Method, device, equipment and medium for optimizing mix proportion of dredged sediment concrete autoclaved building blocks

Abstract

The invention discloses a dredged sediment concrete autoclaved building block mix proportion optimization method, device and equipment and a medium. The method comprises the following steps: acquiring mix proportion key parameters and corresponding compressive strength data of dredged sediment concrete autoclaved building blocks, and constructing a to-be-processed sample set; based on the to-be-processed sample set, adopting Bayesian optimization to search model parameters of the support vector machine model, establishing a compressive strength prediction model, and determining a nonlinear mapping relation function of the mix proportion key parameters and compressive strength data; constructing a multi-objective optimization function according to the nonlinear mapping relation function, the autoclaved building block production material cost and the dredged sediment utilization rate; constructing constraint conditions according to a mix proportion key parameter range, a production material use amount range and the wet volume of each cubic meter of autoclaved building blocks; and based on the multi-objective optimization function and the constraint condition, determining the target mix proportion of the dredged sediment concrete autoclaved building block through a multi-objective particle swarm optimization algorithm. And the determined matching ratio of the dredged sediment concrete autoclaved building block is more reasonable.

Description

Technical Field

[0001] The embodiments of the present invention relate to the technical field of dredging, and particularly to an optimization method, device, equipment and medium for the mix proportion of autoclaved blocks made of dredged sediment concrete. Background Art

[0002] In port dredging, river regulation and offshore engineering, hundreds of millions of tons of dredged sediment are generated every year. Its stacking and landfill not only occupy land resources, but also pose environmental risks such as heavy metal migration and ecological pollution. Preparing autoclaved blocks with dredged sediment as raw materials can achieve the dual goals of solid waste resource utilization and green building material production.

[0003] Due to the complex composition of dredged sediment, which contains clay particles, organic matter and salts, directly using dredged sediment to replace traditional aggregates for preparing autoclaved blocks of concrete may lead to problems such as insufficient compressive strength of autoclaved blocks of concrete, high energy consumption for autoclave curing, and deterioration of long-term durability. It is necessary to prepare autoclaved blocks of concrete through a reasonable mix proportion of autoclaved blocks of dredged sediment concrete.

[0004] Currently, orthogonal experiments and empirical formulas are mostly used to explore the mix proportion of autoclaved blocks of concrete. However, this method requires a large number of specimen preparations and a multi-day curing cycle, and it is difficult to quantify the non-linear relationship between variables involved in preparing autoclaved blocks of concrete with dredged sediment, resulting in an unreasonable mix proportion determined. Summary of the Invention

[0005] The present invention provides an optimization method, device, equipment and medium for the mix proportion of autoclaved blocks of dredged sediment concrete, making the determined mix proportion of autoclaved blocks of dredged sediment concrete more reasonable.

[0006] In a first aspect, an embodiment of the present invention provides an optimization method for the mix proportion of autoclaved blocks of dredged sediment concrete, including:

[0007] Obtain the key parameters of the mix proportion and the corresponding compressive strength data of autoclaved blocks of dredged sediment concrete, and construct a sample set to be processed. The key parameters of the mix proportion at least include the content of dredged sediment, the content of cement, the content of aluminum powder, autoclave pressure, autoclave time and water-binder ratio;

[0008] Based on the sample set to be processed, use Bayesian optimization to search for the model parameters of the support vector machine model, establish a compressive strength prediction model, and determine the non-linear mapping relationship function between the key parameters of the mix proportion and the compressive strength data;

[0009] Construct a multi-objective optimization function according to the non-linear mapping relationship function, the production material cost of autoclaved blocks of dredged sediment concrete and the utilization rate of dredged sediment;

[0010] Construct constraint conditions based on the range of key mix parameters, the range of production material usage, and the wet volume of autoclaved blocks made of dredged sediment concrete per cubic meter.

[0011] Based on the multi-objective optimization function and the constraint conditions, perform multi-objective optimization of the key mix parameters through the multi-objective particle swarm optimization algorithm to determine the target mix proportion of autoclaved blocks made of dredged sediment concrete.

[0012] In a second aspect, an embodiment of the present invention provides an optimization device for the mix proportion of autoclaved blocks made of dredged sediment concrete, including:

[0013] A to-be-processed sample set construction module, configured to obtain the key mix parameters of autoclaved blocks made of dredged sediment concrete and the corresponding compressive strength data, and construct a to-be-processed sample set, where the key mix parameters at least include the dredged sediment content, cement content, aluminum powder content, autoclaving pressure, autoclaving time, and water-binder ratio;

[0014] A non-linear mapping relationship function determination module, configured to, based on the to-be-processed sample set, adopt Bayesian optimization to search for the model parameters of a support vector machine model, establish a compressive strength prediction model, and determine the non-linear mapping relationship function between the key mix parameters and the compressive strength data;

[0015] A multi-objective optimization function construction module, configured to construct a multi-objective optimization function according to the non-linear mapping relationship function, the production material cost of autoclaved blocks made of dredged sediment concrete, and the utilization rate of dredged sediment;

[0016] A constraint condition construction module, configured to construct constraint conditions according to the range of key mix parameters, the range of production material usage, and the wet volume of autoclaved blocks made of dredged sediment concrete per cubic meter;

[0017] A multi-objective optimization module, configured to, based on the multi-objective optimization function and the constraint conditions, perform multi-objective optimization of the key mix parameters through the multi-objective particle swarm optimization algorithm to determine the target mix proportion of autoclaved blocks made of dredged sediment concrete.

[0018] In a third aspect, an embodiment of the present invention provides an electronic device, including:

[0019] At least one processor; and

[0020] A memory communicatively connected to the at least one processor; wherein,

[0021] The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the method described in the first aspect.

[0022] Fourthly, an embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, the method described in the first aspect is implemented.

[0023] In the technical solution of the embodiment of the present invention, key parameters of the mix ratio of autoclaved blocks of dredged sediment concrete and corresponding compressive strength data are obtained, and a sample set to be processed is constructed. The key parameters of the mix ratio at least include the content of dredged sediment, the content of cement, the content of aluminum powder, autoclaving pressure, autoclaving time, and water-binder ratio; based on the sample set to be processed, the model parameters of the Bayesian optimization search support vector machine model are used to establish a compressive strength prediction model, and a non-linear mapping relationship function between the key parameters of the mix ratio and the compressive strength data is determined; according to the non-linear mapping relationship function, the production material cost of autoclaved blocks of dredged sediment concrete and the utilization rate of dredged sediment, a multi-objective optimization function is constructed; according to the range of key parameters of the mix ratio, the range of production material usage, and the wet volume per cubic meter of autoclaved blocks of dredged sediment concrete, constraint conditions are constructed; based on the multi-objective optimization function and the constraint conditions, multi-objective optimization of the key parameters of the mix ratio is carried out through a multi-objective particle swarm optimization algorithm to determine the target mix ratio of autoclaved blocks of dredged sediment concrete. This solution quantifies the non-linear relationship between the key parameters of the mix ratio of autoclaved blocks of dredged sediment concrete and the compressive strength data through the Bayesian optimization support vector machine model, and combines the multi-objective particle swarm optimization algorithm for multi-objective trade-off of compressive strength, material cost, and utilization rate of dredged sediment, making the determined mix ratio of autoclaved blocks of dredged sediment concrete more reasonable.

