Intelligent monitoring method and system for concrete pouring quality, product and medium

Abstract

The invention discloses an intelligent concrete pouring quality monitoring method and system, a product and a medium, and is applied to the field of monitoring and adjustment of control systems. According to the method, a pouring area is divided into a plurality of grid units, an adjacent matrix is established, a diffusion prediction model is established based on the matrix, the coverage range of each pump truck distribution arm and the real-time height difference value between the grid units are obtained, pumping parameters are adjusted according to the height difference value, the model is updated in real time, and the pumping flow is optimized till the height difference value reaches the standard. By implementing the technical scheme provided by the invention, the technical problems that the concrete flowing trend is difficult to master in real time and the pouring coordination of multiple pump trucks is poor in the construction of a mountain viaduct are effectively solved, and the uniform distribution of the concrete in a large-area pouring area is realized.

Description

Technical Field

[0001] The present application relates to the field of monitoring and regulation of a G05B23 control system, and in particular to a method, system, product and medium for intelligently monitoring concrete pouring quality. Background Art

[0002] With the rapid development of my country's transportation infrastructure construction, the number of viaduct projects in mountainous areas is increasing. During the construction of viaducts in mountainous areas, the quality of concrete pouring is directly related to the safety and durability of the bridge structure. Especially for super-large bridges with a span of more than 200 meters and a pier height of more than 90 meters, the concrete pouring construction requires not only one-time molding, but also uniform distribution of concrete in a large pouring area.

[0003] At present, the concrete pouring construction of viaducts in mountainous areas uses the cooperation of multiple concrete pump trucks and tank trucks. Construction workers set up a leveling point every 10 meters in the pouring area, and use a DS3 level to measure the elevation data of each leveling point at fixed time intervals. The surveyor records the elevation data in the construction log, and the on-site construction manager determines the working parameters of the pump truck based on the height difference between adjacent leveling points. The specific adjustment method is: when it is found that the elevation of a certain area is low, increase the pumping pressure and material distribution speed of the pump truck corresponding to the area; when multiple pump trucks are operating in overlapping areas, the construction manager directs each pump truck to take turns pouring operations through a walkie-talkie.

[0004] In the actual construction process, due to the limited number of leveling points and the long measurement interval, it is impossible to grasp the overall height change trend in the pouring area in real time. Especially in foggy weather in mountainous areas, the measurement accuracy is reduced, and it is difficult for construction workers to see the specific conditions of the far-end pouring area. At the same time, the large temperature difference between day and night in mountainous areas will cause changes in concrete properties, and manual observation methods are difficult to timely discover the impact of changes in concrete properties on fluidity, causing the height differences between areas to gradually accumulate, resulting in uneven concrete accumulation. Summary of the invention

[0005] The present application provides a method, system, product and medium for intelligent monitoring of concrete pouring quality, which are used to realize intelligent coordinated control of concrete pouring areas in complex mountainous environments to improve the uniformity of concrete pouring.

[0006] In a first aspect, the present application provides a method for intelligently monitoring the quality of concrete pouring, comprising: dividing a pouring area into a plurality of grid units, and establishing a grid adjacency matrix based on the spatial position relationship of each grid unit; based on the grid adjacency matrix, establishing a diffusion prediction model for predicting the flow characteristics of concrete between grid units, the diffusion prediction model taking the current performance parameters of concrete, the current pouring environment parameters and the real-time height difference between adjacent grid units as input, calculating the flow coefficient of each grid unit in the grid adjacency matrix, and determining the predicted height difference between adjacent grid units after a preset time based on the flow coefficient as output; obtaining the coverage range of each pump truck's distributing arm and the real-time height difference between each grid unit, and determining the main pouring area of ​​each pump truck and its corresponding initial height difference according to the distance relationship between the area where the real-time height difference is greater than a first preset threshold and each pump truck. start pumping flow; for the overlapping pouring area of ​​adjacent pump trucks, when the real-time height difference is greater than the second preset threshold, the pump truck located on the lower side of the overlapping pouring area is switched to the leading pump truck, and the pumping flow of the leading pump truck is increased according to the preset flow adjustment step length until the real-time height difference is less than the second preset threshold; based on the real-time height difference between each grid unit, the flow coefficient in the diffusion prediction model is updated in real time to obtain the predicted height difference between adjacent grid units after a preset time length; based on the predicted height difference output by the diffusion prediction model, the height uniformity index corresponding to different pumping parameter combinations is calculated, and the pumping parameter combination that can make the height difference between adjacent grid units less than the third preset threshold after the preset time length is selected to adjust the pumping flow of each pump truck in real time until the actual height difference and the predicted height difference between each grid unit are both less than the third preset threshold.

[0007] By adopting the above technical solution, the embodiment of the present application divides the pouring area into grid units and establishes an adjacency matrix to construct a diffusion prediction model of concrete flow characteristics. The model can dynamically calculate the flow coefficient based on concrete performance parameters, environmental parameters and real-time height difference, thereby accurately predicting the flow trend of concrete. Combined with the real-time coverage range and height difference data of the pump truck's distributing arm, the system can intelligently determine the main pouring area and pumping parameters of each pump truck, and adaptively adjust the pumping flow of the leading pump truck in the overlapping area. Through the real-time update of the diffusion prediction model and the coordinated control of multiple pump trucks, the uniform distribution of concrete in a large pouring area is achieved, effectively solving the problem of uneven concrete accumulation in the construction of viaducts in mountainous areas.

[0008] In combination with some embodiments of the first aspect, in some embodiments, the establishment of a diffusion prediction model for predicting the flow characteristics of concrete specifically includes: obtaining the slump, initial setting time, external temperature and concrete temperature of concrete as the concrete performance parameters, and obtaining the temperature, humidity and wind speed of the construction environment as the casting environment parameters; constructing a directed graph model based on the grid adjacency matrix, wherein the nodes of the graph represent grid units, and the weights of the edges are determined by the height difference, distance and the concrete performance parameters of adjacent grid units; and calculating the flow coefficient D(i, j, t) between adjacent grid units i and j at time t according to the following formula: ;in, is the basic flow coefficient, i and j represent the numbers of two adjacent grid cells, S is the slump, is the initial setting time, is the concrete temperature, is the ambient temperature, H is the humidity, W is the wind speed, t is the time from the start of pouring, , , , , , and λ are corresponding weight coefficients; the flow coefficient is substituted into the diffusion equation to establish the diffusion prediction model: ; Where K is the global correction coefficient, A(i, j) is the grid adjacency matrix element, and ΔZ(i, j, t) is the height gradient at the current moment.

[0009] By adopting the above technical solution, the embodiment of the present application constructs a diffusion prediction model that takes into account the coupling of multiple factors. The model characterizes the spatial relationship between grid units through a directed graph structure, and introduces performance parameters such as concrete slump, initial setting time, temperature, and environmental parameters such as ambient temperature and humidity, wind speed, etc., to establish a calculation formula for the flow coefficient. Based on the flow coefficient, the diffusion equation can accurately describe the flow characteristics of concrete between grid units.

[0010] In combination with some embodiments of the first aspect, in some embodiments, UWB base stations are deployed at multiple commanding heights in the pouring area, and an inertial navigation unit, a tilt sensor, and a laser rangefinder are installed on the pump truck; the coverage range of each pump truck's fabric arm and the real-time height difference between each grid unit are obtained, specifically including: obtaining the real-time positioning data of the pump truck by the UWB base station, and monitoring the quality of the UWB signal. When it is detected that the signal quality decreases due to mountain obstruction, the positioning scheme is switched from UWB positioning to inertial navigation positioning, and the last set of valid UWB positioning data before the switch is recorded as the initial reference position; obtaining the real-time positioning data of the inertial navigation unit; and The acceleration and angular velocity data are used to calculate the relative displacement relative to the initial reference position, and the inclination data of the fabric arm is obtained through the inclination sensor to calculate the real-time attitude angle of the fabric arm; based on the initial reference position, the relative displacement, the inclination data and the vertical distance data from the fabric arm outlet to the concrete surface measured by the laser rangefinder, the precise spatial position of the fabric arm outlet is determined; according to the spatial position, attitude angle, maximum extension length and rotation angle range of the fabric arm, the coverage range of the fabric arm of each pump truck is calculated; based on the concrete surface elevation data collected by the laser rangefinder, the real-time height difference between each grid unit is calculated.

