Intelligent vibration control equipment based on T-beam construction and control method thereof

Abstract

The invention relates to intelligent vibration control equipment based on T-beam construction and a control method thereof, and belongs to the technical field of construction control systems.The control method comprises the following steps that S1, information of a certain T-beam concrete pouring area is obtained, the change of the area of a concrete mixture area in adjacent images is analyzed, and the quasi-static moment in the concrete paving process is determined; s2, automatically determining approximate center point pixel coordinates of each area, and intelligently avoiding a reinforcing mesh; s3, pixel coordinates of actual vibration point positions are obtained, and driving parameters of point-by-point vibration are obtained according to the mapping relation between the pixel space and the driver space; s4, analyzing the surface structure and distribution characteristics of the image of the outermost ring layer of the action area of the current vibration point location in the vibration process in real time, and obtaining the moment when the vibration quality of the current action area is qualified; s5, the vibrating rod is automatically lifted and moved to the next vibrating area; the method has the beneficial effects that dynamic adaptive optimization of the driving parameters is realized, and the control precision is improved.

Description

Technical Field

[0001] This invention belongs to the field of construction control equipment technology, and specifically relates to an intelligent vibration control device and its control method based on T-beam construction. Background Technology

[0002] As a core load-bearing component in bridge engineering, the construction quality of T-beams directly affects the overall stability and service life of the bridge. Concrete vibration is a crucial step in T-beam construction, aiming to remove air bubbles from the concrete, ensure its dense compaction within the formwork, and prevent quality defects such as honeycomb, pitting, and voids. Currently, manual operation of immersion vibrators is commonly used for T-beam construction, which presents the following problems:

[0003] Vibration parameters rely on the operator's experience, making it difficult to accurately control the vibration frequency, amplitude, and vibration time, which can easily lead to over-vibration (i.e., causing concrete segregation) or under-vibration (i.e., causing insufficient concrete density).

[0004] Manual vibration has a high degree of randomness in the path, and the vibration coverage of key areas such as the web and flange of the T-beam is not comprehensive, and the vibration depth is difficult to control uniformly.

[0005] The construction efficiency is low, the labor intensity is high, and the construction quality of T-beams is unstable due to the differences in the skill level of the operators.

[0006] Existing intelligent vibration equipment is mostly designed for ordinary concrete components and lacks adaptability to the structural characteristics of T-beams (such as thin web, wide flange, and clearly defined critical stress areas). Furthermore, the adjustment of vibration parameters is based on data from a single sensor, which cannot comprehensively reflect the concrete pouring state and results in insufficient control precision. Summary of the Invention

[0007] This invention provides an intelligent vibration control device and its control method based on T-beam construction, which solves the technical problem that the adjustment of vibration parameters cannot comprehensively reflect the concrete pouring state, resulting in insufficient control accuracy. By adjusting the driving parameters in real time according to the concrete workability, the driving parameters are dynamically adapted and optimized, thereby improving the control accuracy.

[0008] To achieve the above objectives, the present invention is implemented through the following technical solution:

[0009] A smart vibration control method based on T-beam construction includes the following steps:

[0010] Acquire real-time images of a concrete pouring area of ​​a T-beam and information on the location of reinforcing bars in the concrete, analyze the changes in the area of ​​the concrete mixture in adjacent images, and determine the quasi-static moment during the concrete paving process.

[0011] Based on the contour features of the concrete mixture area and the position information of the reinforcing bars at the quasi-static moment, the approximate center point pixel coordinates of each area are automatically determined, and the reinforcing mesh is intelligently avoided.

[0012] Based on the specifications for the spacing between adjacent vibration points and the energy efficiency optimization target, the pixel coordinates of the actual vibration points are obtained through orderly screening. Based on the mapping relationship between the pixel space and the driver space, the driving parameters for point-by-point vibration are obtained.

[0013] Vibration is performed according to the driving parameters of point-by-point vibration, and the surface structure and distribution characteristics of the outermost layer of the action area at the current vibration point are analyzed in real time during the vibration process to obtain the moment when the vibration quality of the current action area is qualified.

[0014] After vibration is completed, the vibrator is automatically lifted and moved to the next vibration area.

[0015] Optionally, step S1 includes the following steps:

[0016] Data acquisition: Real-time acquisition of video / image sequences of the T-beam pouring area; synchronous acquisition of prior information on the location of reinforcing bars;

[0017] Image preprocessing and segmentation: image denoising and illumination correction; segmentation of the concrete mixture area and the formwork and rebar areas;

[0018] Rebar location fusion: Maps rebar location information to the image coordinate system; eliminates interference from rebar areas on concrete area calculation;

[0019] Area variation analysis: Calculate the effective concrete area in each frame; construct an area-time series.

[0020] Quasi-static moment determination: Use the area change rate or its derivative to determine whether the paving tends to stabilize.

[0021] Furthermore, the effective concrete area in each frame is calculated as follows:

[0022] During the T-beam concrete pouring process, as time progresses, the concrete will be continuously pumped and spread, and its area within the formwork will gradually expand, reflecting the spreading progress:

[0023] The increasing number of pixels in the effective concrete area of ​​the image over time reflects the dynamic progress of the concrete filling process.

[0024] When the total number of pixels in the effective concrete area of ​​the image stops increasing or changes very little over time, it indicates that the paving is basically completed and has entered the quasi-static stage.

[0025] For determining quasi-static moments, the inter-frame pixel displacement field is estimated using the optical flow method to obtain the quiescent index of the spreading flow field.

[0026] Optionally, in step S2, the contour features of the concrete mixture region at the quasi-static moment are extracted and calculated using connected regions.