[0024] It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of the present invention, nor is it used to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention, and for those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0026] Figure 1 is a flowchart of a method for optimizing the mix ratio of autoclaved blocks of dredged sediment concrete according to Embodiment 1 of the present invention;

[0027] Figure 2 is a flowchart of a method for optimizing the mix ratio of autoclaved blocks of dredged sediment concrete according to Embodiment 2 of the present invention;

[0028] Figure 3It is a schematic diagram of optimizing the model parameters of a compressive strength prediction model provided in the second embodiment of the present invention;

[0029] Figure 4 It is a schematic diagram of performing regression fitting on a training sample set based on a compressive strength prediction model provided in the second embodiment of the present invention;

[0030] Figure 5 It is a schematic diagram of performing regression fitting on a test sample set based on a compressive strength prediction model provided in the second embodiment of the present invention.

[0031] Figure 6 It is a flowchart of a method for optimizing the mix proportion of autoclaved blocks made of dredged bottom mud concrete provided in the third embodiment of the present invention;

[0032] Figure 7 It is a schematic diagram of performing global optimization using a multi-objective particle swarm optimization algorithm provided in the third embodiment of the present invention;

[0033] Figure 8 It is a schematic diagram of the structure of a device for optimizing the mix proportion of autoclaved blocks made of dredged bottom mud concrete provided in the fourth embodiment of the present invention;

[0034] Figure 9 It is a schematic diagram of the structure of an electronic device implementing the embodiments of the present invention. Detailed implementation manners

[0035] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0036] It should be noted that the terms "first", "second", etc. in the present invention are used to distinguish similar objects, and do not necessarily need to describe a specific order or sequence. It should be understood that such used data can be interchanged under appropriate circumstances so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device including a series of steps or units does not necessarily have to be limited to those clearly listed steps or units, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0037] Embodiment 1

[0038] Figure 1 It is a flowchart of a method for optimizing the mix proportion of autoclaved blocks made of dredged sediment concrete according to Embodiment 1 of the present invention. This embodiment is applicable to the situation of optimizing the mix proportion of autoclaved blocks made of dredged sediment concrete. This method can be executed by a device for optimizing the mix proportion of autoclaved blocks made of dredged sediment concrete. This device can be implemented in the form of software and / or hardware and integrated in an electronic device. Further, the electronic device includes but is not limited to: computers, laptops, servers, etc.

[0039] As Figure 1 shown, the method includes:

[0040] S110. Obtain the key parameters of the mix proportion of autoclaved blocks made of dredged sediment concrete and the corresponding compressive strength data, and construct a sample set to be processed. The key parameters of the mix proportion at least include the content of dredged sediment, the content of cement, the content of aluminum powder, the autoclaving pressure, the autoclaving time, and the water-binder ratio.

[0041] The mix proportion of autoclaved blocks made of dredged sediment concrete can be understood as the dosage of each production material involved in preparing autoclaved blocks made of dredged sediment concrete. Among them, the production materials of autoclaved blocks made of dredged sediment concrete can include but are not limited to: dredged sediment, cement, fly ash, quicklime, gypsum, aluminum powder, and water.

[0042] The key parameters of the mix proportion of autoclaved blocks made of dredged sediment concrete can be understood as the key parameters required to determine the mix proportion of autoclaved blocks made of dredged sediment concrete, at least including the content of dredged sediment, the content of cement, the content of aluminum powder, the autoclaving pressure, the autoclaving time, and the water-binder ratio. Among them, the content of dredged sediment is the mass ratio of dredged sediment in the solid materials of the production materials; the content of cement is the mass ratio of cement in the solid materials of the production materials; the content of aluminum powder is the mass ratio of aluminum powder in the solid materials of the production materials; the autoclaving pressure is the pressure value applied during the autoclaving curing process; the autoclaving time is the duration of the autoclaving curing stage; the water-binder ratio is the mass ratio of the water consumption to the binder consumption in the production materials.

[0043] In the case of determining the key parameters of the mix proportion of autoclaved blocks made of dredged sediment concrete, the dosage of each production material involved in autoclaved blocks made of dredged sediment concrete can be determined based on these key parameters, that is, the mix proportion of autoclaved blocks made of dredged sediment concrete is determined.

[0044] The compressive strength data corresponding to the key parameters of the mix proportion of autoclaved blocks made of dredged sediment concrete can be the compressive strength of the autoclaved blocks made of dredged sediment concrete prepared with the mix proportion of autoclaved blocks made of dredged sediment concrete corresponding to these key parameters.

[0045] In practical applications, the key parameters of the mix ratio of autoclaved blocks made of dredged sediment concrete and their corresponding compressive strength data can be used as a sample to be processed, and multiple samples to be processed are obtained to construct a sample set to be processed. Among them, the key parameters of the mix ratio and their corresponding compressive strength data can be the data collected at historical times, which is not limited here.

[0046] S120. Based on the sample set to be processed, use Bayesian optimization to search for the model parameters of the support vector machine model, establish a compressive strength prediction model, and determine the non-linear mapping relationship function between the key parameters of the mix ratio and the compressive strength data.

[0047] The compressive strength prediction model can be a model used to predict the corresponding compressive strength data based on the key parameters of the mix ratio of autoclaved blocks made of dredged sediment concrete. The compressive strength prediction model can be, for example, a support vector machine (SVM) model, and the specific structure of the compressive strength prediction model is not limited. The non-linear mapping relationship function can be a function indicating the non-linear mapping relationship between the key parameters of the mix ratio of autoclaved blocks made of dredged sediment concrete and the compressive strength data.

[0048] In this step, the sample set to be processed can be divided into a training sample set and a test sample set according to a certain ratio and normalized; based on the training sample set, using the key parameters of the mix ratio as the input and the compressive strength data as the output, use the Bayesian optimization algorithm to search for the optimal model parameters of the SVM model and train the compressive strength prediction model, where the model parameters can be the penalty coefficient and kernel parameter of the SVM model; verify the prediction effect of the trained compressive strength prediction model based on the test sample set; through the trained compressive strength prediction model, output the non-linear mapping relationship function between the key parameters of the mix ratio and the compressive strength data, that is, taking a certain key parameter of the mix ratio as the input, the non-linear mapping relationship function can determine the corresponding compressive strength data of the input.

[0049] In one embodiment, the ratio of the number of samples in the training sample set to the number of samples in the test sample set can be 2:1 to 9:1, which is not limited here. Optionally, 4:1 can be used in practical applications.

[0050] S130. Construct a multi-objective optimization function according to the non-linear mapping relationship function, the production material cost of autoclaved blocks made of dredged sediment concrete, and the utilization rate of dredged sediment.

[0051] The production material cost of autoclaved blocks made of dredged sediment concrete can be the sum of the products of the usage amounts of various production materials of autoclaved blocks made of dredged sediment concrete and the corresponding material unit prices. The utilization rate of dredged sediment can be understood as the content of dredged sediment.

[0052] The multi-objective optimization function can be a function used to guide the search process during the multi-objective optimization. Through continuous iteration of the multi-objective optimization, an attempt is made to find the optimal solution of the multi-objective optimization function (which may be the maximum or minimum solution depending on the specific problem), that is, the solution that achieves the best balance among multiple objectives.

[0053] In this step, through the non-linear mapping relationship function, under the key parameters of the mix proportion of different autoclaved aerated concrete blocks made from dredged sediment, an objective function can be constructed with the maximum compressive strength data as the goal; under the material usage of the production materials of autoclaved aerated concrete blocks made from dredged sediment corresponding to the key parameters of the mix proportion, an objective function can be constructed with the lowest production material cost of autoclaved aerated concrete blocks made from dredged sediment as the goal; under different dredged sediment dosages, an objective function can be constructed with the maximum utilization rate of dredged sediment as the goal; the combination of the above multiple objective functions is the multi-objective optimization function. That is, subsequently, through multi-objective optimization, a solution that achieves the best balance among the maximum compressive strength data, the lowest production material cost of autoclaved aerated concrete blocks made from dredged sediment, and the maximum utilization rate of dredged sediment needs to be found.