[0011] By adopting the above technical solution, the embodiment of the present application realizes high-precision positioning of the pump truck and real-time monitoring of the posture of the distributing arm. The system uses a fusion positioning solution of UWB base station and inertial navigation. When the UWB signal quality decreases due to mountain obstruction, it can automatically switch to inertial navigation positioning, ensuring the continuity and reliability of positioning. Combined with the data of the inclination sensor and laser rangefinder, the system can accurately calculate the spatial position and coverage of the distributing arm outlet, providing a guarantee for the precise pouring of concrete.

[0012] In combination with some embodiments of the first aspect, in some embodiments, based on the real-time height difference between each grid unit, the flow coefficient in the diffusion prediction model is updated in real time to obtain the predicted height difference between adjacent grid units after a preset time, specifically including: obtaining the factory performance parameters of each batch of concrete; establishing a correspondence between the grid unit and the concrete batch based on the GPS positioning and unloading records of the tank truck, and mapping the performance parameters of each batch of concrete to the corresponding grid unit; extracting the steel bar layout information from the BIM model, and calculating the steel bar volume ratio and main reinforcement direction of each grid unit; determining the local concrete flow direction based on the real-time height difference between each grid unit; when it is detected that the performance parameter difference of adjacent grid units exceeds the preset performance difference threshold and the difference between the real-time height difference and the predicted height difference of the diffusion prediction model before the preset time exceeds the prediction deviation threshold, when calculating the flow coefficient of the corresponding boundary, the flow coefficient of the diffusion prediction model is corrected to: ; Wherein: Φ(ρ, θ) is the steel bar correction function, ρ is the steel bar volume ratio, and θ is the angle between the local concrete flow direction and the main reinforcement direction; the corrected flow coefficient is substituted into the diffusion prediction model, and the concrete performance parameters, pouring environment parameters and real-time height difference between adjacent grid units at the current moment are input to obtain the predicted height difference between adjacent grid units after the preset time length.

[0013] By adopting the above technical scheme, the embodiment of the present application realizes the accurate correction of the diffusion prediction model at the boundary of the sudden change of concrete performance. The system establishes a dual threshold judgment mechanism by simultaneously monitoring the two key indicators of the performance parameter difference and the height difference prediction deviation of adjacent grid units, which can accurately identify the interface between different batches of concrete. At these interfaces, the applicability of the original diffusion coefficient calculation method will be reduced due to the sudden change of concrete performance. However, this scheme introduces a correction function that considers the steel bar arrangement information at the boundary of the performance mutation, so that the model can more accurately describe the nonlinear flow characteristics of the concrete here. The correction function not only takes into account the influence of the steel bar volume ratio on the effective flow space, but also describes the anisotropic obstruction of the steel mesh to the concrete flow through the relationship between the main reinforcement direction and the angle between the flow direction, thereby achieving more accurate flow prediction at the boundary of the performance mutation, providing a more reliable basis for the optimization and adjustment of pumping parameters.

[0014] In conjunction with some embodiments of the first aspect, in some embodiments, the steel bar correction function Φ(ρ, θ) is specifically: ; Among them: ρ is the steel volume ratio, which indicates the volume ratio of steel bars in a unit volume of concrete; θ is the angle between the local concrete flow direction and the main reinforcement direction; γ is the directional influence coefficient, 0≤γ≤1; (1-ρ) indicates the reduction effect of the steel volume on the effective flow space; It represents the influence of the angle between the flow direction and the main reinforcement direction. When the flow direction is parallel to the main reinforcement, the influence is the smallest, at this time θ=0°, and when it is perpendicular, the influence is the largest, at this time θ=90°.

[0015] By adopting the above technical solution, the embodiment of the present application proposes a correction function that takes into account the influence of steel bar arrangement. This function reflects the reduction effect of steel bars on the effective flow space of concrete through the steel bar volume ratio, and describes the directional influence through the trigonometric function relationship between the flow direction and the angle between the main bar direction, so that the diffusion prediction model can accurately reflect the hindering effect of steel bar arrangement on concrete flow.

[0016] In combination with some embodiments of the first aspect, in some embodiments, the method also includes: based on a vibration sensor array arranged at each grid unit, collecting concrete vibration response characteristics; based on a spectral analysis of the concrete vibration response characteristics, dynamically calculating the density of each grid unit; when it is detected that the density of a local area is lower than a preset density threshold, based on the area size of the low-density area, the density deviation value and the coverage of surrounding pump trucks, adaptively adjusting the pumping flow of the relevant pump trucks until the density meets the standard.

[0017] By adopting the above technical solution, the embodiment of the present application realizes real-time monitoring and intelligent control of concrete density. The system collects the vibration response characteristics of concrete through the deployed vibration sensor array, dynamically evaluates the density based on spectrum analysis, and adaptively adjusts the pumping flow of the relevant pump truck when insufficient density is detected in a local area, thereby ensuring the quality of concrete pouring.

[0018] In combination with some embodiments of the first aspect, in some embodiments, based on the area size of the low-density area, the density deviation value and the coverage of the surrounding pump trucks, the pumping flow of the relevant pump trucks is adaptively adjusted until the density meets the standard, specifically including: calculating the overlap between the coverage range of each pump truck's distribution arm and the low-density area, and selecting the pump truck with the largest overlap as the target pump truck; determining the required supplementary concrete volume based on the area size and the density deviation value of the low-density area and the concrete performance parameters; calculating the target pumping flow based on the supplementary concrete volume and the preset maximum pumping time; gradually increasing the pumping flow from the current value to the target flow by gradually increasing the pumping flow, with the incremental step not exceeding the preset maximum flow change rate, until the density reaches the preset density threshold; when it is detected that the density growth rate exceeds the preset growth rate threshold, reducing the pumping flow to 80% of the current flow value to reduce local stress concentration.

[0019] By adopting the above technical solution, the embodiment of the present application establishes a set of adaptive control mechanism for pumping flow based on density. The system selects the optimal target pump truck by calculating the overlap between the pump truck coverage and the low-density area, and intelligently calculates the target pumping flow based on the supplementary concrete volume. The pumping flow is increased by gradual adjustment, while the density growth rate is monitored. When necessary, the pumping flow is reduced to avoid stress concentration, thereby achieving precise control of concrete density.

[0020] In a second aspect, the present application provides an intelligent monitoring system for concrete pouring quality, comprising: one or more processors and a memory; the memory is coupled to the one or more processors, the memory is used to store computer program code, the computer program code includes computer instructions, and the one or more processors call the computer instructions to enable the system to execute the method described in the first aspect and any possible implementation method of the first aspect.

[0021] In a third aspect, the present application provides a computer program product comprising instructions, which, when executed on a concrete pouring quality intelligent monitoring system, enables the concrete pouring quality intelligent monitoring system to execute the method described in the first aspect and any possible implementation of the first aspect.

[0022] In a fourth aspect, the present application provides a computer-readable storage medium comprising instructions, which, when executed on a concrete pouring quality intelligent monitoring system, enables the concrete pouring quality intelligent monitoring system to execute the method described in the first aspect and any possible implementation of the first aspect.

[0023] It can be understood that the intelligent monitoring system for concrete pouring quality provided in the second aspect, the computer program product provided in the third aspect, and the computer-readable storage medium provided in the fourth aspect are all used to execute the method provided in the embodiment of the present application. Therefore, the beneficial effects that can be achieved can refer to the beneficial effects in the corresponding method, which will not be repeated here.

[0024] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages: 1. Due to the adoption of the diffusion prediction model based on the grid adjacency matrix and the coordinated control mechanism of multiple pump trucks, the technical problems of difficulty in real-time grasp of the concrete flow trend and poor coordination of multiple pump trucks in the construction of mountain viaducts were effectively solved, thereby achieving uniform distribution of concrete in a large area of ​​casting.

[0025] 2. Due to the adoption of the integrated positioning solution of UWB base station and inertial navigation and the collaborative measurement mechanism of multi-sensor of the fabric arm, the technical problems that the positioning signal of the pump truck is easily blocked and the spatial position of the fabric arm is difficult to accurately determine in mountainous environments are effectively solved, thereby realizing the continuous and reliable positioning of the pump truck and the precise spatial positioning of the fabric arm outlet.

