CN116248838A - Bored pile forming quality monitoring system based on computer vision technology - Google Patents

Abstract

The invention discloses a bored pile forming quality monitoring system based on a computer vision technology. Meanwhile, 4 monitoring pipes are arranged around the grouting pipe along different heights, the tail ends of the monitoring pipes are transparent observation bins, miniature cameras, LED unidirectional light sources and conical reflectors are arranged in the tail ends of the monitoring pipes, abnormal events possibly occurring in the grouting process of underwater concrete around the grouting pipe are automatically identified based on a computer vision technology, the lifting speed and the embedding depth of the guide pipe are adjusted, an alarm is sent to remind field technicians, corresponding remedial measures are taken, and pile quality defects such as pile breakage, interlayer and the like caused by too high lifting speed and too shallow embedding depth of the guide pipe are prevented. The invention can greatly improve the automation level of grouting operation of the bored pile, reduce manual operation links, and further greatly accelerate the construction speed and the pile forming efficiency.

Description

Bored pile forming quality monitoring system based on computer vision technology

Technical Field

The invention relates to the technical field of pile foundation construction, in particular to a bored pile forming quality monitoring system and a construction method based on a computer vision technology.

Background

The bored pile is an underground hidden project with high quality requirement and more construction procedures, and needs to be continuously completed in a short time, and is widely applied to the engineering fields of roads, buildings, bridges and the like because the construction noise and vibration are relatively small and the stratum environment requirement on the foundation is low. However, due to the complex geological conditions of the engineering site, uneven mass and energy capability of construction managers, unstable quality of material machines and tools, difficult control of the quality of underwater poured concrete and other factors, the construction quality is difficult to ensure, and the pile forming quality and bearing capacity are greatly influenced. Once construction faults such as hole collapse, reaming, pipe water leakage, pipe blockage, pile interlayer breakage and the like occur, the construction faults cannot be found in time, and the construction faults are reworked after pile forming, so that labor and time are wasted, and construction period and cost control are not facilitated. Therefore, the monitoring and control of the pile forming quality of the bored pile are always the difficulties of engineering construction and the focus of each participating party.

For example, chinese patent document with publication number CN214401796U discloses a bored pile quality control device, and proposes to utilize the ultrasonic principle, through the mode of setting up ultrasonic detection head along the pipe outer wall, with the help of computer imaging system, can see the distribution condition of concrete density on the display screen, and the workman adjusts the vibration intensity in real time according to this, adds concrete. However, this detection scheme has significant drawbacks. Firstly, the principle of the ultrasonic nondestructive detection technology is that ultrasonic waves are emitted based on ultrasonic detection heads and feedback ultrasonic waves are absorbed, a guide pipe is required to be driven in advance before pouring, a reinforcement cage is placed in the guide pipe, concrete is poured, an ultrasonic detection head is vertically arranged on the outer wall of the guide pipe, and the guide pipe is pulled out of a drilled hole after pouring is completed. In actual operation, if the pile body size and length specification are larger, the pouring time is longer, after the pouring is finished, the strength difference is generated along with different concrete setting progress of the pile body height, and then the difficult tube drawing is possibly caused, or the quality problem risk such as broken piles is further increased due to the tube drawing of external force. In addition, the ultrasonic detection result only reveals the density difference of the pile body, but cannot reveal the reason for generating the density difference in detail, so that the method is not beneficial for on-site staff to take remedial measures in a targeted manner.

The Chinese patent document with publication number of CN115511808A discloses an underwater concrete quality detection method based on a convolutional neural network, and provides a basis for reasonable depth and lifting speed of concrete embedded in a conduit by utilizing an industrial camera to acquire an image upper data set of underwater concrete in a concrete pouring process, correcting, enhancing and semantically segmenting image data by means of artificial intelligence technologies such as the convolutional neural network and the like, and finally obtaining the concrete aggregate ratio, so as to master the segregation condition of the concrete in a borehole. The method has the advantages that the route is clear, a new technical thought is provided for the quality control of the device for drilling the cast-in-situ pile, but the realization method and equipment for acquiring the underwater cast-in-situ concrete image are not mentioned, and the method has no feasibility.

