Asphalt mixture ratio calculation method, equipment and medium based on heuristic algorithm

Abstract

The present invention relates to a method, device and medium for calculating asphalt mixture proportion based on a heuristic algorithm. The method comprises the following steps: obtaining target mix proportion curve data and mineral material screening data; initializing a solution space including multiple mix proportion combinations based on the mineral material screening data; taking the value of the objective function of minimizing the synthetic mix proportion curve corresponding to the mix proportion combination and the target mix proportion curve as the optimization target, screening some mix proportion combinations from the current solution space based on a bottom guarantee mechanism, and generating new mix proportion combinations based on an update strategy to form a new solution space and complete one iteration; repeating the iteration multiple times to obtain the mix proportion combination with the highest similarity to the target mix proportion curve. Compared with the prior art, the present invention has the advantages of improving the efficiency of mix proportion calculation and avoiding falling into a local optimum.

Description

Technical Field

[0001] The invention relates to the technical field of civil engineering, and in particular to a method, a device and a medium for calculating the proportion of asphalt mixture based on a heuristic algorithm. Background Art

[0002] Among asphalt mixture proportioning methods, the commonly used methods include empirical proportioning. Empirical proportioning is a ratio relationship based on experience and past practices, the best ratio obtained through laboratory tests and performance tests, or a method of optimizing design based on specific application requirements and material performance indicators. The empirical proportioning method usually lacks scientific and engineering analysis, so experience cannot provide in-depth theoretical support. It may be limited in complex projects and situations with high requirements for quality, performance and sustainability. Although the best ratio can be obtained through laboratory tests and performance tests, it is time-consuming and cost-intensive, especially in engineering projects that require high accuracy, sustainability and performance optimization. It may not meet people's needs. At the same time, it also requires more resources, data and engineering knowledge, and may not be applicable to all project-specific requirements.

[0003] The ultimate goal of the heuristic algorithm-based rapid asphalt mixture ratio design method for optimization is to simply and quickly obtain the optimal asphalt mixture ratio design under different conditions. This method has many advantages, including improving the accuracy, flexibility, efficiency and repeatability of the design. This enables engineers to better design materials that can better meet the differentiated performance requirements to meet project needs, while reducing the risk of errors and saving time and costs. Therefore, the heuristic algorithm-based rapid asphalt mixture ratio design method is particularly important for accurately and quickly determining the asphalt mixture ratio.

[0004] Traditional asphalt mixture proportioning methods have some significant disadvantages, including unsustainability, inefficiency, energy consumption, environmental impact, utilization of recycled materials, and changes in physical and chemical properties that are difficult to cope with. Therefore, many engineering projects have gradually turned to more modern methods, such as model-based and computer simulation methods, to improve the accuracy, efficiency and repeatability of the design. These methods can better meet different performance and mathematical continuity requirements and reduce design costs. However, most of the existing related methods have unsatisfactory accuracy and are prone to local optimality. Summary of the invention

[0005] The purpose of the present invention is to overcome the defects of the above-mentioned prior art and to provide an asphalt mixture ratio calculation method, equipment and medium based on a heuristic algorithm to improve the ratio calculation efficiency.

[0006] The purpose of the present invention can be achieved by the following technical solutions:

[0007] One aspect of the present invention provides a method for calculating asphalt mixture ratio based on a heuristic algorithm, comprising the following steps:

[0008] Obtain target mix ratio curve data and mineral material screening data;

[0009] Initializing a solution space including a plurality of proportion combinations based on the ore screening data;

[0010] Taking the minimization of the value of the objective function as the optimization goal, selecting some proportion combinations from the current solution space based on the bottom guarantee mechanism, and generating new proportion combinations based on the update strategy to form a new solution space, completing one iteration, wherein the objective function is constructed based on the target mix curve and the synthetic mix curve corresponding to the mix combination;

[0011] Repeat the iteration multiple times to obtain the mix ratio combination with the highest similarity to the target mix ratio curve.

[0012] As a preferred technical solution, the solution space is:

[0013] C m =P·C T

[0014] Among them, C m is the solution space composed of m synthetic mix ratio curves, P is the m×n ratio matrix that satisfies uniform distribution, C=[C1,C2,C3,…C n ] is the screening curve of the candidate ore, T represents the transposition, and the i-th row P of the matrix P i =[P i1 ,P i2 ,P i3 ,…P in ],

[0015] As a preferred technical solution, the optimization goal is to minimize the value of the following objective function:

[0016]

[0017] Among them, SSE j represents the objective function value of the jth composite mix ratio curve, where C j =P j1 C1+P j2 C2+P j3 C3+…+P jn C n ,j=1,2,3…m,

[0018] As a preferred technical solution, the process of screening the matching combination based on the bottom guarantee mechanism includes:

[0019] Calculate the value of the objective function corresponding to each ratio combination in the current solution space and perform normalization processing;

[0020] Generate a random probability value, select a ratio combination based on the probability value and the normalized value of the objective function, and repeat this step multiple times.