[0026] 3. Due to the adoption of a density-based pumping flow gradual adjustment mechanism and a stress concentration prevention strategy, the technical problem of insufficient density in local areas and easy stress concentration during concrete pouring has been effectively solved, thereby achieving precise control of concrete density. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 It is a structural diagram of a system architecture to which the intelligent monitoring method for concrete pouring quality in the embodiment of the present application can be applied; Figure 2 This is a flow chart of a method for intelligently monitoring the quality of concrete pouring in an embodiment of the present application; Figure 3 is another flow chart of the method for intelligently monitoring the quality of concrete pouring in an embodiment of the present application; Figure 4 It is an exemplary hardware structure diagram of the control server in the intelligent monitoring system for concrete pouring quality in the embodiment of the present application. DETAILED DESCRIPTION

[0028] The terms used in the following embodiments of the present application are only for the purpose of describing specific embodiments, and are not intended to be used as limitations to the present application. As used in the specification and appended claims of the present application, the singular expressions "one", "a kind of", "said", "above", "the" and "this" are intended to also include plural expressions, unless there is a clear indication to the contrary in the context. It should also be understood that the term "and / or" used in the present application refers to and includes any or all possible combinations of one or more listed items.

[0029] In the following, the terms "first" and "second" are used for descriptive purposes only and are not to be understood as suggesting or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features, and in the description of the embodiments of the present application, unless otherwise specified, "plurality" means two or more.

[0030] Figure 1 It is a structural diagram of a system architecture that can be applied to the intelligent monitoring method for concrete pouring quality in the embodiment of the present application.

[0031] See also Figure 1 , the concrete pouring quality intelligent monitoring system includes: The control server 100 is used to receive various data collected by other devices in the system and perform unified scheduling and control of the entire system; A plurality of UWB base stations 110 are arranged at commanding heights in the casting area to provide positioning signals; Multiple concrete pump trucks 120, each of which is equipped with: an inertial navigation unit for acquiring the motion state data of the pump truck; an inclination sensor for measuring the attitude angle of the distributing arm; a laser rangefinder for measuring the distance between the end of the distributing arm and the casting surface; and a pumping control unit for adjusting the pumping parameters; A plurality of concrete tank trucks 130, each tank truck is equipped with: an inertial navigation unit for acquiring motion state data of the tank truck; an inclination sensor for monitoring the tilt state of the tank truck; UAV 140, equipped with a laser radar system, is used to obtain three-dimensional point cloud data of the construction site; Environmental monitoring unit 150, used to collect environmental parameters such as temperature, humidity, wind speed, etc. at the construction site; The control server 100 exchanges data with each device unit through a wireless network. The data from the UWB base station 110, the inertial navigation unit and the laser rangefinder are used to achieve high-precision positioning. The data collected in real time by the environmental monitoring unit 150 is used to assist construction decision-making. The three-dimensional point cloud data obtained by the drone 140 during regular patrols is used to update the terrain information of the construction site.

[0032] The following describes the intelligent monitoring method for concrete pouring quality in the embodiment of the present application in combination with the above exemplary system architecture diagram: See also Figure 2 , is a flow chart of a method for intelligently monitoring the quality of concrete pouring in an embodiment of the present application. The method comprises the following steps: S201, dividing the casting area into a plurality of grid units, and establishing a grid adjacency matrix based on the spatial position relationship of each grid unit; Among them, the grid unit represents a regular small area obtained by dividing the entire casting area according to a preset size, which is used to achieve refined management of the casting area; the grid adjacency matrix refers to a two-dimensional matrix used to describe the spatial adjacency relationship between grid units, which is used to represent the connectivity and relative position relationship between each grid unit.

[0033] Before starting the concrete pouring operation, the pouring area needs to be gridded and the topological relationship between grids needs to be established. Specifically, the boundary range of the pouring area is first determined based on the three-dimensional point cloud data obtained by the drone, and then the area is divided into several square grid units of equal size. The side length of each grid unit is set according to the construction requirements, usually 1-2 meters. For each grid unit, the adjacent grid units around it are determined, and the corresponding connection relationship is established in the grid adjacency matrix. If two grid units are adjacent, the corresponding matrix element value is 1, otherwise it is 0.

[0034] In some embodiments, the grid division and adjacency matrix construction of the pouring area can be achieved in a variety of ways: Optionally, the boundary coordinate points of the pouring area can be determined by RTK measurement, and then the Delaunay triangulation algorithm can be used to divide the area into regular triangular grids, and finally the grid adjacency matrix can be constructed based on the node connection relationship of the triangular grid; Optionally, the pouring area can also be divided into regular rectangular grids based on the orthophotos collected by the drone, and the grid adjacency matrix can be automatically generated by scanning the four-neighborhood relationship of each grid. It is understandable that other grid division methods and adjacency matrix construction methods can also be used, which are not limited here.

[0035] S202, based on the grid adjacency matrix, establishing a diffusion prediction model for predicting the flow characteristics of concrete between grid units, the diffusion prediction model taking the current performance parameters of concrete, the current pouring environment parameters and the real-time height difference between adjacent grid units as input, calculating the flow coefficient of each grid unit in the grid adjacency matrix, and determining the predicted height difference between adjacent grid units after a preset time period based on the flow coefficient as output; Among them, the diffusion prediction model refers to a mathematical model used to describe the flow characteristics of concrete between grid units under the action of gravity, and is used to predict the flow trend of concrete; the flow coefficient refers to a parameter that characterizes the flow ability of concrete under specific conditions, and is used to quantify the flow characteristics of concrete; performance parameters include characteristic indicators such as the slump, initial setting time and temperature of concrete; pouring environment parameters include external conditions such as ambient temperature, humidity and wind speed.

[0036] After executing step S201 to complete the grid division, a mathematical model that can accurately predict the flow characteristics of concrete can be established. Specifically, first, a diffusion equation that takes into account the effect of gravity is constructed based on fluid mechanics theory, and the flow process of concrete between grid units is simplified to a diffusion process. Then, concrete performance parameters and environmental parameters are introduced to correct the flow coefficient, and a calculation formula for the flow coefficient is established. Finally, the corrected flow coefficient is substituted into the diffusion equation, and the height difference between each grid unit after a preset time length is obtained by numerical solution.

[0037] In some embodiments, the diffusion prediction model can be established in a variety of ways: Optionally, a basic diffusion equation can be established based on Fick's diffusion law, and then a complete diffusion prediction model can be constructed by introducing correction terms that consider concrete performance and environmental impact, and the finite difference method can be used to solve it; Optionally, a machine learning method can also be used to collect a large amount of historical pouring data to establish a training set, and a deep neural network can be used to fit the flow characteristics of concrete to achieve accurate prediction of height differences. It is understandable that other mathematical models or prediction methods can also be used to predict the flow characteristics of concrete, which is not limited here.

[0038] Preferably, in some embodiments, establishing a diffusion prediction model for predicting concrete flow characteristics may specifically include: Acquire the concrete slump, initial setting time, external temperature and concrete temperature as the concrete performance parameters, and acquire the temperature, humidity and wind speed of the construction environment as the casting environment parameters; Constructing a directed graph model based on the grid adjacency matrix, wherein the nodes of the graph represent grid units, and the weights of the edges are determined by the height difference, distance and the concrete performance parameters of adjacent grid units; Specifically, the steps of constructing a directed graph model may include: 1) Treat each grid unit as a node in the graph, and the node attributes include the spatial coordinates of the grid unit and the current concrete height; 2) Determine the connection relationship between nodes based on the grid adjacency matrix A(i, j). If A(i, j) = 1, a bidirectional edge is established between nodes i and j. 3) Calculate the edge weight w(i, j): ; Where f(Δh, d) is a function of height difference and distance: f(Δh, d) = |Δh| / d; Δh is the height difference between nodes i and j, and d is the Euclidean distance between the two nodes; is the influence function of concrete performance parameters: ;in , , is the corresponding weight coefficient.

[0039] The flow coefficient D(i, j, t) between adjacent grid cells i and j at time t can then be calculated according to the following formula: ; in, is the basic flow coefficient, i and j represent the numbers of two adjacent grid cells, S is the slump, is the initial setting time, is the concrete temperature, is the ambient temperature, H is the humidity, W is the wind speed, t is the time from the start of pouring, , , , , , and λ are the corresponding weight coefficients; Substituting the flow coefficient into the diffusion equation, the diffusion prediction model is established: ; Among them, K is the global correction coefficient, A(i, j) is the grid adjacency matrix element, and ΔZ(i, j, t) is the height gradient at the current moment.

[0040] S203, obtaining the coverage of the distributing arm of each pump truck and the real-time height difference between each grid unit, and determining the main pouring area of ​​each pump truck and its corresponding initial pumping flow rate according to the distance relationship between the area where the real-time height difference is greater than the first preset threshold and each pump truck; Among them, the coverage range of the placing arm indicates the maximum working area that the placing arm of the pump truck can reach, which is used to determine the effective operating range of the pump truck; the real-time height difference refers to the height difference between the concrete surfaces of adjacent grid units, which is used to characterize the uniformity of pouring; the first preset threshold represents the height difference critical value that triggers the dispatch of the pump truck, which is used to identify areas that need priority pouring; the main pouring area refers to the priority pouring area assigned to each pump truck, which is used to realize the coordinated operation of multiple pump trucks.