The bored pile construction has considerable concealment, pile forming quality is influenced by various aspects such as technical conditions, personnel capacity, management level and the like, quality problems that broken piles, mud clamping and the like are difficult to perceive on the ground in time easily occur in actual construction, actual bearing capacity is influenced, upper structure safety is endangered, and the post reworking workload is large, so that construction cost is increased, construction period is prolonged and adverse social influence is generated.

Disclosure of Invention

The invention provides a bored pile forming quality monitoring system and a construction method based on artificial intelligence, which realize the function of acquiring real-time underwater concrete pouring condition image data, automatically judge the current underwater concrete pouring quality based on artificial intelligence, automatically adjust the conduit burial depth and the lifting speed, greatly improve the automation level of the bored pile grouting operation, reduce the manual operation links, and further improve the construction speed and the pile forming efficiency.

In order to achieve the above purpose, the technical scheme adopted by the invention is as follows:

a bored pile forming quality monitoring system based on a computer vision technology comprises a grouting guide pipe, wherein a plurality of monitoring pipes are symmetrically arranged around the grouting guide pipe and are fixed around the grouting guide pipe through a support;

the lower end of the monitoring tube is a high-hardness transparent observation bin, the other end of the monitoring tube is an independent power supply and signal enhancement module and a data jack, the inside of the monitoring tube is connected with elements at the two ends through a lead, a data wire is inserted into the data jack at the top end of the monitoring tube and is connected with a data processing and control system terminal, the data processing and control system terminal is connected with a control system of a lifting machine tool, the lifting machine tool is connected with a support through a sling, and the transparent observation bin is composed of a high-definition camera, a unidirectional luminous LED annular lamp belt and a conical reflector;

under the operating condition, the unidirectional luminous LED annular lamp belt emits light to illuminate concrete slurry around the transparent observation bin, and the high-definition camera is matched with the conical reflector to collect underwater concrete image data to the terminal of the data processing and control system.

The invention further discloses the following technology:

preferably, the grouting guide pipe is fixed by the subsection pipe joint through a screwed joint, the monitoring pipe is formed by splicing a first monitoring pipe joint and a second monitoring pipe joint, each pipe joint is fixed by the screwed joint, and the screwed joint of the grouting guide pipe and the monitoring pipe is located at the same height.

Preferably, one end of the first monitoring pipe joint is a high-hardness transparent observation bin, and the other end of the first monitoring pipe joint is an independent power supply and signal enhancement module and a data jack;

one end of the second monitoring pipe joint is the same as the first monitoring pipe joint, an independent power supply and signal enhancement module and a data jack are arranged, and the other end of the second monitoring pipe joint is a data plug;

the first monitoring pipe joint and the second monitoring pipe joint are internally connected with elements at two ends through wires;

when the first monitoring pipe joint is tightly connected with the second monitoring pipe joint through the threaded interface, the plug is embedded into the data jack to achieve the function of transmitting data between the pipe joints.

Preferably, the signal enhancement module, the data jack, the data plug and the inner wall of the monitoring tube are filled with sealing resin glue.

Preferably, the support is divided into a first support fixed on the lowermost segmented pipe section, a second support on the intermediate pipe section and a third support on the top.

The first support and the second support are reserved with circular limiting grooves for installing the monitoring pipe, the circular limiting grooves play a limiting role, a layer of flexible buffer layer is further arranged on the contact surface of the groove body and the monitoring pipe, the limiting retaining structure is arranged under the first support, and the transparent observation bin of the first monitoring pipe joint penetrates through the first support and is connected with the limiting retaining structure in a butt-inserting mode.

