cable body 10 is provided with 2, and the length of 2 first steel cable bodies 10 equals with the length of bridge body 7 to 2 first steel cable bodies 10 all erect between first mounting 9, are convenient for make inspection robot 5 walk through first steel cable body 10 like this, thereby ensure inspection robot detection achievement's normal clear. Limiting plate 12 and first mounting 9 formula structure as an organic whole, and be threaded connection between limiting plate 12 and the inside first spacing groove 14 of abutment 8 and first locking lever 15, and limiting plate 12 is provided with 2, 2 limiting plates 12 are about the longitudinal center line symmetric distribution of first mounting 9 simultaneously, when damage appears in first mounting 9 like this, can be timely dismantle the maintenance to first mounting 9, bring conveniently for staff's operation, further guarantee going on continuously of detection achievement.
EXAMPLE five
As shown in fig. 9, 10 and 11, the fitting device required to be fitted with a robot for pier detection includes a bridge body 7, an abutment 8, a first fixing member 9, a first cable body 10, a second fixing member 16, a second cable body 17, a guide rail 18, a pier 11, a clamp groove 19, a second limit groove 20 and a second lock lever 21, the abutment 8 is disposed at the bottom of the bridge body 7, the first fixing member 9 is disposed at the outer side of the abutment 8, the first cable body 10 is erected between the first fixing members 9, the pier 11 is mounted at the bottom of the abutment 8, the second fixing member 16 is mounted on the pier 11, the second cable body 17 is disposed at the outer side of the second fixing member 16, the second cable body 17 is connected with the first cable body 10 through the guide rail 18, the clamp groove 19 is disposed inside the second fixing member 16, and the clamp groove 19 is connected with the pier 11, the pier 11 and the second fixing member 16 are both provided with a second limiting groove 20, the inside of the second limiting groove 20 is connected with a second locking rod 21, and the robot body 5 is arranged on the first steel cable body 10.
First steel cable body 10 is provided with 2, and the length of 2 first steel cable bodies 10 equals with the length of bridge body 7 to 2 first steel cable bodies 10 are all erect between first mounting 9, can make robot 5 walk, thereby guarantee the normal clear to bridge detection achievement.
The second fixing member 16 forms a disassembly structure through the clamping groove 19 and the pier 11, the second fixing member 16 comprises a first mounting plate 1601, an elastic spring 1602 and a second mounting plate 1603, the first mounting plate 1601 and the second mounting plate 1603 are connected through the elastic spring 1602, and the side section of the second mounting plate 1603 is in an L shape, so that when the second fixing member 16 is damaged, the second fixing member 16 can be disassembled and assembled conveniently, and the second fixing member 16 with a telescopic structure can enable the robot body 5 to better enter a next detection point for detection.
The guide rail 18 is in an S shape, the connection part of the guide rail 18, the second steel cable body 17 and the first steel cable body 10 is smooth, and the second steel cable body 17 is in a circular ring shape, so that the stroke of the robot body 5 can be smoother, and the stable operation of bridge pier detection is further ensured.
Various other modifications and changes may be made by those skilled in the art based on the above-described technical solutions and concepts, and all such modifications and changes should fall within the scope of the claims of the present invention.
Claims (10)
1. An intelligent bridge detection method is characterized by comprising the following specific steps:
s1: acquiring basic information of the bridge according to the bridge environment information, the traffic information around the bridge and the bridge;
s2: acquiring a detected bridge detection image and information on site, and analyzing by using an image fusion detection method according to the acquired detection image and information to obtain a bridge detection result;
s3: and analyzing and generating a maintenance suggestion of the bridge to be detected according to the bridge environment information, the traffic information around the bridge, the basic bridge information and the bridge detection result.
2. The bridge detection method according to claim 1, wherein the field acquisition of the detected bridge detection image and information, and the analysis by the image fusion detection method according to the acquired detection image and information, to obtain the bridge detection result, comprises the following steps:
s21: controlling an ultrasonic generator to send ultrasonic waves to a detection area for excitation; acquiring infrared temperature data of a detected area of the bridge by using a thermosensitive sensor to form an infrared image, and qualitatively judging bridge diseases to obtain a bridge detection qualitative result;
the bridge detected area comprises all bottom surface structures, bottom structures and auxiliary structures;
s22: collecting video image data of a bridge detected area, and carrying out quantitative judgment on a bridge disease area through a neural network algorithm to obtain a bridge detection quantitative result;
s23: and performing row matrix coupling calculation on the bridge detection qualitative result and the bridge detection quantitative result to further obtain a final bridge detection result.