[0021] As a preferred technical solution, during the screening process, the probability of a matching combination being selected is inversely proportional to the value of the corresponding objective function.

[0022] As a preferred technical solution, the process of generating a new ratio combination based on the update strategy includes:

[0023] For a matching combination in the current solution space, a crossover operation is performed with a preset probability.

[0024] As a preferred technical solution, the crossover operation includes exchanging the corresponding proportions of any two mineral materials within the same combination.

[0025] As a preferred technical solution, the mineral material screening data includes candidate mineral material types, particle sizes and screening curve information.

[0026] Another aspect of the present invention provides an electronic device, comprising: one or more processors and a memory, wherein the memory stores one or more programs, and the one or more programs include instructions for executing the above-mentioned asphalt mixture ratio calculation method based on the heuristic algorithm.

[0027] Another aspect of the present invention provides a computer-readable storage medium, comprising one or more programs for execution by one or more processors of an electronic device, wherein the one or more programs include instructions for executing the above-mentioned asphalt mixture proportion calculation method based on heuristic algorithm.

[0028] Compared with the prior art, the present invention has the following beneficial effects:

[0029] (1) Improving the efficiency of mix ratio calculation: The present application randomly generates multiple mix ratio combinations, and performs multiple iterations with the optimization goal of minimizing the value of the objective function between the synthetic mix ratio curve corresponding to the mix ratio combination and the target mix ratio curve. Finally, a mixture ratio close to the target mix ratio curve is obtained. The final designed asphalt mixture gradation can be accurately predicted through design parameters, which is conducive to quickly optimizing the design ratio of the asphalt mixture and obtaining the expected effect. It has a positive significance for improving the accuracy and speed of the mixture ratio.

[0030] (2) Avoiding falling into local optimality: This application introduces a bottom-line mechanism and an optimization strategy in the iterative process. By introducing randomness, local optimal solutions can be avoided, thereby increasing the possibility of finding a global optimal solution. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 It is a schematic diagram of a flow chart of a method for calculating asphalt mixture ratio based on a heuristic algorithm in an embodiment;

[0032] Figure 2 Schematic diagram of commonly used mixing ratio types in the embodiments;

[0033] Figure 3 This is the design diagram of the best material ratio for each gear in the embodiment;

[0034] Figure 4 The optimal solution gradation curve and the gradation median comparison chart are output in the embodiment.

[0035] Figure 5 It is the first generation iterative evolution grading curve in the embodiment.

[0036] Figure 6 It is the initial iterative evolution error value curve in the embodiment.

[0037] Figure 7 It is the third generation iterative evolution grading curve in the embodiment.

[0038] Figure 8 It is the third generation iterative evolution error value curve in the embodiment.

[0039] Fig. 9 It is the 46th generation iterative evolution grading curve in the embodiment.

[0040] Fig.10 It is the forty-sixth generation iterative evolution error value curve in the embodiment.

[0041] Fig.11 This is the fitness evolution curve of the forty-sixth generation in the embodiment.

[0042] Fig.12 This is the 100th generation iterative evolution matching curve in the embodiment.

[0043] Fig.13 It is the 100th generation iterative evolution error value curve in the embodiment.

[0044] Fig.14 It is the optimization result and the gradation median curve in the embodiment. DETAILED DESCRIPTION

[0045] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0046] Example 1

[0047] In view of the problems existing in the above-mentioned prior art, this embodiment provides a method for calculating asphalt mixture ratio based on a heuristic algorithm, see Figure 1 , the method comprises the following steps:

[0048] Step S0: Before the formal calculation, the target mix ratio of mineral aggregates (i.e., target mix ratio curve data) and the key factors affecting the gradation of asphalt mixture are determined in advance according to the requirements of the road project for materials, and the target mix ratio curve is used as the optimization final value. Among them, the key factors include aggregate type and particle size, aggregate ratio, target gradation, and other parameter ratios. The types of mix ratios usually involved in road projects are: dense mix, discontinuous gradation, and open gradation, such as Figure 2 shown.

