Method for optimizing mix proportion of solid waste coal gangue light foam thermal insulation material and related equipment
By combining support vector machines and the firefly algorithm, global optimization of the mix proportion of lightweight foam insulation material made from solid waste coal gangue was achieved, solving the problems of insufficient optimization efficiency and accuracy in existing technologies, and improving the efficiency and accuracy of material performance prediction and optimization models.
Patent Information
- Application Number
- CN202511070108.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-31
- Publication Date
- 2025-11-28
AI Technical Summary
In existing technologies, it is difficult to achieve global optimization of the mix proportion of lightweight foam insulation materials made from solid waste coal gangue, resulting in insufficient optimization efficiency and accuracy.
We employ a machine learning model based on support vector machines to predict material properties, and combine it with the firefly algorithm to construct a mix ratio optimization model. We use the "brightness-attraction" mechanism to conduct large-scale global exploration and fine local mining. The optimization objective is to maximize the amount of solid waste mixed in, and we use material property data and constraints to delineate the feasible region.
It improves the efficiency and accuracy of mix design optimization, achieves global optimization of lightweight foam insulation material made from solid waste coal gangue, and enhances the accuracy of material performance prediction and construction efficiency.
Smart Images
Figure CN121034487A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of engineering materials technology, specifically to a method for optimizing the mix proportion of lightweight foam insulation material made from solid waste coal gangue and related equipment. Background Technology
[0002] Foamed concrete is a lightweight, microporous, energy-saving material made by introducing foam prepared with a foaming agent into a slurry made of cementitious materials, admixtures, and water in a specific ratio. This forms a uniform and stable fluid foam mixture, which is then cast, molded, and cured. Using metakaolin and coal gangue as raw materials to prepare this lightweight foamed insulation material not only effectively solves the resource waste and environmental pressure caused by the large-scale accumulation of coal gangue, but also reduces the environmental pollution caused by cement production by decreasing cement usage, thus achieving a win-win situation for both economic benefits and environmental protection.
[0003] In the manufacturing process of lightweight foam insulation materials made from solid waste coal gangue, the mix design directly shapes and determines the final performance of the material. Therefore, optimizing the mix design is particularly important. Related technologies typically employ simulated annealing algorithms, neural network models, or adaptive adjustments to the Paul Mey formula to determine the optimal mix design.
[0004] However, simulated annealing algorithms may face the challenge of slow convergence when dealing with large-scale problems, and their performance is highly dependent on the setting of initial temperature and cooling rate; the training process of neural networks may be time-consuming and resource-intensive, and the accuracy of their prediction results largely depends on the quality and representativeness of the training data; adaptive adjustments to the Paul Mee formula are difficult to achieve comprehensive optimization of the mix proportions, which limits further improvement of material properties. Summary of the Invention
[0005] In view of this, it is necessary to provide a method and related equipment for optimizing the mix proportion of lightweight foam insulation material made from solid waste coal gangue, so as to solve the problem that the existing technology is difficult to optimize the mix proportion of lightweight foam insulation material made from solid waste coal gangue globally, which reduces the optimization efficiency and the accuracy of the optimization results.
[0006] To address the aforementioned technical problems, in a first aspect, the present invention provides a method for optimizing the mix proportion of lightweight foam insulation material made from solid waste coal gangue, comprising: The material performance prediction model is used to predict the material performance based on the raw material usage data in the target material, and the corresponding material performance data is obtained. The target material is a lightweight foam insulation material made from solid waste coal gangue, and the preset material performance prediction model is a machine learning model based on support vector machine. Based on the material performance data and the constraints corresponding to the process and quality of the raw materials in the target material, the constraints corresponding to the mix proportion optimization model to be constructed are determined. Based on the preset optimization objective and the constraints, a mix proportion optimization model is constructed using the firefly algorithm, and the optimal mix proportion of the target material is output based on the mix proportion optimization model.
[0007] In one possible implementation, the material performance data includes the compressive strength, flexural strength, and dry density of the target material, and the limiting conditions include preset material ratio limiting conditions and total weight limiting conditions of the insulation material. The constraints for determining the mix proportion optimization model to be constructed, based on the material performance data and the constraints corresponding to the process and quality of the raw materials in the target material, include: The first constraint function corresponding to the compressive strength, the second constraint function corresponding to the flexural strength, the third constraint function corresponding to the dry density, the fourth constraint function corresponding to the preset material ratio limit condition, and the fifth constraint function corresponding to the total weight limit condition of the thermal insulation material are determined as the constraint conditions.
[0008] In one possible implementation, the preset optimization objective includes maximizing the amount of solid waste mixed in; the step of constructing a mix proportion optimization model using the firefly algorithm based on the preset optimization objective and the constraints includes: Several individual fireflies are generated according to the constraints, and the initial position of each individual firefly corresponds to a set of initial mixing ratios. In the current iteration step, based on the preset raw material usage data, the real-time position vector of each individual firefly is determined, and each component of the real-time position vector represents the usage of each raw material in the preset raw material usage data. Based on the real-time position vector, the brightness of each individual firefly is determined, and the brightness is positively correlated with the amount of solid waste mixed in. The attraction between the individual fireflies is determined based on the real-time position vector and brightness of each individual firefly. Based on the optimization objective, the position and attraction of the real-time position vector of the individual firefly are updated, wherein each position update corresponds to an update of the matching ratio. Each firefly individual after the update is sorted in a non-dominated order to form a new Pareto front. If the preset iteration criteria are met, the iteration is terminated, and the optimal solution in the current Pareto front is output to obtain the mix ratio optimization model; If the preset iteration criteria are not met, the process returns to the step of determining the real-time position vector of each firefly individual based on the preset raw material usage data in the current iteration step. Each component of the real-time position vector represents the usage amount of each raw material in the preset raw material usage data.