Preferably, the driving motor is arranged in the third support, the mounting groove is arranged at the lower end of the third support, the top of the second monitoring pipe joint is inserted into the mounting groove and is connected with the driving motor, the system control driving motor drives the monitoring pipe to rotate during operation, the contact surface of the limiting retaining structure and the transparent observing bin is also provided with a flexible silica gel strip, and the monitoring pipe rubs with the outer wall of the transparent observing bin during rotation operation.

Compared with the prior art, the invention has the beneficial effects that:

1. the invention is derived from experience summary in the process of drilling and grouting pile engineering practice, originally designs the detection tube arranged around the grouting tube, realizes the function of acquiring real-time high-definition images of underwater concrete in the grouting process, realizes the self-cleaning function of the outer surface of the observation bin in a spin friction mode, effectively solves the problems of blurred vision and the like caused by large viscosity wall built-up of concrete slurry and wall-protecting slurry, and ensures the reliability and timeliness of image data acquisition.

2. According to the invention, an image recognition technology route based on artificial intelligence is adopted, and the obtained real-time image of the underwater poured concrete can be used for more accurately distinguishing abnormal event types affecting pile forming quality, such as hole collapse, boulder, slurry groundwater infiltration and the like, compared with single density distribution information obtained through ultrasonic detection, so that a basis is provided for the targeted taking of remedial measures by field technicians.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are needed in the description of the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.

FIG. 1 is a schematic diagram of the working process of a bored pile forming quality monitoring system based on a computer vision technology in the invention;

FIG. 2 is a schematic diagram of the installation construction of a conduit and a monitoring pipe of a bored pile forming quality monitoring system based on a computer vision technology in the present invention;

FIG. 3 is a schematic view of a first monitoring pipe section detail construction of a bored pile forming quality monitoring system based on a computer vision technology in the present invention;

FIG. 4 is a schematic view of a second monitoring pipe section detail construction of a bored pile forming quality monitoring system based on a computer vision technology in the present invention;

FIG. 5 is a schematic diagram showing the detailed construction of a third support on top of a bored pile forming quality monitoring system based on computer vision technology according to the present invention;

FIG. 6 is a schematic view of the construction of a first support at the bottom of a bored pile forming quality monitoring system based on computer vision technology in accordance with the present invention;

FIG. 7 is a schematic diagram showing the detailed construction of a second support in the middle of a bored pile forming quality monitoring system based on computer vision technology in the present invention;

FIG. 8 is a schematic diagram showing the detailed construction of a first support in the middle of a bored pile forming quality monitoring system based on computer vision technology in the present invention;

FIG. 9 is a schematic side view of a transparent observation cabin structure of the bottom of a monitoring pipe of a bored pile forming quality monitoring system based on a computer vision technology;

fig. 10 is a schematic cross section of a supplementary view of a monitoring tube-conduit-support construction of a bored pile forming quality monitoring system based on computer vision technology in the present invention.

Reference numerals in the drawings: the system comprises a guide pipe 1, a monitoring pipe 2, a support 3, a concrete slurry 4, slurry 5, a lifting tool 6, a data processing and control system terminal 7, a threaded interface 8, a segmented pipe section 1-1, a first monitoring pipe section 2-1, a second monitoring pipe section 2-2, an independent power supply and signal enhancement module 2-3, a wire 2-4, a sealing resin adhesive 2-5, a data socket 2-6, a data plug 2-7, a first support 3-1, a second support 3-2, a third support 3-3, a limit groove 3-4, a flexible buffer layer 3-5, a limit retaining structure 3-6, a transparent observing bin 2-1-1, a high-definition camera 2-1-3, a unidirectional luminous LED annular lamp belt 2-1-4 conical reflector 3-3-1 driving motor 3-3-2 mounting grooves and a flexible silica gel strip 3-6-1.

The specific embodiment is as follows:

the present invention will be described in further detail with reference to the following examples, so that the technical means, the creation characteristics, the achievement of the purpose and the effect achieved by the present invention can be easily understood. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.