[0027] The method for calculating the reinforcement position information of the concrete mixture area at the quasi-static moment is to calculate it through the pixel coordinates of the center point of the connected region, which is used to locate the target;

[0028] Automatically determine the approximate center point pixel coordinates of each region and judge whether the centroid falls within the reinforcement area:

[0029] Check the centroid coordinates of connected regions in the rebar mask image:

[0030] If the value is 0 or not on the rebar, accept this point as the final center point; if the value is 1 or on the rebar, the center point needs to be adjusted to avoid the rebar.

[0031] Optionally, in step S2, the strategy for intelligently avoiding the reinforcing bars is an iterative exclusion method:

[0032] Starting from the center of mass;

[0033] If it is on the reinforcing bar, then move along the gradient direction or away from the reinforcing bar;

[0034] The distance field of the steel bar mask image is used as the repulsive potential field;

[0035] To get away from the reinforcing bars as quickly as possible, move in small steps along the direction of the gradient until you enter the effective area.

[0036] Optionally, in step S3, for the specification requirements of the spacing between adjacent vibration points and the energy efficiency optimization target, the contribution of each candidate point to the overall compaction efficiency is measured by introducing an energy efficiency weight function, and the calculation is performed by combining the local state of the concrete with image features.

[0037] The ordered screening strategy arranges the candidate point set generated by the initial uniform grid in descending order according to the energy efficiency weight function, and removes subsequent points whose neighborhood distance is less than the minimum allowable spacing between adjacent vibration points, so as to ensure that the minimum spacing constraint is met and high energy efficiency points are retained first.

[0038] Furthermore, the driving parameters for point-by-point vibration are based on adaptive vibration intensity calculation using image gradient and curvature. By controlling the amplitude through grayscale values ​​and combining local geometric features, the processing of key areas of the structure is enhanced.

[0039] Optionally, in step S4, the driving parameters for point-by-point vibration are used to perform vibration evolution from point to field, defining a spatiotemporal influence kernel centered on the position coordinates of the vibration point in the connected region.

[0040] The surface structure and distribution characteristics of the outermost layer of the action area are based on the position coordinates of the vibration point in the connected region where the current vibration point is located. The effective action radius is taken as the maximum influence radius of the vibration action, resulting in an outermost ring area with the distribution characteristics of the image within the ring area.

[0041] The moment when the vibration quality of the current working area is qualified is determined by the concrete construction quality control or intelligent vibration system, which judges whether sufficient vibration has been completed in a specific area, thereby achieving the quality qualification standard.

[0042] Optionally, in step S5, the automated continuous execution of the vibration operation continues, as follows:

[0043] Automatic lifting of vibrator: The control system drives the actuator to lift the vibrator vertically from the concrete smoothly and at a uniform speed, avoiding the formation of voids inside the concrete or disturbance of already compacted areas due to excessive lifting speed;

[0044] Equipment posture adjustment and path planning: Based on the pixel coordinates of the determined next vibration point, and combined with the mapping relationship between pixel space and driver physical space, the target position and posture that the vibratory rod end effector needs to move are calculated.

[0045] Automatically shift to the next vibration area: Drive the robotic arm or mobile platform to precisely transport the vibratory rod to the preparatory position directly above the next point to be vibrated, preparing for the next round of insertion and vibration;

[0046] Status Reset and Cycle Preparation: The system synchronously updates the operation status, including recording the information of completed vibration points, verifying equipment operating parameters, and preparing to receive image feedback for the next cycle, ensuring that subsequent vibration processes can be seamlessly connected and continuously optimized in a closed loop.

[0047] A smart vibration control device for T-beam construction includes:

[0048] The image acquisition module is used to acquire real-time images of the concrete pouring area of ​​the T-beam and information on the location of the reinforcing bars in the concrete.

[0049] The quasi-static determination module is connected to the image acquisition module and is used to analyze the changes in the area of ​​the concrete mixture in adjacent real-time images to determine the quasi-static moment in the concrete paving process.

[0050] The vibration point planning module is connected to the image acquisition module and the quasi-static determination module. It is used to automatically determine the pixel coordinates of the approximate center point of each area and avoid the steel mesh based on the contour features of the concrete mixture area and the position information of the steel bars at the quasi-static moment. Then, it selects the pixel coordinates of the actual vibration point based on the specification requirements of the spacing between adjacent vibration points and the energy efficiency optimization target. Furthermore, it generates the driving parameters for point-by-point vibration based on the mapping relationship between pixel space and driver space. The driving parameters are dynamically adjusted in real time according to the workability of concrete.

[0051] The vibration execution module is signal-connected to the vibration point planning module and is used to execute the vibration operation according to the driving parameters of point-by-point vibration.

[0052] The quality judgment module is connected to the image acquisition module and the vibration execution module. It is used to analyze the surface structure and distribution characteristics of the outermost layer of the image of the current vibration point during the vibration process, and to determine the moment when the vibration quality of the current area is qualified.

[0053] The movement control module is connected to the vibration execution module and the quality judgment module. It is used to control the vibration execution module to lift the vibrator and move it to the next vibration area after the vibration quality in the current working area is qualified.

[0054] The beneficial effects of this invention are:

[0055] 1. This invention integrates real-time image recognition and rebar location information to accurately locate the quasi-static moment and effective vibration center point of the concrete paving area, effectively avoiding interference from the rebar mesh, preventing missed vibration, over-vibration, or vibration blind spots, significantly improving concrete density and structural uniformity, and enhancing vibration accuracy and uniformity.

