Homogenizing and mixing preparation method and device for flow-state filling material

Through the three-dimensional model division and parameter optimization of fluid filling materials, the problem of not being able to comprehensively consider load and spatial characteristics in the existing technology is solved, and the uniform distribution of fluid filling materials and the satisfaction of engineering requirements is achieved, and the engineering quality is improved and economic benefits.

CN119017536AInactive Publication Date: 2024-11-26JIANGSU YUNJIN PAVEMENT ENGINEERING CO LTD
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202411229637.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-09-03
Publication Date
2024-11-26
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The existing fluid filling preparation technology cannot comprehensively consider the load, spatial characteristics and material construction processing parameters of the backfill area, making it difficult for the filling material to meet the fluidity, uniform density and engineering requirements at the same time.

Method used

By obtaining the three-dimensional model of the backfill area, dividing it into large and small areas models, optimizing the screening and homogenizing mixing parameters, predicting the filling rate and load satisfaction, using the optimal homogenizing mixing parameters to mix material building materials, and achieving fluid filling Even distribution of material construction.

Benefits of technology

The fluidity and compactness of fluid filling materials in the backfill area are achieved, which meets the load and space needs of different regions, and improves the quality, stability and economic benefits of the project.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119017536A_ABST
    Figure CN119017536A_ABST
Patent Text Reader

Abstract

The invention discloses a homogenizing and mixing preparation method and device for a flow state filling material, and relates to the technical field of flow state filling materials, and the method comprises the following steps: obtaining a backfill area three-dimensional model, dividing the backfill area three-dimensional model into a large area model and a plurality of small area models, and analyzing the large area load requirement. And building raw materials are collected, screening parameters are optimized according to the small-area model, and the building raw materials with the optimal particle size are obtained through processing. And the load satisfaction degree is analyzed by combining the small-area model prediction filling rate and the large-area load requirement according to the large-area model and the small-area model optimization homogenization mixing parameters. And mixing the building raw materials by adopting the optimal homogenization mixing parameters to obtain the flow-state filling material. The technical problem that in the prior art, when the flow state filling material is prepared, the load, spatial characteristics and building material processing parameters of a backfill area cannot be comprehensively considered, and consequently the filling material cannot meet the requirements for fluidity, compactness uniformity and engineering at the same time is solved, and the technical effect of improving the engineering quality, stability and economic benefits is achieved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the technical field of fluidized filling materials, and in particular to a homogenizing, mixing and preparing method and device for fluidized filling materials. Background Art

[0002] In the scenario of fluid filling material preparation, ensuring the performance of the filling material and meeting the requirements of the backfill area is particularly prominent. Optimizing the preparation method to achieve better results has become a crucial link in the application of fluid filling materials. Traditional fluid filling material preparation is often extensive and lacks specificity. It only relies on experience or simple process parameters, and does not adequately analyze the load and spatial characteristics of the backfill area. It lacks in-depth research and optimization of the characteristics of the building material raw materials, and it is difficult to accurately determine the screening and homogenization mixing parameters. In the preparation process, there are unreasonable situations, resulting in uneven fluidity and density of the filling material, which cannot meet the engineering needs well.

[0003] In the current relevant technologies, there is a technical problem that the load, spatial characteristics and material processing parameters of the backfill area cannot be comprehensively considered when preparing fluidized filling materials, resulting in the filling materials being difficult to simultaneously meet fluidity, uniform density and engineering requirements. Summary of the invention

[0004] The present application provides a homogenized mixing preparation method and device for fluidized filling materials, which obtains a three-dimensional model of the backfill area, divides it into a large area and multiple small area models, and analyzes the load requirements of the large area. The building materials are collected, and the screening parameters are optimized according to the small area model, and the building materials with the optimal particle size are processed. The homogenized mixing parameters are optimized according to the large and small area models, and the load satisfaction is analyzed in combination with the predicted filling rate of the small area model and the load requirements of the large area. The building materials are mixed with the optimal homogenized mixing parameters to obtain fluidized filling materials, which realizes the uniform distribution of fluidity and density of the fluidized filling materials in the backfill area, meets the load and space requirements of different areas, and achieves the technical effect of improving the quality, stability and economic benefits of the project.

[0005] The present application provides a homogenizing and mixing preparation method of a fluidized filling material, comprising:

[0006] Acquire a regional three-dimensional model of the backfill area to be backfilled; divide the regional three-dimensional model to obtain a large area model and multiple small area models, and analyze to obtain the large area load requirement information; collect building materials to be homogenized to prepare fluidized filling materials, optimize the screening parameters of the building materials according to the multiple small area models, obtain the optimal screening parameters, screen the building materials to obtain screened building materials with optimal particle size; optimize the homogenization mixing parameters of the screened building materials combined with the mixed fillers according to the large area model and the multiple small area models, and obtain the optimal homogenization mixing parameters, wherein the filling rate of the small area model is predicted according to the homogenization mixing parameters combined with the multiple small area models, and the load satisfaction is analyzed and optimized in combination with the large area load requirement information; use the optimal homogenization mixing parameters to mix the building materials with fillers to obtain the prepared fluidized filling materials.

[0007] The present application also provides a homogenizing and mixing preparation device for fluidized filling materials, comprising:

[0008] A regional three-dimensional model acquisition module, the regional three-dimensional model acquisition module is used to acquire a regional three-dimensional model of a backfill area to be backfilled; a large-area load requirement information acquisition module, the large-area load requirement information acquisition module is used to divide the regional three-dimensional model, obtain a large-area model and multiple small-area models, and analyze to obtain the large-area load requirement information; an optimal screening parameter acquisition module, the optimal screening parameter acquisition module is used to collect building materials to be homogenized to prepare fluidized filling materials, optimize the screening parameters of the building materials according to the multiple small-area models, obtain the optimal screening parameters, screen the building materials, and obtain the optimal Screened building material with optimal particle size; an optimal homogenized mixing parameter acquisition module, which is used to optimize the homogenized mixing parameters of the screened building material combined with the mixed filler according to the large area model and the multiple small area models, and obtain the optimal homogenized mixing parameters, wherein the filling rate of the small area model is predicted according to the homogenized mixing parameters in combination with the multiple small area models, and the load satisfaction is analyzed and optimized in combination with the large area load requirement information; a fluidized filling material preparation module, which is used to adopt the optimal homogenized mixing parameters to mix the building material with fillers to obtain prepared fluidized filling materials.