As shown in fig. 1, in the bored pile forming quality monitoring system based on the computer vision technology, the grouting guide pipe 1 is fixed by screwing the segmented pipe joint 1-1 through the threaded interface 8, and 4 monitoring pipes 2 are symmetrically arranged around the grouting guide pipe. The monitoring pipe 2 is fixed around the grouting guide 1 by a support 3.

As shown in fig. 2, the 4 monitoring pipes 2 are formed by splicing a first monitoring pipe joint 2-1 and a second monitoring pipe joint 2-2, the first monitoring pipe joint 2-1 and the second monitoring pipe joint 2-2 have optional different length specifications, and can be adapted according to the length specifications of the segmented pipe joint 1-1, so that the threaded interfaces of the grouting pipe 1 and the monitoring pipe 2 are positioned at the same height, thereby facilitating pipe joint installation before construction, inspection and pipe joint disassembly in the pipe lifting process during construction.

As shown in fig. 3-4, one end of the first monitoring pipe joint 2-1 is a transparent observation bin 2-1 with high hardness, and the other end is an independent power supply and signal enhancement module 2-3 and a data jack 2-6.

One end of the second monitoring pipe joint 2-2 is identical to the first monitoring pipe joint 2-1, an independent power supply and signal enhancement module 2-3 and a data jack 2-6 are arranged, and the other end is a data plug 2-7. The first monitoring pipe joint 2-1 and the second monitoring pipe joint 2-2 are internally connected with elements at two ends through leads 2-4. When the first monitoring pipe joint 2-1 and the second monitoring pipe joint 2-2 are tightly connected together through the threaded interface 8, the plug 2-7 can be tightly and reliably embedded into the data jack 2-6, and the function of transmitting data between the pipe joints is achieved. The data jack 2-6 at the top end is inserted with a data wire and is connected with the data processing and control system terminal 7, the data processing and control system terminal 7 is connected with the control system of the lifting machine tool 6, the lifting machine tool is connected with the support 3 through a sling, and particularly, the independent power supply and signal enhancement module 2-3, the data jack 2-6, the data plug 2-7 and the inner wall of the monitoring tube 2 are filled by adopting sealing resin glue 2-5, so that the functions of water resistance and fixation are achieved.

As shown in FIG. 3, the transparent observation bin 2-1-1 is internally composed of a high-definition camera 2-1-2, a unidirectional luminous LED annular lamp strip 2-1-3 and a conical reflector 2-1-4. In the working state, the unidirectional luminous LED annular lamp belt 2-1-3 emits light to illuminate concrete slurry around the transparent observation bin 2-1-1, and the high-definition camera 2-1-2 is matched with the conical reflector 2-1-4 to collect underwater concrete image data and upload the underwater concrete image data to the data processing and control system terminal 7.

As shown in fig. 5, 6 and 7, the support 3 is divided into a first support 3-1 fixed to the lowermost segment, a second support 3-2 on the middle segment, and a third support 3-3 on the top.

As shown in fig. 6, 7, 8, 9 and 10, 4 circular limiting grooves 3-4 for installing the monitoring tube 2 are reserved in the first support 3-1 and the second support 3-2, and besides the limiting function of the circular limiting grooves 3-4, a layer of flexible buffer layer 3-5 is further arranged on the contact surface of the groove body and the monitoring tube, so that the monitoring tube body 2 is allowed to rotate around the axis of the monitoring tube body in the limiting grooves 3-4 and friction resistance is reduced. The first support 3-1 is further characterized in that besides the limiting groove 3-4, a limiting supporting structure 3-6 is arranged right below the first support 3-1, so that the monitoring pipe 2 can rotate smoothly around the shaft and bear self-weight load, and falling is prevented. The contact surface of the limiting supporting structure 3-6 and the transparent observing bin 2-1-1 is also provided with a flexible silica gel strip 3-6-1, the monitoring tube 2 rubs with the outer wall of the transparent observing bin 2-1-1 during self-transportation, slurry residues adhered to the outer surface of the observing bin during lifting are cleaned timely, and the visual field of a camera is ensured to be clear.