[0054] S140. Construct constraint conditions based on the range of key parameters of the mix proportion, the range of production material usage, and the wet volume of each cubic meter of autoclaved aerated concrete blocks made from dredged sediment.

[0055] In this step, according to the actual application requirements, respective value ranges can be set for each parameter in the key parameters of the mix proportion except the water-binder ratio; respective usage ranges can be set for each material in the production materials of autoclaved aerated concrete blocks made from dredged sediment except water; a corresponding volume range can be set for the wet volume of each cubic meter of autoclaved aerated concrete blocks made from dredged sediment. The wet volume of autoclaved aerated concrete blocks made from dredged sediment is the sum of the usage amounts of each material included in the production materials of autoclaved aerated concrete blocks made from dredged sediment and the ratio of the material density; the combination of the above value range, usage range, and volume range is the constraint condition. Among them, the constraint condition can be a constraint condition used to guide the search process during the multi-objective optimization, and the above value range, usage range, and volume range are not limited.

[0056] S150. Based on the multi-objective optimization function and the constraint condition, perform multi-objective optimization of the key parameters of the mix proportion through the multi-objective particle swarm optimization algorithm to determine the target mix proportion of autoclaved aerated concrete blocks made from dredged sediment.

[0057] In this step, the multi-objective particle swarm optimization algorithm can be used to search for the optimal balance among multiple objectives indicated by the multi-objective optimization function. That is, the maximum compressive strength data under different key parameters of the mix ratio of autoclaved blocks made from dredged sediment concrete, the lowest production material cost of autoclaved blocks made from dredged sediment concrete under the production material consumption corresponding to different key parameters of the mix ratio, and the maximum utilization rate of dredged sediment under different dredged sediment dosages are used as search objectives. Based on the constraint conditions, that is, the respective value ranges corresponding to each parameter except the water-binder ratio among the above key parameters of the mix ratio, the respective dosage ranges corresponding to each material except water in the production materials of autoclaved blocks made from dredged sediment concrete, and the volume range corresponding to the wet volume of each cubic meter of autoclaved blocks made from dredged sediment concrete as search constraints, continuous search iterations are carried out to obtain the key parameters of the mix ratio that can achieve the best balance among multiple objectives. The mix ratio of autoclaved blocks made from dredged sediment concrete corresponding to the key parameters of the mix ratio that achieve the best balance among multiple objectives is determined as the target mix ratio of autoclaved blocks made from dredged sediment concrete. Among them, the multi-objective particle swarm optimization algorithm (Multi-Objective Particle Swarm Optimization, MOPSO) can be understood as an algorithm for finding a set of Pareto optimal solutions to balance the above multiple objectives.

[0058] In the technical solution of the embodiment of the present invention, the key parameters of the mix ratio of autoclaved blocks made from dredged sediment concrete and the corresponding compressive strength data are obtained, and a sample set to be processed is constructed. The key parameters of the mix ratio at least include the dredged sediment dosage, cement dosage, aluminum powder dosage, autoclaving pressure, autoclaving time, and water-binder ratio. Based on the sample set to be processed, the model parameters of the Bayesian optimization search support vector machine model are used to establish a compressive strength prediction model, and the nonlinear mapping relationship function between the key parameters of the mix ratio and the compressive strength data is determined. A multi-objective optimization function is constructed according to the nonlinear mapping relationship function, the production material cost of autoclaved blocks made from dredged sediment concrete, and the utilization rate of dredged sediment. Constraint conditions are constructed according to the range of key parameters of the mix ratio, the range of production material consumption, and the wet volume of each cubic meter of autoclaved blocks made from dredged sediment concrete. Based on the multi-objective optimization function and the constraint conditions, multi-objective optimization of the key parameters of the mix ratio is carried out through the multi-objective particle swarm optimization algorithm to determine the target mix ratio of autoclaved blocks made from dredged sediment concrete. This solution quantifies the nonlinear relationship between the key parameters of the mix ratio of autoclaved blocks made from dredged sediment concrete and the compressive strength data through the Bayesian optimization support vector machine model, and combines the multi-objective trade-off of compressive strength, material cost, and utilization rate of dredged sediment by the multi-objective particle swarm optimization algorithm, making the determined mix ratio of autoclaved blocks made from dredged sediment concrete more reasonable.

[0059] Embodiment 2

[0060] Figure 2It is a flowchart of a method for optimizing the mix proportion of autoclaved blocks made of dredged sediment concrete according to Embodiment 2 of the present invention. This embodiment further refines, based on Embodiment 1 above, the process of using Bayesian optimization search to support vector machine model parameters for the to-be-processed sample set, establishing a compressive strength prediction model, and determining the non-linear mapping relationship function between the key mix proportion parameters and the compressive strength data; and further refines the construction of a multi-objective optimization function according to the non-linear mapping relationship function, the production material cost of autoclaved blocks made of dredged sediment concrete, and the utilization rate of dredged sediment; and further refines the construction of constraint conditions according to the range of key mix proportion parameters, the range of production material usage, and the wet volume of per cubic meter of autoclaved blocks made of dredged sediment concrete.

[0061] As Figure 2 shown, the method includes:

[0062] S110. Obtain the key mix proportion parameters and corresponding compressive strength data of autoclaved blocks made of dredged sediment concrete, and construct a to-be-processed sample set. Among them, the key mix proportion parameters at least include the content of dredged sediment, the content of cement, the content of aluminum powder, autoclaving pressure, autoclaving time, and water-binder ratio. The to-be-processed sample set includes a training sample set and a test sample set.

[0063] S121. Select Gaussian process as the surrogate model for Bayesian optimization, and use expected improvement as the acquisition function.

[0064] The role of the surrogate model can be understood as using a relatively simple and easy-to-calculate function to approximate the process of Bayesian optimization, so as to quickly search for the possible optimal model parameters of the support vector machine model in the model parameter search space of the support vector machine model. Among them, the model parameter search space is the search space of the model parameters of the support vector machine model.

[0065] Optionally, the expression of the kernel function of the above Gaussian process is as follows, where x i and x j are respectively the i-th and j-th model parameter samples in the model parameter search space, and σ 2 is the variance parameter corresponding to the model parameter.

[0066]

[0067] The acquisition function can be used to guide the selection of model parameters to be evaluated in the model parameter search space. The expected improvement (EI) acquisition function can measure the expected improvement obtained by evaluating the model parameters under the current known information.

[0068] Optionally, the expression of the acquisition function is as follows, where x is a sample of model parameters, μ(x) is the predicted mean corresponding to the model parameters, σ(x) is the standard deviation corresponding to the model parameters, and f best is the currently observed optimal model parameter, ξ is the exploration weight taken as 0.01, Φ(Z) is the cumulative distribution function of the standard normal distribution (representing the probability that a standard normal random variable is less than or equal to Z), and φ(Z) is the probability density function of the standard normal distribution.

[0069] EI(x) = (μ(x) - f best - ξ)Φ(Z) + σ(x)φ(Z)

[0070] S122. Randomly sample multiple groups of model parameters in the model parameter search space of the support vector machine model, train the support vector machine model through the training sample set, calculate the root mean square error, and construct an initial data set.