[0056] 2. This invention eliminates the need for manual intervention throughout the entire process, from image acquisition and vibration point planning to the generation and execution of driving parameters. This significantly reduces reliance on operator experience, improves the level of construction standardization and work efficiency, and achieves full automation and intelligence. It adjusts key driving parameters such as vibration frequency, time, and lifting speed in real time based on the concrete's workability, ensuring vibration quality while also considering energy efficiency, extending equipment life and reducing energy consumption, and dynamically optimizing vibration parameters. Attached Figure Description

[0057] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0058] Figure 1 This is a schematic diagram of the device structure of the present invention;

[0059] Figure 2 This is a schematic diagram of the workflow of the present invention. Detailed Implementation

[0060] The embodiments of this application will now be described in detail with reference to the accompanying drawings.

[0061] Example 1

[0062] like Figure 1 As shown, this embodiment provides an intelligent vibration control device based on T-beam construction, including:

[0063] The image acquisition module is used to acquire real-time images of the concrete pouring area of ​​the T-beam and information on the location of the reinforcing bars in the concrete.

[0064] The quasi-static determination module is connected to the image acquisition module and is used to analyze the changes in the area of ​​the concrete mixture in adjacent real-time images to determine the quasi-static moment in the concrete paving process.

[0065] The vibration point planning module is connected to the image acquisition module and the quasi-static determination module. It is used to automatically determine the pixel coordinates of the approximate center point of each area and avoid the steel mesh based on the contour features of the concrete mixture area and the position information of the steel bars at the quasi-static moment. Then, it selects the pixel coordinates of the actual vibration point based on the specification requirements of the spacing between adjacent vibration points and the energy efficiency optimization target. Based on the mapping relationship between pixel space and driver space, it generates driving parameters for point-by-point vibration. The driving parameters include vibration frequency, vibration time, vibrator lifting speed, insertion position and vibration radius. The driving parameters are dynamically adjusted in real time according to the workability of concrete.

[0066] The vibration execution module is signal-connected to the vibration point planning module and is used to execute the vibration operation according to the driving parameters of point-by-point vibration.

[0067] The quality judgment module is connected to the image acquisition module and the vibration execution module. It is used to analyze the surface structure and distribution characteristics of the outermost layer of the image of the current vibration point during the vibration process, and to determine the moment when the vibration quality of the current area is qualified.

[0068] The movement control module is connected to the vibration execution module and the quality judgment module. It is used to control the vibration execution module to lift the vibrator and move it to the next vibration area after the vibration quality in the current working area is qualified.

[0069] By acquiring the location information of the reinforcing bars through the image acquisition module and combining it with the concrete contour characteristics at the quasi-static moment, the vibration point planning module can automatically avoid the reinforcing mesh and select points according to the standard spacing and energy efficiency targets. This solves the problems of easy omission, accidental contact with reinforcing bars, and non-compliance with spacing in traditional manual planning, greatly improving the accuracy and rationality of vibration point locations. It also avoids insufficient concrete density or structural damage caused by point deviations, achieving intelligent and precise planning of vibration points.

[0070] The driving parameters (vibration frequency, time, and lifting speed) can be adjusted in real time according to the workability of concrete. Compared with traditional fixed parameter vibration, it can adapt to the characteristics of concrete with different grades and slumps, as well as the dynamic changes in the state of concrete during the pouring process. It ensures sufficient vibration (avoiding under-vibration and missed vibration) and prevents defects such as aggregate segregation and surface slurry caused by over-vibration, thereby improving the stability of T-beam concrete construction quality and realizing dynamic adaptation and optimization of driving parameters.

[0071] The quality judgment module analyzes the image features of the outermost layer of the vibration area to determine the qualified moment in real time. In conjunction with the movement control module, it realizes an automated closed loop of "vibration-judgment-movement". No manual supervision is required for judgment. It solves the problems of inconsistent quality standards and lag caused by traditional reliance on experience judgment. It ensures that each vibration point meets the qualified standard, reduces the rework rate, and enables real-time closed-loop control of vibration quality.

[0072] Example 2

[0073] Based on Example 1, such as Figure 2 As shown, this embodiment provides an intelligent vibration control method based on T-beam construction, including the following steps:

[0074] Step S1: Obtain real-time images of the concrete pouring area of ​​a T-beam and information on the location of the reinforcing bars in the concrete, analyze the changes in the area of ​​the concrete mixture in adjacent images, and determine the quasi-static moment in the concrete paving process.

[0075] Step S2: Based on the contour features of the concrete mixture area and the position information of the reinforcing bars at the quasi-static moment, automatically determine the approximate center point pixel coordinates of each area and intelligently avoid the reinforcing bar mesh;

[0076] Step S3: Based on the specifications for the spacing between adjacent vibration points and the energy efficiency optimization target, perform orderly screening to obtain the pixel coordinates of the actual vibration points. Based on the mapping relationship between the pixel space and the driver space, obtain the driving parameters for point-by-point vibration. The driving parameters include vibration frequency, vibration time, vibrator lifting speed, insertion position, and vibration radius. The driving parameters are dynamically adjusted in real time according to the workability of the concrete.

[0077] Step S4: Vibrate according to the driving parameters of point-by-point vibration, and at the same time analyze the surface structure and distribution characteristics of the outermost layer of the action area of ​​the current vibration point in real time during the vibration process to obtain the moment when the vibration quality of the current action area is qualified.

[0078] Step S5: After vibration is completed, the vibrator is automatically lifted and moved to the next vibration area.

[0079] By integrating real-time image recognition with rebar location information, the quasi-static moment and effective vibration center point of the concrete paving area are accurately located, effectively avoiding interference from the rebar mesh, preventing missed vibration, over-vibration, or vibration blind spots, significantly improving concrete density and structural uniformity, and thus enhancing vibration accuracy and uniformity.

[0080] From image acquisition and vibration point planning to driving parameter generation and execution, the entire process requires no manual intervention, significantly reducing reliance on operator experience, improving construction standardization and operational efficiency, and achieving full automation and intelligence.