3. The bridge detection method of claim 2, wherein the bridge detection qualitative results are a bridge disease area and a bridge health area;
and the bridge detection quantitative result is the type, size and specific position of the bridge diseases.
4. The bridge detection method of claim 2, wherein the step S21 specifically comprises the steps of:
s211: the ultrasonic generator sends an excitation signal to the detection area, and a controller on the detection robot records the time for sending the excitation signal;
s212: a thermosensitive sensor on the detection robot acquires a feedback signal of the detected surface of the bridge after receiving an excitation signal of the ultrasonic generator, acquires infrared temperature data of a detected area of the bridge to form an infrared image, and a controller on the detection robot records the time of receiving the feedback signal;
s213: and judging the crack depth and the crack size of the bridge according to the time from sending the excitation signal to collecting the feedback signal and the infrared image, so as to identify the cracks of the bridge and obtain a bridge detection qualitative result.
5. The bridge detection method of claim 2, wherein the step of acquiring the infrared temperature data of the surface of the bridge to be detected by using the thermal sensor to form an infrared image comprises the following specific steps:
s2121: the thermal sensor captures the surface temperature of the bridge measured area before and after receiving the excitation signal and acquires infrared imaging data;
s2122: and generating an infrared image of the detected area of the bridge according to the acquired imaging data.
6. The bridge detection method of claim 5, wherein the calculation formula of the crack depth of the bridge in the bridge detection qualitative result is as follows:
wherein alpha is thermal diffusion efficiency, rho is the density of the material in the measured area of the bridge, c is specific heat, d is the crack depth of the bridge, tmIs the optimum detection time attmMeasured maximum temperature difference DeltaTmQ is pulsed excitation;
and judging the bridge disease area and the healthy bridge area by using the infrared image.
7. The bridge detection method of claim 3, wherein the step S22 specifically comprises the steps of:
s221: acquiring the existing healthy bridge image and various bridge disease images, and marking the images with labels;
s222: the collected image marked with the label is used as a training sample, various bridge disease images marked with the label are used as positive training samples, and the healthy bridge image marked with the label is used as a negative training sample;
s223: manufacturing a neural network model, and training the neural network model by adopting a positive training sample and a negative training sample;
and S224, inputting the visual image data of the disease area collected by the camera into the trained neural network model to obtain a bridge detection quantitative result.
8. The system for realizing the intelligent bridge detection method of any one of claims 1 to 7 is characterized by comprising the following steps:
a collection end: the system is used for acquiring detection information, environmental information and traffic information of a bridge to be detected and acquiring basic information of the bridge;
a background analysis server: the bridge maintenance system is used for analyzing bridge disease information according to the relevant information collected by the collection end, determining the position of a bridge disease to obtain a bridge detection result, formulating a bridge renovation and maintenance plan according to the analysis result of the data analysis module on the detection information, generating a bridge maintenance suggestion, and sending the bridge detection result and the bridge maintenance suggestion to the client;
a client: and receiving and displaying the bridge detection result and the bridge maintenance suggestion.
9. The intelligent bridge detection system of claim 8, wherein the background analysis server comprises:
a data receiving module: receiving detected bridge detection information, environment information and traffic information acquired by an acquisition end, and acquiring basic bridge information;
bridge detection analysis module: the system comprises a collecting terminal, a data acquisition terminal and a data processing terminal, wherein the collecting terminal is used for collecting image information and detection data of a bridge to be detected, so as to determine a disease area and a health area of the bridge to be detected and determine the disease type of the disease area;
the environment information analysis module: analyzing environmental information acquired by a sensor to generate an environmental analysis result;
a traffic information analysis module: recording the traffic information of the bridge to be detected by using a camera device on the bridge;
the bridge maintenance suggestion generation module: generating bridge renovation and maintenance results according to the analysis results of the data analysis module, the environmental information analysis module and the traffic information module;
a data sending module: and sending the bridge detection result and the bridge maintenance suggestion to the client.