[0071] In one embodiment, the model parameter search space is composed of the value range of the penalty coefficient of the support vector machine model and the value range of the radial basis kernel function parameter. Exemplarily, set the value range of the penalty coefficient of the support vector machine model as C ∈ [10 -1 , 10 4 , and set the value range of the radial basis kernel function parameter of the support vector machine model, i.e., the kernel parameter, as γ ∈ [10 -4 , 10 2 . The above two value ranges constitute the model parameter search space.

[0072] In this step, within the model parameter search space of the SVM, multiple groups of model parameters can be randomly sampled, such as 20 groups of model parameters. Each group of model parameters includes a penalty coefficient and a kernel parameter, and set the goal of Bayesian optimization as the minimum root mean square error (Root Mean Squared Error, RMSE) of ten-fold cross-validation; use the mix ratio key parameters as the input and the compressive strength data as the output through the training sample set, combine the multiple groups of randomly sampled model parameters, train the support vector machine model, calculate the root mean square error of each group of model parameters under ten-fold cross-validation, and thus construct an initial data set.

[0073] S123. Perform multiple Bayesian optimization iterations based on the initial data set. In each iteration, select the next group of model parameters by maximizing the expected improvement value. When the root mean square error obtained from consecutive multiple iterations reaches the first iteration termination condition, terminate the iteration, and select the group of model parameters with the minimum root mean square error as the optimal model parameter.

[0074] In this step, multiple Bayesian optimization iterations are performed based on the initial dataset. For example, 50 Bayesian optimization iterations are performed. In each iteration, the next set of model parameters to be evaluated is selected by maximizing the EI function. When the improvement amplitude of the root mean square error obtained from consecutive multiple iterations, such as 10 consecutive iterations, is less than 1%, it is considered that the model optimization has tended to be stable, and the iteration is terminated in advance. The set of model parameters with the smallest root mean square error of cross-validation is selected as the optimal model parameters. Among them, the first iteration termination condition can be that the improvement amplitude of the root mean square error obtained from consecutive multiple iterations is less than 1%.

[0075] S124. Combine the optimal model parameters with the test sample set to verify the prediction results of the trained support vector machine model.

[0076] In this step, the trained support vector machine model is run under the above optimal model parameters. Using the key mix parameters of each test sample in the test sample set as the input, the predicted value of the compressive strength data is output and inverse normalization processing is performed to obtain the prediction results of the trained support vector machine model. The predicted value of the compressive strength data is compared with the actual value of the compressive strength data of the corresponding test sample in the test sample set to verify the prediction results of the trained support vector machine model.

[0077] In one embodiment, the prediction results of the trained support vector machine model are verified by at least the following parameters: the root mean square error and the goodness of fit corresponding to each test sample in the test sample set.

[0078] The root mean square error RMSE corresponding to each test sample can be determined by the following formula: where N is the number of test samples, y i is the actual value of the compressive strength data, and y i ' is the predicted value of the compressive strength data.

[0079] The goodness of fit R 2 of the trained support vector machine model can be determined by the following formula: where is the mean value of the actual values of each compressive strength data.

[0080] S125. When the prediction results meet the set conditions, use the trained support vector machine model as the compressive strength prediction model and output the nonlinear mapping relationship function between the key mix parameters and the compressive strength data.

[0081] When the prediction result meets the set conditions, such as when the root mean square error corresponding to the prediction result is lower than the first threshold and the goodness of fit exceeds the second threshold, the trained support vector machine model is used as the compressive strength prediction model, and the nonlinear mapping relationship function between the key parameters of the mix ratio and the compressive strength data is output. The above first threshold, second threshold, and set conditions are not limited.

[0082] S131. Construct the first objective function of the nonlinear mapping relationship function, which indicates finding a set of key parameters of the mix ratio to maximize the corresponding compressive strength data.

[0083] S132. Construct the second objective function of the production material cost of the autoclaved block made of dredged sediment concrete, which indicates finding a set of usage amounts of the production materials of the autoclaved block made of dredged sediment concrete to minimize the corresponding production material cost of the autoclaved block made of dredged sediment concrete.

[0084] S133. Construct the third objective function of the dredged sediment utilization rate, which indicates finding a dredged sediment admixture amount to maximize the corresponding dredged sediment utilization rate.

[0085] S134. Construct a multi-objective optimization function according to the first objective function, the second objective function, and the third objective function.

[0086] The execution order of the above S131 to S133 is not limited. The following is an explanation of S131 - S134. The multi-objective optimization function can be expressed by the following formula:

[0087]

[0088] Among them, the first objective function of the nonlinear mapping relationship function is f1, X i is the key parameter of the mix ratio, is the compressive strength data predicted by the nonlinear mapping relationship function based on X i where X1 to X6 are the dredged sediment admixture amount, cement admixture amount, aluminum powder admixture amount, autoclaving pressure, autoclaving time, and water-binder ratio respectively; the second objective function of the production material cost of the autoclaved block made of dredged sediment concrete is f2, Y i is the usage amount of the production materials of the autoclaved block made of dredged sediment concrete per cubic meter, c i is the material unit price of Y i Y1 to Y7 are the usage amounts of dredged sediment, cement, fly ash, quicklime, gypsum, aluminum powder, and water respectively; the third objective function of the dredged sediment utilization rate is f3.

[0089] S141. Construct the value constraint, and the value constraint indicates that each parameter except the water-binder ratio in the key parameters of the mix ratio satisfies its corresponding value range.

[0090] S142. Construct the dosage constraint, where the dosage constraint indicates that each material in the autoclaved aerated concrete block production materials made from dredged sediment, except water, meets its corresponding dosage range. The autoclaved aerated concrete block production materials made from dredged sediment at least include dredged sediment, cement, fly ash, quicklime, gypsum, aluminum powder, and water.

[0091] S143. Construct the volume constraint, where the volume constraint indicates that the wet volume of each cubic meter of autoclaved aerated concrete block made from dredged sediment is 1, and the wet volume of the autoclaved aerated concrete block made from dredged sediment is the sum of the dosages of each material included in the autoclaved aerated concrete block production materials made from dredged sediment divided by the material density.

[0092] The execution order of the above S141 to S143 is not limited. The following explains S141 - S143.

[0093] The value constraint can be expressed as:

[0094]

[0095] Among them, the units of X1, X2, and X3 are %; the unit of X4 is MPa; the unit of X5 is hours.

[0096] The dosage constraint can be expressed as:

[0097]

[0098] Among them, the units of Y1 to Y6 are all kg.

[0099] The volume constraint can be expressed as:

[0100]

[0101] Among them, V is the wet volume of each cubic meter of autoclaved aerated concrete block made from dredged sediment, and ρ i is the density of Y i .

[0102] The combination of the above value constraint, dosage constraint, and volume constraint is the constraint condition.

[0103] S150. Based on the multi - objective optimization function and the constraint condition, through the multi - objective particle swarm optimization algorithm, perform multi - objective optimization of the key parameters of the mix ratio to determine the target mix ratio of the autoclaved aerated concrete block made from dredged sediment.

[0104] In the technical solution of the embodiment of the present invention, by using the Gaussian process as the surrogate model of Bayesian optimization and the expected improvement as the acquisition function, multiple Bayesian optimization iterations are performed in the model parameter search space of the support vector machine model to obtain the optimal model parameters. The nonlinear mapping relationship function between the mix ratio key parameters and the compressive strength data is output by the support vector machine model under the optimal model parameters, improving the prediction accuracy of the compressive strength data; by constructing the objective functions corresponding to the nonlinear mapping relationship function, the production material cost of autoclaved blocks made of dredged sediment, and the utilization rate of dredged sediment respectively, search targets are provided for subsequent multi-objective optimization; by constructing value constraints, dosage constraints, and volume constraints, constraint conditions are provided for subsequent multi-objective optimization.