[0009] The homogenized mixing preparation method and device of fluidized filling materials proposed in this application are to first obtain a three-dimensional model of the backfill area, divide it into a large area and multiple small area models, and analyze the load requirements of the large area. Collect building materials, optimize screening parameters according to the small area model, and process to obtain building materials with optimal particle size. Optimize homogenized mixing parameters based on large and small area models, and analyze load satisfaction in combination with the predicted filling rate of the small area model and the load requirements of the large area. Use the optimal homogenized mixing parameters to mix the building materials to obtain fluidized filling materials, so as to achieve uniform distribution of fluidity and density of the fluidized filling materials in the backfill area, meet the load and space requirements of different areas, and achieve the technical effect of improving project quality, stability and economic benefits. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the technical solution of the embodiment of the present invention, the accompanying drawings of the embodiment of the present invention will be briefly introduced below. A flow chart is used in the present application to illustrate the operations performed by the device according to the embodiment of the present application. It should be understood that the previous or following operations are not necessarily performed precisely in order. On the contrary, various steps can be processed in reverse order or simultaneously as needed. At the same time, other operations can also be added to these processes, or one or more operations can be removed from these processes.

[0011] Figure 1 A schematic flow chart of a homogenizing and mixing preparation method of a fluid filling material provided in an embodiment of the present application;

[0012] Figure 2 A schematic diagram of the structure of a homogenizing, mixing and preparing device for fluidized filling material provided in an embodiment of the present application.

[0013] Explanation of reference numerals: regional three-dimensional model acquisition module 10, large-area load requirement information acquisition module 20, optimal screening parameter acquisition module 30, optimal homogenization mixing parameter acquisition module 40, fluidized filling material preparation module 50. DETAILED DESCRIPTION

[0014] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below.

[0015] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.

[0016] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments, but it is understood that "some embodiments" may be the same subset or different subsets of all possible embodiments, and may be combined with each other without conflict, and the terms "first\second" involved are merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "including" and "having" and any variations are intended to cover non-exclusive inclusions, for example, a process, method, device, product, or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those generally understood by technicians in the technical field of this application. The terms used herein are for the purpose of describing the embodiments of the present application only.

[0017] The present application provides a method for preparing a fluidized filling material by homogenizing and mixing. Figure 1 As shown, the method includes:

[0018] Step S100, obtaining a regional three-dimensional model of a backfill area to be backfilled. Specifically, when the road construction is completed and the foundation pit needs to be backfilled, high-precision 3D scanning equipment is prepared, including laser scanners, photogrammetry equipment, etc. The scanning equipment is placed in a suitable position to ensure that the internal space of the foundation pit can be fully covered, and the foundation pit is scanned from different angles and positions to obtain the most complete and accurate surface data. During the scanning process, the equipment will emit a laser beam or capture an image, and record the reflected signal or image information. The information contains data such as the distance, position and shape of each point inside the foundation pit. After the scanning is completed, the large amount of raw data obtained is transferred to the computer, and the 3D modeling software is used to process and analyze the data. The data is subjected to point cloud splicing, denoising and optimization operations through algorithms to remove possible errors and interference. The initial 3D model of the foundation pit is constructed based on the processed data. The model preliminarily presents the approximate shape and structure of the foundation pit. By supplementing the missing parts, smoothing the surface, correcting the inaccurate areas and other operations, the model can more accurately reflect the actual situation of the foundation pit, and obtain a complete and accurate regional 3D model of the foundation pit after road construction, which provides basic data and visual reference for subsequent backfilling work.

[0019] In a possible implementation, a regional three-dimensional model of a backfill area to be backfilled is obtained, and step S100 further includes: step S110, scanning the backfill area to be backfilled to obtain backfill scanning data. Specifically, prepare professional scanning equipment, such as a ground laser scanner or a scanning equipment carried by a drone, and place the scanning equipment around the backfill area to ensure that the entire range to be scanned can be covered. Start the scanning equipment, and the equipment will emit a laser beam or other detection signal to measure the surface of the backfill area point by point. During the scanning process, the equipment will record the spatial coordinates, reflection intensity and other information of each measurement point. As the equipment moves and rotates, a large amount of point data is continuously collected, thereby forming rich backfill scanning data.

[0020] Step S120, constructing a regional three-dimensional model of the backfill area based on the backfill scan data. Specifically, the backfill scan data obtained is imported into a special three-dimensional modeling software, and the software processes and analyzes the data. First, the data is preprocessed, including removing noise points, erroneous data, and duplicate data, etc. The data obtained by scanning at different positions and angles are unified into a coordinate system through point cloud registration technology to form a complete point cloud set. A preliminary three-dimensional surface model is generated based on the point cloud data using a surface reconstruction algorithm, and the preliminary model is optimized and refined, such as increasing the grid density, smoothing the surface, repairing holes, etc., to improve the accuracy and quality of the model, and adding attributes such as material and color to the model to make it more realistic and intuitive, thereby constructing a complete and accurate regional three-dimensional model of the backfill area.

[0021] Step S200, dividing the regional three-dimensional model to obtain a large regional model and a plurality of small regional models, and analyzing to obtain the large regional load requirement information. Specifically, a complete regional three-dimensional model is obtained through the previous scanning and modeling steps, such as a cube or cylindrical foundation pit model, and the model division begins. For parts with regular main shapes and occupying a large space, such as the main body of the foundation pit in the shape of a cube or a cylinder, it is defined as a large regional model. For parts such as irregular depressions caused by excavation on the side walls of the foundation pit, they are identified and separated through specific algorithms and judgment criteria. The irregular parts are divided into multiple small regional models. In the division process, methods such as geometric shape analysis and spatial position judgment are used. For example, whether the depression belongs to a small regional model is determined based on its area, depth, shape complexity and other characteristics. After the division is completed, the load requirement information of the large regional model is analyzed, involving mechanical calculation and structural analysis. By considering the future use scenarios of the foundation pit, the possible weight of the upper structure, the impact of the surrounding environment and other factors, professional engineering mechanics theories and calculation formulas are used to determine the load size and distribution that the large regional model needs to bear, and the load changes under different working conditions are evaluated to ensure that comprehensive and accurate large regional load requirement information is obtained, which provides an important basis for subsequent filling material preparation and filling work.