As shown in fig. 5, the third support 3-3 is characterized in that 4 circular limiting grooves are replaced by mounting grooves 3-3-2 which are provided with internal threads, are connected with the driving motor 3-3-1 and can drive the monitoring pipe 2 to rotate around the shaft. When the camera is in operation, the system controls the driving motor 3-3-1 to drive the monitoring tube 2 to rotate at a certain speed, and the self-cleaning function of the outer wall of the transparent observation bin 2-1-1 is realized by friction with the flexible silica gel strip 3-6-1 arranged on the fixed welding support 3-1, so that the visual field of the camera is ensured to be clear.

In the embodiment, the construction method of the bored pile forming quality monitoring system based on the computer vision technology comprises the following steps:

step 1: the preparation work before grouting comprises drilling, hole washing, equipment positioning, steel reinforcement cage installation hanging and the like.

Step 2: the method as described in claim 2, the length specification of the end monitoring tube 2-1 and the extension relay tube 2-2 with proper lengths is selected, the grouting guide tube 1 and the monitoring tube 2 are assembled on the ground and are electrified for inspection, and video and control signals are debugged.

Step 3: and (3) hoisting and placing the grouting guide pipe 1 and the monitoring pipe 2, and completing the first bucket grouting after calculating the grouting amount according to the specification.

Step 4: the driving motor 3-3-1 drives the monitoring tube 2 to rotate, and meanwhile, the high-definition camera 2-1-2 and the unidirectional luminous LED annular lamp belt 2-1-3 in the tail end monitoring tube 2-1 are electrified to run, so that real-time image data of the pouring condition near the outlet of the grouting guide tube 1 are acquired and recorded. Judging the proportion change of concrete aggregate and mortar around the transparent observation bin 2-1-1, and identifying abnormal events such as hole collapse, boulder, slurry groundwater infiltration and the like which affect pile forming quality. Once the abnormality occurs, the lifting speed and the burial depth of the guide pipe are immediately adjusted, and the field technicians are reminded to take corresponding remedial measures in time.

The abnormal event identification specifically includes:

step 1: after distortion correction and clipping treatment are carried out on video images acquired by a high-definition camera (2-1-2), the acquired images are optimized and amplified by using a generated countermeasure network (DE-GAN and the like), and the images are divided into a training set, a verification set and a test set after labeling.

Step 2: and extracting and dividing R, G and B three-channel color information, gray level co-occurrence matrix statistics and the like of an image mud region and an aggregate region as key characteristic information, and performing semantic division and label output (aggregate region, slurry/mud region and the like) on an obtained underwater concrete pouring image after model training and model parameter acquisition are completed by using a Support Vector Machine (SVM).

Step 3: and obtaining and recording the proportion change of the components such as concrete aggregate, slurry/slurry and the like in the pouring process according to the image output after the segmentation.

Step 4: the processed image data is input to a deep learning recognition network (YOLOv 5, etc.), and the current perfusion quality is recognized and the type of the abnormal event is judged and output.

The foregoing has shown and described the basic principles, principal features and advantages of the invention. It will be understood by those skilled in the art that the present invention is not limited to the foregoing embodiments, and that the foregoing invention and description are merely illustrative of the principles of this invention, and various changes and modifications may be made without departing from the spirit and scope of the invention, which is defined in the appended claims. The scope of the invention is defined by the appended claims and equivalents thereof.

Claims (6)

1. Drilling bored concrete pile formation quality monitored control system based on computer vision technique, its characterized in that: the grouting device comprises a grouting guide pipe (1), wherein a plurality of monitoring pipes (2) are symmetrically arranged around the grouting guide pipe, and the monitoring pipes (2) are fixed around the grouting guide pipe (1) through a support (3);