[0105] This solution combines Bayesian optimization of SVM with cross-validation to optimize the kernel function parameters, solves the problem of insufficient expression ability of traditional regression models for the nonlinear relationship between concrete compressive strength and complex mix ratio parameters, significantly improves the prediction accuracy, and provides reliable input for optimization.

[0106] By setting the goal of maximizing the utilization rate of dredged sediment, this solution promotes the efficient resource utilization of dredged sediment, reduces the consumption of natural aggregates, reduces the risk of heavy metal pollution, saves the cost per cubic meter of concrete, and achieves a double breakthrough in environmental and economic benefits.

[0107] The following is an exemplary description of some contents of this embodiment.

[0108] Figure 3 It is a schematic diagram of optimizing the model parameters of a compressive strength prediction model provided by the second embodiment of the present invention. The optimization result shows that the optimal penalty coefficient C is 1 and the kernel parameter γ (i.e., gamma) is 100.

[0109] Figure 4 It is a schematic diagram of performing regression fitting on the training sample set based on the compressive strength prediction model provided by the second embodiment of the present invention. Figure 5 It is a schematic diagram of performing regression fitting on the test sample set based on the compressive strength prediction model provided by the second embodiment of the present invention. Among them, the goodness of fit R 2 corresponding to the training sample set (i.e., the training set) is 0.974, and the R 2 corresponding to the test sample set (i.e., the test set) is 0.972. The fitting results are good, and the error between the predicted value (i.e., the predicted strength) and the actual value (i.e., the true strength) of the compressive strength data is small.

[0110] Embodiment Three

[0111] Figure 6It is a flowchart of a method for optimizing the mix proportion of autoclaved blocks made of dredged sediment concrete according to Embodiment 3 of the present invention. This embodiment is a further refinement on the basis of the above-mentioned Embodiment 1, aiming to perform multi-objective optimization of the key mix proportion parameters through a multi-objective particle swarm optimization algorithm based on the multi-objective optimization function and the constraint conditions, so as to determine the target mix proportion of autoclaved blocks made of dredged sediment concrete.

[0112] As Figure 6 shown, the method includes:

[0113] S110. Obtain the key mix proportion parameters of autoclaved blocks made of dredged sediment concrete and the corresponding compressive strength data, and construct a sample set to be processed. Among them, the key mix proportion parameters at least include the content of dredged sediment, the content of cement, the content of aluminum powder, autoclaving pressure, autoclaving time, and water-binder ratio.

[0114] S120. Based on the sample set to be processed, adopt Bayesian optimization to search for the model parameters of the support vector machine model, establish a compressive strength prediction model, and determine the non-linear mapping relationship function between the key mix proportion parameters and the compressive strength data.

[0115] S130. Construct a multi-objective optimization function according to the non-linear mapping relationship function, the production material cost of autoclaved blocks made of dredged sediment concrete, and the utilization rate of dredged sediment.

[0116] S140. According to the range of key mix proportion parameters, the range of production material usage, and the wet volume per cubic meter of autoclaved blocks made of dredged sediment concrete, construct constraint conditions.

[0117] S151. Based on the multi-objective optimization function and the constraint conditions, construct a multi-objective particle swarm optimization algorithm.

[0118] Regard the multiple objectives indicated by the multi-objective optimization function, namely the maximum compressive strength data, the lowest production material cost of autoclaved blocks made of dredged sediment concrete, and the maximum utilization rate of dredged sediment, as the search objectives, and regard the constraint conditions, namely the value ranges of the parameters in the key mix proportion parameters except the water-binder ratio, the usage ranges of the materials in the production materials of autoclaved blocks made of dredged sediment concrete except water, and the volume range of the wet volume per cubic meter of autoclaved blocks made of dredged sediment concrete, as search constraints, and integrate them into the multi-objective particle swarm optimization algorithm as the objectives and constraints for algorithm iteration.

[0119] S152. Perform particle swarm initialization and coding. Use the key mix proportion parameters as particles, encode the key mix proportion parameters and the corresponding production material usage into multi-dimensional vectors, and initialize the position and velocity of each particle.

[0120] In this step, particle swarm initialization and encoding are performed. Each particle represents a mix proportion scheme, which is encoded as a 13-dimensional vector Z = [X1,..., X6, Y1,..., Y7], that is, it is encoded as a multi-dimensional vector composed of the key mix proportion parameters X1 to X6 and the dosages Y1 to Y7 of the corresponding autoclaved aerated concrete blocks produced from dredged bottom mud. The size of the particle swarm is N (such as 50), and the position Z of each particle is randomly initialized. i and the velocity V i , and the initial position of each particle is uniformly distributed within the parameter feasible region and satisfies the constraint conditions.

[0121] S153. Adopt a linear decreasing strategy to dynamically adjust the global and local search capabilities, set the individual cognitive factor and the social cognitive factor, and introduce a random perturbation term to enhance the search ability.

[0122] The linear decreasing strategy can dynamically adjust the global and local search capabilities by using the following inertia weight adjustment formula:

[0123]

[0124] where t is the current iteration number; w(t) is the inertia weight corresponding to t; w max is the maximum value of the inertia weight; w min is the minimum value of the inertia weight; T max is the maximum number of iterations. Optionally, w max is 0.9, w min is 0.4, and T max is 300.

[0125] The random perturbation term can be expressed as follows:

[0126]

[0127] where c1(t) and c2(t) are the variables affected by random perturbation at the current iteration number t, c1 and c2 are the basic values of random perturbation, c1 represents the individual cognitive factor, c2 represents the social cognitive factor, and both can be taken as 1.8; r1 and r2 can be random terms.

[0128] S154. Set up an external archive to store non-dominated solutions. During the multi-objective optimization of the key mix proportion parameters, add the particles that meet the archive addition conditions to the archive, and sort the solutions in the archive according to the values of the multi-objective optimization function to determine the solutions to be retained.

[0129] In this step, an external initialization empty archive is set up with a capacity limit of 150, which is used to store non-dominated solutions (i.e., Pareto optimal solutions). For each newly generated particle, if it meets one of the following conditions, it is added to the archive: it is not dominated by any solution in the archive; it is non-dominated with the archive and can improve the diversity of the solution set. The solutions in the archive are sorted according to the values of the multi-objective optimization function, and the individual crowding distance is calculated. The solutions with a lower crowding degree are preferentially retained to maintain the distribution. Optionally, the archive addition conditions include at least one of the following: the new particle is not dominated by any solution in the archive; the new particle is non-dominated with the solutions in the archive and can improve the diversity of the solution set.

[0130] S155. If the particle does not improve its dominance level for consecutive iterations, mutation is triggered, and a Gaussian perturbation is applied to the position of the particle, and the perturbation amplitude decays with the number of iterations.

[0131] In this step, if the particle does not improve its dominance level for consecutive iterations, that is, it is not selected as the global optimal or individual optimal, mutation is triggered, and a Gaussian perturbation is applied to the position of the particle, and the perturbation amplitude decays with the number of iterations.

[0132] The perturbation formula of the applied Gaussian perturbation can be:

[0133]

[0134] where t is the current number of iterations, and T max is the maximum number of iterations; Z max and Z min are the upper and lower limits of the position value of the particle respectively; N represents the Gaussian distribution; Z new and Z old are the particle positions after and before applying the Gaussian perturbation to the particle respectively.

[0135] S156. Update the velocity and position of the particle until the second iteration termination condition is reached, and determine the target mix ratio of autoclaved aerated concrete blocks for dredged sediment based on the non-dominated solution set in the external archive.