[0009] In one possible implementation, determining the attraction between individual fireflies based on their real-time position vector and brightness includes: Based on the real-time position vector and brightness of each individual firefly, the relative positions and relative brightness among the fireflies are determined respectively. The attraction is calculated based on the relative position and the relative brightness.
[0010] In one possible implementation, updating the position and attractiveness of the firefly individual's real-time position vector based on the optimization objective includes: In each iteration step, the position and attractiveness of the firefly individual are updated according to the movement method of the step size factor corresponding to the individual firefly, so that the individual firefly searches in the direction corresponding to the optimization target; If the updated position exceeds the preset boundary, the position will be corrected to the preset boundary.
[0011] In one possible implementation, the step size factor is obtained by adjusting it using a chaotic mapping operator.
[0012] In one possible implementation, the step of obtaining the preset material property prediction model includes: Obtain the original support vector machine model, wherein the kernel function of the original support vector machine model includes the RBF kernel function; The parameters of the original support vector machine model are tuned using a Bayesian optimization-based parameter tuning strategy to obtain the preset material property prediction model.
[0013] Secondly, the present invention also provides a device for optimizing the mix proportion of lightweight foam insulation material from solid waste coal gangue, comprising: The acquisition unit is used to predict the corresponding material performance data based on the raw material usage data in the target material using a preset material performance prediction model. The target material is a lightweight foam insulation material made from solid waste coal gangue, and the preset material performance prediction model is a machine learning model based on support vector machine. The determining unit is used to determine the constraints corresponding to the mix proportion optimization model to be constructed based on the material performance data and the constraints corresponding to the process and quality of the raw materials in the target material. The optimization unit is used to construct a mix proportion optimization model based on the preset optimization objective and the constraints, and to output the optimal mix proportion of the target material based on the mix proportion optimization model.
[0014] Thirdly, the present invention also provides an electronic device, including a memory and a processor, wherein, The memory is used to store programs; The processor, coupled to the memory, is used to execute the program stored in the memory to implement the steps in the method for optimizing the mix proportion of lightweight foam insulation material from solid waste coal gangue as described in any of the above implementations.
[0015] Fourthly, the present invention also provides a computer-readable storage medium for storing a computer-readable program or instruction, which, when executed by a processor, can implement the steps in the method for optimizing the mix proportion of lightweight foam insulation material from solid waste coal gangue as described in any of the above implementations.
[0016] The beneficial effects of this invention are: The present invention provides a method for optimizing the mix proportion of lightweight foam insulation material made from solid waste coal gangue. This method utilizes a pre-set material performance prediction model to predict the material's performance based on raw material usage data. The target material is lightweight foam insulation material made from solid waste coal gangue. The pre-set material performance prediction model is a machine learning model based on support vector machines. By utilizing the nonlinear prediction model of the pre-set material performance prediction model, the relationship between material performance and composition can be accurately reflected, achieving rapid mapping between raw material usage data and material performance data, thus improving the accuracy of material performance prediction based on raw material usage data. Based on the material performance data and the constraints of the target material, the constraints corresponding to the mix proportion optimization model to be constructed are determined. The method uses the material performance data and the constraints of the target material to jointly... By defining the feasible region and thus determining the constraints, the original high-dimensional continuous space was pruned into an effective subspace, reducing the number of invalid iterations in subsequent model construction, improving the convergence speed, and thereby improving the construction efficiency and accuracy of the mix proportion optimization model. Based on the preset optimization objectives and constraints, the firefly algorithm was used to construct the mix proportion optimization model, which outputs the optimal mix proportion of the target material. Since the firefly algorithm uses a "brightness-attraction" mechanism, it can perform large-scale global exploration in the early stage and fine local mining in the later stage with the help of decay step size, avoiding getting trapped in local optima. It achieves both global and local search, improving the optimization efficiency of the mix proportion. At the same time, it realizes the global optimization of the mix proportion of lightweight foam insulation material made from solid waste coal gangue, improving the accuracy of the mix proportion optimization results. Attached Figure Description
[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0018] Figure 1 This is a schematic flowchart of an embodiment of the method for optimizing the mix proportion of lightweight foam insulation material made from solid waste coal gangue provided by the present invention. Figure 2 A schematic diagram of the high-speed solid waste coal gangue lightweight foam insulation material mix ratio optimization system based on grating array vibration sensing provided by the present invention; Figure 3 A schematic diagram illustrating the process of optimizing the mix proportion of the lightweight foam insulation material made from solid waste coal gangue provided by this invention. Figure 4 A schematic diagram of an embodiment of the device for optimizing the mix proportion of lightweight foam insulation material for solid waste coal gangue provided by the present invention; Figure 5 A schematic diagram of an embodiment of the electronic device provided by the present invention. Detailed Implementation
[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.
[0020] In the description of the embodiments of the present invention, unless otherwise stated, "a plurality of" means two or more.
[0021] The terms "first," "second," etc., used in the embodiments of this invention are for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Therefore, a technical feature defined with "first" or "second" may explicitly or implicitly include at least one of that feature.
[0022] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.
[0023] This invention provides a method for optimizing the mix proportion of lightweight foam insulation material made from solid waste coal gangue and related equipment, which will be described below.