10. An intelligent bridge detection robot is characterized in that the detection method of any one of claims 1 to 7 is adopted, the detection system of any one of claims 8 to 9 is used, and the specific structure is as follows: the detection robot comprises walking wheels, a steel cable detection device, a camera, a walking speed control device and a protection device, wherein the walking wheels are installed at the bottom end of the detection robot, the steel cable detection device is arranged between the walking wheels, the camera is installed on the detection robot, the walking speed control device is arranged obliquely above the camera, and the protection device is arranged on the outer side of the detection robot; protection device and inspection robot's connected mode is for bonding, and protection device is the rubber material, inspection robot shanghai is provided with temperature and humidity sensor, baroceptor and air quality sensor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011322256.6A CN112627023A (en) | 2020-11-23 | 2020-11-23 | Intelligent bridge detection method and system and intelligent bridge detection robot |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011322256.6A CN112627023A (en) | 2020-11-23 | 2020-11-23 | Intelligent bridge detection method and system and intelligent bridge detection robot |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112627023A true CN112627023A (en) | 2021-04-09 |
Family
ID=75303980
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011322256.6A Pending CN112627023A (en) | 2020-11-23 | 2020-11-23 | Intelligent bridge detection method and system and intelligent bridge detection robot |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112627023A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114775457A (en) * | 2022-03-16 | 2022-07-22 | 浙江广厦建设职业技术大学 | Detection device for repairing ancient bridge and detection method thereof |
| CN115272276A (en) * | 2022-08-12 | 2022-11-01 | 东南大学 | Suspension bridge main cable subsurface disease identification method and device based on infrared light camera shooting |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2002168816A (en) * | 2000-12-04 | 2002-06-14 | Penta Ocean Constr Co Ltd | Bridge diagnosing method and apparatus therefor |
| JP2004012454A (en) * | 2002-06-11 | 2004-01-15 | Toyo High Mech Kk | Method and device for detecting exfoliation inside concrete structure |
| CN101713756A (en) * | 2009-12-24 | 2010-05-26 | 首都师范大学 | Non-contact ultrasonic thermal-excitation infrared imaging nondestructive detection method and system |
| CN105021699A (en) * | 2015-07-16 | 2015-11-04 | 无锡市崇安区科技创业服务中心 | Bridge pier crack nondestructive detection device |
| CN105200918A (en) * | 2015-09-10 | 2015-12-30 | 舟山市智海技术开发有限公司 | Bridge detection structure and bridge detection method adopting same |
| KR20160086506A (en) * | 2015-01-09 | 2016-07-20 | 주식회사 에스원 | Tracing System of Falling of Bridge Rail and Falling Tracing Method thereof |
| CN205861679U (en) * | 2016-07-26 | 2017-01-04 | 云南省公路科学技术研究院 | A kind of device of pier stud RC axial loading column under bridge water |
| CN107818227A (en) * | 2017-11-17 | 2018-03-20 | 厦门大学 | A kind of bridge machinery approaches to IM and system based on BIM |
| CN109989342A (en) * | 2019-04-15 | 2019-07-09 | 南安市弈星机械科技有限公司 | A kind of road and bridge construction concrete NDT system |
| CN110533629A (en) * | 2019-07-10 | 2019-12-03 | 湖南交工智能技术有限公司 | A kind of detection method and detection device of Bridge Crack |
| CN210603956U (en) * | 2019-12-06 | 2020-05-22 | 广州交投工程检测有限公司 | Bridge detection device convenient to long distance detects |
| CN111595859A (en) * | 2020-05-22 | 2020-08-28 | 三峡大学 | Bridge and culvert damage detection method and damage detection and management system |
-
2020
- 2020-11-23 CN CN202011322256.6A patent/CN112627023A/en active Pending
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2002168816A (en) * | 2000-12-04 | 2002-06-14 | Penta Ocean Constr Co Ltd | Bridge diagnosing method and apparatus therefor |
| JP2004012454A (en) * | 2002-06-11 | 2004-01-15 | Toyo High Mech Kk | Method and device for detecting exfoliation inside concrete structure |
| CN101713756A (en) * | 2009-12-24 | 2010-05-26 | 首都师范大学 | Non-contact ultrasonic thermal-excitation infrared imaging nondestructive detection method and system |
| KR20160086506A (en) * | 2015-01-09 | 2016-07-20 | 주식회사 에스원 | Tracing System of Falling of Bridge Rail and Falling Tracing Method thereof |