[0022] In a possible implementation, the regional three-dimensional model is divided to obtain a large regional model and a plurality of small regional models, and the large regional load requirement information is obtained by analysis. Step S200 further includes: Step S210, collecting a set of sample regional three-dimensional models, dividing the large regional model and the small regional model in each sample regional three-dimensional model, and obtaining a set of sample regional model division results. Specifically, a large number of sample regional three-dimensional models are collected from a plurality of actual engineering cases, and the models cover backfill areas of various types and shapes. For each sample regional three-dimensional model, a professional engineer or a specific algorithm is used to manually or automatically divide the large regional model (such as a regular main part) and the small regional model (such as an irregular concave or protruding part) in the model according to preset rules and standards, and the division result of each sample model is recorded to form a set of sample regional model division results.

[0023] Step S220, using the sample regional 3D model set and the sample regional model division result set, training a regional model divider, dividing the regional model, and obtaining a large regional model and a plurality of small regional models. Specifically, the collected sample regional 3D model set and the corresponding division result set are input into a machine learning algorithm, such as a neural network. The algorithm trains a regional model divider that can automatically divide a new regional 3D model by learning the features and rules in the sample data. When a new regional model to be divided is input, the regional model divider can quickly and accurately divide it into a large regional model and a plurality of small regional models.

[0024] Step S230, backfill load analysis is performed on the large-area model to obtain large-area load requirement information. Specifically, after obtaining the large-area model, relevant theories and methods of structural mechanics and soil mechanics are used for analysis, taking into account factors such as the geological conditions where the large-area model is located, the weight and distribution of the superstructure, and the properties of the backfill material, and by establishing a mathematical model and performing numerical calculations, the load size, direction, and distribution of the large-area model under different working conditions are predicted, and the results of the comprehensive analysis are used to obtain large-area load requirement information, which provides a basis for subsequent material selection and construction plan formulation.

[0025] In a possible implementation, backfill load analysis is performed on the large-area model to obtain large-area load requirement information, and step S230 further includes: step S231, obtaining a sample large-area model set in the same backfill project, and obtaining a sample large-area load requirement information set according to the load requirement after backfilling. Specifically, large-area models are collected from multiple completed backfill engineering projects of the same type, and the models have similar engineering backgrounds and conditions. According to the actual load requirements of each backfill project after backfilling, corresponding load requirement data are obtained, including but not limited to the maximum pressure that can be tolerated, the allowable deformation, etc. The large-area models and the corresponding load requirement information are sorted and classified to form a sample large-area model set and a sample large-area load requirement information set.

[0026] Step S232, using the sample large area model set and the sample large area load requirement information set to train the backfill load analyzer. Specifically, the acquired sample large area model set is used as input data, and the corresponding sample large area load requirement information set is used as expected output data, and is input into a specific machine learning algorithm or analysis model, such as a neural network model based on deep learning or a traditional statistical regression model. By repeatedly adjusting the parameters of the model, the model can learn the inherent relationship and law between the characteristics of the sample large area model and the load requirement information, thereby completing the training of the backfill load analyzer.

[0027] Step S233, using the backfill load analyzer, perform backfill load analysis on the large area model to obtain large area load requirement information. Specifically, the large area model to be analyzed is input into the trained backfill load analyzer, and the analyzer will analyze and calculate the input large area model according to the previously learned patterns and rules, predict the load of the large area model after backfilling, and finally output the large area load requirement information, including the specific load value and distribution that may be borne, and the impact assessment on the structural stability, etc., to provide an important reference basis for subsequent engineering design and construction.

[0028] Step S300, collecting building materials to be homogenized to prepare fluidized filling materials, optimizing screening parameters of the building materials according to the multiple small area models to obtain optimal screening parameters, screening the building materials to obtain screened building materials with optimal particle size. Specifically, raw materials for homogenizing and preparing fluid filling materials, such as sand and gravel, are collected from designated raw material collection sites. During the collection process, the quality of the raw materials is strictly controlled to ensure that their physical and chemical properties meet the requirements of the project, while ensuring that the collected quantity can meet the needs of the project. Detailed information on multiple small area models that have been divided, including the shape, size, and spatial position of each small area, is obtained. The screening parameters of the raw materials are optimized according to the specific conditions of the small area model. Taking sand and gravel as an example, assuming that the space of the small area model is relatively narrow, in order to allow sand and gravel to smoothly enter any small area model, preliminary screening parameters are set, such as using a diameter of less than 25 mm as the screening standard. This parameter is continuously adjusted and verified through a combination of mathematical simulation and actual experiments. In the simulation process, a distribution model of sand and gravel particles after passing through different screening parameters is constructed using computer software to analyze its filling effect in the small area model. In the actual experiment, a certain amount of sand is screened according to the set screening parameters. The stone is screened and the screened sand and gravel are tried to be filled into the small area model to observe the compactness and uniformity of the filling. Through multiple simulations and experiments, the filling effects under different screening parameters are comprehensively evaluated, and the parameters are further optimized and adjusted until the screening parameters that can make the sand and gravel achieve the best filling effect in all small area models are determined, that is, the optimal screening parameters. Once the optimal screening parameters are obtained, for example, the diameter is finally determined to be less than 25mm, professional screening equipment is used to screen the large amount of collected building materials. The screen aperture of the screening equipment is set according to the optimal screening parameters. Under the vibration and screening action of the equipment, the building materials are separated according to the particle size. After screening, what is obtained is the screening building material that meets the optimal particle size requirements, that is, the diameter is less than 25mm. The screened building material has the particle size characteristics that can efficiently and evenly fill each small area model, providing an ideal material basis for subsequent homogenization mixing and final filling construction.