[0081] Based on the concrete workability (such as slump and fluidity), the key driving parameters of vibration frequency, time and lifting speed are adjusted in real time to ensure vibration quality while taking into account energy efficiency, extend equipment life and reduce energy consumption, and dynamically optimize vibration parameters.

[0082] Example 3

[0083] Based on Embodiment 2, step S1 consists of the following steps:

[0084] Data acquisition: Real-time acquisition of video / image sequences of the T-beam pouring area; synchronous acquisition of prior information on the location of reinforcing bars (e.g., BIM model, laser scanning, or embedded sensors);

[0085] Image preprocessing and segmentation: image denoising and illumination correction; segmentation of the concrete mixture area and the formwork and rebar areas;

[0086] Rebar location fusion: Maps rebar location information to the image coordinate system; eliminates interference from rebar areas on concrete area calculation;

[0087] Area variation analysis: Calculate the effective concrete area in each frame; construct an area-time series.

[0088] Quasi-static moment determination: Use the area change rate or its derivative to determine whether the paving tends to stabilize.

[0089] Furthermore, the effective concrete area in each frame is calculated as follows:

[0090] ;

[0091] in, For the first The total number of pixels in the effective concrete region of a frame image (i.e., the area per unit pixel).

[0092] In the first Frame-corrected binary mask image, It is a binary mask. To correct the binary mask, This means iterating through and accumulating all pixels in the entire image. and Here are the pixel coordinates, and each pixel has the following value:

[0093] 1: Indicates that a certain location belongs to the effective concrete area (i.e., the non-reinforced area);

[0094] 0: Indicates that a certain location does not belong to the effective concrete area (such as formwork, air, or reinforcing steel).

[0095] During the T-beam concrete pouring process, as time progresses, the concrete is continuously pumped and spread, and its area within the formwork gradually expands, reflecting the spreading progress:

[0096] when The value increases over time, reflecting the dynamic progress of the concrete filling process;

[0097] when The value stops increasing or changes very little over time, indicating that the paving is basically complete and has entered a quasi-static stage.

[0098] For the area-time series to be constructed:

[0099] Pixel area construction Each sequence The time length corresponding to each pixel area sequence is frame;

[0100] Constructing the first-order difference , , Indicates the flow rate of a pixel;

[0101] First-order difference After smoothing, the derivative is obtained. : ;in, The area over a time period of time is Frame rate of change (unit: pixels² / second); For the first The concrete area of ​​the frame; For the first The concrete area of ​​the frame; This represents the number of time offset steps. The frame rate for image acquisition.

[0102] Used for calculating area In time length The instantaneous rate of change of the frame, i.e. the rate of area growth, is important in concrete paving monitoring. The sequence is shifted or low-pass filtered to enhance stability.

[0103] For determining quasi-static moments, optical flow is used to estimate the pixel displacement field between frames. The stabilization index (SFSI) of the paving flow field was obtained:

[0104] ;

[0105] in, For a time length of The silence index of the frame, with a value range of [0,1]; The duration is The pixel area of ​​the effective concrete region in the frame of the image; The current concrete area; Let be the optical flow field vector, representing each point in the image. Motion speed (unit: pixels / frame); The magnitude (module) of the velocity at that point; This is a velocity threshold used to distinguish between flow and stillness; The indicator function returns 1 if the speed threshold is met, and 0 if the speed threshold is not met.

[0106] Indicator Function express, ; It can count the number (or area) of all pixels that meet the conditions. This yields a percentage value representing the proportion of the static area to the total concrete area.

[0107] To sense microscopic flow behavior and identify quasi-static mechanical states, even if the area remains unchanged, as long as the internal pixels no longer move, it can be determined to be stationary, preventing premature termination of vibration due to stagnant area but still internal flow, thus avoiding misjudgment.

[0108] Example 4

[0109] Based on Example 2, in step S2, the contour features of the concrete mixture region at the quasi-static moment are calculated using connected component extraction, i.e., for the concrete mixture region... Perform connected component labeling to obtain Connected regions :

[0110] ;

[0111] in, For the first A pixel in a connected component, that is, a single concrete block; These are the pixel coordinates in the image. It is a column (horizontal direction). It is a line (vertical direction); For the entire pixel space of the image, for example: That is, all possible pixel positions; For a labelmap, each pixel is assigned an integer label indicating which connected region it belongs to; For regional indexes; This is a binary mask for the concrete mixture, where... This indicates that the point is within the concrete area; otherwise, it is 0. This is an algorithm in image processing used to identify and label all continuous foreground regions (i.e. connected components with a value of 1) in an image.

[0112] Input binary image 1 represents concrete, and 0 represents background or non-concrete; traverse the image, using 4-neighbor or 8-neighbor connection rules to merge all adjacent 1 pixels into a single connected component; output: an integer image of the same size as the original image. , of which, background pixels ( Each connected concrete block is marked as 0, and each connected concrete block is assigned a unique positive integer label (such as 1, 2, 3, ...). It is a topologically connected subset, representing an independent concrete region, used for subsequent calculations of area, centroid, and shape features.

[0113] The specific calculation for the reinforcement location information in the concrete mixture area at the quasi-static moment is as follows:

[0114] ;

[0115] in, For the first The pixel coordinates of the center point of each connected region; For each pixel within the region; It applies to all pixels within the region. Summation of coordinates; It applies to all pixels within the region. Summation of coordinates;

[0116] For the first Connected regions The x-coordinates of all pixels in ( The average of the coordinates is used to calculate the geometric center (centroid) of the connected region. Quantity;

[0117] For the first Connected regions The x-coordinates of all pixels in ( The average of the coordinates is used to calculate the geometric center (centroid) of the connected region. Quantity.