[0024] The method for optimizing the mix proportion of lightweight foam insulation material for solid waste coal gangue provided in this application embodiment can be applied to scenarios that require the disposal of solid waste such as coal gangue and to achieve synergistic optimization of insulation, mechanics, cost, and environmental protection.
[0025] The execution entity of the method for optimizing the mix proportion of lightweight foam insulation material from solid waste coal gangue in this application embodiment can be the device for optimizing the mix proportion of lightweight foam insulation material from solid waste coal gangue provided in this application embodiment, or different types of electronic devices such as server equipment, physical host, or user equipment (UE) that integrate the device for optimizing the mix proportion of lightweight foam insulation material from solid waste coal gangue. The device for optimizing the mix proportion of lightweight foam insulation material from solid waste coal gangue can be implemented in hardware or software. The UE can be a terminal device such as a smartphone, tablet computer, laptop computer, handheld computer, desktop computer, or personal digital assistant (PDA).
[0026] Figure 1 This is a schematic flowchart of an embodiment of the method for optimizing the mix proportion of lightweight foam insulation material made from solid waste coal gangue provided by the present invention. Figure 1 As shown, the method for optimizing the mix proportion of lightweight foam insulation material made from solid waste coal gangue includes: S101. Using a preset material performance prediction model, prediction is made based on the raw material usage data in the target material to obtain the corresponding material performance data. The target material is a lightweight foam insulation material made from solid waste coal gangue, and the preset material performance prediction model is a machine learning model based on support vector machine.
[0027] The preset material performance prediction model can be obtained by adjusting or improving the model parameters of the machine learning model based on Support Vector Machine (SVM) so that the material performance prediction model can accurately predict material performance data under different raw material usage data. Since SVM can map low-dimensional input (such as raw material usage data) to high-dimensional space through kernel function, it can effectively capture the nonlinear relationship between material performance and complex mix ratios, accurately reflect the relationship between material performance and raw material composition. Moreover, SVM can avoid overfitting for small samples. Compared with neural networks, it does not need to rely on a large number of samples, ensuring the accuracy and reliability of model prediction.
[0028] The target material is a lightweight foam insulation material made from solid waste coal gangue, which comprises metakaolin, coal gangue, fly ash, alkali activator, hydrogen peroxide, water, calcium stearate, and polypropylene fiber. The raw material usage data refers to the mass of each raw material in the target material, used to characterize the proportions of the raw materials, such as the mixing ratio.
[0029] Specifically, by using a machine learning model based on support vector machines, the material performance data of the target material corresponding to the raw material usage data is predicted. By utilizing the nonlinear prediction model of the preset material performance prediction model, the relationship between material performance and composition can be accurately reflected, realizing the rapid mapping between raw material usage data and material performance data. This improves the accuracy of material performance prediction based on raw material usage data and provides an evaluation benchmark for subsequent mix design optimization.
[0030] S102. Based on the material performance data and the constraints corresponding to the process and quality of the raw materials in the target material, determine the constraints corresponding to the mix proportion optimization model to be constructed.
[0031] Among them, the limiting conditions of the target material refer to the process limiting conditions of the lightweight foam insulation material made from solid waste coal gangue, such as the water-cement ratio of the target material being within the normal range.
[0032] The mix proportion optimization model to be constructed can be a multi-objective constrained optimization mathematical model. It uses raw material usage data as decision variables, determines constraints based on performance material data, and combines these with the set optimization objectives to form the mix proportion optimization model. The mix proportion is determined by solving the constructed mix proportion optimization model.
[0033] The constraints are to use the mix proportion optimization model to be constructed to find the optimal solution and delineate the feasible region to ensure the reliability of the mix proportion optimization results.
[0034] By using material property data (compressive strength, flexural strength, dry density, etc.) and target material constraints (specification limits, total weight, water-to-binder ratio, etc.) to jointly define the feasible region, the original high-dimensional continuous space is clipped into an "effective subspace", which significantly reduces the number of invalid iterations of the firefly algorithm and improves the convergence speed.
[0035] Specifically, by utilizing material performance data and the constraints of the target material, the feasible region is jointly defined, thereby determining the constraints. This allows the original high-dimensional continuous space to be truncated into an effective subspace, reducing the number of invalid iterations in subsequent model construction, improving the convergence speed, and thus enhancing the construction efficiency and accuracy of the mix ratio optimization model.
[0036] In one specific implementation, the mix proportion optimization model to be constructed includes the following: Decision variable x = [w1,w2,…,w8]: w1,w2,…,w8 are the absolute mass or volume of metakaolin, coal gangue, fly ash, alkali activator, hydrogen peroxide, water, calcium stearate, and polypropylene fiber in each raw material. The objective function is F(x) = f1(x), where f1(x) represents maximizing the amount of solid waste mixed in; The constraints are determined based on the material performance data and the limitations of the target material.
[0037] S103. Based on the preset optimization objective and the constraints, a mix proportion optimization model is constructed using the firefly algorithm, and the optimal mix proportion of the target material is output based on the mix proportion optimization model.
[0038] The preset optimization objective is a pre-defined optimization goal, such as maximizing the amount of solid waste mixed in.
[0039] The optimal mix proportion is the optimal solution output by the mix proportion optimization model, that is, the mix proportion of raw material usage in a set of target materials output by the mix proportion optimization model.