[0041] After executing step S201 to complete the grid division, each pump truck can be reasonably scheduled according to the height difference between the grid units to achieve uniform pouring. Specifically, first obtain the spatial position and posture data of the cloth arm through the sensor system on each pump truck, and calculate its coverage range in combination with the maximum extension length and rotation angle range of the cloth arm. Then, based on the laser ranging data, calculate the real-time height difference between each grid unit, and identify the area where the height difference exceeds the first preset threshold. Finally, according to the distance relationship between these areas and each pump truck, divide the main pouring area for each pump truck according to the principle of proximity, and set the initial pumping flow rate according to the area size and height difference.

[0042] In some embodiments, the dispatching and allocation of pump trucks can be achieved in a variety of ways: Optionally, a minimum spanning tree algorithm can be used, with the pump truck position as the root node and the grid cells with height differences exceeding the threshold as leaf nodes, to construct a minimum distance tree, assign grid cells on the same subtree to the corresponding pump truck, and determine the pumping flow rate based on the size of the subtree; Optionally, a clustering algorithm can be used to cluster grid cells with height differences exceeding the threshold, and then calculate the distance matrix between each cluster center and the pump truck, and use the Hungarian algorithm to solve the optimal match to achieve the optimal allocation of pump trucks and pouring areas. It is understandable that other scheduling algorithms can also be used to achieve the matching of pump trucks and pouring areas, which is not limited here.

[0043] S204, for the overlapping pouring area of ​​adjacent pump trucks, when the real-time height difference is greater than a second preset threshold, the pump truck located at the lower side of the overlapping pouring area is switched to the leading pump truck, and the pumping flow of the leading pump truck is increased according to a preset flow adjustment step length until the real-time height difference is less than the second preset threshold; Among them, the overlapping pouring area represents the intersection area of ​​the coverage range of the distribution arms of two or more pump trucks, which is used to realize the coordinated pouring of multiple pump trucks; the second preset threshold represents the height difference critical value that triggers the switching of pump trucks, which is used to control the pouring uniformity of the overlapping area; the leading pump truck refers to the pump truck that undertakes the main pouring task in the overlapping area, which is used to coordinate the pouring rhythm of multiple pump trucks; the flow adjustment step represents the incremental size of each adjustment of the pumping flow, which is used to achieve a smooth change in the pumping flow.

[0044] After executing step S203 to determine the main pouring area of ​​each pump truck, it is necessary to focus on the overlapping pouring areas of adjacent pump trucks. Specifically, first calculate the intersection of the coverage range of each pump truck's distribution arm to determine the range of the overlapping pouring area. Then monitor the height difference in the overlapping area in real time. When it is detected that the height difference exceeds the second preset threshold, the pump truck covering the lower height area is set as the leading pump truck. Finally, the pumping flow of the leading pump truck is gradually increased according to the preset flow adjustment step, while reducing the pumping flow of other pump trucks in the area until the height difference is reduced below the threshold.

[0045] In some embodiments, the coordinated control of overlapping areas can be achieved in a variety of ways: optionally, a proportional integral control algorithm is used, the height difference is used as a feedback signal, the steady-state error is eliminated through the integral term, and the flow increment required by the leading pump truck is calculated in real time to achieve a smooth transition of the overlapping area; optionally, based on a fuzzy control method, the height difference and the change trend are used as input variables, and a fuzzy rule library is designed to guide the flow adjustment to achieve adaptive control of the leading pump truck. It is understandable that other control algorithms can also be used to achieve coordinated pouring of overlapping areas, which is not limited here.

[0046] S205, based on the real-time height difference between each grid unit, the flow coefficient in the diffusion prediction model is updated in real time to obtain the predicted height difference between adjacent grid units after a preset time period; Among them, real-time update means dynamically adjusting model parameters according to the latest monitoring data to improve prediction accuracy; preset duration represents the time span of prediction, which is used to determine the time range of prediction; predicted height difference refers to the height difference between adjacent grid cells at a certain moment in the future predicted by the model, which is used to guide the optimization adjustment of pumping parameters.

[0047] After executing step S202 to obtain the diffusion prediction model, the diffusion prediction model can be continuously optimized while executing step S204 to perform pump truck scheduling and flow control. Specifically, first collect the real-time height difference data between each grid unit, compare it with the model prediction value, and calculate the prediction error. Then, based on the error information, use the parameter optimization algorithm to perform real-time correction on the flow coefficient to improve the prediction accuracy of the model. Finally, use the updated flow coefficient to recalculate the height difference prediction result after the preset time length to provide a basis for subsequent pumping parameter optimization.

[0048] In some embodiments, real-time updating of model parameters can be achieved in a variety of ways: optionally, using a Kalman filter algorithm, taking the measured height difference as the observed value, and the predicted height difference as the state value, the filter is used to estimate the optimal flow coefficient in real time to improve the prediction accuracy of the model; optionally, using an online learning method to construct a recursive neural network model, taking the historical data sequence as input, and continuously optimizing the network parameters through the back propagation algorithm to achieve adaptive updating of the model. It is understandable that other parameter optimization methods can also be used to achieve real-time updating of the diffusion prediction model, which is not limited here.

[0049] Since different batches of concrete may have performance differences that lead to abnormal boundary flow performance, in some embodiments, the factory performance parameters of each batch of concrete can be obtained first; based on the GPS positioning and unloading records of the tank truck, a corresponding relationship between the grid unit and the concrete batch is established, and the performance parameters of each batch of concrete are mapped to the corresponding grid unit; the steel bar layout information is extracted from the BIM model, and the steel bar volume ratio and main bar direction of each grid unit are calculated; the local concrete flow direction is determined based on the real-time height difference between the grid units; when it is detected that the performance parameter difference of adjacent grid units exceeds the preset performance difference threshold and the difference between the real-time height difference and the predicted height difference of the diffusion prediction model before the preset time exceeds the prediction deviation threshold, the flow coefficient of the diffusion prediction model is corrected to the following when calculating the flow coefficient of the corresponding boundary: ; Where: Φ (ρ, θ) is the steel bar correction function, ρ is the steel bar volume ratio, and θ is the angle between the local concrete flow direction and the main reinforcement direction; Substitute the corrected flow coefficient into the diffusion prediction model, input the concrete performance parameters, pouring environment parameters and the real-time height difference between adjacent grid units at the current moment, and obtain the predicted height difference between adjacent grid units after the preset time. If the performance parameter difference between adjacent grid units does not exceed the preset performance difference threshold, continue to update the flow coefficient of the original diffusion prediction model and execute the subsequent steps.

[0050] Among them, the reinforcement correction function Φ (ρ, θ) can be specifically: ; Among them: ρ is the steel volume ratio, which indicates the volume ratio of steel bars in a unit volume of concrete; θ is the angle between the local concrete flow direction and the main reinforcement direction; γ is the directional influence coefficient, 0≤γ≤1; (1-ρ) indicates the reduction effect of the steel volume on the effective flow space; It represents the influence of the angle between the flow direction and the main reinforcement direction. When the flow direction is parallel to the main reinforcement, the influence is the smallest, at this time θ=0°, and when it is perpendicular, the influence is the largest, at this time θ=90°.

[0051] S206. Based on the predicted height difference output by the diffusion prediction model, calculate the height uniformity index corresponding to different pumping parameter combinations, select the pumping parameter combination that can make the height difference between adjacent grid units less than the third preset threshold after the preset time, and adjust the pumping flow of each pump truck in real time until the actual height difference and the predicted height difference between each grid unit are both less than the third preset threshold.

[0052] Among them, the pumping parameter combination represents a collection of working parameters such as pumping flow and pumping pressure of multiple pump trucks, which is used to realize coordinated control of multiple pump trucks; the height uniformity index refers to a quantitative index used to evaluate the flatness of the concrete surface in the pouring area, which is used to measure the pouring quality; the third preset threshold represents the target value of pouring quality control, which is used to judge whether the expected pouring effect is achieved.

[0053] After executing step S205 to obtain the updated prediction results, the pumping parameters can be optimized to achieve uniform pouring. Specifically, first construct a candidate combination set containing parameters such as the pumping flow rate and pumping pressure of each pump truck, and use the diffusion prediction model to calculate the predicted height difference corresponding to each combination. Then, the height uniformity index is calculated based on the predicted height difference, and the optimal parameter combination that can make the height difference between adjacent grid units less than the third preset threshold after a preset time is selected. Finally, the optimal parameter combination is sent to each pump truck for execution, and the actual effect is continuously monitored until the expected goal is achieved.