[0118] To calculate the first Connected regions The centroid (geometric center) coordinates, i.e., the center point considering rebar avoidance; reflect the lateral center position of the area in the image. If the area is biased to the right, this value will be larger; if it is biased to the left, it will be smaller; it is a commonly used feature descriptor in image processing for locating targets.

[0119] The approximate center point pixel coordinates of each region are automatically determined by checking whether the centroid falls within the area of ​​the reinforcing steel reinforcement.

[0120] examine :

[0121] If the value is 0 (not on the rebar) → accept this point as the final center point; if the value is 1 (on the rebar) → the center point needs to be adjusted to avoid the rebar.

[0122] In the rebar mask image In the middle, query the center point Is it located in a reinforced concrete area? It is a key criterion used to determine whether the centroid falls on the reinforcing bar. It is a scalar value that takes the value of 0 or 1, which represents whether the pixel where the center point is located is marked as a reinforcing bar in the reinforcing bar mask.

[0123] The strategy for intelligently avoiding steel bars is an iterative exclusion method:

[0124] Starting from the center of mass;

[0125] If on the reinforcing bars, move along the gradient direction (away from the reinforcing bars);

[0126] Using rebar mask images Distance field As a repulsive potential field;

[0127] in, Euclidean distance transformation (EDT) for the reinforced area. It is a mask for non-reinforced areas; or Point The Euclidean distance to the nearest rebar pixel; inside the rebar, The farther away from the steel bars, The larger;

[0128] To move away from the rebar as quickly as possible, follow the gradient direction. Move in small steps until entering the effective area;

[0129] Where, the gradient direction for: ;

[0130] for At point The gradient vector is a two-dimensional vector; yes coordinates The partial derivatives of, in The rate of change of distance caused by a unit change in direction; yes coordinates The partial derivatives of, in The rate of change of distance caused by a unit change in direction. Gradient direction. The direction of fastest increase is the direction that gets further and further away from the reinforcing bar, which is the best path for escaping the reinforcing bar.

[0131] Furthermore, within the effective area, a multi-area coordinated avoidance method is adopted to avoid the steel reinforcement mesh. If multiple concrete areas are adjacent and share the steel reinforcement mesh, a repulsive potential field can be introduced to prevent the center points from getting too close. For all determined center points... Add exclusion terms:

[0132] ;

[0133] in, For the first A collaborative optimization objective function for each connected region is used to find the optimal center point; The original objective function is (i.e., safety distance + centroid approximation). The repulsion strength coefficient (a positive number) controls the magnitude of the repulsion force; This applies to all previously determined center points. Summation (in order of processing) For the first The final determination of the center point of each region;

[0134] It is an exponential function; It is a point The square of the distance to the existing center point; The radius (standard deviation) of the repulsive effect determines the size of the repulsion range;

[0135] The goal is to get the point as close as possible to the original centroid and as far away from the reinforcing steel as possible; This means that if a point is too close to an existing center point, it is penalized (its value decreases), thus avoiding overlap; multi-region cooperative avoidance is for the first... When selecting the center point for a connected component, avoiding excessive proximity to other already determined center points is crucial in the original objective function. Based on this, a repulsive term is subtracted to form a repulsive potential field.

[0136] Example 5

[0137] Based on Example 2, in step S3, for the specification requirements of the spacing between adjacent vibration points and the energy efficiency optimization target, an energy efficiency weighting function is introduced. The contribution of each candidate point to the overall compaction efficiency is measured, taking into account the local conditions of the concrete (e.g., slump). gas content Through image features (e.g., grayscale gradient) Texture entropy Indirect calculations are performed:

[0138] ;

[0139] in, For point The overall energy efficiency weight (the closer to 1, the more important vibration is needed). This is the first normalization adjustment factor; This is the second normalization adjustment factor; For the image at points Gradient magnitude (intensity of grayscale change) at the location; The parameters that control the steepness of the Sigmoid function are... Used to control response sensitivity; the larger the value, the steeper the slope and the higher the discrimination. The gradient threshold represents the criterion for judging significant edges. For point Texture entropy (information complexity) of the local region; This represents the maximum texture entropy value in the image.

[0140] For structure sensitivity weights based on image gradients, image gradients It reflects the local rate of change of pixel grayscale; in concrete surface images, high gradient areas correspond to: aggregate concentration areas (particle boundaries), air bubble aggregation areas, and uneven pouring interfaces. These areas are more prone to voids or segregation and require stronger vibration.

[0141] Use the Sigmoid function to map gradient values ​​to the [0,1] interval: when Low weight → smooth area, no need for strong vibration; when The weight approaches 1 → strong edge region, which is processed first.

[0142] For information richness weights based on texture entropy, texture entropy To measure the degree of disorder or information content of a local image, high-entropy regions represent: disordered distribution (e.g., aggregate mixture), areas with poor concrete flow, and areas with potential defects, while low-entropy regions are relatively uniform (e.g., areas dominated by slurry). In reverse mapping, the lower the entropy, the higher the weight, because low entropy means a simple structure that may lack density and therefore requires more vibration to promote fusion.

[0143] The ordered screening strategy is based on sorted in descending order And remove those whose neighborhood distance is less than For subsequent points, ensure that the minimum spacing constraint is met and prioritize the retention of high-efficiency points:

[0144] ;

[0145] in, This is the final set of actual vibration points (i.e., the output result). A candidate point set generated for the initial uniform grid (e.g., points are arranged in a 0.4m × 0.4m grid). An energy efficiency weighting function for each point measures the importance of vibration at that point; This refers to the minimum allowable spacing between adjacent vibration points; The greedy selection algorithm selects points step by step according to weight priority while satisfying spatial constraints.