[0040] Specifically, based on the preset optimization objectives and the aforementioned constraints, a mix proportion optimization model is constructed using the firefly algorithm. Since the firefly algorithm operates through a "brightness-attraction" mechanism, it can perform large-scale global exploration in the early stages and fine local mining in the later stages by using a decay step size, thus avoiding getting trapped in local optima. This achieves both global and local dual search, improving the optimization efficiency of the mix proportion. At the same time, it realizes the global optimization of the mix proportion of lightweight foam insulation material made from solid waste coal gangue, improving the accuracy of the mix proportion optimization results and providing an optimization scheme for the material preparation process.
[0041] Understandably, through the synergy of the SVM model and the Firefly algorithm, the SVM model provides accurate performance predictions, while the Firefly algorithm, through a global optimization strategy, quickly finds the optimal solution, improving the optimization efficiency of the mix ratio and the accuracy of the optimization results.
[0042] In summary, the mix proportion optimization method for lightweight foam insulation material made from solid waste coal gangue provided in this embodiment of the invention utilizes a preset material performance prediction model to predict the material performance based on the raw material usage data in the target material, thereby obtaining the corresponding material performance data. The target material is lightweight foam insulation material made from solid waste coal gangue, and the preset material performance prediction model is a machine learning model based on support vector machines. By utilizing the nonlinear prediction model of the preset material performance prediction model, the relationship between material performance and composition can be accurately reflected, achieving rapid mapping between raw material usage data and material performance data, thus improving the accuracy of material performance prediction based on raw material usage data. Based on the material performance data and the constraints of the target material, the constraints corresponding to the mix proportion optimization model to be constructed are determined, and the material performance data and the constraints of the target material are used to... The feasible region is defined by the combination of factors, thereby determining the constraints and realizing the pruning of the original high-dimensional continuous space into an effective subspace. This reduces the number of invalid iterations in subsequent model construction, improves the convergence speed, and thus improves the construction efficiency and accuracy of the mix proportion optimization model. Based on the preset optimization objectives and constraints, the firefly algorithm is used to construct the mix proportion optimization model, which outputs the optimal mix proportion of the target material. Since the firefly algorithm uses the "brightness-attraction" mechanism, it can perform large-scale global exploration in the early stage and fine local mining in the later stage with the help of decay step size, avoiding getting trapped in local optima. It realizes both global and local search, improves the optimization efficiency of the mix proportion, and realizes global optimization of the mix proportion of lightweight foam insulation material from solid waste coal gangue, improving the accuracy of the mix proportion optimization results.
[0043] In some embodiments of the present invention, the material performance data includes the compressive strength, flexural strength, and dry density of the target material, and the limiting conditions include preset material ratio limiting conditions and total weight limiting conditions of the insulation material; step S102 includes: S201. The first constraint function corresponding to the compressive strength, the second constraint function corresponding to the flexural strength, the third constraint function corresponding to the dry density, the fourth constraint function corresponding to the preset material ratio limit condition, and the fifth constraint function corresponding to the total weight limit condition of the thermal insulation material are determined as the constraint conditions.
[0044] The preset material ratio constraint can be that the ratio of water to the total of other raw materials is between a specified minimum ratio and a maximum ratio. The corresponding expression for the fourth constraint function is as follows:
[0045] W w For the amount of water used, W p + W m +W f + W a This is the sum of the amounts of other raw materials used. R 1 min For the minimum ratio, R 1 max This is the maximum ratio.
[0046] The first constraint function corresponding to the compressive strength can be the constraint function corresponding to the compressive strength at 3 days, 7 days, and 28 days being greater than the specified minimum compressive strength at 3 days, 7 days, and 28 days, respectively. The expression of the corresponding first constraint function is as follows:
[0047]
[0048]
[0049] , , These represent the compressive strength at 3 days, 7 days, and 28 days of age, respectively. , , These represent the minimum compressive strength at 3 days, 7 days, and 28 days of age, respectively.
[0050] The second constraint function corresponding to the flexural strength can be one where the flexural strength is greater than the minimum flexural strength. The expression for the corresponding second constraint function is as follows:
[0051] Indicates flexural strength. This indicates the minimum flexural strength.
[0052] The third constraint function corresponding to the dry density can be one where the dry density is greater than the minimum dry density, and its expression is as follows:
[0053] Indicates dry density, This indicates the minimum dry density.
[0054] The total weight limit for thermal insulation material is that the total weight of raw materials must be between the specified minimum and maximum weights. The corresponding fifth constraint function expression is as follows:
[0055] R 2 min Minimum weight,R 2 max This is the maximum weight.
[0056] Specifically, the first constraint function corresponding to compressive strength, the second constraint function corresponding to flexural strength, the third constraint function corresponding to dry density, the fourth constraint function corresponding to the preset material ratio limit, and the fifth constraint function corresponding to the total weight limit of thermal insulation material are used as constraints. The combined effect of these five constraints can control multiple indicators such as performance, process, and cost within an acceptable range, reduce the search space, and make the final output of the optimal mix ratio more accurate and efficient.