[0049] Step S1: Use relevant software and heuristic algorithms to establish a data model of the above-mentioned factors and aggregate ratio optimization method. The main significance of the data model is that it describes how to use heuristic algorithms to solve a complex optimization problem. Through iterative evolution, the optimal solution to the problem is gradually found. The algorithm initially performs some initialization settings, including the setting of parameters such as the solution space size, update probability, single solution dimension, solution space range and number of iterations, and then enters the main program to loop in order to find the optimal solution. Further, this algorithm uses a bottom-line mechanism to use fitness to select a single solution data. The fitness value is usually realized through the objective function of the problem, which indicates the performance or pros and cons of any single solution in the solution space. It is guaranteed that after the heuristic algorithm iteration is completed, the final code finds the location and value of the optimal solution and displays it on the screen.

[0050] Specifically, according to the aggregate particle size, the type of raw material n constituting the mixture is determined, and a screening experiment is performed to obtain the screening curve C = [C1, C2, C3, ... C n ] (i.e., obtaining the ore screening data), and then generating a ratio matrix P of m×n quantity satisfying uniform distribution, P i =[P i1 ,P i2 ,P i3 ,…P in ]; 1; i = 1, 2, 3 ... m. After that, P·CT , and obtain m synthetic mix ratio curves C m , forming a solution space, where C j =P j1 C1+P j2 C2+P j3 C3+…+P jn C n ,j=1,2,3…m, Calculate C j With the average The sum of squares of errors (SSE) of the data points between them is calculated by the formula: Take "minimum SSE" as the optimization goal.

[0051] In this step, some algorithm parameters are first defined, including the matching size, matching string length, update probability, solution space size, and number of iterations.

[0052] Randomly generate matching combinations, which are sets of multiple solutions, where each single solution is a decimal, in order to calculate the objective function value.

[0053] Calculate the objective function value of the initial ratio.

[0054] A graph of the initial proportions of the calling function, including the distribution of proportions in the solution space and the change in the objective function value.

[0055] Initialize and record the optimal ratio and solution ratio of each generation, and record the default optimal ratio.

[0056] Enter the main loop and execute the iterative process of the heuristic algorithm. In each generation:

[0057] Compute the objective function value of a single solution from the previous generation.

[0058] The fitness (i.e., the objective function value) is calculated based on the objective function value.

[0059] Duplicate a single solution to create a new generation of solution space.

[0060] Perform optimization strategy operations.

[0061] Select the best single solution and keep it.

[0062] Update the ratio, calculate the objective function value of the solution space, and record the optimal value and the corresponding solution.

[0063] After the loop is completed, the location and value of the optimal solution are found and displayed.

[0064] Step S2: Use the guarantee mechanism in the heuristic algorithm to select an excellent ratio combination. The process of seeking the optimal solution mainly includes the guarantee mechanism and the update strategy. The guarantee mechanism calculates the fitness and replicates the solution with a lower fitness value with a higher probability.

[0065] Among them, the selection strategy of excellent matching combinations will play a decisive role in the performance of the algorithm. In order to transform the existing solution space into the next generation solution space, this method selects excellent single solutions from the old solution space for replication. The selection is based on the fitness value of the single solution. The single solution with a low fitness value is replicated to reproduce it; the single solution with a high fitness value is replaced by the single solution with a low fitness value to make it disappear.

[0066] In this embodiment, the method of fitness ratio selection is adopted, in which the expected value of any single solution being selected is related to the ratio of its fitness and the average fitness of the solution space, and a bottom guarantee mechanism is adopted. First, the fitness of each single solution is calculated, and then the proportion of this fitness in the total fitness of the solution space is calculated, which represents the probability of the single solution being selected in the selection process. The selection process embodies the idea of ​​"survival of the fittest" in the biological evolution process, and ensures that good genes are inherited by the next generation of single solutions.

[0067] The guarantee mechanism makes it easier for single solutions with the same fitness to be selected, thereby increasing the representativeness of these single solutions in the next generation and helping the heuristic algorithm to find better solutions faster. Although the guarantee mechanism tends to select single solutions with lower fitness values, it still provides a certain chance for single solutions with higher fitness values ​​to be selected to maintain the diversity of the ratio and prevent the local optimal solution of the gradient too early. At the same time, the guarantee mechanism introduces a certain degree of randomness, so that the selection of a single solution is not completely deterministic, which helps to avoid the algorithm taking the same strategy in each generation, so as to better explore the space.