[0054] In some embodiments, for a pump truck located in an overlapping pouring area, when receiving instructions to adjust the pumping flow rate by executing step S204 and executing step S206 at the same time, the flow rate can be adjusted only according to the instruction to adjust the pumping flow rate by executing step S206.

[0055] In some embodiments, the prediction calculation based on the candidate parameter combination can be implemented in the following manner: First, for each pump truck, the pumping flow and pumping pressure are discretized and sampled within their respective reasonable ranges to obtain a set of parameter combinations for a single pump truck. For example, the pumping flow can take a value every 10 m³ / h in the range of 30-120 m³ / h, and the pumping pressure can take a value every 1 MPa in the range of 4-12 MPa, thereby constructing a candidate set containing multiple sets of parameter combinations.

[0056] Then, through the Cartesian product operation, the parameter combination sets of each pump truck are combined into a complete multi-pump truck parameter combination set. For each set of parameter combinations, it is substituted into the diffusion prediction model and the predicted height difference is calculated through the following steps: 1) Based on the real-time height of each grid unit and the pumping flow rate in the parameter combination, calculate the concrete increment of each grid unit within the preset time; 2) Using the flow coefficient of the diffusion prediction model, calculate the height change caused by concrete flow within a preset time; Specifically, the step of calculating the height change using the flow coefficient of the diffusion prediction model includes: determining the adjacent units of each grid unit according to the grid adjacency matrix; for each pair of adjacent grid units (i, j), based on the current flow coefficient k_ij and the height difference Δh_ij, the following diffusion equation is used to calculate the flow volume q_ij per unit time: q_ij = k_ij *Δh_ij; multiplying the flow volume per unit time by the preset time length Δt to obtain the total flow volume Q_ij within the preset time length: Q_ij = q_ij * Δt; for each grid unit i, accumulating the flow volume between all adjacent grid units j to obtain the net height change ΔH_i of the grid unit due to concrete flow: ΔH_i = Σ(Q_ij) / A_i; where A_i is the area of ​​the grid unit i. It can be understood that this calculation method is based on a simplified diffusion model. In practical applications, nonlinear characteristics such as viscosity and yield stress of concrete can also be considered to modify the calculation formula, which is not limited here.

[0057] 3) Superimpose the height changes caused by concrete increment and flow to obtain the predicted height of each grid unit after a preset time; 4) Calculate the predicted height difference between adjacent grid cells.

[0058] Finally, the prediction results of all parameter combinations are evaluated, and the optimal combination that can make the height difference meet the requirements is selected. It can be understood that other methods can also be used to implement the prediction calculation of parameter combinations, which are not limited here.

[0059] In some embodiments, the height uniformity index can be calculated based on the predicted height difference in the following manner: First, for the prediction results of each set of pumping parameter combinations, the height uniformity index is calculated using the following steps: 1) Calculate the root mean square error (RMSE) of the predicted height difference of all adjacent grid unit pairs in the entire casting area: RMSE = sqrt(Σ(Δh_ij²) / N); where Δh_ij is the predicted height difference between grid unit i and its adjacent unit j, and N is the total number of all adjacent unit pairs. 2) Calculate the local height difference coefficient (LHC): For each grid unit i, calculate the maximum height difference max(|Δh_ij|) between it and all adjacent units j; LHC = Σmax(|Δh_ij|) / M; where M is the total number of grid units. This indicator reflects the degree of height mutation in the local area. 3) Calculate the global height standard deviation (GHS): First, calculate the average h_avg of the predicted heights of all grid units; then calculate the standard deviation: GHS = sqrt(Σ(h_i - h_avg)² / M); where h_i is the predicted height of grid unit i. This indicator reflects the overall level of flatness. 4) Comprehensive calculation of height uniformity index (HUI): HUI = w1*(1-RMSE / RMSE_max) + w2*(1-LHC / LHC_max) + w3*(1-GHS / GHS_max); where w1, w2, w3 are weight coefficients, and w1+w2+w3=1; RMSE_max, LHC_max, GHS_max are the maximum allowable values ​​of each indicator. The value range of the height uniformity index HUI is 0-1, and the larger the value, the more uniform the predicted casting effect. It is understandable that other statistical indicators or calculation methods can also be used to evaluate height uniformity, which is not limited here.

[0060] In some embodiments, the selection of the optimal parameter combination can be achieved in the following ways: first, the prediction results of all candidate parameter combinations are screened to eliminate combinations that do not meet the constraints: 1) Eliminate parameter combinations where the predicted height difference between any adjacent grid cells exceeds a third preset threshold; 2) Eliminate parameter combinations where the pumping flow or pressure exceeds the safe operating range of the equipment; 3) Eliminate parameter combinations that may cause local stress concentration, such as combinations where the flow differences between adjacent pump trucks are too large. Then, the remaining feasible parameter combinations are evaluated by multi-objective optimization: 1) Calculate the height uniformity objective function J1: J1 = α1*RMSE + α2*LHC + α3*GHS; α1, α2, α3 are weight coefficients; 2) Calculate the energy consumption objective function J2: J2 = Σ(β1*Q_i + β2*P_i); Q_i and P_i are the pumping flow and pressure of pump truck i, respectively, and β1 and β2 are weight coefficients; 3) Calculate the adjustment amplitude objective function J3: J3 = Σ|Q_i - Q_i_current| / Q_i_max; Q_i_current is the current flow, and Q_i_max is the maximum allowable flow; 4) Construct a comprehensive evaluation function: J = w1*J1 + w2*J2 + w3*J3; w1, w2, w3 are the weights of each objective. Finally, the parameter combination with the smallest value of the comprehensive evaluation function J is selected as the optimal solution. It is understandable that other methods may be used to select the optimal parameter combination, which is not limited here.

[0061] In the embodiment of the present application, due to the use of a diffusion prediction model based on a grid adjacency matrix and a multi-pump truck collaborative control mechanism, the concrete flow coefficient can be calculated in real time and the pumping parameters can be dynamically adjusted, which effectively solves the technical problems of difficulty in real-time grasp of the concrete flow trend and poor coordination of multi-pump truck casting in the construction of viaducts in mountainous areas, thereby achieving uniform distribution of concrete in large casting areas.

[0062] In the above embodiment, the intelligent monitoring system for concrete pouring quality can achieve uniform pouring through the diffusion prediction model and the coordinated control of multiple pump trucks. In practical applications, when executing the above monitoring method, due to the complex terrain of the mountainous area, the pump truck positioning signal is easily blocked and the spatial position of the fabric arm is difficult to accurately determine, which may lead to inaccurate measurement of the pump truck position and the fabric arm posture, thereby affecting the quality of concrete pouring. In the following embodiment, the fusion positioning solution of UWB base station and inertial navigation and the collaborative measurement mechanism of the multi-sensor fabric arm can be used to solve this technical problem, improve the positioning accuracy of the pump truck position and the measurement accuracy of the fabric arm posture.

[0063] See also Figure 3 , is another flow chart of the intelligent monitoring method for concrete pouring quality in an embodiment of the present application.

[0064] S301, obtain the real-time positioning data of the pump truck from the UWB base station and monitor the UWB signal quality. When it is detected that the signal quality has decreased due to mountain obstruction, switch the positioning scheme from UWB positioning to inertial navigation positioning, and record the last set of valid UWB positioning data before the switch as the initial reference position; Among them, UWB base station refers to the ultra-wideband positioning base station deployed at the commanding heights of the casting area, which is used to provide high-precision positioning signals; signal quality refers to the strength and stability indicators of the UWB positioning signal, which is used to evaluate the positioning accuracy; the initial reference position refers to the last reliable position data before switching the positioning scheme, which is used as the reference point for subsequent inertial navigation positioning.

[0065] In mountainous construction environments, UWB signals are easily blocked due to complex terrain, so a reliable positioning scheme switching mechanism needs to be established. Specifically, the pump truck is first located in real time through multiple UWB base stations, while monitoring signal quality indicators such as signal strength and multipath effect. When it is detected that the signal quality drops below the preset threshold due to mountain blockage, the system automatically switches the positioning scheme to inertial navigation mode, and uses the last set of high-quality UWB positioning data before the switch as the initial reference position for inertial navigation.