[0057] In some embodiments of the present invention, the preset optimization objective includes maximizing the amount of solid waste mixed in; such as Figure 2 As shown, step S103 includes: S301. Generate several individual fireflies according to the constraints, and the initial position of each individual firefly corresponds to a set of initial mixing ratios. S302. In the current iteration step, based on the preset raw material usage data, determine the real-time position vector of each individual firefly, wherein each component of the real-time position vector represents the usage of each raw material in the preset raw material usage data. S303. Based on the real-time position vector, determine the brightness of each individual firefly, wherein the brightness is positively correlated with the amount of solid waste mixed in. S304. Based on the real-time position vector and brightness of each individual firefly, determine the attraction between the individual fireflies; S305. Based on the optimization objective, update the position and attraction of the real-time position vector of the individual firefly, wherein each position update corresponds to an update of the matching ratio. S306. Perform a non-dominated sort on each updated firefly individual to form a new Pareto front. S307. If the preset iteration criteria are met, the iteration is terminated, and the optimal solution in the current Pareto front is output to obtain the mix ratio optimization model. S308. If the preset iteration criteria are not met, return to step S302.
[0058] The solid waste blending amount refers to the total amount of coal gangue and fly ash used in the mixture. The optimization objective is to maximize the solid waste blending amount.
[0059] Specifically, several firefly individuals are generated according to the constraints, and the initial position of each firefly individual corresponds to an initial mix ratio. In the current iteration step, the real-time position vector of each firefly individual is determined according to the preset raw material usage data, and each component of the real-time position vector represents the usage of each raw material in the preset raw material usage data. Based on the real-time position vector, the brightness of each firefly individual is determined, and the brightness is positively correlated with the amount of solid waste mixed in. Based on the real-time position vector and the brightness of each firefly individual, the attraction between firefly individuals is determined. Based on the optimization objective, the position and attraction corresponding to the real-time position vector of the firefly individual are updated. Each position update corresponds to an update of the mix ratio. The updated firefly individuals are non-dominated and sorted to form a new Pareto front, realizing a closed-loop search of the position, brightness, and attraction of the firefly individuals. In addition, the Pareto front improves the search efficiency. At the same time, the brightness is directly correlated with the amount of solid waste mixed in, so that the calculation process evolves towards the direction of the maximum amount of solid waste mixed in, thereby improving the utilization rate of solid waste.
[0060] In one specific implementation, the firefly algorithm is used to establish a mix ratio optimization model as follows: (1) Initialize the population, that is, determine the initial parameters and the initial position of individual fireflies according to the constraints, that is, determine the initial matching ratio; (2) Based on the specific parameter combination of the current iteration step, accurately identify and calculate the real-time position of each member in the firefly swarm; (3) Determine the attraction value at the current iteration number based on the relative position and relative brightness. The darker fireflies are attracted by the fireflies with greater attraction. (4) Update the position and attraction of firefly particles, i.e. update the mix ratio; (5) Update the Pareto frontier; (6) If the stopping criterion has been met, the calculation is terminated; otherwise, the process jumps to step (2). Each iteration and update position is the update of the mix ratio.
[0061] The initialization formula for the population is:
[0062] t represents the number of iterations. This indicates the current position of an individual firefly. It's about attractiveness. It is a random item.
[0063]
[0064] in This represents the maximum attraction when r=0, and is usually taken as 1.
[0065] The distance between two individual fireflies is:
[0066] This indicates that any two fireflies i and j are in x i and x j The Cartesian distance is used to determine the real-time location of individual fireflies.
[0067] In one specific implementation, the updated individual fireflies are sorted non-dominated to form a new Pareto front, specifically as follows: The population size is set to 10, meaning that within a given range of raw material quantities, firefly particles are randomly distributed in space, generating 10 mix proportions according to random rules; the number of raw material types corresponds to the particle dimension, and their quantities correspond to the spatial coordinates of the particles; the steps for calculating the position and brightness of each firefly particle using the Pareto front solution set are as follows: Sorting: Sort the non-dominated solutions according to the objective function values.
[0068] Boundary solution processing: For the maximum and minimum values (boundary solutions) of each objective, directly assign them the maximum crowding distance.
[0069] Distance calculation: For each solution, calculate its distance to its neighboring solutions in the target space, using the following formula:
[0070] In the formula and These are the values of the adjacent solutions of solution i on the objective k. and These are the maximum and minimum values of the objective k, respectively. The optimal attraction is the maximum amount of solid waste to be mixed in, satisfying the constraints.
[0071] After establishing the mix proportion optimization model, the optimal solution is restored, that is, the calculated optimal parameter combination and its position are restored to the maximum solid waste admixture and the corresponding mix proportion, and the optimal mix proportion is output.
[0072] In some embodiments of the present invention, step S304 includes: S401. In each iteration step, the position and attractiveness of the firefly individual are updated according to the movement method of the step size factor corresponding to the individual firefly, so that the individual firefly searches in the direction corresponding to the optimization target. S402. If the updated position exceeds the preset boundary, then the position is corrected to the preset boundary.
[0073] Specifically, in each iteration step, a step-size factor-driven movement method is used to update the position and attractiveness of individual fireflies according to the position update equation and the attraction equation, respectively. This ensures that the fireflies search in the direction corresponding to the optimization target. If the updated position exceeds a preset boundary, it is forcibly pulled back to the boundary to ensure that the solution of the mix ratio optimization model is within the feasible mix ratio space and satisfies the constraints. Understandably, in this embodiment, while maintaining search diversity, it ensures that the mix ratio generated in each iteration step satisfies engineering constraints.
