Pre-treatment method for artificial granite manufacturing by optimizing aggregate gradation interface performance

Abstract

The present invention discloses a method for pre-treatment of artificial granite manufacturing by optimizing aggregate gradation interface performance, comprising the following steps: (1) establishing a data set for characterizing artificial granite aggregates and selecting decision-making data; (2) obtaining powder gradation and material internal parameters and optimizing machine learning data sets; (3) obtaining the adaptation conditions for the synergistic enhancement of the artificial granite aggregate gradation-resin system, establishing a large data set platform for the intelligent production of artificial granite, and directly generating an adapted artificial granite formula in reverse according to customized requirements. The present invention is based on the data set of experimental and LBM-DEM computer simulation pre-mixed analysis results, and uses machine learning to complete the automatic decision-making process. It combines integrated equipment to complete the selection and preparation of powders at all levels, predict and monitor the complex flow behavior of the mixture in the mold, and simulate the mixing process in advance, thereby inferring the appropriate amount of additives such as multi-coupling agents in production, quantitatively controlling the hydrophobic modification process of aggregates, and using highly intelligent production of artificial granite.

Description

Technical Field

[0001] The invention belongs to the technical field of artificial granite manufacturing, and in particular relates to an artificial granite manufacturing pre-treatment method utilizing optimized aggregate gradation interface performance. Background Art

[0002] Artificial granite is currently the most widely used and largest amount of artificial stone. It uses natural marble crushed pieces and stone powder as the main raw materials, unsaturated polyester resin as the binder, and is formed by high pressure and vibration. It is then cured at room temperature and made into panels through cutting, grinding, polishing and other processes. In the sample preparation and trial production stage of artificial granite, the proportion of the aggregate system is often adjusted by experience. It often takes multiple trials and adjustments to get a more ideal proportion formula. The trial production process has a long cycle and high cost. In the face of large aggregate products and actual production processes, the proportion of artificial granite raw materials has always followed the empirical formula. The production process is relatively backward, resulting in serious product homogeneity and over-reliance on the experience of production technicians. While the production efficiency is low, there is still a lack of quantitative, scientific and intelligent key technology research that can guide the production of artificial granite.

[0003] The inability to understand how the artificial granite aggregate formula affects the overall mechanical properties of artificial granite products from the inside has long troubled technical personnel in the artificial granite industry. At present, there is still an urgent need to explore standardized methods for obtaining ideal artificial granite formulas, break through traditional production processes and improve traditional processes through intelligent manufacturing. This is one of the effective ways to essentially reduce the production cost of artificial granite and improve production efficiency. Summary of the invention

[0004] The purpose of the present invention is to provide a pre-treatment method for manufacturing artificial granite by optimizing the aggregate gradation interface performance in view of the shortcomings of the prior art.

[0005] In order to achieve the above object, the present invention adopts the following technical solutions:

[0006] A method for pre-processing artificial granite by optimizing aggregate gradation interface performance comprises the following steps:

[0007] (1) Establishment of data set for characterization and selection decision of artificial granite aggregate;

[0008] ① Pre-treatment process of artificial granite material selection: Comprehensively analyze its chemical composition, three-dimensional microscopic morphology, particle size and gradation composition to obtain the characterization data of artificial granite aggregate;

[0009] ②Through experiments and simulations, the accumulation theory of granite aggregates is studied, characterized and classified to obtain artificial granite aggregates with stable structure and uniform particle size distribution, which are then classified by label type and included in the aggregate information dataset for machine learning;

[0010] (2) Obtaining powder gradation and material internal parameters and optimizing machine learning data sets;

[0011] ① Write a special simulation program for artificial granite based on C++ language experiments and lattice Boltzmann-discrete element method to examine the internal condition parameters of powder gradation and artificial stone materials;

[0012] ② On this basis, the parameter settings required for simulation are obtained, and the compatibility index parameters are obtained through experiments and simulations. After calibration, the key control parameters are summarized to form a matching labeled optimal machine learning data set;

[0013] (3) An intelligent system for synergistic enhancement of the artificial granite aggregate grading-resin system;

[0014] ① Select appropriate machine learning training models to seek the ideal balance between aggregate gradation and resin system to enhance mechanical properties, reduce expansion and cost; factors such as gradation system, additives and resin types to the ideal adaptation conditions of rheology, viscosity, stress distribution, vacuum pressure and temperature of materials during processing;

[0015] ② Establish a big data set platform for the intelligent production of artificial granite, and directly generate suitable artificial granite formulas based on customized requirements.