[0029] In a possible implementation, raw materials for preparing fluidized filling materials by homogenization are collected, and screening parameters of the raw materials are optimized according to the multiple small area models to obtain optimal screening parameters, and the raw materials are screened to obtain screened raw materials with optimal particle size. Step S300 further includes: Step S310, obtaining screening parameter space. Specifically, various parameter ranges that affect the screening effect of the raw materials are clarified, such as the size of the mesh aperture, the vibration frequency, the screening time, etc., and the parameter value ranges together constitute the screening parameter space.

[0030] Step S320, constructing a screening function for optimizing screening parameters, as shown in the following formula: Among them, SCF is the screening fitness, M is the number of multiple small area models, and w i is the weight assigned according to the size of the i-th small area model, C z is the maximum particle size of the sieving building material after sieving according to the sieving parameters, C i is the area size in the i-th small area model. Specifically, the purpose of the screening function is to quantitatively evaluate the pros and cons of different screening parameters, that is, the screening fitness. SCF represents the screening fitness. The larger its value, the more ideal the corresponding screening parameter. M represents the total number of small area models, and w i is the weight assigned according to the size of the i-th small area model. The larger the small area model, the higher its corresponding weight, to reflect its importance in the overall evaluation. i is the area size in the i-th small area model, reflecting the spatial size of the small area, C z is the maximum particle size of the screening building material after screening according to the screening parameters, The relative relationship between each small area model and the particle size of the raw material after screening is calculated, and a certain scaling and adjustment is performed through the logarithmic function, multiplied by the weight w i And sum it up, taking into account the conditions of all small area models, and derive an overall fitness evaluation value.

[0031] Step S330, randomly generate a first screening parameter in the screening parameter space, and obtain a first particle size of the screened building material after the first screening parameter is screened. Specifically, in the determined multidimensional screening parameter space, a set of initial parameter combinations is randomly generated by a random number generation algorithm, recorded as the first screening parameters, and the parameters are used to perform actual screening operations on the building material, and the maximum particle size of the building material after screening is measured and recorded, which is defined as the first particle size.

[0032] Step S340: Calculate a first screening fitness based on the first particle size, the region sizes in the plurality of small region models, and the screening function.z The area size C in each small area model i Substitute them into the screening function together, and according to the weight w corresponding to each small area model i , perform corresponding mathematical operations to calculate the fitness value under this set of first screening parameters, that is, the first screening fitness SCF.

[0033] Step S350, continue to perform iterative optimization based on the screening function in the screening parameter space until convergence, output the screening parameters with the largest screening fitness, and obtain the optimal screening parameters. Specifically, after the first screening fitness is calculated for the first time, a new set of screening parameters is randomly generated in the screening parameter space again, and the above-mentioned screening operation, particle size measurement, and fitness calculation process are repeated. By continuously generating new parameters, calculating fitness and comparing, the set of screening parameters that maximize the screening fitness "SCF" is gradually found. During the iterative process, when the improvement of fitness is no longer significant or reaches a pre-set convergence condition (for example, the change in fitness is less than a certain threshold or reaches the maximum number of iterations), the iteration is stopped, and the set of screening parameters with the largest fitness at this time is finally output, and it is determined as the optimal screening parameters that can make the building material best fit the small area model.

[0034] Step S400, based on the large area model and the multiple small area models, the homogenized mixing parameters of the screened building material combined with the mixed filler are optimized, and the optimal homogenized mixing parameters are obtained, wherein the filling rate of the small area model is predicted based on the homogenized mixing parameters combined with the multiple small area models, and the load satisfaction is analyzed and optimized in combination with the large area load requirement information. Specifically, detailed data of the large-area model and multiple small-area models are obtained, including information such as shape, size, and spatial position. Taking into account the load requirements after backfilling, the homogenization mixing parameters of the screened building materials combined with the mixed fillers are optimized. The homogenization mixing parameters include filler component parameters (such as filler components of different types and proportions) and mixing control parameters (such as stirring time, stirring speed, etc.). For the optimization of filler component parameters, different combination ratios are tried by analyzing the physical and mechanical properties of different filler components and their roles in the backfill structure. For example, the proportion of filler components with higher strength is increased to improve the load-bearing capacity after backfilling. When optimizing the mixing control parameters, the stirring time and stirring speed and other conditions are changed to observe their effects on the mixing effect. Longer stirring time and appropriate stirring speed can be used. It can help to mix various components more evenly, thereby improving the overall performance. During the optimization process, the filling rate of the small area model is predicted based on the homogenized mixing parameters combined with multiple small area models. By establishing a mathematical model or using simulation software, the homogenized mixing parameters and the characteristics of the small area model are input to calculate the expected filling rate. Combined with the large area load requirement information, it is analyzed whether the current homogenized mixing parameters can meet the load requirements. If the predicted filling rate is low or the load satisfaction is insufficient, continue to adjust the homogenized mixing parameters, and repeat the above optimization, prediction and analysis steps until the optimal homogenized mixing parameters that can simultaneously improve the backfill load and the filling rate of the small area model are obtained, ensuring that the structure after backfill has sufficient strength and stability, while avoiding the appearance of cavities in the small area model, and improving the overall fluidity and filling effect.

[0035] In a possible implementation, the homogenization mixing parameters of the screened building materials combined with the mixed fillers are optimized according to the large area model and the multiple small area models, and the optimal homogenization mixing parameters are obtained, wherein the filling rate of the small area model is predicted according to the homogenization mixing parameters combined with the multiple small area models, and the load satisfaction is analyzed and optimized in combination with the large area load requirement information. Step S400 further includes: Step S410, obtaining the homogenization mixing parameter space, wherein each homogenization mixing parameter includes a filler component parameter and a mixing control parameter. Specifically, the various parameter ranges that can affect the homogenization mixing effect of the building materials are clarified, the filler component parameters may cover the types and proportions of different fillers, and the mixing control parameters may include stirring time, stirring speed, order of addition, etc. All possible value combinations of the parameters constitute the homogenization mixing parameter space.