[0066] In some embodiments, intelligent switching of positioning schemes can be achieved in a variety of ways: optionally, a signal quality assessment algorithm can be used to calculate the weighted scores of indicators such as signal strength, carrier signal-to-noise ratio, and geometric precision factor, and trigger switching when the score is lower than a threshold, and use the Kalman filter algorithm to smooth the last several groups of UWB data to obtain a reliable initial reference position; optionally, based on a deep learning method, a neural network model can be trained to identify signal quality degradation patterns, predict possible signal obstructions in advance, and achieve seamless switching to inertial navigation mode, and use a Bayesian estimation method to fuse UWB data at multiple times to determine the optimal initial reference position. It is understandable that other methods can also be used to achieve the switching of positioning schemes and the determination of the initial reference position, which are not limited here.

[0067] S302, acquiring acceleration and angular velocity data of the inertial navigation unit, calculating the relative displacement relative to the initial reference position, and acquiring the inclination data of the fabric arm through the inclination sensor, and calculating the real-time attitude angle of the fabric arm; Among them, the inertial navigation unit refers to the inertial measurement device installed on the pump truck, which is used to obtain motion state data; the acceleration and angular velocity data refer to the physical quantities that describe the motion characteristics of the pump truck, which are used to calculate the displacement change; the relative displacement represents the change in spatial position relative to the initial reference position, which is used to update the position of the pump truck; the attitude angle refers to the directional angle of the fabric arm in space, which is used to determine the fabric direction.

[0068] After switching to the inertial navigation mode, it is necessary to continuously track the position of the pump truck through multi-sensor data fusion. Specifically, first obtain the three-axis acceleration and angular velocity data from the inertial navigation unit, and calculate the displacement change of the pump truck relative to the initial reference position through integration operation. At the same time, use the inclination sensor to measure the pitch angle and rotation angle of the fabric arm, and combine the kinematic model of the robot arm to calculate the real-time attitude angle of the fabric arm in space.

[0069] In some embodiments, accurate calculation of position and attitude can be achieved in a variety of ways: Optionally, a zero-speed correction algorithm can be used to eliminate the acceleration integral error by detecting the stationary state of the pump truck, and the quaternion method can be used to represent the attitude to avoid universal joint lock, and finally a high-precision position solution can be obtained by fusing multi-source data through an extended Kalman filter; Optionally, an end-to-end neural network model can be built based on a deep learning method to directly estimate the displacement and attitude from the raw sensor data, and the weights of different sensors can be adaptively adjusted through the attention mechanism to achieve robust state estimation. It is understandable that other methods can also be used to achieve the calculation of position and attitude, which are not limited here.

[0070] S303, determining the precise spatial position of the material distribution arm outlet based on the initial reference position, the relative displacement, the inclination data, and the vertical distance data from the material distribution arm outlet to the concrete surface measured by the laser rangefinder; Among them, the laser rangefinder refers to the distance measuring device installed at the end of the fabric arm, which is used to measure the height of the discharge port; the vertical distance data refers to the height value from the discharge port to the concrete surface, which is used to determine the location of the pouring point; the spatial position indicates the specific location of the fabric arm discharge port in the three-dimensional coordinate system, which is used to guide precise pouring.

[0071] After acquiring data from various sensors, the precise position of the discharging port of the discharging arm needs to be determined through multi-source data fusion. Specifically, the current position of the pump truck is first determined based on the initial reference position and the relative displacement calculated by inertial navigation, and then the spatial configuration of the robotic arm is calculated in combination with the inclination data of the discharging arm. Finally, the height coordinates of the discharging port are corrected through the vertical distance data measured by the laser rangefinder to achieve sub-meter positioning accuracy.

[0072] In some embodiments, the precise calculation of the position of the discharge port can be achieved in a variety of ways: Optionally, the forward kinematics model of the robot arm can be used to establish the coordinate transformation relationship of the cloth arm through the DH parameter method, the end position can be calculated by combining the real-time angle of each joint and the arm length, and the laser ranging data can be used for height compensation to obtain the spatial coordinates of the discharge port; Optionally, based on visual SLAM technology, a local three-dimensional map can be constructed through a binocular camera installed on the cloth arm, and the real-time tracking and positioning of the discharge port can be achieved by combining feature point matching and pose graph optimization algorithms. It is understandable that other methods can also be used to determine the position of the discharge port, which is not limited here.

[0073] S304, calculating the coverage of the distributing arm of each pump truck according to the spatial position, attitude angle, maximum extension length and rotation angle range of the distributing arm; Among them, the maximum extension length refers to the farthest distance that the fabric arm can extend, which is used to determine the working radius; the rotation angle range refers to the angle limit that the fabric arm can rotate, which is used to determine the scanning range; the coverage range refers to the set of all spatial points that the fabric arm can reach, which is used to plan the casting area.

[0074] After determining the position of the discharge port, it is necessary to calculate the effective working range of each pump truck. Specifically, first determine the working origin based on the spatial position of the fabric arm, then determine the current working direction based on the attitude angle, calculate the radial coverage range based on the maximum extension length of the fabric arm, and finally determine the circumferential coverage range based on the rotation angle limit, thereby obtaining a complete three-dimensional working space.

[0075] In some embodiments, the calculation of the coverage range can be achieved in a variety of ways: optionally, a geometric modeling method can be used to simplify the cloth arm into a multi-segment connecting rod structure, and a large number of configurations can be generated in the joint angle space through Monte Carlo sampling, mapped to the Cartesian space to obtain a discrete set of reachable points, and finally a continuous coverage area can be constructed through a convex hull algorithm; optionally, the movement of the cloth arm can be described as a parametric equation based on an analytical geometry method, and the analytical expression of the workspace can be obtained by solving the boundary equation under the constraint conditions to achieve accurate calculation of the coverage range. It is understandable that other methods can also be used to calculate the coverage range, which are not limited here.

[0076] S305, dividing the casting area into a plurality of grid units, and establishing a grid adjacency matrix based on the spatial position relationship of each grid unit; Before step S304 is completed, step S305 may be performed simultaneously. Step S305 is similar to step S201 and will not be described in detail here.

[0077] S306, based on the grid adjacency matrix, establishing a diffusion prediction model for predicting the flow characteristics of concrete between grid units, the diffusion prediction model taking the current performance parameters of concrete, the current pouring environment parameters and the real-time height difference between adjacent grid units as input, calculating the flow coefficient of each grid unit in the grid adjacency matrix, and determining the predicted height difference between adjacent grid units after a preset time period based on the flow coefficient as output; Similar to step S202, details will not be repeated here.

[0078] S307, calculating the real-time height difference between each grid unit based on the concrete surface elevation data collected by the laser rangefinder; Among them, elevation data represents the three-dimensional terrain information of the concrete surface, which is used to characterize the undulating state of the casting surface; the real-time height difference refers to the height difference between adjacent grid units, which is used to evaluate the uniformity of casting; the grid unit represents the basic calculation unit that divides the casting area, which is used to achieve refined control.

[0079] After executing step S305 to divide the grid units, the height difference between each grid unit can be calculated. Specifically, firstly, the laser rangefinder on the fabric arm performs scanning measurement within the working range to obtain discrete elevation point cloud data of the concrete surface. Then, the point cloud data is mapped to the pre-divided grid units through the spatial interpolation algorithm, and the average elevation of each grid is calculated. Finally, the height difference between adjacent grid units is calculated to guide the subsequent pumping control.

[0080] In some embodiments, the calculation of height difference can be achieved in a variety of ways: Optionally, the Kriging interpolation algorithm can be used to build a variation function model based on the spatial correlation of the measurement points, and the elevation value of the grid node can be obtained by optimal linear unbiased estimation, and finally the height gradient of the adjacent grids can be calculated by finite difference; Optionally, based on the deep learning method, a point cloud processing network can be used to directly reconstruct a continuous surface model from the original ranging data, and the local features can be extracted by the graph convolution network to calculate the height difference. It is understandable that other methods can also be used to calculate the height difference, which are not limited here.

[0081] S308, determining the main pouring area of ​​each pump truck and its corresponding initial pumping flow rate according to the distance relationship between the area where the real-time height difference is greater than the first preset threshold and each pump truck; S309: For the overlapping pouring area of ​​adjacent pump trucks, when the real-time height difference is greater than a second preset threshold, the pump truck located at the lower side of the overlapping pouring area is switched to the leading pump truck, and the pumping flow of the leading pump truck is increased according to a preset flow adjustment step length until the real-time height difference is less than the second preset threshold; S310, based on the real-time height difference between each grid unit, updating the flow coefficient in the diffusion prediction model in real time, and obtaining the predicted height difference between adjacent grid units after a preset time period; S311, based on the predicted height difference output by the diffusion prediction model, calculate the height uniformity index corresponding to different pumping parameter combinations, select the pumping parameter combination that can make the height difference between adjacent grid units less than the third preset threshold after the preset time, and adjust the pumping flow of each pump truck in real time until the actual height difference and the predicted height difference between each grid unit are both less than the third preset threshold; Steps S308-S311 are similar to steps S203-S206 and will not be described again here.