[0136] The velocity of the particle can be updated by the following formula:

[0137] V i (t + 1) = w(t)V i (t) + c1(t)r1(P best - Z i (t)) + c2(t)r2(G best - Z i (t))

[0138] where V i (t + 1) and V i(t) represents the particle velocities corresponding to the (t + 1)-th iteration and the t-th iteration respectively; Z i (t) represents the particle position corresponding to the t-th iteration; w(t) represents the inertia weight corresponding to t; c1(t) and c2(t) are variables affected by random perturbations at the current iteration t, and r1 and r2 can be random terms; P best is the individual historical optimum; G best is the global guide selected from the archive according to the crowding probability.

[0139] The update of the particle position can be expressed by the following formula:

[0140] Z i (t + 1) = Z i (t) + V i (t + 1)

[0141] where, Z i (t + 1) is the particle position corresponding to the (t + 1)-th iteration.

[0142] Optionally, during the position update process, if the particle position exceeds the feasible region, reflection correction is performed, and the specific correction method is not limited.

[0143] The velocities and positions of the particles are updated in the above manner until the second iteration termination condition is reached. For example, when the iteration reaches 300 times, the non-dominated solution set in the external archive is determined as the final Pareto front solution, and each Pareto front solution corresponds to a set of key parameters of the mix ratio of the autoclaved aerated concrete block made of dredged sediment. The target mix ratio of the autoclaved aerated concrete block made of dredged sediment can be determined through the key parameters.

[0144] Figure 7 is a schematic diagram of global optimization by a multi-objective particle swarm optimization algorithm provided in Embodiment 3 of the present invention. Through the MOPSO algorithm, with a population size of 50 and a maximum iteration number of 300, global optimization is performed with the compressive strength, material cost, and sediment utilization rate of the autoclaved aerated concrete block made of dredged sediment as the objectives. After 300 iterations, the target mix ratio of the autoclaved aerated concrete block made of dredged sediment is obtained.

[0145] As Figure 7 shown, as the compressive strength increases, the material cost increases accordingly. Among them, the compressive strength of the autoclaved aerated concrete block made of dredged sediment takes values between 1.5 and 7.5 MPa, the material cost takes values between 105 and 165 yuan, and the sediment utilization rate takes values between 25% and 75%. When the autoclaved aerated concrete block made of dredged sediment reaches the compressive strength target, the lowest economic cost is 124 yuan / m 3, the utilization rate of sediment is 63%. At this time, the amount of dredged sediment used in autoclaved blocks of dredged sediment concrete per unit volume is 1093 kg, the amount of cement is 243 kg, the amount of fly ash is 87 kg, the amount of quicklime is 330 kg, the amount of gypsum is 35 kg, the amount of aluminum powder is 10 kg, the amount of water is 626 kg, and the water-binder ratio is 0.35.

[0146] The technical solution of the embodiment of the present invention adopts an improved multi-objective particle swarm optimization algorithm (MOPSO), introduces a dynamic inertia weight and a crowding degree sorting mechanism, effectively balances the global exploration and local development capabilities, improves the iterative convergence speed, and improves the uniformity of the Pareto front distribution, realizing the multi-objective optimal trade-off of compressive strength, cost and sediment utilization rate.

[0147] Embodiment 4

[0148] Figure 8 is a structural schematic diagram of an optimization device for the mix proportion of autoclaved blocks of dredged sediment concrete provided according to Embodiment 4 of the present invention. This embodiment is applicable to the situation of optimizing the mix proportion of autoclaved blocks of dredged sediment concrete. As Figure 8 shown, the specific structure of the device includes:

[0149] A to-be-processed sample set construction module 81, configured to obtain key mix proportion parameters of autoclaved blocks of dredged sediment concrete and corresponding compressive strength data, and construct a to-be-processed sample set, where the key mix proportion parameters at least include the content of dredged sediment, the content of cement, the content of aluminum powder, autoclaving pressure, autoclaving time, and water-binder ratio;

[0150] A non-linear mapping relationship function determination module 82, configured to, based on the to-be-processed sample set, adopt a Bayesian optimization search support vector machine model to model parameters, establish a compressive strength prediction model, and determine a non-linear mapping relationship function between the key mix proportion parameters and the compressive strength data;

[0151] A multi-objective optimization function construction module 83, configured to construct a multi-objective optimization function according to the non-linear mapping relationship function, the production material cost of autoclaved blocks of dredged sediment concrete, and the sediment utilization rate;

[0152] A constraint condition construction module 84, configured to construct constraint conditions according to the key mix proportion parameter range, the production material usage range, and the wet volume of per cubic meter of autoclaved blocks of dredged sediment concrete;

[0153] A multi-objective optimization module 85, configured to perform multi-objective optimization of the key mix proportion parameters based on the multi-objective optimization function and the constraint conditions, and determine the target mix proportion of autoclaved blocks of dredged sediment concrete through a multi-objective particle swarm optimization algorithm.

[0154] The device for optimizing the mix proportion of autoclaved blocks made of dredged sediment concrete provided in this embodiment obtains the key parameters of the mix proportion of autoclaved blocks made of dredged sediment concrete and the corresponding compressive strength data through the to-be-processed sample set construction module, and constructs the to-be-processed sample set. The key parameters of the mix proportion at least include the content of dredged sediment, the content of cement, the content of aluminum powder, the autoclaving pressure, the autoclaving time, and the water-binder ratio. The non-linear mapping relationship function determination module determines the model parameters of the Bayesian optimization search support vector machine model based on the to-be-processed sample set, establishes a compressive strength prediction model, and determines the non-linear mapping relationship function between the key parameters of the mix proportion and the compressive strength data. The multi-objective optimization function construction module constructs a multi-objective optimization function according to the non-linear mapping relationship function, the production material cost of autoclaved blocks made of dredged sediment concrete, and the utilization rate of dredged sediment. The constraint condition construction module constructs constraint conditions according to the range of key parameters of the mix proportion, the range of production material usage, and the wet volume of per cubic meter of autoclaved blocks made of dredged sediment concrete. The multi-objective optimization module performs multi-objective optimization of the key parameters of the mix proportion through the multi-objective particle swarm optimization algorithm based on the multi-objective optimization function and the constraint conditions, and determines the target mix proportion of autoclaved blocks made of dredged sediment concrete. This solution quantifies the non-linear relationship between the key parameters of the mix proportion and the compressive strength data of autoclaved blocks made of dredged sediment concrete through the Bayesian optimization support vector machine model, and combines the multi-objective particle swarm optimization algorithm for multi-objective trade-off of compressive strength, material cost, and utilization rate of dredged sediment, making the determined mix proportion of autoclaved blocks made of dredged sediment concrete more reasonable.

[0155] Further, the to-be-processed sample set includes a training sample set and a test sample set. The non-linear mapping relationship function determination module 82 is specifically used for:

[0156] Select a Gaussian process as the surrogate model for Bayesian optimization and use the expected improvement as the acquisition function;

[0157] Randomly sample multiple groups of model parameters in the model parameter search space of the support vector machine model, train the support vector machine model through the training sample set, calculate the root mean square error, and construct an initial data set;

[0158] Perform multiple Bayesian optimization iterations based on the initial data set. In each iteration, select the next group of model parameters by maximizing the expected improvement value. When the root mean square error obtained from consecutive multiple iterations reaches the first iteration termination condition, terminate the iteration, and select the group of model parameters with the smallest root mean square error as the optimal model parameters;

[0159] Verify the prediction results of the trained support vector machine model through the test sample set combined with the optimal model parameters;

[0160] When the prediction result meets the set conditions, the trained support vector machine model is used as the compressive strength prediction model, and the non-linear mapping relationship function between the key parameters of the mix proportion and the compressive strength data is output.