[0036] Step S420, constructing a homogenization mixing function for optimizing the homogenization mixing parameters of the screened building material combined with the mixed filler, as shown in the following formula: Among them, HMF is the hybrid fitness, w1 and w2 are the load weight and filling weight, F h F is the backfill load information of backfilling after the sieved building material is homogenized and mixed according to the homogenization mixing parameters, d Information on large area load requirements, K i is the filling rate of the backfilled i-th small area after the screened building material is homogenized and mixed according to the homogenization mixing parameters. Specifically, the homogenization mixing function aims to comprehensively evaluate the effects of different homogenization mixing parameters. HMF represents the mixing fitness, which is used to measure the pros and cons of the homogenization mixing parameters. w1 and w2 are the load weight and filling weight, respectively, reflecting the relative importance of backfill load and filling rate in the overall evaluation. F h is the actual backfill load information after mixing and backfilling according to the current homogenized mixing parameters, F d is the load requirement information for large areas, through Calculate the satisfaction degree of backfill load, K i is the filling rate of the ith small area, through Comprehensively consider the filling conditions of multiple small areas.

[0037] Step S430, randomly generating first averaged mixing parameters in the averaged mixing parameter space. Specifically, in the determined averaged mixing parameter space, a group of initial parameter combinations is randomly generated by a random algorithm and marked as first averaged mixing parameters.

[0038] Step S440, based on the first homogenized mixing parameter and the optimal particle size, combined with the large area model and the multiple small area models, backfill load prediction and backfill filling rate prediction are performed, and the first mixing fitness is calculated based on the homogenized mixing function. Specifically, the first homogenized mixing parameter and the previously determined optimal particle size are used, combined with the characteristics of the large area model and the multiple small area models, and a mathematical model or simulation software is used to predict the load after backfilling and the filling rate of each small area, and the predicted backfill load and filling rate data are substituted into the homogenized mixing function, and the calculation is performed according to the calculation rule of the function to obtain the mixing fitness value under this set of first homogenized mixing parameters, that is, the first mixing fitness.

[0039] Step S450, according to the homogenized mixing function, continue to iteratively optimize the homogenized mixing parameters in the homogenized mixing space until convergence, and obtain the homogenized mixing parameters with the maximum mixing fitness as the optimal homogenized mixing parameters. Specifically, after obtaining the first mixing fitness, randomly generate new homogenized mixing parameters in the homogenized mixing parameter space again, and repeat the process of backfill load prediction, backfill filling rate prediction and mixing fitness calculation, and continuously compare the newly generated mixing fitness with the previous maximum value. If the new fitness is larger, update the optimal value and the corresponding homogenized mixing parameter, and continue the iterative process until the change in the mixing fitness is no longer obvious or reaches the preset convergence condition (for example, the change in fitness is less than a specific threshold or reaches the maximum number of iterations), then output the homogenized mixing parameter corresponding to the maximum mixing fitness, and identify it as the optimal homogenized mixing parameter.

[0040] In a possible implementation, according to the first homogenized mixing parameter and the optimal particle size, combined with the large area model and multiple small area models, backfill load prediction and backfill filling rate prediction are performed, and based on the homogenized mixing function, the first mixing fitness is calculated, and step S440 further includes: step S441, according to the historical data of backfilling of the same building materials, a sample particle size set, a sample homogenized mixing parameter set, a sample large area model set and a sample small area model set are collected, and the sample backfill load information set and the sample filling rate set are annotated. Specifically, the project data of backfilling using the same building materials in the past are collected, and the sample building material particle size data in different projects are extracted from the historical data to form a sample particle size set; similarly, the corresponding sample homogenized mixing parameter data are obtained to form a sample homogenized mixing parameter set; and the relevant data of the sample large area model and the sample small area model are extracted to form sets respectively, and the historical backfill projects are analyzed and annotated to obtain the corresponding sample backfill load information to form a sample backfill load information set; and the sample filling rate is determined to form a sample filling rate set.

[0041] Step S442, using the sample particle size set, sample homogenization mixing parameter set, and sample large-area model set as input, and using the sample backfill load information set as supervision, to train the backfill load prediction branch. Specifically, the collected data of the sample particle size set, sample homogenization mixing parameter set, and sample large-area model set are input into a machine learning model. The goal of the model is to predict the backfill load, and the labeled sample backfill load information set is used as supervision data, that is, the correct result of the expected model output. By continuously adjusting the parameters of the model, the model can accurately predict the backfill load based on the input sample particle size, homogenization mixing parameters, and large-area model information.

[0042] Step S443, using the sample particle size set, sample averaged mixing parameter set, and sample small area model set as input, using the sample filling rate set as supervision, training the backfill filling rate prediction branch, and combining the backfill load prediction branch to obtain a backfill predictor. Specifically, using the sample particle size set, sample averaged mixing parameter set, and sample small area model set as input data, and the sample filling rate set as supervision data, a machine learning branch specifically used to predict the backfill filling rate is trained. After the training is completed, the backfill load prediction branch and the backfill filling rate prediction branch are combined to form a backfill predictor that can simultaneously predict the backfill load and filling rate.

[0043] Step S444, using the backfill predictor, backfill prediction is performed on the first homogenized mixing parameter, optimal particle size, large area model and multiple small area models to obtain first backfill load information and multiple first filling rate information, and the first mixing fitness is calculated based on the homogenized mixing function. Specifically, the data of the new first homogenized mixing parameter, the previously determined optimal particle size, the large area model and the multiple small area models are input into the trained backfill predictor, and the backfill predictor will output the corresponding first backfill load information and multiple first filling rate information. The prediction result is substituted into the homogenized mixing function, and the calculation is performed according to the operation rules of the function, so as to obtain the first mixing fitness, which is used to evaluate the effect of the current homogenized mixing parameters and models, etc.