[0068] Specifically, the guarantee mechanism first needs to calculate the fitness value of each single solution. Then, the probability of being selected is calculated for each single solution. The probability is usually inversely proportional to the fitness of the single solution, that is, the single solution with lower fitness has a greater probability of being selected.

[0069] This can be achieved by following these steps:

[0070] Normalize the fitness values ​​of all individual solutions in the population so that their sum is equal to 1. This can be achieved by dividing the fitness value of each individual solution by the sum of the fitness values ​​of all individual solutions. Create a floor selection mechanism whose size is the same as the number of individual solutions in the solution space. The size of each region of the floor mechanism is proportional to the selection probability of each individual solution. Generate a random probability value (usually a random number between 0 and 1) and then select it individually on the floor mechanism. The selection method is to determine whether a single solution is selected by comparing the random probability value with the boundaries of each region on the floor mechanism. If the random probability value belongs to a certain region, then the corresponding single solution will be selected. Repeat the above selection process until a sufficient number of single solutions are selected to construct the next generation of solution space. Generally, individual solutions with lower fitness values ​​have a greater probability of being selected, and therefore are more likely to pass their genetic information to the next generation.

[0071] Step S3, cross-optimize the single solution within the ratio combination to ensure that the single solution within the ratio has a lower fitness value, and continuously copy the single solution within the appropriate ratio combination to implement the survival of the fittest system to ensure that the final output result is the optimal solution. The update strategy adopts the intra-group optimization strategy, that is, any two ratios in the group are crossed with a certain probability.

[0072] The performance of S2 is improved by the new ratio combination generated by replication, but it cannot generate new single solutions. In order to generate new single solutions, the heuristic algorithm imitates the gene recombination process of sexual reproduction in nature, performs random permutation optimization on chromosomes (strings), inherits the original good genes to the next generation of single solutions, and generates new single solutions containing more complex gene structures. Whether a single solution in the heuristic algorithm mutates can be randomly selected or determined by selection operator calculation. The optimized mothers are all new generation single solutions generated by replication from the previous generation. The operation of the optimization strategy can ensure that the optimization process of the heuristic algorithm can converge to the global optimal point and play an important role in improving the convergence speed of the optimization process.

[0073] In order to ensure the quality and diversity of solutions, an optimization strategy is adopted to evolve within a group of aggregate ratio combinations to prevent the algorithm gradient from reaching the local optimal solution, improve the global search capability, give full play to the advantages of direct search and random search, and continuously establish and improve individual solutions. The optimization strategy operation helps to generate better solutions, which can produce results close to more optimal solutions in optimization problems.

[0074] Step S4, repeat step S3 until the solution space converges to the global optimal solution.

[0075] By running the completed heuristic algorithm, many different grading curves are generated with random proportions in a global scope. The root mean square difference between the solved grading curve and the target grading is the smallest, which means that the final grading curve image is basically similar to the target grading curve. At this time, it means that a set of data closest to the target gradation of asphalt mixture has been found, and outputting this set of data is the optimal solution.

[0076] The following is an explanation with reference to specific embodiments.

[0077] Step 0, according to the requirements of road engineering for materials, determine the target mix ratio of mineral aggregates and the key factors affecting the gradation of asphalt mixture, and use the target mix ratio curve as the optimization final value. Usually, the mix ratio types involved in road engineering are: dense mix, discontinuous gradation and open gradation.

[0078] This example takes AC-13 asphalt mixture as an example. In actual engineering, this type of mixture usually divides the crushed aggregate into four grades, namely 0-3, 3-5, 5-10, and 10-15 (mm). AC-13 asphalt mixture is commonly used in road construction such as highways, expressways, and airport runways. It generally has high strength and durability and can withstand the impact of traffic loads and climate change on the road surface. Its specific uses and performance characteristics may vary depending on the region and project requirements. It provides good structural, water-resistant, anti-gap and anti-rutting properties, and is suitable for road construction under most climate conditions. It is a more commonly used mix in road construction.

[0079] This example is optimized with the median of the upper and lower limits of the AC-13 asphalt mixture gradation range as the target.

[0080] The key factors affecting asphalt mixture gradation mainly include the following:

[0081] Aggregate type and aggregate size distribution: The combination of different types and sizes of aggregates will affect the stability and strength characteristics of the asphalt mixture.

[0082] Asphalt viscosity and binder content: The viscosity and binder content of asphalt determine the bonding performance and stability of the mixture.

[0083] Asphalt quality and properties: The quality and properties of asphalt directly affect the durability and displacement characteristics of the mixture.