[0161] Furthermore, the model parameter search space is composed of the value range of the penalty coefficient of the support vector machine model and the value range of the parameters of the radial basis kernel function.

[0162] The prediction result of the trained support vector machine model is verified by at least the following parameters: the root mean square error and the goodness of fit corresponding to each test sample in the test sample set.

[0163] Furthermore, the multi-objective optimization function construction module 83 is specifically used for:

[0164] Construct the first objective function of the non-linear mapping relationship function, indicating to find a set of key parameters of the mix proportion to maximize the corresponding compressive strength data.

[0165] Construct the second objective function of the production material cost of the autoclaved block made of dredged sediment concrete, indicating to find a set of usage amounts of the production materials of the autoclaved block made of dredged sediment concrete to minimize the corresponding production material cost of the autoclaved block made of dredged sediment concrete.

[0166] Construct the third objective function of the dredged sediment utilization rate, indicating to find a dredged sediment admixture amount to maximize the corresponding dredged sediment utilization rate.

[0167] Construct a multi-objective optimization function according to the first objective function, the second objective function and the third objective function.

[0168] Furthermore, the constraint conditions include value constraints, usage constraints and volume constraints. The constraint condition construction module 84 is specifically used for:

[0169] Construct the value constraint, and the value constraint indicates that each parameter in the key parameters of the mix proportion except the water-binder ratio satisfies its corresponding value range.

[0170] Construct the usage constraint, and the usage constraint indicates that each material in the production materials of the autoclaved block made of dredged sediment concrete except water satisfies its corresponding usage range. The production materials of the autoclaved block made of dredged sediment concrete at least include dredged sediment, cement, fly ash, quicklime, gypsum, aluminum powder and water.

[0171] Construct the volume constraint, and the volume constraint indicates that the wet volume of each cubic meter of autoclaved block made of dredged sediment concrete is 1, and the wet volume of the autoclaved block made of dredged sediment concrete is the sum of the usage amounts of each material included in the production materials of the autoclaved block made of dredged sediment concrete and the ratio of the material density.

[0172] Further, the multi-objective optimization module 85 is specifically configured to:

[0173] Construct a multi-objective particle swarm optimization algorithm based on the multi-objective optimization function and the constraint conditions;

[0174] Perform particle swarm initialization and encoding. Use the mix ratio key parameters as particles, encode the mix ratio key parameters and the corresponding production material dosages into multi-dimensional vectors, and initialize the position and velocity of each particle;

[0175] Adopt a linear decreasing strategy to dynamically adjust the global and local search capabilities, set the individual cognitive factor and the social cognitive factor, and introduce a random perturbation term to enhance the search ability;

[0176] Set up an external archive to store non-dominated solutions. During the multi-objective optimization of the mix ratio key parameters, add the particles that meet the archive addition conditions to the archive, and sort the solutions in the archive according to the values of the multi-objective optimization function to determine the solutions to be retained;

[0177] If a particle does not improve its domination level for consecutive iterations, trigger mutation, apply Gaussian perturbation to the position of the particle, and the perturbation amplitude decays with the number of iterations;

[0178] Update the velocity and position of the particles. When the second iteration termination condition is reached, determine the target mix ratio of the autoclaved block made of dredged sediment concrete based on the non-dominated solution set in the external archive.

[0179] Further, the archive addition conditions at least include one of the following: the new particle is not dominated by any solution in the archive; the new particle and the solutions in the archive are mutually non-dominated and can improve the diversity of the solution set.

[0180] The autoclaved block mix ratio optimization device provided by the embodiments of the present invention can execute the autoclaved block mix ratio optimization method provided by any embodiment of the present invention, and has the corresponding functional modules and beneficial effects for executing the method.

[0181] Embodiment Five

[0182] Figure 9 It is a schematic structural diagram of an electronic device implementing the embodiments of the present invention. The electronic device is intended to represent various forms of digital computers, such as, laptop computers, desktop computers, workbenches, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The components shown herein, their connections and relationships, and their functions are only examples and are not intended to limit the implementation of the present invention described and / or claimed herein.

[0183] As Figure 9As shown, the electronic device 10 includes at least one processor 11 and a memory communicatively connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc. Among them, the memory stores a computer program executable by the at least one processor. The processor 11 can execute various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 into the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. The input / output (I / O) interface 15 is also connected to the bus 14.

[0184] Multiple components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a disk, an optical disc, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0185] The processor 11 can be various general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the processor 11 include but are not limited to a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as the method for optimizing the mix ratio of autoclaved aerated concrete blocks with dredged sediment.

[0186] In some embodiments, the method for optimizing the mix ratio of autoclaved aerated concrete blocks with dredged sediment can be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as the storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed onto the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the method for optimizing the mix ratio of autoclaved aerated concrete blocks with dredged sediment described above can be executed. Alternatively, in other embodiments, the processor 11 can be configured to execute the method for optimizing the mix ratio of autoclaved aerated concrete blocks with dredged sediment in any other appropriate manner (for example, by means of firmware).

[0187] The various embodiments of the systems and techniques described above in this document can be implemented in digital electronic circuitry, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems-on-chip (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special-purpose or general-purpose programmable processor that receives data and instructions from a storage system, at least one input device, and at least one output device, and transmits the data and instructions to the storage system, the at least one input device, and the at least one output device.

[0188] The computer programs for implementing the methods of the present invention can be written in any combination of one or more programming languages. These computer programs can be provided to a processor of a general purpose computer, a special purpose computer, or other programmable data processing apparatus, such that the computer programs, when executed by the processor, cause the functions / operations specified in the flowchart and / or block diagram to be implemented. The computer programs can be executed entirely on the machine, partly on the machine, as a stand-alone software package partly on the machine and partly on a remote machine or entirely on the remote machine or server.

[0189] In the context of the present invention, a computer-readable storage medium can be a tangible medium that can contain or store a computer program for use by or in connection with an instruction execution system, apparatus, or device. The computer-readable storage medium can include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. Alternatively, the computer-readable storage medium can be a machine-readable signal medium. More specific examples of the machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0190] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or a trackball) through which the user can provide input to the electronic device. Other kinds of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0191] The systems and techniques described herein can be implemented in a computing system including backend components (e.g., as a data server), or a computing system including middleware components (e.g., an application server), or a computing system including frontend components (e.g., a user computer having a graphical user interface or a web browser through which the user can interact with an implementation of the systems and techniques described herein), or a computing system including any combination of such backend components, middleware components, or frontend components. The components of the system can be interconnected to each other by digital data communication in any form or medium (e.g., a communication network). Examples of communication networks include: local area network (LAN), wide area network (WAN), blockchain network, and the Internet.

[0192] The computing system can include a client and a server. The client and the server are generally far from each other and usually interact through a communication network. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or a cloud host, which is a host product in the cloud computing service system and solves the defects of difficult management and weak business scalability existing in traditional physical hosts and VPS services.

[0193] It should be understood that various forms of the processes shown above can be used, with steps reordered, added, or deleted. For example, the steps recited in the present invention can be executed in parallel, sequentially, or in a different order, as long as the desired results of the technical solution of the present invention can be achieved, and this is not limited herein.