[0146] Candidate point set A regular mesh is generated on the concrete region with a fixed step size, and a weighting function is applied. For each point Calculate its importance (combining image gradient and texture entropy); a higher value indicates a greater need for focused vibration; minimum spacing Ensure that the vibration does not overlap or become too dense. It involves real-time monitoring to detect a decrease in the slump of a certain area, which can then be dynamically improved. Therefore, it will be selected first in subsequent iterations.

[0147] By generating an initial candidate point set Calculate the energy efficiency weight for each point. Ordered screening under normative constraints This allows us to obtain the pixel coordinates of the actual vibration points.

[0148] Pixel space refers to a discrete coordinate system on the image or control plane, denoted as . The actuator space refers to the actual physical coordinates or control parameter space that a physical actuator (such as piezoelectric ceramics, electromagnetic vibrators, and robotic arm ends) can reach, denoted as . ;

[0149] The mapping relationship between pixel space and driver space is assumed to exist as a mapping function. Map pixel coordinates to driving parameters:

[0150] ;

[0151] in, Indicates the first line, number The corresponding position or parameter vector of column pixels in driver space (physical space); It is a mapping function that maps two-dimensional pixel coordinates. Converted into actual physical control parameters.

[0152] The driving parameters for point-by-point vibration compaction are based on adaptive vibration intensity calculation using image gradient and curvature. The amplitude is controlled by grayscale values, combined with local geometric features, to enhance the processing of key structural areas (such as edges and corners).

[0153] ;

[0154] in, For pixels The corresponding vibration amplitude (driving parameters); The grayscale value, depth value, or density requirement of the input image at this point; Image gradient (reflecting the rate of change of intensity). The gradient length required for the grayscale value, depth value, or density of the input image at that point; The maximum grayscale value of the image; This represents the maximum gradient magnitude across the entire graph (used for normalization). Let be the curvature of the contour lines in the image at a certain point. Let be the magnitude of curvature of the contour lines in the image at a certain point; The weighting parameter balances grayscale and gradient. is the enhancement coefficient for the curvature term; This is the global amplitude scaling factor (determined by device capabilities).

[0155] As a basic strength item, the brighter the image, the stronger the vibration, which directly reflects the required strength of the target area (e.g., the severity of defects). As an edge enhancement term, it automatically enhances the amplitude at locations of abrupt changes in grayscale (edges). Edges often correspond to structural boundaries, cracks, or material transition zones, and require special attention.

[0156] The basic strength term reflects the impact of the grayscale value of each pixel in the input image on the vibration intensity. A higher grayscale value means that the area may require stronger vibration. For example, areas with more severe defects on the concrete surface or less tight bonding between 3D printed layers often require special attention during concrete vibration. Edges and corners are more prone to voids or cracks, and increasing the vibration intensity can effectively improve the compaction of these areas.

[0157] For edge enhancement, this part increases the vibration intensity in edge areas by calculating the magnitude of the image gradient. A large gradient means drastic changes in grayscale, which usually corresponds to material boundaries or structural transition zones. These areas are often stress concentration points or places prone to quality problems. During concrete vibration, edges and corners often require special attention because these areas are more likely to form voids or cracks. Increasing the vibration intensity can effectively improve the compaction of these areas.

[0158] The geometric complexity term, specifically the curvature term, considers the local geometric complexity of the material surface. High curvature regions indicate significant bending or abrupt changes in the surface, which can lead to uneven energy distribution during vibration, affecting the final quality. For concrete structures with complex geometries, appropriately increasing the vibration intensity in areas with high curvature (such as corners or bends) can help better fill the material and reduce internal voids.

[0159] Taking into account the grayscale information, edge characteristics, and geometric complexity of the material surface, a global amplitude scaling factor is used. The vibration parameters are dynamically adjusted. This strategy not only improves the targeting and efficiency of the vibration process but also effectively enhances the overall quality and uniformity of the material. In this way, the system can more intelligently identify areas requiring focused treatment and apply appropriate vibration intensity to ensure optimal treatment results.

[0160] Example 6

[0161] Based on Example 2, in step S4, vibration is performed using point-by-point vibration driving parameters, employing a vibration evolution from point to field, defining a point-based vibration evolution. The spatiotemporal influence core centered on:

[0162] ;

[0163] in, Representing the The vibration points in the connected areas Location, The driving parameters of the frame and the vibration point , used to describe the first The spatial and temporal influence range and intensity of the vibration points in a connected area on the density and strength of the surrounding concrete. For the first The coordinates of the location of the vibration point in a connected region; For the first The moment when the vibration point in a connected region is determined to be qualified for vibration (i.e., the time when vibration is stopped). For the first The moment when vibration begins at the vibration point in a connected region; The radius of spatial influence (determined by the driving parameters); It is a time decay scale (reflecting the duration of vibrational energy). It is an exponential function; It is the Heaviside step function.

[0164] The superposition of Gaussian kernel functions in two orthogonal directions controls the attenuation of the effects in the spatial and temporal dimensions, respectively. For spatial terms, it indicates the distance from the vibration point. From the moment of departure, the influence diminishes exponentially with the square of the distance; the greater the distance, the smaller the influence. For the time term, it represents the influence at the moment of completion. Centered on the timeline, they are distributed in a bell shape. The aviside step function is the trigger condition.

[0165] A three-dimensional spatiotemporal influence field is described, starting from the vibration point. In the beginning, in time Initially, the impact is small, but it gradually increases as vibration continues. It reaches its maximum effect at a certain time, and then decays over time (due to energy dissipation). It spreads outward in space, but is limited by the material properties and vibration capacity.

[0166] The surface structure and distribution characteristics of the outermost layer of the image of the area of ​​action are based on the assumption that the current vibration point is located at... The effective radius of action is The outermost layer is a ring-shaped region, and the image distribution characteristics are present within this ring-shaped region:

[0167] ;

[0168] in, For the first One vibration point at The outermost set of the frame's spatial layers is used to determine the first... The vibration points in the connected areas The geometric extent of the outermost ring within the effective region of a frame, and the image distribution characteristics within that ring.