[0084] Void Content and Compactness: The void content and compactness in a mix affect the strength, durability, and resistance to water loss of the mix.

[0085] Additives and modified materials: Adding appropriate amounts of additives and modified materials, such as rubber particles, asphalt plasticizers, etc., to the mixture can improve the performance and stability of the mixture.

[0086] The above factors have a comprehensive impact on the grading of asphalt mixture. Since the main component of asphalt mixture is mineral aggregate, which accounts for about 95% of the total volume, the selection and proportion of mineral aggregate is particularly important for asphalt mixture. By reasonably controlling and adjusting these factors, we can obtain asphalt mixture that meets the needs of different projects.

[0087] The mineral mixture in asphalt mixture is composed of coarse aggregate, fine aggregate and filler. The total is 100%, which means that aggregates of various particle sizes are mixed in a certain proportion and the total is 100%.

[0088] There are many types of mineral aggregates commonly used in engineering. In actual projects, engineers will choose the appropriate type of mineral aggregate according to specific geographical and engineering requirements. At the same time, the mineral aggregates usually used in actual projects are screened coarsely and are relatively difficult to proportion. In addition, in order to be more convenient and quick in actual projects, the construction party generally does not finely screen the mineral aggregates. In order to improve the quality of the engineering proportion, it is necessary to use the above-mentioned asphalt mixture proportion calculation method based on heuristic algorithm for design.

[0089] Step 1, using relevant software and heuristic algorithms to establish a data model of the above factors and aggregate ratio optimization methods, and define a series of variables: including the size of the solution space, the dimension of a single solution, the number of iterations, and the crossover probability. These variables will be used to define the algorithm parameters and problem settings.

[0090] Usually AC-13 is composed of four grades of aggregates. Assume that the proportions of the four grades in the asphalt mixture are X1, X2, X3, and X4 respectively, so X1+X2+X3+X4=100%. In this example, a 100×4 matrix is ​​set as the solution space, which means that there are 100 single solutions in each iteration solution space. In the actual AC-13 asphalt mixture project, the crushed aggregates are usually divided into four grades, so the dimension of each single solution is 4.

[0091] A matrix A is defined to represent the mass percentage a(%) of the four grades of aggregates constituting AC-13 asphalt mixture passing through the sieve holes [square hole sieve (mm)], as described in Table 1.

[0092] Define a matrix with 1 row and 10 columns to represent the median of the target gradation. In this example, the median mass percentages of various mineral materials are taken to obtain the matrix B.

[0093] The data of each generation of solution space are stored in a 100×10 matrix, which represents the mass percentage of a single solution in the solution space that passes through the standard sieve [square hole sieve (mm)]. The matrix is ​​obtained by the 100×4 ratio of the proportion of the four-grade material in the AC-13 asphalt mixture × the sieve residue percentage of each of the four-grade material.

[0094] Table 1 Mass percentage a (%) of AC-13 passing through sieve hole [square hole sieve (mm)]

[0095]

[0096]

[0097] The initial generation ratio combination is updated in a loop, and the specific update operation is as follows.

[0098] To make X1, X2, X3, and X4 randomly distributed under the premise that their sum is equal to 100, perform the following operations: Suppose a random variable a i Satisfying uniform distribution on [0-1], let X1 = (100 / 4)*a1, which represents the random proportion of the first-level aggregate. represents the random proportion of the second-tier aggregate, let Represents the random proportion of the third-level aggregate. Since X1+X2+X3+X4=100 must be ensured, X4=100-X1-X2-X3. At this point, the proportions of the four-level aggregates in the first solution of the first-generation solution space have been updated, and the proportions of the four-level aggregates in the second solution of the first-generation solution space are updated. The cycle is stopped after one hundred operations. At this time, all individual solutions in the initial solution space have been updated. Some data are listed here, as described in Table 2.

[0099] Table 2 Partial initialization ratio values ​​(rows 1-3, 54-57, 98-100) where particle size is in mm

[0100]

[0101] Use the function to calculate the function value of each ratio and assign the function value to the solution space. The specific calculation method is as follows:

[0102] Determine the types of raw materials that make up the mixture 4, conduct screening experiments, determine the screening curve C = [C1, C2, C3, C4] for each raw material, and then generate a 100×4 ratio matrix P that satisfies uniform distribution, P i =[P i1 ,P i2 ,P i3 ,P i ]; Afterwards, through P.C. T , and obtain 100 synthetic mix ratio curves C m , forming a solution space, where C j =P j1 C1+P j2 C2+P j3 C3+…+P 100,4 C4,j=1,2,3…100. Calculate C j and The sum of squared errors SSE of the data points between them. Calculated by the formula: Take "minimum SSE" as the optimization goal.