[0082] S312, collecting concrete vibration response characteristics based on the vibration sensor array arranged at each grid unit; Among them, the vibration sensor array refers to multiple vibration sensors regularly arranged in the pouring area, which are used to collect vibration signals; the vibration response characteristics refer to the vibration waveform characteristics generated by concrete under the action of external force, which are used to characterize the flow state of concrete; the response characteristics include parameters such as amplitude, frequency and phase, which are used to evaluate the density of concrete.

[0083] When the concrete pouring is completed or during the execution of step S311, the pouring quality can be evaluated. Specifically, a vibration sensor array is first arranged at the characteristic position of the grid unit, and the vibration response signal of the concrete is synchronously collected at a high sampling rate. Then, the collected original signal is preprocessed, including filtering and noise reduction, baseline correction and other operations, and time domain and frequency domain characteristic parameters are extracted. Finally, these characteristic parameters are used as input data for concrete density evaluation.

[0084] In some embodiments, the collection and processing of vibration response features can be achieved in a variety of ways: optionally, a distributed data acquisition system can be used to achieve multi-point synchronous sampling through a wireless sensor network, and signal noise reduction and feature extraction can be performed using wavelet transform, and finally the feature dimension can be reduced through principal component analysis to improve computing efficiency; optionally, based on the edge computing architecture, signal preprocessing and feature extraction can be completed at the sensor node end, and automatic feature learning can be achieved through a lightweight neural network, and finally the extracted features can be transmitted to the central processing unit through the network. It is understandable that other methods can also be used to achieve the acquisition of vibration response features, which are not limited here.

[0085] S313, dynamically calculating the density of each grid unit based on the spectrum analysis of the concrete vibration response characteristics; Among them, spectrum analysis refers to the mathematical method of frequency domain decomposition of vibration signals, which is used to extract frequency characteristics; density refers to the degree of filling of internal voids in concrete, which is used to evaluate the pouring quality; dynamic calculation refers to the real-time updated calculation process, which is used to achieve online monitoring.

[0086] After obtaining the vibration response characteristics, the density of concrete needs to be evaluated through signal processing methods. Specifically, firstly, the vibration signal is subjected to fast Fourier transform to obtain the spectrum distribution of the signal. Then the energy distribution characteristics of the spectrum are analyzed, including parameters such as main frequency, energy ratio, and harmonic ratio. Finally, based on these spectrum characteristic parameters, the density value of concrete is calculated through the pre-established mapping relationship. For example, the vibration signal can be first subjected to a fast Fourier transform to obtain a spectrum, and then the following characteristic parameters can be extracted: Main frequency feature: The main frequency f0 is determined by finding the frequency point corresponding to the maximum energy in the spectrum, and the main frequency energy ratio η0=E0 / Et is calculated, where E0 is the main frequency energy and Et is the total energy; Energy ratio feature: The spectrum is divided into a low frequency band (0-f1), a medium frequency band (f1-f2) and a high frequency band (above f2), and the energy ratio of each frequency band is calculated ηL=EL / Et, ηM=EM / Et, ηH=EH / Et; Harmonic ratio feature: The energy En at the nth-order harmonic frequency point fn=n*f0 of the main frequency is extracted, and the harmonic ratio γn=En / E0 is calculated; Band energy feature: The normalized energy spectral density is calculated in each characteristic frequency band [fa, fb]: , f∈

fa,fb

[0087] Based on the above characteristic parameters, a density calculation model is established: ρ=w0*η0+wL*ηL+wM*ηM+wH*ηH+Σ(wn*γn)+wp*P(f), Among them, w0, wL, wM, wH, wn, and wp are weight coefficients of each characteristic parameter, which are obtained through experimental calibration. The model comprehensively considers the frequency domain characteristics of the vibration signal and can accurately reflect the density of concrete.

[0088] In some embodiments, dynamic calculation of density can be achieved in a variety of ways: Optionally, a time-frequency analysis method can be used to obtain the time-varying spectrum of the signal through short-time Fourier transform, extract the energy change of the characteristic frequency band by combining empirical mode decomposition, and finally establish a mapping relationship from spectrum features to density through support vector regression; Optionally, based on a deep learning method, a one-dimensional convolutional neural network can be used to directly learn the density features from the original vibration signal, and a recurrent neural network can be used to capture the timing dependency to achieve end-to-end prediction of density. It is understandable that other methods can also be used to achieve the calculation of density, which is not limited here.

[0089] S314: When it is detected that the density of a local area is lower than a preset density threshold, based on the area size of the low-density area, the density deviation value and the coverage of surrounding pump trucks, the pumping flow of the relevant pump trucks is adaptively adjusted until the density meets the standard.

[0090] Among them, the preset density threshold represents the standard value for judging whether the density is qualified, which is used to trigger the control operation; the density deviation value refers to the difference between the actual density and the target value, which is used to determine the supplementary amount; the adaptive adjustment represents the process of automatic adjustment according to the real-time status, which is used to realize intelligent control.

[0091] When it is found that the density of a local area is insufficient, remedial measures need to be taken to improve the pouring quality. Specifically, first determine the scope and degree of the low-density area and calculate the volume of concrete that needs to be supplemented. Then analyze the coverage of pump trucks around the area and select the most suitable pump truck for supplementary pouring. Finally, increase the pumping flow rate through gradual adjustment, while monitoring the trend of density changes to ensure that new problems will not be caused by excessive supplementation.

[0092] In some embodiments, adaptive regulation of pumping flow can be achieved in a variety of ways: optionally, a fuzzy control algorithm can be used, density deviation and change rate can be used as input variables, and the flow adjustment amount can be calculated by fuzzy rule reasoning to achieve smooth flow control; optionally, based on the reinforcement learning method, the pumping flow regulation can be modeled as a Markov decision process, and the regulation strategy can be optimized by a deep learning algorithm to achieve intelligent flow control. It is understandable that other methods can also be used to achieve pumping flow regulation, which is not limited here.

[0093] Preferably, in some embodiments, the overlap between the coverage range of each pump truck's distribution arm and the low-density area can be calculated, and the pump truck with the largest overlap can be selected as the target pump truck; the required supplementary concrete volume can be determined based on the area size and density deviation value of the low-density area in combination with the concrete performance parameters; the target pumping flow rate can be calculated based on the supplementary concrete volume and the preset maximum pumping time; the pumping flow rate of the target pump truck can be gradually increased from the current value to the target flow rate by gradually increasing the pumping flow rate, with the incremental step not exceeding the preset maximum flow rate change rate, until the density reaches the preset density threshold; when it is detected that the density growth rate exceeds the preset growth rate threshold, the pumping flow rate is reduced to 80% of the current flow value to reduce local stress concentration.

[0094] In the embodiment of the present application, due to the adoption of the fusion positioning scheme of UWB base station and inertial navigation and the collaborative measurement mechanism of multiple sensors of the fabric arm, it is possible to automatically switch to inertial navigation positioning when the UWB signal quality decreases due to mountain obstruction, effectively solving the technical problem that the pump truck positioning signal is easily obstructed and the spatial position of the fabric arm is difficult to accurately determine in mountainous environments, thereby achieving continuous and reliable positioning of the pump truck position and accurate spatial positioning of the fabric arm outlet. And due to the adoption of a step-by-step adjustment mechanism for pumping flow based on density and a stress concentration prevention strategy, it is possible to dynamically evaluate the density and intelligently control the pumping flow according to the vibration response characteristics, effectively solving the technical problem that the local area density is insufficient and stress concentration is easily generated during the concrete pouring process, thereby achieving accurate regulation of the concrete density.

[0095] The following introduces the control server in the intelligent monitoring system for concrete pouring quality provided in the embodiment of the present application. Figure 4 It is a schematic diagram of an exemplary hardware structure of the control server provided in an embodiment of the present application.

[0096] In some embodiments, the control server 100 includes a processor, a memory and a network interface connected by a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data. The network interface of the computer device is used to communicate with other external terminals or servers through a network connection. In some embodiments, the network interface can be a wired network interface, and in some embodiments, the network interface can also be a wireless network interface. When the computer program is executed by the processor, the method in the embodiment of the present application is implemented.

[0097] Those skilled in the art will understand that Figure 4 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0098] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application.