[0016] The present invention further illustrates that the calculation process of the artificial granite special simulation program includes:

[0017] 1) Import the 3D reconstruction model of artificial granite aggregate;

[0018] 2) Initialize the flow field information and the position, translational velocity and angular velocity of the artificial granite aggregate;

[0019] 3) At the starting point t0 of each time step, input the flow field information and the particle position, translational velocity and angular velocity;

[0020] 4) Determine whether the grid is covered by particles and mark it; if so, calculate the solid content of the grid control volume and go to step 5); if not, use LBM to solve the flow field at the next moment, that is, t f The flow field at the moment, and then go to step 5);

[0021] 5) At t f The flow field information and particle coverage information at the moment are used to calculate the force and moment acting on the particle at this moment;

[0022] 6) Determine the particle size at t f Is there a collision at this moment? If there is a collision, calculate the inter-particle force according to the DEM and determine whether it is greater than the maximum number of sub-cycles. If it is less than the maximum number of sub-cycles, return to step 5); if it is greater than the maximum number of sub-cycles, go to step 7);

[0023] 7) At t f The flow field information and particle coverage information at the moment are used to calculate the force and moment acting on the particle at this moment;

[0024] 8) Determine whether the grid is covered by particles and mark it; if so, output the calculation result; if not, return to step 4).

[0025] Advantages of the present invention:

[0026] The present invention is based on the data set of experimental and LBM-DEM computer simulation pre-mixed analysis results, completes the automatic decision-making process through machine learning, and combines integrated equipment to complete the selection and deployment of powders at all levels, predict and monitor the complex flow behavior of the mixture in the mold, simulate the mixing process in advance, and thus reversely infer the appropriate amount of additives such as multi-coupling agents in production, quantitatively control the hydrophobic modification process of aggregates, and use highly intelligent production of artificial granite. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 It is a schematic diagram of the calculation process of the artificial granite LBM-DEM coupling simulation framework in one embodiment of the present invention. DETAILED DESCRIPTION

[0028] The present invention is further described below in conjunction with the accompanying drawings. Example

[0029] A method for pre-processing artificial granite by optimizing aggregate gradation interface performance comprises the following steps:

[0030] (1) Establishment of data set for characterization and selection decision of artificial granite aggregate;

[0031] ① Pre-treatment process of artificial granite material selection: Comprehensively analyze its chemical composition, three-dimensional microscopic morphology, particle size and gradation composition to obtain the characterization data of artificial granite aggregate;

[0032] ②Through experiments and simulations, the accumulation theory of granite aggregates is studied, characterized and classified to obtain artificial granite aggregates with stable structure and uniform particle size distribution, which are then classified by label type and included in the aggregate information dataset for machine learning;

[0033] (2) Obtaining powder gradation and material internal parameters and optimizing machine learning data sets;

[0034] ① Write a special simulation program for artificial granite based on C++ language experiments and lattice Boltzmann-discrete element method to examine the internal condition parameters of powder gradation and artificial stone materials;

[0035] ② On this basis, the parameter settings required for simulation are obtained, and the compatibility index parameters are obtained through experiments and simulations. After calibration, the key control parameters are summarized to form a matching labeled optimal machine learning data set;

[0036] (3) An intelligent system for synergistic enhancement of the artificial granite aggregate grading-resin system;

[0037] ① Select appropriate machine learning training models to seek the ideal balance between aggregate gradation and resin system to enhance mechanical properties, reduce expansion and cost; factors such as gradation system, additives and resin types to the ideal adaptation conditions of rheology, viscosity, stress distribution, vacuum pressure and temperature of materials during processing;

[0038] ② Establish a big data set platform for the intelligent production of artificial granite, and directly generate suitable artificial granite formulas based on customized requirements.