[0044] Step S500, using the optimal homogenization mixing parameters, the building materials are mixed with fillers to obtain the prepared fluid filling materials. Specifically, it is ensured that the optimal homogenization mixing parameters obtained through a series of previous optimizations and calculations have been obtained, and the parameters include the specific component ratio of the fillers and the control conditions in the mixing process, such as stirring time, speed, etc. The prepared building materials, such as sand, gravel, cement, etc., are put into the mixing device according to the proportion and order specified in the optimal homogenization mixing parameters, and the mixing device is started. The stirring operation is performed according to the stirring time and speed set in the optimal homogenization mixing parameters. During the stirring process, various building materials are gradually fully contacted and mixed, and the stirring continues until the time specified in the optimal homogenization mixing parameters is reached, ensuring that the building materials are in a uniform state in terms of physical and chemical properties. After sufficient mixing and stirring, the obtained mixture is the prepared fluid filling material, which has the required fluidity and uniformity and can meet the subsequent backfilling construction requirements. During the entire mixing process, the operation is strictly carried out in accordance with the optimal homogenization mixing parameters, and the mixed filling materials are quality tested to ensure that their performance meets expectations.

[0045] The embodiment of the present application adopts the method of obtaining a three-dimensional model of the backfill area, dividing it into a large area and multiple small area models, and analyzing the load requirements of the large area. The building materials are collected, the screening parameters are optimized according to the small area model, and the building materials with the optimal particle size are processed. The homogenization mixing parameters are optimized according to the large and small area models, and the load satisfaction is analyzed in combination with the predicted filling rate of the small area model and the load requirements of the large area. The building materials are mixed with the optimal homogenization mixing parameters to obtain fluidized filling materials. By achieving uniform distribution of fluidity and density of the fluidized filling materials in the backfill area, the load and space requirements of different regions are met, and the technical effect of improving the quality, stability and economic benefits of the project is achieved.

[0046] In the above, refer to Figure 1 A homogenization and mixing preparation method of a fluid filling material according to an embodiment of the present invention is described in detail. Figure 2 A homogenizing, mixing and preparing device for fluidized filling materials according to an embodiment of the present invention is described.

[0047] A homogenizing and mixing preparation device for fluidized filling materials according to an embodiment of the present invention is used to solve the technical problem that the prior art cannot comprehensively consider the load, spatial characteristics and material processing parameters of the backfill area when preparing fluidized filling materials, resulting in the filling materials being difficult to simultaneously meet the fluidity, uniform density and engineering requirements. By achieving uniform distribution of fluidity and density of fluidized filling materials in the backfill area, the load and space requirements of different regions are met, and the technical effect of improving the quality, stability and economic benefits of the project is achieved. A homogenizing and mixing preparation device for fluidized filling materials includes: a regional three-dimensional model acquisition module 10, a large-area load requirement information acquisition module 20, an optimal screening parameter acquisition module 30, an optimal homogenizing and mixing parameter acquisition module 40, and a fluidized filling material preparation module 50.

[0048] The regional three-dimensional model acquisition module 10 is used to acquire a regional three-dimensional model of a backfill area to be backfilled;

[0049] The large-area load requirement information acquisition module 20 is used to divide the regional three-dimensional model, obtain a large-area model and a plurality of small-area models, and analyze and obtain the large-area load requirement information;

[0050] The optimal screening parameter acquisition module 30 is used to collect building materials to be homogenized to prepare fluidized filling materials, optimize the screening parameters of the building materials according to the multiple small area models, obtain the optimal screening parameters, and screen the building materials to obtain the screened building materials with the optimal particle size;

[0051] The optimal homogenized mixing parameter acquisition module 40 is used to optimize the homogenized mixing parameters of the screened building material combined with the mixed filler according to the large area model and the multiple small area models, and obtain the optimal homogenized mixing parameters, wherein the filling rate of the small area model is predicted according to the homogenized mixing parameters combined with the multiple small area models, and the load satisfaction is analyzed and optimized in combination with the large area load requirement information;

[0052] The fluid filling material preparation module 50 is used to adopt the optimal homogenization mixing parameters to mix the building material raw materials to obtain the prepared fluid filling material.

[0053] The specific configuration of the regional 3D model acquisition module 10 will be described in detail below. As described above, the regional 3D model of the backfill area to be backfilled is acquired, and the regional 3D model acquisition module 10 further includes: a backfill scan data acquisition unit, the backfill scan data acquisition unit is used to scan the backfill area to be backfilled and obtain backfill scan data; a regional 3D model construction unit, the regional 3D model construction unit is used to construct the regional 3D model of the backfill area according to the backfill scan data.

[0054] The specific configuration of the large area load requirement information acquisition module 20 will be described in detail below. As described above, the regional three-dimensional model is divided to obtain a large area model and multiple small area models, and the large area load requirement information is obtained by analysis. The large area load requirement information acquisition module 20 further includes: a division result set acquisition unit, the division result set acquisition unit is used to collect a sample area three-dimensional model set, divide the large area model and the small area model in each sample area three-dimensional model, and obtain a sample area model division result set; a regional model division unit, the regional model division unit is used to use the sample area three-dimensional model set and the sample area model division result set to train a regional model divider, divide the regional model, and obtain a large area model and multiple small area models; a backfill load analysis unit, the backfill load analysis unit is used to perform backfill load analysis on the large area model to obtain large area load requirement information.

[0055] Among them, backfill load analysis is performed on the large area model to obtain large area load requirement information, and the backfill load analysis unit further includes: a sample large area load requirement information set acquisition subunit, the sample large area load requirement information set acquisition subunit is used to obtain a sample large area model set in the same backfill project, and obtain the sample large area load requirement information set according to the load requirement after backfilling; a backfill load analyzer training subunit, the backfill load analyzer training subunit is used to adopt the sample large area model set and the sample large area load requirement information set to train the backfill load analyzer; a large area load requirement information acquisition unit, the large area load requirement information acquisition unit is used to adopt the backfill load analyzer to perform backfill load analysis on the large area model to obtain large area load requirement information.