[0169] For any point coordinate in the plane (e.g., pixel position on a two-dimensional image); The radius of influence of the vibration action; The width of the annular region is used to control the thickness of the outermost ring; exist Location to vibrating point Euclidean distance.

[0170] This means that when concrete is vibrated, the vibration energy propagates outward, forming a [structure / structure]. The image shows the circular influence zone centered on the vibratory energy. This zone is the outermost ring, and the image distribution characteristics reflect whether the vibration effect is adequate. It also shows the location of the most likely defects such as air bubbles, segregation, and incomplete compaction within the influence zone.

[0171] The moment when the vibration quality of the current working area is qualified is when it appears in the concrete construction quality control or intelligent vibration system. It is used to determine whether a certain area (such as a concrete pouring area) has been sufficiently vibrated to meet the quality qualification standard.

[0172] The specific time for passing the test is:

[0173] ;

[0174] in, The moment when the vibration quality meets the standard; This is the current vibration time; The minimum vibration time threshold; For time Accumulated vibrational energy; The energy threshold; It is the rate of change of energy over time (instantaneous energy growth rate). A small positive number indicates a standard that tends towards stability.

[0175] This means that the vibration time should not be too short. Even if the energy is sufficient, if the vibration time is insufficient, air bubbles will not be fully expelled and the concrete will not be compacted.

[0176] Sufficient energy must be accumulated for vibration to accumulate; energy reflects the intensity and sustained effect of vibration.

[0177] The rate of increase in vibration energy is already very slow, indicating that the concrete is close to saturation.

[0178] The point at which vibration quality is deemed acceptable indicates that the air bubbles inside the concrete have been largely expelled and the particles are tightly packed. Continuing vibration further reduces energy absorption efficiency and slows down energy growth, thus avoiding over-vibration, which can lead to aggregate segregation, bleeding, and reduced strength.

[0179] The system achieves a three-in-one vibration quality control system integrating time, energy, and dynamic response to determine when the vibration quality is qualified. This is a key technical criterion in modern intelligent construction (such as intelligent vibrators and BIM+IoT).

[0180] Example 7

[0181] Based on the embodiment, in step S5, the automated continuous execution of the vibration operation continues, and the specific process is as follows:

[0182] Automatic lifting of vibrator: The control system drives the actuator to lift the vibrator vertically from the concrete smoothly and at a uniform speed, avoiding the formation of voids or disturbance of already compacted areas inside the concrete due to excessively fast lifting.

[0183] Equipment posture adjustment and path planning: Based on the pixel coordinates of the determined next vibration point, and combined with the mapping relationship between pixel space and driver physical space, the target position and posture that the vibratory rod end effector needs to move are calculated.

[0184] Automatically shift to the next vibration area: Drive the robotic arm or mobile platform to precisely transport the vibrator to the preparatory position directly above the next point to be vibrated, preparing for the next round of insertion and vibration.

[0185] Status Reset and Cycle Preparation: The system synchronously updates the operation status, including recording the information of completed vibration points, verifying equipment operating parameters, and preparing to receive image feedback for the next cycle, ensuring that subsequent vibration processes can be seamlessly connected and continuously optimized in a closed loop.

[0186] It enables safe exit after single-point vibration ends, and also ensures the efficiency, continuity and automation of multi-point continuous vibration operation. It is a key link in realizing the "perception-decision-execution-feedback" closed loop of the entire intelligent vibration control method.

[0187] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope described in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A smart vibration control method based on T-beam construction, characterized in that, Includes the following steps: Acquire real-time images of a concrete pouring area of ​​a T-beam and information on the location of reinforcing bars in the concrete, analyze the changes in the area of ​​the concrete mixture in adjacent images, and determine the quasi-static moment during the concrete paving process. Based on the contour features of the concrete mixture area and the position information of the reinforcing bars at the quasi-static moment, the approximate center point pixel coordinates of each area are automatically determined, and the reinforcing mesh is intelligently avoided. Based on the specifications for the spacing between adjacent vibration points and the energy efficiency optimization target, the pixel coordinates of the actual vibration points are obtained through orderly screening. Based on the mapping relationship between the pixel space and the driver space, the driving parameters for point-by-point vibration are obtained. Vibration is performed according to the driving parameters of point-by-point vibration, and the surface structure and distribution characteristics of the outermost layer of the action area at the current vibration point are analyzed in real time during the vibration process to obtain the moment when the vibration quality of the current action area is qualified. After vibration is completed, the vibrator is automatically lifted and moved to the next vibration area.

2. The intelligent vibration control method based on T-beam construction according to claim 1, characterized in that, Step S1 consists of the following steps: Data acquisition: Real-time acquisition of video / image sequences of the T-beam pouring area; synchronous acquisition of prior information on the location of reinforcing bars; Image preprocessing and segmentation: image denoising and illumination correction; Separate the concrete mixture area from the formwork and reinforcement area; Rebar location fusion: Maps rebar location information to the image coordinate system; eliminates interference from rebar areas on concrete area calculation; Area variation analysis: Calculate the effective concrete area in each frame; construct an area-time series. Quasi-static moment determination: Use the area change rate or its derivative to determine whether the paving tends to stabilize.

3. The intelligent vibration control method based on T-beam construction according to claim 2, characterized in that, The effective concrete area is calculated for each frame as follows: During the T-beam concrete pouring process, as time progresses, the concrete will be continuously pumped and spread, and its area within the formwork will gradually expand, reflecting the spreading progress: The increasing number of pixels in the effective concrete area of ​​the image over time reflects the dynamic progress of the concrete filling process. When the total number of pixels in the effective concrete area of ​​the image stops increasing or changes very little over time, it indicates that the paving is basically completed and has entered the quasi-static stage. For determining quasi-static moments, the inter-frame pixel displacement field is estimated using the optical flow method to obtain the quiescent index of the spreading flow field.