[0194] The above specific embodiments do not constitute a limitation on the protection scope of the present invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. An optimization method for the mix proportion of autoclaved blocks made of dredged sediment concrete, characterized in that, Including: Obtain the key parameters of the mix proportion and the corresponding compressive strength data of the autoclaved block made of dredged sediment concrete, and construct a sample set to be processed. The key parameters of the mix proportion include at least the content of dredged sediment, the content of cement, the content of aluminum powder, the autoclaving pressure, the autoclaving time, and the water-binder ratio. Based on the sample set to be processed, use Bayesian optimization to search for the model parameters of the support vector machine model, establish a compressive strength prediction model, and determine the non-linear mapping relationship function between the key parameters of the mix proportion and the compressive strength data. Construct a multi-objective optimization function according to the non-linear mapping relationship function, the production material cost of the autoclaved block made of dredged sediment concrete, and the utilization rate of dredged sediment. Construct constraint conditions according to the range of key parameters of the mix proportion, the range of production material usage, and the wet volume of per cubic meter of autoclaved block made of dredged sediment concrete. Based on the multi-objective optimization function and the constraint conditions, perform multi-objective optimization of the key parameters of the mix proportion through the multi-objective particle swarm optimization algorithm to determine the target mix proportion of the autoclaved block made of dredged sediment concrete.

2. The method according to claim 1, characterized in that, The sample set to be processed includes a training sample set and a test sample set. Based on the sample set to be processed, use Bayesian optimization to search for the model parameters of the support vector machine model, establish a compressive strength prediction model, and determine the non-linear mapping relationship function between the key parameters of the mix proportion and the compressive strength data, including: Select the Gaussian process as the surrogate model of Bayesian optimization and use the expected improvement as the acquisition function. Randomly sample multiple groups of model parameters in the model parameter search space of the support vector machine model, train the support vector machine model through the training sample set, and calculate the root mean square error to construct an initial data set. Perform multiple Bayesian optimization iterations based on the initial data set. In each iteration, select the next group of model parameters by maximizing the expected improvement value. When the root mean square error obtained from consecutive multiple iterations reaches the first iteration termination condition, terminate the iteration, and select the group of model parameters with the smallest root mean square error as the optimal model parameters. Verify the prediction results of the trained support vector machine model through the test sample set combined with the optimal model parameters. When the prediction results meet the set conditions, use the trained support vector machine model as the compressive strength prediction model and output the non-linear mapping relationship function between the key parameters of the mix proportion and the compressive strength data.

3. The method according to claim 2, wherein The model parameter search space is composed of the value range of the penalty coefficient of the support vector machine model and the value range of the radial basis kernel function parameter. The prediction results of the trained support vector machine model are verified by at least the following parameters: the root mean square error and the goodness of fit corresponding to each test sample in the test sample set.

4. The method according to claim 1, wherein Construct a multi-objective optimization function according to the non-linear mapping relationship function, the production material cost of the autoclaved block made of dredged sediment concrete, and the utilization rate of dredged sediment, including: Construct the first objective function of the non-linear mapping relationship function, indicating to find a set of key parameters of the mix proportion such that the corresponding compressive strength data is the largest. Construct the second objective function of the production material cost of the autoclaved block made of dredged sediment concrete, indicating to find a set of production material usage of the autoclaved block made of dredged sediment concrete such that the corresponding production material cost of the autoclaved block made of dredged sediment concrete is the smallest. Construct the third objective function for the utilization rate of dredged sediment, indicating to find an amount of dredged sediment admixture such that the corresponding utilization rate of dredged sediment is maximized; Construct a multi-objective optimization function according to the first objective function, the second objective function and the third objective function.

5. The method according to claim 1, wherein The constraint conditions include value constraints, dosage constraints and volume constraints. According to the range of key parameters of the mix ratio, the range of production material dosages, and the wet volume of autoclaved blocks of dredged sediment concrete per cubic meter, the constraint conditions are constructed as follows: Construct the value constraint, which indicates that each parameter in the key parameters of the mix ratio except the water-binder ratio satisfies its corresponding value range; Construct the dosage constraint, which indicates that each material in the production materials of autoclaved blocks of dredged sediment concrete except water satisfies its corresponding dosage range. The production materials of autoclaved blocks of dredged sediment concrete at least include dredged sediment, cement, fly ash, quicklime, gypsum, aluminum powder and water; Construct the volume constraint, which indicates that the wet volume of autoclaved blocks of dredged sediment concrete per cubic meter is 1. The wet volume of autoclaved blocks of dredged sediment concrete is the sum of the dosages of each material included in the production materials of autoclaved blocks of dredged sediment concrete and the ratio of the material density.

6. The method according to claim 1, characterized in that, Based on the multi-objective optimization function and the constraint conditions, perform multi-objective optimization of the key parameters of the mix ratio through the multi-objective particle swarm optimization algorithm to determine the target mix ratio of autoclaved blocks of dredged sediment concrete, including: Based on the multi-objective optimization function and the constraint conditions, construct a multi-objective particle swarm optimization algorithm; Perform particle swarm initialization and encoding. Use the key parameters of the mix ratio as particles, encode the key parameters of the mix ratio and the corresponding production material dosages as multi-dimensional vectors, and initialize the position and velocity of each particle; Adopt a linear decreasing strategy to dynamically adjust the global and local search capabilities, set the individual cognitive factor and the social cognitive factor, and introduce a random perturbation term to enhance the search ability; Set an external archive to store non-dominated solutions. During the multi-objective optimization of the key parameters of the mix ratio, add the particles that meet the archive addition conditions to the archive, and sort the solutions in the archive according to the values of the multi-objective optimization function to determine the solutions to be retained; If the particle does not improve the dominance level for consecutive iterations, trigger mutation, apply Gaussian perturbation to the position of the particle, and the perturbation amplitude decays with the number of iterations; Update the velocity and position of the particle. When the second iteration termination condition is reached, determine the target mix ratio of autoclaved blocks of dredged sediment concrete based on the non-dominated solution set in the external archive.

7. The method according to claim 6, characterized in that, The archive addition conditions at least include one of the following: the new particle is not dominated by any solution in the archive; the new particle and the solutions in the archive are mutually non-dominated and can improve the diversity of the solution set.

8. An autoclaved block mix proportion optimization device for dredged sediment concrete, characterized in that, Include: A to-be-processed sample set construction module, which is used to obtain the key parameters of the mix ratio of autoclaved blocks of dredged sediment concrete and the corresponding compressive strength data, and construct a to-be-processed sample set. The key parameters of the mix ratio at least include the amount of dredged sediment admixture, the amount of cement admixture, the amount of aluminum powder admixture, the autoclaving pressure, the autoclaving time and the water-binder ratio; A non-linear mapping relationship function determination module, which is used to establish a compressive strength prediction model by using Bayesian optimization to search for the model parameters of a support vector machine model based on the to-be-processed sample set, and determine the non-linear mapping relationship function between the key parameters of the mix ratio and the compressive strength data; A multi-objective optimization function construction module, which is used to construct a multi-objective optimization function according to the non-linear mapping relationship function, the production material cost of the autoclaved block made of dredged sediment concrete, and the utilization rate of dredged sediment; A constraint condition construction module, which is used to construct constraint conditions according to the key parameter range of the mix ratio, the production material usage range, and the wet volume of per cubic meter of autoclaved block made of dredged sediment concrete; A multi-objective optimization module, which is used to perform multi-objective optimization of the key parameters of the mix ratio through a multi-objective particle swarm optimization algorithm based on the multi-objective optimization function and the constraint conditions, and determine the target mix ratio of the autoclaved block made of dredged sediment concrete.

9. An electronic device, characterized in that, Comprising: At least one processor; And A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the method according to any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the method according to any one of claims 1-7.

Images