[0056] The specific configuration of the optimal screening parameter acquisition module 30 will be described in detail below. As described above, the building material to be homogenized to prepare the fluidized filling material is collected, the screening parameters of the building material are optimized according to the multiple small area models, the optimal screening parameters are obtained, the building material is screened to obtain the screening building material with the optimal particle size, and the optimal screening parameter acquisition module 30 further includes: a screening parameter space acquisition unit, the screening parameter space acquisition unit is used to acquire the screening parameter space; a screening function construction unit, the screening function construction unit is used to construct a screening function for optimizing the screening parameters, as shown in the following formula: Among them, SCF is the screening fitness, M is the number of multiple small area models, and w i is the weight assigned according to the size of the i-th small area model, C z is the maximum particle size of the sieving building material after sieving according to the sieving parameters, C i is the area size in the i-th small area model; a first particle size acquisition unit, the first particle size acquisition unit is used to randomly generate a first screening parameter in the screening parameter space, and obtain the first particle size of the screened building material after screening with the first screening parameter; a first screening fitness calculation unit, the first screening fitness calculation unit is used to calculate and obtain the first screening fitness based on the screening function according to the first particle size and the area sizes in the multiple small area models; continue to iteratively optimize based on the screening function in the screening parameter space until convergence, output the screening parameter with the largest screening fitness, and obtain the optimal screening parameter.

[0057] The specific configuration of the optimal homogenized mixing parameter acquisition module 40 will be described in detail below. As described above, according to the large area model and the multiple small area models, the homogenized mixing parameters of the screened building material combined with the mixed filler are optimized, and the optimal homogenized mixing parameters are obtained, wherein the filling rate of the small area model is predicted according to the homogenized mixing parameters combined with the multiple small area models, and the load satisfaction is analyzed and optimized in combination with the large area load requirement information. The optimal homogenized mixing parameter acquisition module 40 further includes: a homogenized mixing parameter space acquisition unit, the homogenized mixing parameter space acquisition unit is used to acquire a homogenized mixing parameter space, wherein each homogenized mixing parameter includes a filler component parameter and a mixing control parameter; a homogenized mixing function construction unit, the homogenized mixing function construction unit is used to construct a homogenized mixing function that optimizes the homogenized mixing parameters of the screened building material combined with the mixed filler, as shown in the following formula: Among them, HMF is the hybrid fitness, w1 and w2 are the load weight and filling weight, F h F is the backfill load information of backfilling after the sieved building material is homogenized and mixed according to the homogenization mixing parameters, d Information on large area load requirements, K i The filling rate of the i-th small area after the screened building material is homogenized and mixed according to the homogenized mixing parameters; a first homogenized mixing parameter generation unit, the first homogenized mixing parameter generation unit is used to randomly generate a first homogenized mixing parameter in the homogenized mixing parameter space; a first mixing fitness calculation unit, the first mixing fitness calculation unit is used to predict the backfill load and the backfill filling rate according to the first homogenized mixing parameter and the optimal particle size, combined with the large area model and multiple small area models, and calculate the first mixing fitness based on the homogenized mixing function; an iterative optimization unit, the iterative optimization unit is used to continue to iteratively optimize the homogenized mixing parameters in the homogenized mixing space according to the homogenized mixing function until convergence, and obtain the homogenized mixing parameters with the largest mixing fitness as the optimal homogenized mixing parameters.

[0058] Among them, according to the first homogenized mixing parameter and the optimal particle size, combined with the large area model and multiple small area models, backfill load prediction and backfill filling rate prediction are performed, and based on the homogenized mixing function, a first mixing fitness is calculated, and the first mixing fitness calculation unit further includes: a set annotation subunit, the set annotation subunit is used to collect a sample particle size set, a sample homogenized mixing parameter set, a sample large area model set and a sample small area model set according to the backfill historical data of similar building materials, and annotate and obtain a sample backfill load information set and a sample filling rate set; a backfill load prediction branch training subunit, the backfill load prediction branch training subunit is used to adopt the sample particle size set, the sample homogenized mixing parameter set, the sample large area model set and the sample small area model set The set is used as input, and the sample backfill load information set is used as supervision to train the backfill load prediction branch; a backfill predictor acquisition subunit, the backfill predictor acquisition subunit is used to use the sample particle size set, the sample equalization mixing parameter set, and the sample small area model set as input, and use the sample filling rate set as supervision to train the backfill filling rate prediction branch, and combine the backfill load prediction branch to obtain a backfill predictor; a backfill prediction subunit, the backfill prediction subunit is used to use the backfill predictor to perform backfill prediction on the first equalization mixing parameter, optimal particle size, large area model and multiple small area models, obtain first backfill load information and multiple first filling rate information, and calculate and obtain the first mixing fitness based on the equalization mixing function.

[0059] A homogenizing, mixing and preparing device for fluidized filling material provided in an embodiment of the present invention can execute a homogenizing, mixing and preparing method for fluidized filling material provided in any embodiment of the present invention, and has functional modules and beneficial effects corresponding to the execution method.

[0060] Although the present application makes various references to certain modules in the device according to the embodiments of the present application, any number of different modules may be used and run on the user terminal and / or server, and the various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.

[0061] The above specific implementation manner does not constitute a limitation to the protection scope of the present application. It should be understood by those skilled in the art that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application should be included in the protection scope of the present application. In some cases, the actions or steps recorded in the present application can be performed in an order different from that in the embodiment and can still achieve the desired results. In addition, the process depicted in the accompanying drawings does not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

Claims

1. A homogenizing and mixing preparation method of fluidized filling material, characterized in that: The method comprises: Obtaining a three-dimensional model of the backfill area to be backfilled; Dividing the regional three-dimensional model to obtain a large regional model and a plurality of small regional models, and analyzing to obtain the load requirement information of the large regional model; Collecting building materials to be homogenized to prepare fluidized filling materials, optimizing screening parameters of the building materials according to the multiple small area models to obtain optimal screening parameters, and screening the building materials to obtain sieved building materials with optimal particle size; According to the large area model and the multiple small area models, the homogenization mixing parameters of the screened building material combined with the mixed filler are optimized, and the optimal homogenization mixing parameters are obtained, wherein the filling rate of the small area model is predicted according to the homogenization mixing parameters combined with the multiple small area models, and the load satisfaction is analyzed and optimized in combination with the load requirement information of the large area; The optimal homogenization mixing parameters are used to mix the building material raw materials to obtain the prepared fluidized filling material.