| CN105021699A (en) * | 2015-07-16 | 2015-11-04 | 无锡市崇安区科技创业服务中心 | Bridge pier crack nondestructive detection device |
| CN105200918A (en) * | 2015-09-10 | 2015-12-30 | 舟山市智海技术开发有限公司 | Bridge detection structure and bridge detection method adopting same |
| CN205861679U (en) * | 2016-07-26 | 2017-01-04 | 云南省公路科学技术研究院 | A kind of device of pier stud RC axial loading column under bridge water |
| CN107818227A (en) * | 2017-11-17 | 2018-03-20 | 厦门大学 | A kind of bridge machinery approaches to IM and system based on BIM |
| CN109989342A (en) * | 2019-04-15 | 2019-07-09 | 南安市弈星机械科技有限公司 | A kind of road and bridge construction concrete NDT system |
| CN110533629A (en) * | 2019-07-10 | 2019-12-03 | 湖南交工智能技术有限公司 | A kind of detection method and detection device of Bridge Crack |
| CN210603956U (en) * | 2019-12-06 | 2020-05-22 | 广州交投工程检测有限公司 | Bridge detection device convenient to long distance detects |
| CN111595859A (en) * | 2020-05-22 | 2020-08-28 | 三峡大学 | Bridge and culvert damage detection method and damage detection and management system |
Non-Patent Citations (5)
| Title |
|---|
| 何嘉辉等: "航空发动机叶片裂纹检测技术及应用分析", 《内燃机与配件》 * |
| 刘奕等: "《城市安全与应急管理》", 31 January 2012, 中国城市出版社 * |
| 夏纪真: "《无损检测导论》", 31 August 2016, 中山大学出版社 * |
| 杨建华等: "基于机器学习的RC桥梁病害检测方法", 《北方交通》 * |
| 贾宇等: "水工混凝土构件裂缝检测方法及发展趋势", 《无损检测》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114775457A (en) * | 2022-03-16 | 2022-07-22 | 浙江广厦建设职业技术大学 | Detection device for repairing ancient bridge and detection method thereof |
| CN115272276A (en) * | 2022-08-12 | 2022-11-01 | 东南大学 | Suspension bridge main cable subsurface disease identification method and device based on infrared light camera shooting |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Jing et al. | Developments, challenges, and perspectives of railway inspection robots | |
| CN111553902B (en) | Highway road surface safety monitoring system based on big data | |
| CN112627023A (en) | Intelligent bridge detection method and system and intelligent bridge detection robot | |
| Song et al. | Automatic segmentation and quantification of global cracks in concrete structures based on deep learning | |
| CN110789566B (en) | Track defect monitoring method and monitoring equipment based on axle box acceleration signal | |
| CN108288055A (en) | Block of bow collector of electric locomotive based on depth network and placement test and arc method for measuring | |
| CN111368423A (en) | Rapid detection and evaluation system and method for bearing capacity of vehicle-mounted bridge | |
| CN112556760A (en) | Bridge crack monitoring system | |
| CN115452723A (en) | Bridge defect detection equipment and defect identification method for bridge engineering | |
| CN111259770A (en) | Rapid cable force testing system and method based on unmanned aerial vehicle platform and deep learning under complex background | |
| CN113638294A (en) | Intelligent detection system technology for road maintenance diseases | |
| CN115439643A (en) | Road disease size analysis and management method based on monocular measurement | |
| Masoudi et al. | An investigation of rail failure due to wear using statistical pattern recognition techniques | |
| TWI656260B (en) | Automatic track detection device | |
| CN116631187A (en) | Intelligent acquisition and analysis system for case on-site investigation information | |
| Jeong et al. | New bridge inspection approach with joint UAV and DIC system | |
| Jiang et al. | A visual inspection and diagnosis system for bridge rivets based on a convolutional neural network | |
| CN116519703A (en) | System and method for detecting carbon slide plate image of collector shoe based on line scanning 3D image | |
| CN112254701B (en) | Vehicle-mounted bridge head vehicle jump detection and analysis system and method thereof | |
| Seo et al. | Machine learning approach to visual bridge inspection with drones | |
| Huang et al. | Machine inspection equipment for tunnels: a review | |
| Brintha et al. | FOD-YOLO NET: Fasteners fault and object detection in railway tracks using deep yolo network 1 | |
| Nichoha et al. | Development of modern methods and directions of rapid diagnostics of railway tracks defects by television methods | |
| Song et al. | Design of cable detection robot and image detection method | |
| Srinivasan et al. | Development of a railroad track inspection system based on visual perception using labview |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210409 |
|
| RJ01 | Rejection of invention patent application after publication |