[0074] In one specific implementation, each firefly particle represents a potential solution, and its brightness is positively correlated with the objective function value. Fireflies calculate attraction based on differences in brightness and distance, updating their positions accordingly and moving towards brighter fireflies. Through iterative processes, the fireflies continuously approach the globally optimal position, ultimately finding the optimal or near-optimal solution to the problem. In the algorithm, each firefly maintains its historical optimal position ptopi, and the optimal positions of all particles are compared to obtain the globally optimal position wtop. The equation used in the firefly algorithm is: Attraction Equation:
[0075] Position update equation: ( )
[0076] i Indicates the first i One particle, d The first particle represents the second particle. d dimension, t Indicates the first t generation. It is the step size factor, used to control the step size of the firefly's movement. It is a random perturbation term, usually a random number vector generated from a Gaussian distribution, a uniform distribution, or other distributions.
[0077] In each iteration, each particle updates its position and attraction according to the position update equation and the attraction equation, that is, updates the mix proportion of lightweight foam insulation material for solid waste and coal gangue, so that it searches and moves towards the direction of maximum solid waste content. ,but ;if ,but .
[0078] In some embodiments of the present invention, the step size factor is obtained by adjusting it using a chaotic mapping operator.
[0079] Specifically, by introducing a chaotic mapping operator to dynamically adjust the step size factor during the iteration process of the firefly algorithm, the randomness and exploration ability of the algorithm can be increased, which helps to escape local optima, improve the probability of finding the global optimum and the convergence accuracy of the algorithm.
[0080] In one specific implementation, to avoid getting stuck in periodic behavior, a chaotic mapping operator mapping step size factor is introduced while still balancing global search and local fine-grained search. To improve the convergence accuracy and flexibility of the algorithm, the expression is as follows:
[0081] In the formula, This is the current iteration step size factor; is the initial step size factor; i is the current iteration number; The total number of iterations is used to control the decay rate of the step size, ensuring that the change in step size is reasonably distributed with the number of iterations.
[0082] In some embodiments of the present invention, the step of obtaining the preset material property prediction model includes: S501. Obtain the original support vector machine model, wherein the kernel function of the original support vector machine model includes the RBF kernel function; S502. The parameters of the original support vector machine model are tuned using a parameter tuning strategy based on Bayesian optimization to obtain the preset material performance prediction model.
[0083] Among these, tuning the parameters of the original support vector machine model can be done by tuning parameters such as kernel width γ and penalty coefficient C.
[0084] The parameter tuning strategy based on Bayesian optimization is as follows: using cross-validation accuracy and / or mean squared error as the objective function, evaluate the SVM performance corresponding to each (C, γ) pair on the validation set; update the Gaussian process surrogate model based on the cross-validation results until the maximum number of iterations is reached or the objective function converges; output the optimal (C, γ) combination and use it to train the final SVM model, where the training termination criterion is the maximum number of iterations or the performance improvement of two consecutive iterations being less than a set threshold.
[0085] Specifically, the kernel function of SVM is the RBF kernel function. Since the RBF kernel implicitly maps the original low-dimensional mix proportion space (such as 8-dimensional raw material usage) to an infinite-dimensional feature space, it makes the originally inseparable linear relationship separable, thereby accurately capturing the strong nonlinear coupling of raw material usage in the target material, significantly improving the prediction accuracy of compressive strength, flexural strength, and dry density. The parameters of the original support vector machine model are optimized using a parameter tuning strategy based on Bayesian optimization, overcoming the sensitivity problem of parameter selection in SVM, and providing a reliable performance evaluation benchmark for the mix proportion optimization of lightweight foam insulation material made from solid waste coal gangue.
[0086] In one specific implementation, such as Figure 3 The diagram shows the process of optimizing the mix proportion of lightweight foam insulation material made from solid waste coal gangue. By using SVM prediction and the Firefly algorithm in synergistic calculation, the solution from raw material usage to the optimal mix proportion can be obtained quickly and accurately.
[0087] The following are application examples of this invention. A company needs to design lightweight foam insulation materials made from solid waste coal gangue with a dry density of 300-350 kg / m³ and a compressive strength of 0.6-0.7 MPa, and a dry density of 350-400 kg / m³ and a compressive strength of 0.8-0.9 MPa. The raw materials selected include metakaolin, coal gangue, fly ash, alkali activator, hydrogen peroxide, water, calcium stearate, and polypropylene fiber. The quality indicators of each raw material are as follows: cement is grade P.042.5, with 3-day and 28-day compressive strengths of 23.7 MPa and 47 MPa, respectively; metakaolin is an aluminosilicate material; fly ash is grade I; coal gangue is grade I; the alkali activator is prepared by adding sodium hydroxide flakes to sodium silicate water glass; the water glass modulus is 1.2; the foaming agent is 30% H2O2 diluted with water; the polypropylene fiber content is 1.5%; and the foam stabilizer content is 0.3%. Lightweight foam insulation materials made from solid waste coal gangue were designed using the "Standard Specification for Foamed Concrete" (JGT266-2011) and the two methods in the embodiments of this application, as shown in Tables 1 and 2. From Tables 1 and 2, it can be seen that the optimized method in the application examples of this invention has a higher maximum solid waste ratio than the traditional method. Table 1. Application Example 1 of the Invention (dry density 300-350 kg / m³) 3 Unit: kg / m³
[0088] Table 2. Application Example 2 of the Invention (dry density 350-400 kg / m³) 3 Unit: kg / m³
[0089] To better implement the mix proportion optimization method for lightweight foam insulation material made from solid waste coal gangue in this embodiment of the invention, based on the existing method, the corresponding method is as follows: Figure 4 As shown, this embodiment of the invention also provides a device for optimizing the mix proportion of lightweight foam insulation material made from solid waste coal gangue. The device 400 for optimizing the mix proportion of lightweight foam insulation material made from solid waste coal gangue includes: The acquisition unit 401 is used to predict the corresponding material performance data based on the raw material usage data in the target material using a preset material performance prediction model. The target material is a lightweight foam insulation material made from solid waste coal gangue, and the preset material performance prediction model is a machine learning model based on support vector machine. The determining unit 402 is used to determine the constraints corresponding to the mix proportion optimization model to be constructed based on the material performance data and the constraints corresponding to the process and quality of the raw materials in the target material. The optimization unit 403 is used to construct a mix proportion optimization model based on the preset optimization objective and the constraints, using the firefly algorithm, and to output the optimal mix proportion of the target material based on the mix proportion optimization model.