[0103] Use the function to draw the initial generation grading curve and iterative evolution error value.

[0104] Step 2, in the iterative process, use the replication and crossover operations of the genetic algorithm to optimize the solution space, as follows: After the initial function update is completed, start the main program loop. Continue the loop operation to calculate the objective function value of the single solution within the i-1 generation matching combination. According to the calculated function value, further calculate the fitness of the i-1 generation, determine the fitness level, and the single solution with a lower fitness value is copied, and the single solution with a higher fitness value is replaced by the one with a lower fitness value. After the end, create the i-th generation matching combination, and then calculate the fitness of the i-th generation. When i≤100, the same steps are looped, the higher fitness value is replaced, and the lower fitness value is used to copy and cross operations to generate a new matching combination, and output the optimal value.

[0105] The SSE (and variance) of the current ratio is calculated using the function and stored in the solution space.

[0106] The fitness value of each ratio is calculated using the function. According to the algorithm explanation, the function value itself is directly used as the fitness value.

[0107] Use the function to generate new population individuals and store them in the new solution space. The specific implementation method is as follows:

[0108] Calculate the probability p of a single solution being selected i , calculated as follows: single solution fitness / total solution space fitness. Using the cumulative distribution function, the generated p i Gradually accumulate and generate a new solution space P i , P i Each element position in corresponds to p i For example: p1 = 0.1, p2 = 0.4, p3 = 0.2, p4 = 0.3, then the new solution space is P1 = 0.1, P2 = 0.5, P3 = 0.7, P4 = 1, a random variable R i Satisfy uniform distribution on [0-1], randomly generate R i , and sort the generated random numbers in descending order and store them in a 100×1 matrix (since the number of single solutions in the above population is 100, 100 rows are generated to represent: 100 rows represent generating 100 random numbers corresponding to a single solution, and then comparing each random number with the probability p i Compare and determine whether to copy. Column 1 indicates that the generated random numbers correspond to the single solutions.

[0109] Comparison R i With P i If the random number R i <P i , which means selecting the single solution and copying it to the corresponding position in the new solution space. i >P i , it means that the single solution is not selected for copying. At this point, the calculation of whether the first solution in the solution space is copied is completed, and the calculation of whether the second solution is copied begins. The cycle stops after one hundred operations, and the optimization of individual solutions in the solution space is achieved.

[0110] Eventually, the new solution space will contain a single solution that is selected with a probability proportional to the fitness value. This selection process is an important step in the heuristic algorithm to create the next generation of single solutions, thereby gradually improving the fitness of the population during evolution.

[0111] After Step 2 is copied, multiple new single solutions will be generated. In order to ensure that the population size is fixed, some single solutions need to be replaced. By comparing the fitness values ​​calculated above, it is determined whether a single solution should be replaced.

[0112] Step 3: Use the replication operation in the genetic algorithm to select excellent ratio combinations, and cross-optimize the individual solutions within the ratio combination to ensure that the individual solutions within the ratio have a lower fitness value, and continuously replicate the individual solutions within the appropriate ratio combination to implement the survival of the fittest system to ensure that the final output result is the optimal solution. Randomly generate variable c i , let c i Satisfies uniform distribution on [0-1], set the crossover probability q i =0.4, if c i <q i , then perform a crossover operation, the specific operations are as follows:

[0113] The function is used to perform crossover operation on the new population. The mineral aggregate ratio of AC-13 asphalt mixture is divided into four aggregates, and the crossover operation is performed by randomly selecting crossover sites. The specific operation method is as follows: Assume that the four aggregates are X1, X2, X3, and X4, representing 0-3, 3-5, 5-10, and 10-15 (mm) aggregates. Any two Xs are selected for crossover and exchange. After crossover, the results are X1, X4, X3, and X2, representing 0-3, 3-5, 5-10, and 10-15 (mm) aggregates.

[0114] The replication and crossover operations may select more single solutions with lower fitness values, thereby increasing the representativeness of these single solutions in the next generation, and the selection helps the heuristic algorithm to find better solutions faster. Although the replication method tends to select single solutions with lower fitness values, it still provides a certain opportunity for single solutions with higher fitness values ​​to be selected to maintain the diversity of the ratio and prevent the local optimal solution of the gradient from appearing too early.