[0099] As used in the above embodiments, the term "when..." may be interpreted to mean "if..." or "after..." or "in response to determining..." or "in response to detecting...", depending on the context. Similarly, the phrases "upon determining..." or "if (the stated condition or event) is detected" may be interpreted to mean "if determining..." or "in response to determining..." or "upon detecting (the stated condition or event)" or "in response to detecting (the stated condition or event)", depending on the context.

[0100] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium may be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more available media integrated. The available medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state hard disk), etc.

Claims

1. A method for intelligent monitoring of concrete pouring quality, characterized in that: include: The pouring area is divided into a plurality of grid units, and a grid adjacency matrix is ​​established based on the spatial position relationship of each grid unit; Based on the grid adjacency matrix, a diffusion prediction model for predicting the flow characteristics of concrete between grid units is established, wherein the diffusion prediction model takes the current performance parameters of concrete, the current pouring environment parameters and the real-time height difference between adjacent grid units as input, calculates the flow coefficient of each grid unit in the grid adjacency matrix, and determines the predicted height difference between adjacent grid units after a preset time period based on the flow coefficient as output; Obtain the coverage of the distributing arm of each pump truck and the real-time height difference between each grid unit, and determine the main pouring area of ​​each pump truck and its corresponding initial pumping flow rate according to the distance relationship between the area where the real-time height difference is greater than the first preset threshold and each pump truck; For the overlapping pouring area of ​​adjacent pump trucks, when the real-time height difference is greater than a second preset threshold, the pump truck located at the lower side of the overlapping pouring area is switched to the leading pump truck, and the pumping flow of the leading pump truck is increased according to a preset flow adjustment step length until the real-time height difference is less than the second preset threshold; Based on the real-time height difference between each grid unit, the flow coefficient in the diffusion prediction model is updated in real time to obtain the predicted height difference between adjacent grid units after a preset time period; Based on the predicted height difference output by the diffusion prediction model, the height uniformity index corresponding to different pumping parameter combinations is calculated, and the pumping parameter combination that can make the height difference between adjacent grid units less than the third preset threshold after the preset time period is selected to adjust the pumping flow of each pump truck in real time until the actual height difference and the predicted height difference between each grid unit are both less than the third preset threshold.

2. The method according to claim 1, characterized in that The establishment of a diffusion prediction model for predicting concrete flow characteristics specifically includes: Acquire the concrete slump, initial setting time, external temperature and concrete temperature as concrete performance parameters, and acquire the temperature, humidity and wind speed of the construction environment as the casting environment parameters; Constructing a directed graph model based on the grid adjacency matrix, wherein the nodes of the graph represent grid units, and the weights of the edges are determined by the height difference, distance and the concrete performance parameters of adjacent grid units; The flow coefficient D(i, j, t) between adjacent grid cells i and j at time t is calculated according to the following formula: ； in, is the basic flow coefficient, i and j represent the numbers of two adjacent grid cells, S is the slump, is the initial setting time, is the concrete temperature, is the ambient temperature, H is the humidity, W is the wind speed, t is the time from the start of pouring, , , , , , and λ are the corresponding weight coefficients; Substituting the flow coefficient into the diffusion equation, the diffusion prediction model is established: ； Among them, K is the global correction coefficient, A(i, j) is the grid adjacency matrix element, and ΔZ(i, j, t) is the height gradient at the current moment.

3. The method according to claim 1, characterized in that UWB base stations are deployed at multiple commanding heights in the pouring area, and the pump truck is equipped with an inertial navigation unit, a tilt sensor, and a laser rangefinder; the acquisition of the coverage range of each pump truck's placing arm and the real-time height difference between each grid unit specifically includes: Obtain the real-time positioning data of the pump truck from the UWB base station and monitor the UWB signal quality. When it is detected that the signal quality has decreased due to mountain obstruction, switch the positioning scheme from UWB positioning to inertial navigation positioning, and record the last set of valid UWB positioning data before the switch as the initial reference position; Acquire the acceleration and angular velocity data of the inertial navigation unit, calculate the relative displacement relative to the initial reference position, and obtain the inclination data of the fabric arm through the inclination sensor to calculate the real-time attitude angle of the fabric arm; Determine the precise spatial position of the material discharging port of the material discharging arm based on the initial reference position, the relative displacement, the inclination data, and the vertical distance data from the material discharging port of the material discharging arm to the concrete surface measured by the laser rangefinder; Calculate the coverage of each pump truck’s dispensing arm based on its spatial position, attitude angle, maximum extension length and rotation angle range; Based on the concrete surface elevation data collected by the laser rangefinder, the real-time height difference between each grid unit is calculated.

4. The method according to claim 1, characterized in that: The method of updating the flow coefficient in the diffusion prediction model in real time based on the real-time height difference between each grid unit to obtain the predicted height difference between adjacent grid units after a preset time period specifically includes: Obtain the factory performance parameters of each batch of concrete; Based on the tank truck GPS positioning and unloading records, the correspondence between grid units and concrete batches is established, and the performance parameters of each batch of concrete are mapped to the corresponding grid units; Extract reinforcement layout information from the BIM model and calculate the reinforcement volume ratio and main reinforcement direction of each grid unit; Determining the local concrete flow direction based on the real-time height difference between the grid cells; When it is detected that the performance parameter difference of adjacent grid cells exceeds the preset performance difference threshold and the difference between the real-time height difference and the predicted height difference of the diffusion prediction model before the preset time exceeds the prediction deviation threshold, the flow coefficient of the diffusion prediction model is corrected to: ； Where: Φ(ρ, θ) is the reinforcement correction function, ρ is the reinforcement volume ratio, θ is the angle between the local concrete flow direction and the main reinforcement direction; The corrected flow coefficient is substituted into the diffusion prediction model, and the concrete performance parameters, pouring environment parameters and real-time height difference between adjacent grid units at the current moment are input to obtain the predicted height difference between adjacent grid units after a preset time.

5. The method according to claim 4, characterized in that The steel bar correction function Φ(ρ,θ) is specifically: ； Where: ρ is the steel volume ratio, which indicates the volume ratio of steel bars in a unit volume of concrete; θ is the angle between the local concrete flow direction and the main reinforcement direction; γ is the directional influence coefficient, 0≤γ≤1; (1-ρ) indicates the reduction effect of the steel volume on the effective flow space; It represents the influence of the angle between the flow direction and the main reinforcement direction. When the flow direction is parallel to the main reinforcement, the influence is the smallest, at this time θ=0°, and when it is perpendicular, the influence is the largest, at this time θ=90°.

6. The method according to claim 1, characterized in that The method further comprises: Based on the vibration sensor array arranged at each grid unit, the vibration response characteristics of concrete are collected; Dynamically calculating the density of each grid unit based on the frequency spectrum analysis of the concrete vibration response characteristics; When it is detected that the density of a local area is lower than the preset density threshold, the pumping flow of the relevant pump trucks is adaptively adjusted until the density meets the standard based on the area size of the low-density area, the density deviation value and the coverage of surrounding pump trucks.

7. The method according to claim 6, characterized in that Based on the area size of the low-density area, the density deviation value and the coverage of surrounding pump trucks, the pumping flow rate of the relevant pump trucks is adaptively adjusted until the density reaches the standard, specifically including: Calculate the overlap between the coverage of each pump truck's distributing arm and the low-density area, and select the pump truck with the largest overlap as the target pump truck; Determine the required volume of additional concrete based on the area size and density deviation value of the low-density area and the concrete performance parameters; Calculate the target pumping flow rate based on the added concrete volume and the preset maximum pumping time; The pumping flow rate of the target pump truck is gradually increased from the current value to the target flow rate by gradually increasing the pumping flow rate, and the incremental step length does not exceed the preset maximum flow rate change rate, until the density reaches the preset density threshold value; When it is detected that the density growth rate exceeds the preset growth rate threshold, the pumping flow rate is reduced to 80% of the current flow rate value to reduce local stress concentration.

8. An intelligent monitoring system for concrete pouring quality, characterized in that: The intelligent monitoring system for concrete pouring quality includes: one or more processors and a memory; the memory is coupled to the one or more processors, the memory is used to store computer program code, the computer program code includes computer instructions, and the one or more processors call the computer instructions to enable the system to execute the method described in any one of claims 1-7.

9. A computer program product comprising instructions, characterized in that When the computer program product runs on a concrete pouring quality intelligent monitoring system, the concrete pouring quality intelligent monitoring system is enabled to execute the method according to any one of claims 1 to 7.

10. A computer-readable storage medium comprising instructions, characterized in that: When the instruction is executed on the intelligent monitoring system for concrete pouring quality, the intelligent monitoring system for concrete pouring quality executes the method according to any one of claims 1 to 7.

Images