[0090] The solid waste coal gangue lightweight foam insulation material mix ratio optimization device 400 provided in the above embodiments can realize the technical solutions described in the above embodiments of the solid waste coal gangue lightweight foam insulation material mix ratio optimization method. The specific implementation principles of each module or unit can be found in the corresponding content in the above embodiments of the solid waste coal gangue lightweight foam insulation material mix ratio optimization method, which will not be repeated here.
[0091] like Figure 5 As shown, the present invention also provides an electronic device 500. The electronic device 500 includes a processor 501, a memory 502, and a display 503. Figure 5 Only some components of the electronic device 500 are shown, but it should be understood that it is not required to implement all the components shown, and more or fewer components may be implemented instead.
[0092] In some embodiments, processor 501 may be a central processing unit (CPU), microprocessor, or other data processing chip, used to run program code stored in memory 502 or process data, such as the method for optimizing the mix proportion of lightweight foam insulation material made from solid waste coal gangue in this invention.
[0093] In some embodiments, processor 501 may be a single server or a group of servers. The server group may be centralized or distributed. In some embodiments, processor 501 may be local or remote. In some embodiments, processor 501 may be implemented on a cloud platform. In one embodiment, the cloud platform may include a private cloud, public cloud, hybrid cloud, community cloud, distributed cloud, inter-cloud, multi-cloud, or any combination thereof.
[0094] In some embodiments, memory 502 may be an internal storage unit of electronic device 500, such as a hard disk or memory of electronic device 500. In other embodiments, memory 502 may also be an external storage device of electronic device 500, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc. equipped on electronic device 500.
[0095] Furthermore, the memory 502 may include both internal storage units of the electronic device 500 and external storage devices. The memory 502 is used to store application software and various types of data installed on the electronic device 500.
[0096] In some embodiments, display 503 may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, or an OLED (Organic Light-Emitting Diode) touchscreen. Display 503 is used to display information from electronic device 500 and to display a visual user interface. Components 501-503 of electronic device 500 communicate with each other via a system bus.
[0097] In one embodiment, when the processor 501 executes the solid waste coal gangue lightweight foam insulation material mix ratio optimization program stored in the memory 502, the following steps can be implemented: The material performance prediction model is used to predict the material performance based on the raw material usage data in the target material, and the corresponding material performance data is obtained. The target material is a lightweight foam insulation material made from solid waste coal gangue, and the preset material performance prediction model is a machine learning model based on support vector machine. Based on the material performance data and the constraints corresponding to the process and quality of the raw materials in the target material, the constraints corresponding to the mix proportion optimization model to be constructed are determined. Based on the preset optimization objective and the constraints, a mix proportion optimization model is constructed using the firefly algorithm, and the optimal mix proportion of the target material is output based on the mix proportion optimization model.
[0098] It should be understood that when the processor 501 executes the solid waste coal gangue lightweight foam insulation material mix ratio optimization program in the memory 502, in addition to the functions mentioned above, it can also perform other functions, as can be found in the description of the corresponding method embodiments above.
[0099] Furthermore, this embodiment of the invention does not specifically limit the type of electronic device 500 mentioned. Electronic device 500 can be a mobile phone, tablet computer, personal digital assistant (PDA), wearable device, laptop computer, or other portable electronic device. Exemplary embodiments of portable electronic devices include, but are not limited to, portable electronic devices running iOS, Android, Microsoft, or other operating systems. The aforementioned portable electronic device can also be other portable electronic devices, such as a laptop computer with a touch-sensitive surface (e.g., a touch panel). It should also be understood that in some other embodiments of the invention, electronic device 500 may not be a portable electronic device, but rather a desktop computer with a touch-sensitive surface (e.g., a touch panel).
[0100] Accordingly, this application also provides a computer-readable storage medium for storing computer-readable programs or instructions. When the programs or instructions are executed by a processor, they can implement the steps or functions in the method for optimizing the mix proportion of lightweight foam insulation material from solid waste coal gangue provided in the above-described method embodiments.
[0101] Those skilled in the art will understand that all or part of the processes of the methods described in the above embodiments can be implemented by a computer program instructing related hardware (such as a processor, controller, etc.), and the computer program can be stored in a computer-readable storage medium. The computer-readable storage medium may be a disk, optical disk, read-only memory, or random access memory, etc.
[0102] The above provides a detailed description of the method, apparatus, electronic equipment, and storage medium for optimizing the mix proportion of lightweight foam insulation material made from solid waste coal gangue provided by this invention. Specific examples have been used to illustrate the principles and implementation methods of this invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of this invention. At the same time, those skilled in the art will recognize that there will be changes in the specific implementation methods and application scope based on the ideas of this invention. Therefore, the content of this specification should not be construed as a limitation of this invention.