[0115] Step 4: Repeat Step 2 and 3 until the solution space converges to the global optimal solution.

[0116] After the copy and crossover operations are completed, the function calculation results are used to update the solution space and calculate the new objective function value, namely SSE.

[0117] The function is used to calculate the fitness of a single solution in the solution space. According to the algorithm interpretation, the function value itself is directly used as the fitness value.

[0118] Determine the fitness value of the new solution and the old solution. If the fitness value of the new solution is smaller than that of the old solution, copy the new solution into the solution space and replace the old solution with a larger fitness value. If the fitness value of the new solution is larger than that of the old solution, do not copy it. The solution space is updated while ensuring that the size of the solution space remains unchanged, and the function value of the updated solution space is calculated.

[0119] Use the function to draw the gradation curve and iterative error value of the updated solution space.

[0120] Find the value and position of the optimal function value (minimum function value) of each generation, and record the optimal function value of each generation, such as Figure 3 , Table 3, and Table 4, where Table 3 only lists part of the data.

[0121] Table 3 Optimal ratio values ​​of each generation of each batch of materials (rows 1-3, 54-57, 98-100) where the particle size unit is mm

[0122]

[0123] Table 4 Optimal ratio of each material

[0124]

[0125] The function is used to convert the proportion of the single solution with the best fitness in each generation into the mass percentage a(%) of the 4-grade aggregate passing through the sieve hole [square hole sieve (mm)]. In order to record and further analyze the actual value of the best solution and draw the grading curve, the specific operation is as follows:

[0126] It is known that the proportion of four grades of aggregate in the mixture is X1, X2, X3, and X4. Suppose the required content of asphalt mixture at a certain particle size (grading sieve percentage) is aM(i) The particle size content (grading sieve residue percentage) of the four grades of aggregate in the original grading is a X1(i) 、a X2(i) 、a x3(i) 、a x4(i) , then a M(i) =a X1(i) .X1+a X2(i) .X2+a x3(i) .X3+a x4(i) .X4,

[0127] See also Figure 5 , Figure 6 It is the evolutionary grading curve and evolutionary error value after the first iteration. Figure 7 , Figure 8 is the third iterative evolution grading curve and iterative evolution error value.

[0128] See also Figure 9-11 They are respectively the evolutionary grading curve, evolutionary error value curve and fitness evolution curve after forty-six iterations. Tables 5 and 6 are partial data after the forty-sixth iteration.

[0129] Table 5 Partial ratio values ​​of the 46th generation (rows 1-3, 54-57, 98-100) where the particle size is in mm

[0130]

[0131] Table 6 Optimal ratio data for the 46th generation

[0132]

[0133] Output the best ratio: Figure 4 As shown in the figure, after comparison, the median curve of AC-13 mixture target gradation is basically consistent with the optimal gradation curve obtained by heuristic algorithm.

[0134] After the cycle is completed, the function value and position of the optimal ratio are found, and the minimum error, the optimal ratio of each material and the corresponding optimal gradation value are displayed. This information is related to the best single solution, and the function is used to draw a graph to represent the relationship between the optimal ratio and the target ratio. See Table 7 and Table 8 for some data after the 100th iteration. Figure 12-14 Schematic diagrams of the 100th iteration grading curve, iterative evolution error, and final target grading and grading median curve respectively.

[0135] Table 7: The first 100th generation of partial ratio values ​​(rows 1-3, 54-57, 98-100) where the particle size is in mm

[0136]

[0137]

[0138] Table 8 Optimal ratio data of the 100th generation

[0139]

[0140] In summary, this method uses a rapid asphalt mixture ratio design method based on a heuristic algorithm, which can accurately predict the final designed asphalt mixture gradation through design parameters, which is conducive to quickly optimizing the design ratio of the asphalt mixture and obtaining the expected effect; it has a positive significance for improving the accuracy and speed of the mixture ratio. By using the bottom-line mechanism and optimization strategy method in the heuristic algorithm and introducing randomness, the local optimal solution can be effectively avoided, thereby increasing the possibility of finding the global optimal solution.

[0141] Example 2

[0142] Based on Example 1, this embodiment provides an electronic device, including: one or more processors and a memory, wherein the memory stores one or more programs, and the one or more programs include instructions for executing the asphalt mixture ratio calculation method based on the heuristic algorithm as described in Example 1.

[0143] Example 3

[0144] Based on Example 1, this embodiment provides a computer-readable storage medium, including one or more programs for execution by one or more processors of an electronic device, wherein the one or more programs include instructions for executing the asphalt mixture proportion calculation method based on the heuristic algorithm as described in Example 1.