2. The homogenizing and mixing preparation method of fluidized filling material according to claim 1, characterized in that: Obtain a regional 3D model of the backfill area to be backfilled, including: Scan the backfill area to be backfilled to obtain backfill scanning data; A regional three-dimensional model of the backfill area is constructed based on the backfill scan data.

3. The homogenizing and mixing preparation method of fluidized filling material according to claim 1, characterized in that: The regional three-dimensional model is divided to obtain a large regional model and a plurality of small regional models, and the load requirement information of the large regional model is obtained by analysis, including: Collecting a set of sample area three-dimensional models, dividing a large area model and a small area model in each sample area three-dimensional model, and obtaining a set of sample area model division results; Using the sample region three-dimensional model set and the sample region model division result set, training a region model divider, dividing the region model to obtain a large region model and a plurality of small region models; A backfill load analysis is performed on the large-area model to obtain large-area load requirement information.

4. The homogenizing and mixing preparation method of fluidized filling material according to claim 3, characterized in that: Perform backfill load analysis on the large area model, including: Obtain a sample large area model set in the same backfill project, and obtain a sample large area load requirement information set based on the backfill post-load requirement; Using the sample large area model set and the sample large area load requirement information set, training a backfill load analyzer; The backfill load analyzer is used to perform backfill load analysis on the large-area model to obtain large-area load requirement information.

5. The homogenizing and mixing preparation method of fluidized filling material according to claim 1, characterized in that: According to the multiple small area models, the screening parameters of the building material are optimized to obtain the optimal screening parameters, including: Get the screening parameter space; Construct a screening function for screening parameter optimization as follows: Among them, SCF is the screening fitness, M is the number of multiple small area models, and w i is the weight assigned according to the size of the i-th small area model, C z is the maximum particle size of the sieving building material after sieving according to the sieving parameters, C i is the area size in the i-th small area model; Randomly generate a first screening parameter in the screening parameter space, and obtain a first particle size of the sieved building material after screening with the first screening parameter; According to the first particle size, combined with the region sizes in the multiple small region models, and based on the screening function, a first screening fitness is calculated; Continue to perform iterative optimization based on the screening function in the screening parameter space until convergence, output the screening parameters with the maximum screening fitness, and obtain the optimal screening parameters.

6. The homogenizing and mixing preparation method of fluidized filling material according to claim 1, characterized in that: According to the large area model and the plurality of small area models, the homogenization mixing parameters of the screened building material combined with the mixed filler are optimized, including: Acquire a homogenized mixing parameter space, wherein each homogenized mixing parameter includes a filler component parameter and a mixing control parameter; A homogenization mixing function is constructed to optimize the homogenization mixing parameters of the sieved building material combined with the mixed filler, as shown in the following formula: Among them, HMF is the hybrid fitness, w1 and w2 are the load weight and filling weight, F h F is the backfill load information of backfilling after the sieved building material is homogenized and mixed according to the homogenization mixing parameters, d Information on large area load requirements, K i is the filling rate of the i-th small area after the sieved building material is homogenized and mixed according to the homogenization mixing parameters; Randomly generating a first averaged mixing parameter in the averaged mixing parameter space; According to the first homogenized mixing parameter and the optimal particle size, in combination with the large area model and the plurality of small area models, backfill load prediction and backfill filling rate prediction are performed, and based on the homogenized mixing function, a first mixing fitness is calculated; According to the homogenized mixing function, the iterative optimization of the homogenized mixing parameters is continued in the homogenized mixing space until convergence, and the homogenized mixing parameters with the maximum mixing fitness are obtained as the optimal homogenized mixing parameters.

7. The homogenizing and mixing preparation method of fluidized filling material according to claim 6, characterized in that: According to the first homogenization mixing parameter and the optimal particle size, in combination with the large area model and the plurality of small area models, backfill load prediction and backfill filling rate prediction are performed, including: According to the backfill history data of the same building material, the sample particle size set, sample homogenization mixing parameter set, sample large area model set and sample small area model set are collected, and the sample backfill load information set and sample filling rate set are marked; The sample particle size set, the sample averaging mixing parameter set, and the sample large area model set are used as inputs, and the sample backfill load information set is used as supervision to train a backfill load prediction branch; The sample particle size set, the sample averaging mixing parameter set, and the sample small area model set are used as inputs, and the sample filling rate set is used as supervision to train a backfill filling rate prediction branch, and a backfill predictor is obtained by combining the backfill load prediction branch; The backfill predictor is used to perform backfill prediction on the first homogenized mixing parameter, optimal particle size, large area model and multiple small area models to obtain first backfill load information and multiple first filling rate information, and the first mixing fitness is calculated based on the homogenized mixing function.

8. A homogenizing and mixing preparation device for fluidized filling materials, characterized in that: The device is used to implement the homogenization mixing preparation method of a fluid filling material according to any one of claims 1 to 7, and the device comprises: A regional three-dimensional model acquisition module, the regional three-dimensional model acquisition module is used to acquire a regional three-dimensional model of a backfill area to be backfilled; A large-area load requirement information acquisition module, the large-area load requirement information acquisition module is used to divide the regional three-dimensional model, obtain a large-area model and a plurality of small-area models, and analyze and obtain the large-area load requirement information; An optimal screening parameter acquisition module, the optimal screening parameter acquisition module is used to collect building materials to be homogenized to prepare fluidized filling materials, optimize the screening parameters of the building materials according to the multiple small area models, obtain the optimal screening parameters, and screen the building materials to obtain the screened building materials with the optimal particle size; An optimal homogenized mixing parameter acquisition module, the optimal homogenized mixing parameter acquisition module is used to optimize the homogenized mixing parameters of the screened building material combined with the mixed filler according to the large area model and the multiple small area models, and obtain the optimal homogenized mixing parameters, wherein the filling rate of the small area model is predicted according to the homogenized mixing parameters combined with the multiple small area models, and the load satisfaction is analyzed and optimized in combination with the large area load requirement information; A fluid filling material preparation module is used to adopt the optimal homogenization mixing parameters to mix the building material raw materials to obtain the prepared fluid filling material.

Citation Information

Cited By

  • Water conservancy project three-dimensional parameterization design method based on deep learning

    CN120372751A