Claims
1. A method for optimizing the mix proportion of lightweight foam insulation material made from solid waste coal gangue, characterized in that, include: The material performance prediction model is used to predict the material performance based on the raw material usage data in the target material, and the corresponding material performance data is obtained. The target material is a lightweight foam insulation material made from solid waste coal gangue, and the preset material performance prediction model is a machine learning model based on support vector machine. Based on the material performance data and the constraints corresponding to the process and quality of the raw materials in the target material, the constraints corresponding to the mix proportion optimization model to be constructed are determined. Based on the preset optimization objective and the constraints, a mix proportion optimization model is constructed using the firefly algorithm, and the optimal mix proportion of the target material is output based on the mix proportion optimization model.
2. The method for optimizing the mix proportion of lightweight foam insulation material for solid waste coal gangue according to claim 1, characterized in that, The material performance data includes the compressive strength, flexural strength, and dry density of the target material, and the limiting conditions include preset material ratio limiting conditions and total weight limiting conditions of the insulation material. The constraints for determining the mix proportion optimization model to be constructed, based on the material performance data and the constraints corresponding to the process and quality of the raw materials in the target material, include: The first constraint function corresponding to the compressive strength, the second constraint function corresponding to the flexural strength, the third constraint function corresponding to the dry density, the fourth constraint function corresponding to the preset material ratio limit condition, and the fifth constraint function corresponding to the total weight limit condition of the thermal insulation material are determined as the constraint conditions.
3. The method for optimizing the mix proportion of lightweight foam insulation material for solid waste coal gangue according to claim 1, characterized in that, The preset optimization objective includes maximizing the amount of solid waste mixed in; the process of constructing a mix proportion optimization model using the firefly algorithm based on the preset optimization objective and the constraints includes: Several individual fireflies are generated according to the constraints, and the initial position of each individual firefly corresponds to a set of initial mixing ratios. In the current iteration step, based on the preset raw material usage data, the real-time position vector of each individual firefly is determined, and each component of the real-time position vector represents the usage of each raw material in the preset raw material usage data. Based on the real-time position vector, the brightness of each individual firefly is determined, and the brightness is positively correlated with the amount of solid waste mixed in. The attraction between the individual fireflies is determined based on the real-time position vector and brightness of each individual firefly. Based on the optimization objective, the position and attraction of the real-time position vector of the individual firefly are updated, wherein each position update corresponds to an update of the matching ratio. Each firefly individual after the update is sorted in a non-dominated order to form a new Pareto front. If the preset iteration criteria are met, the iteration is terminated, and the optimal solution in the current Pareto front is output to obtain the mix ratio optimization model; If the preset iteration criteria are not met, the process returns to the step of determining the real-time position vector of each firefly individual based on the preset raw material usage data in the current iteration step. Each component of the real-time position vector represents the usage amount of each raw material in the preset raw material usage data.
4. The method for optimizing the mix proportion of lightweight foam insulation material for solid waste coal gangue according to claim 3, characterized in that, Determining the attraction between individual fireflies based on their real-time position vector and brightness includes: Based on the real-time position vector and brightness of each individual firefly, the relative positions and relative brightness among the fireflies are determined respectively. The attraction is calculated based on the relative position and the relative brightness.
5. The method for optimizing the mix proportion of lightweight foam insulation material for solid waste coal gangue according to claim 3, characterized in that, The step of updating the position and attraction of the individual firefly based on the optimization objective includes: In each iteration step, the position and attractiveness of the firefly individual are updated according to the movement method of the step size factor corresponding to the individual firefly, so that the individual firefly searches in the direction corresponding to the optimization target; If the updated position exceeds the preset boundary, the position will be corrected to the preset boundary.
6. The method for optimizing the mix proportion of lightweight foam insulation material for solid waste coal gangue according to claim 5, characterized in that, The step size factor is obtained by adjusting it using a chaotic mapping operator.
7. The method for optimizing the mix proportion of lightweight foam insulation material for solid waste coal gangue according to claim 1, characterized in that, The steps for obtaining the preset material property prediction model include: Obtain the original support vector machine model, wherein the kernel function of the original support vector machine model includes the RBF kernel function; The parameters of the original support vector machine model are tuned using a Bayesian optimization-based parameter tuning strategy to obtain the preset material property prediction model.
8. A device for optimizing the mix proportion of lightweight foam insulation material from solid waste coal gangue, characterized in that, include: The acquisition unit is used to predict the corresponding material performance data based on the raw material usage data in the target material using a preset material performance prediction model. The target material is a lightweight foam insulation material made from solid waste coal gangue, and the preset material performance prediction model is a machine learning model based on support vector machine. The determining unit is used to determine the constraints corresponding to the mix proportion optimization model to be constructed based on the material performance data and the constraints corresponding to the process and quality of the raw materials in the target material. The optimization unit is used to construct a mix proportion optimization model based on the preset optimization objective and the constraints, and to output the optimal mix proportion of the target material based on the mix proportion optimization model.
9. An electronic device, characterized in that, It includes a memory and a processor, wherein the memory is used to store programs; The processor, coupled to the memory, is used to execute the program stored in the memory to implement the steps in the method for optimizing the mix proportion of lightweight foam insulation material from solid waste coal gangue as described in any one of claims 1 to 7.
10. A computer-readable storage medium, characterized in that, Used to store computer-readable programs or instructions, which, when executed by a processor, can implement the steps in the method for optimizing the mix proportion of lightweight foam insulation material from solid waste coal gangue as described in any one of claims 1 to 7.