[0145] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed by the present invention, and these modifications or replacements should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A method for calculating asphalt mixture proportion based on heuristic algorithm, characterized in that: The steps include: Obtain target mix ratio curve data and mineral material screening data; Initializing a solution space including a plurality of proportion combinations based on the ore screening data; Taking the minimization of the value of the objective function as the optimization goal, selecting some proportion combinations from the current solution space based on the bottom guarantee mechanism, and generating new proportion combinations based on the update strategy to form a new solution space, completing one iteration, wherein the objective function is constructed based on the target mix curve and the synthetic mix curve corresponding to the mix combination; Repeat the iteration multiple times to obtain the ratio combination with the highest similarity to the target ratio curve. In the optimization process, a data model is established based on a heuristic algorithm, and the bottom-line guarantee mechanism in the heuristic algorithm is used to select an excellent ratio combination, including the following steps: Randomly generate matching combinations, that is, a set of multiple solutions; Calculate the objective function value of the initial ratio; Initialize the optimal ratio and solution ratio of each generation, and record the optimal ratio; Enter the main loop and execute the iterative process of the heuristic algorithm. In each generation: Calculate the objective function value of a single solution of the previous generation; Calculate the fitness based on the objective function value; Copy a single solution to create a new generation of solution space; Carry out optimization strategy operations; Select the best single solution and keep it; Update the ratio, calculate the objective function value of the solution space, and record the optimal value and the corresponding solution; After the loop is completed, the location and value of the optimal solution are found. Among them, the bottom guarantee mechanism is used to realize the fitness ratio selection, including the following steps: Normalize the fitness values ​​of all individual solutions in the population; Construct a guaranteed mechanism selection method, whose size is the same as the number of individual solutions in the solution space. The size of each region of the guaranteed mechanism is proportional to the selection probability of each individual solution. Generate a random probability value and select it separately on the guaranteed mechanism. The selection method is to determine whether a single solution is selected by the random probability value and the boundaries of each region on the guaranteed mechanism. If the random probability value belongs to a certain region, then the corresponding single solution will be selected. Repeat the selection process until a sufficient number of single solutions are selected to construct the next generation of solution space.

2. The asphalt mixture ratio calculation method based on heuristic algorithm according to claim 1 is characterized in that: The solution space is: ， in, is the solution space composed of synthetic mix ratio curves, To satisfy the uniform distribution of the number of m×n proportional matrices, is the screening curve of the candidate ore. Represents transpose, matrix No. OK , .

3. The asphalt mixture ratio calculation method based on heuristic algorithm according to claim 2 is characterized in that: The optimization goal is to minimize the value of the following objective function: ， Among them, represents the objective function value of the synthetic mix ratio curve, among which, 。 4. The asphalt mixture ratio calculation method based on heuristic algorithm according to claim 1 is characterized in that: The process of selecting the matching combination based on the bottom guarantee mechanism includes: Calculate the value of the objective function corresponding to each ratio combination in the current solution space and perform normalization processing; Generate a random probability value, select a ratio combination based on the probability value and the normalized value of the objective function, and repeat this step multiple times.

5. The asphalt mixture ratio calculation method based on heuristic algorithm according to claim 4 is characterized in that: During the screening process, the probability of a matching combination being selected is inversely proportional to the value of the corresponding objective function.

6. The asphalt mixture ratio calculation method based on heuristic algorithm according to claim 1 is characterized in that: The process of generating a new matching combination based on the update strategy includes: For a matching combination in the current solution space, a crossover operation is performed with a preset probability.

7. The asphalt mixture ratio calculation method based on heuristic algorithm according to claim 6 is characterized in that: The crossover operation includes exchanging the corresponding proportions of any two mineral materials within the same combination.

8. The asphalt mixture ratio calculation method based on heuristic algorithm according to claim 1 is characterized in that: The mineral material screening data includes candidate mineral material types, particle sizes and screening curve information.

9. An electronic device, characterized in that: include: One or more processors and a memory, wherein the memory stores one or more programs, and the one or more programs include instructions for executing the asphalt mixture ratio calculation method based on the heuristic algorithm as described in any one of claims 1-8.

10. A computer-readable storage medium, characterized in that: It comprises one or more programs for execution by one or more processors of an electronic device, wherein the one or more programs comprise instructions for executing the asphalt mixture proportion calculation method based on the heuristic algorithm as described in any one of claims 1-8.

Images