the lower end of the monitoring tube (2) is a high-hardness transparent observation bin (2-1-1), the other end of the monitoring tube is an independent power supply and signal enhancement module (2-3) and a data jack (2-6), the inside of the monitoring tube (2) is connected with elements at two ends through a lead (2-4), a data wire is inserted into the data jack (2-6) at the top end of the monitoring tube and is connected with a data processing and control system terminal (7), the data processing and control system terminal (7) is connected with a control system of a lifting machine tool (6), the lifting machine tool is connected with a support (3) through a sling, and the transparent observation bin (2-1-1) is internally composed of a high-definition camera (2-1-2), a unidirectional luminous LED annular lamp belt (2-1-3) and a conical reflector (2-1-4);

the unidirectional luminous LED annular lamp strip (2-1-3) emits light to illuminate concrete slurry around the transparent observation bin (2-1-1), and the high-definition camera (2-1-2) is matched with the conical reflector (2-1-4) to collect underwater concrete image data to the data processing and control system terminal (7).

2. A bored pile forming quality monitoring system based on computer vision technology as set forth in claim 1, wherein: the grouting guide pipe (1) is fixed by a subsection pipe joint (1-1) through a threaded interface (8) in a screwed mode, the monitoring pipe (2) is formed by assembling a first monitoring pipe joint (2-1) and a second monitoring pipe joint (2-2), the pipe joints are fixed by the threaded interface (8) in a screwed mode, and the threaded interfaces of the grouting guide pipe (1) and the monitoring pipe (2) are located at the same height.

3. A bored pile forming quality monitoring system based on computer vision technology as set forth in claim 2, wherein: one end of the first monitoring pipe joint (2-1) is a high-hardness transparent observation bin (2-1-1), and the other end of the first monitoring pipe joint is an independent power supply and signal enhancement module (2-3) and a data jack (2-6);

one end of the second monitoring pipe joint (2-2) is the same as the first monitoring pipe joint (2-1), an independent power supply and signal enhancement module (2-3) and a data jack (2-6) are arranged, and a data plug (2-7) is arranged at the other end;

the first monitoring pipe joint (2-1) and the second monitoring pipe joint (2-2) are internally connected with elements at two ends through leads (2-4);

when the first monitoring pipe joint (2-1) is tightly connected with the second monitoring pipe joint (2-2) through the threaded interface (8), the plug (2-7) is embedded into the data jack (2-6) to play a role in transferring data between the pipe joints.

4. A bored pile forming quality monitoring system based on computer vision technology as set forth in claim 3, wherein: the signal enhancement module (2-3), the data jack (2-6), the data plug (2-7) and the inner wall of the monitoring tube (2) are filled by adopting sealing resin glue (2-5).

5. A bored pile forming quality monitoring system based on computer vision technology as set forth in claim 4, wherein: the support (3) is divided into a first support (3-1) fixed on the lowest segmented pipe joint, a second support (3-2) on the middle pipe joint and a third support (3-3) at the top;

the device is characterized in that a circular limiting groove (3-4) for installing the monitoring pipe (2) is reserved in the first support (3-1) and the second support (3-2), a layer of flexible buffer layer (3-5) is further arranged on the contact surface of the groove body and the monitoring pipe except for the limiting effect of the circular limiting groove (3-4), a limiting retaining structure (3-6) is arranged right below the first support (3-1), and a transparent observation bin (2-1-1) of the first monitoring pipe joint (2-1) penetrates through the first support (3-1) and is connected with the limiting retaining structure (3-6) in an opposite-inserting mode.

6. A bored pile forming quality monitoring system based on computer vision technology as set forth in claim 5, wherein: the device is characterized in that a driving motor (3-3-1) is arranged in the third support (3-3), a mounting groove (3-3-2) is formed in the lower end of the third support (3-3), the top of the second monitoring pipe joint (2-2) is inserted into the mounting groove (3-3-2) and connected with the driving motor (3-3-1), the system controls the driving motor (3-3-1) to drive the monitoring pipe (2) to rotate when in operation, a flexible silica gel strip (3-6-1) is further arranged on the contact surface of the limiting retaining structure (3-6) and the transparent observation bin (2-1-1), and the outer wall of the transparent observation bin (2-1-1) is rubbed when the monitoring pipe (2) rotates and operates.

Images