[0039] The calculation process of the above artificial granite special simulation program is as follows: Figure 1 As shown, including:

[0040] 1) Import the 3D reconstruction model of artificial granite aggregate;

[0041] 2) Initialize the flow field information and the position, translational velocity and angular velocity of the artificial granite aggregate;

[0042] 3) At the starting point t0 of each time step, input the flow field information and the particle position, translational velocity and angular velocity;

[0043] 4) Determine whether the grid is covered by particles and mark it; if so, calculate the solid content of the grid control volume and go to step 5); if not, use LBM to solve the flow field at the next moment, that is, t f The flow field at the moment, and then go to step 5);

[0044] 5) At t f The flow field information and particle coverage information at the moment are used to calculate the force and moment acting on the particle at this moment;

[0045] 6) Determine the particle size at t f Is there a collision at this moment? If there is a collision, calculate the inter-particle force according to the DEM and determine whether it is greater than the maximum number of sub-cycles. If it is less than the maximum number of sub-cycles, return to step 5); if it is greater than the maximum number of sub-cycles, go to step 7);

[0046] 7) At t f The flow field information and particle coverage information at the moment are used to calculate the force and moment acting on the particle at this moment;

[0047] 8) Determine whether the grid is covered by particles and mark it; if so, output the calculation result; if not, return to step 4).

[0048] Obviously, the above embodiments are only examples for clearly explaining the present invention, and are not intended to limit the implementation of the present invention. For ordinary technicians in the relevant technical field, other different forms of changes or modifications can be made based on the above description; it is not necessary and impossible to list all the implementation methods here; and the obvious changes or modifications derived from them are still within the protection scope of the present invention.

Claims

1. A method for pre-processing artificial granite by optimizing aggregate gradation interface performance, characterized in that The following steps are involved: (1) Establishment of data set for characterization and selection decision of artificial granite aggregate; ① Pre-treatment process of artificial granite material selection: Comprehensively analyze its chemical composition, three-dimensional microscopic morphology, particle size and gradation composition to obtain the characterization data of artificial granite aggregate; ②Through experiments and simulations, the accumulation theory of granite aggregates is studied, characterized and classified to obtain artificial granite aggregates with stable structure and uniform particle size distribution, which are then classified by label type and included in the aggregate information data set for machine learning; (2) Obtaining powder gradation and material internal parameters and optimizing machine learning data sets; ① Write a special simulation program for artificial granite based on C++ language experiments and lattice Boltzmann-discrete element method to examine the internal condition parameters of powder gradation and artificial stone materials; ② On this basis, the parameter settings required for simulation are obtained, and the compatibility index parameters are obtained through experiments and simulations. After calibration, the key control parameters are summarized to form a matching labeled optimal machine learning data set; (3) Obtain the adaptation conditions for the synergistic effect enhancement of the aggregate gradation and resin system of artificial granite, and establish a big data set platform for the intelligent production of artificial granite; ① Select appropriate machine learning training models to seek the ideal balance between aggregate gradation and resin system to synergistically enhance mechanical properties, reduce expansion and cost; the gradation system, additives and resin type factors have the ideal adaptation conditions for the rheology, viscosity, stress distribution, vacuum pressure and temperature of the material during processing; ② Establish a big data set platform for the intelligent production of artificial granite, and directly generate suitable artificial granite formulas based on customized requirements.

2. The method for pre-processing artificial granite by optimizing aggregate gradation interface performance according to claim 1, characterized in that: The calculation process of the artificial granite special simulation program includes: 1) Import the 3D reconstruction model of artificial granite aggregate; 2) Initialize the flow field information and the position, translational velocity and angular velocity of the artificial granite aggregate; 3) At the starting point t0 of each time step, input the flow field information and the particle position, translational velocity and angular velocity; 4) Determine whether the grid is covered by particles and mark it; if so, calculate the solid content of the grid control volume and go to step 5); if not, use LBM to solve the flow field at the next moment, that is, t f The flow field at the moment, and then go to step 5); 5) At t f The flow field information and particle coverage information at the moment are used to calculate the force and moment acting on the particle at this moment; 6) Determine the particle size at t f Is there a collision at this moment? If there is a collision, calculate the inter-particle force according to the DEM and determine whether it is greater than the maximum number of sub-cycles. If it is less than the maximum number of sub-cycles, return to step 5); if it is greater than the maximum number of sub-cycles, go to step 7); 7) At t f The flow field information and particle coverage information at the moment are used to calculate the force and moment acting on the particle at this moment; 8) Determine whether the grid is covered by particles and mark it; if so, output the calculation result; if not, return to step 4).

Images