4. The intelligent vibration control method based on T-beam construction according to claim 1, characterized in that, In step S2, the contour features of the concrete mixture region at the quasi-static moment are extracted and calculated using connected regions. The method for calculating the reinforcement position information of the concrete mixture area at the quasi-static moment is to calculate it through the pixel coordinates of the center point of the connected region, which is used to locate the target; Automatically determine the approximate center point pixel coordinates of each region and judge whether the centroid falls within the reinforcement area: Check the centroid coordinates of connected regions in the rebar mask image: If the value is 0 or not on the rebar, accept this point as the final center point; if the value is 1 or on the rebar, the center point needs to be adjusted to avoid the rebar.

5. The intelligent vibration control method based on T-beam construction according to claim 1, characterized in that, In step S2, the strategy for intelligently avoiding the reinforcing bars is an iterative exclusion method: Starting from the center of mass; If it is on the reinforcing bar, then move along the gradient direction or away from the reinforcing bar; The distance field of the steel bar mask image is used as the repulsive potential field; To get away from the reinforcing bars as quickly as possible, move in small steps along the direction of the gradient until you enter the effective area.

6. The intelligent vibration control method based on T-beam construction according to claim 1, characterized in that, In step S3, for the standard requirements of the spacing between adjacent vibration points and the energy efficiency optimization target, the contribution of each candidate point to the overall compaction efficiency is measured by introducing an energy efficiency weight function, and the calculation is performed by combining the local state of concrete with image features. The ordered screening strategy arranges the candidate point set generated by the initial uniform grid in descending order according to the energy efficiency weight function, and removes subsequent points whose neighborhood distance is less than the minimum allowable spacing between adjacent vibration points, so as to ensure that the minimum spacing constraint is met and high energy efficiency points are retained first.

7. The intelligent vibration control method based on T-beam construction according to claim 6, characterized in that, The driving parameters for point-by-point vibration are based on adaptive vibration intensity calculation using image gradient and curvature. By controlling the amplitude through grayscale values ​​and combining local geometric features, the processing of key areas of the structure is enhanced.

8. The intelligent vibration control method based on T-beam construction according to claim 1, characterized in that, In step S4, the driving parameters for point-by-point vibration are used to perform vibration, which is to adopt the vibration evolution from point to field and define the spatiotemporal influence kernel centered on the position coordinates of the vibration point in the connected region. The surface structure and distribution characteristics of the outermost layer of the action area are based on the position coordinates of the vibration point in the connected region where the current vibration point is located. The effective action radius is taken as the maximum influence radius of the vibration action, resulting in an outermost ring area with the distribution characteristics of the image within the ring area. The moment when the vibration quality of the current working area is qualified is determined by the concrete construction quality control or intelligent vibration system, which judges whether sufficient vibration has been completed in a specific area, thereby achieving the quality qualification standard.

9. The intelligent vibration control method based on T-beam construction according to claim 1, characterized in that, In step S5, the automated continuous execution of the vibration operation continues, and the specific process is as follows: Automatic lifting of vibrator: The control system drives the actuator to lift the vibrator vertically from the concrete smoothly and at a uniform speed, avoiding the formation of voids inside the concrete or disturbance of already compacted areas due to excessive lifting speed; Equipment posture adjustment and path planning: Based on the pixel coordinates of the determined next vibration point, and combined with the mapping relationship between pixel space and driver physical space, the target position and posture that the vibratory rod end effector needs to move are calculated. Automatically shift to the next vibration area: Drive the robotic arm or mobile platform to precisely transport the vibratory rod to the preparatory position directly above the next point to be vibrated, preparing for the next round of insertion and vibration; Status Reset and Cycle Preparation: The system synchronously updates the operation status, including recording the information of completed vibration points, verifying equipment operating parameters, and preparing to receive image feedback for the next cycle, ensuring that subsequent vibration processes can be seamlessly connected and continuously optimized in a closed loop.

10. An intelligent vibration control device for T-beam construction, used to execute the intelligent vibration control method for T-beam construction according to any one of claims 1-9, characterized in that, include: The image acquisition module is used to acquire real-time images of the concrete pouring area of ​​the T-beam and information on the location of the reinforcing bars in the concrete. The quasi-static determination module is connected to the image acquisition module and is used to analyze the changes in the area of ​​the concrete mixture in adjacent real-time images to determine the quasi-static moment in the concrete paving process. The vibration point planning module is connected to the image acquisition module and the quasi-static determination module. It is used to automatically determine the pixel coordinates of the approximate center point of each area and avoid the steel mesh based on the contour features of the concrete mixture area and the position information of the steel bars at the quasi-static moment. Then, it selects the pixel coordinates of the actual vibration point based on the specification requirements of the spacing between adjacent vibration points and the energy efficiency optimization target. Furthermore, it generates the driving parameters for point-by-point vibration based on the mapping relationship between pixel space and driver space. The driving parameters are dynamically adjusted in real time according to the workability of concrete. The vibration execution module is signal-connected to the vibration point planning module and is used to execute the vibration operation according to the driving parameters of point-by-point vibration. The quality judgment module is connected to the image acquisition module and the vibration execution module. It is used to analyze the surface structure and distribution characteristics of the outermost layer of the image of the current vibration point during the vibration process, and to determine the moment when the vibration quality of the current area is qualified. The movement control module is connected to the vibration execution module and the quality judgment module. It is used to control the vibration execution module to lift the vibrator and move it to the next vibration area after the vibration quality in the current working area is qualified.

Images