Magnesium alloy dissolution behavior control method and system driven by AI prediction

Abstract

The invention provides an AI prediction driven magnesium alloy dissolution behavior control method and system, and relates to the technical field of new material design, and the method comprises the steps: carrying out the parameter collection of a use water area environment of a sonar buoy, and obtaining an environment data set; by taking the environment data set as a conditional constraint and an alloy component parameter threshold value as an optimization space, alloy component parameter optimization is carried out based on the component fitness evaluation function until convergence, and an optimal component parameter is output; and preparing the sonar buoy according to the optimal component parameters. By means of the method, the technical problem that a traditional magnesium-based alloy production method cannot accurately set magnesium-based alloy component parameters according to the sonar buoy using environment can be solved, and scientificity and accuracy of magnesium-based alloy component parameter setting can be improved through environment data driving and alloy component optimization design; the controllable dissolution rate and stable corrosion performance of the sonar buoy in a specific environment are realized, so that the performance of the buoy is effectively improved, and the service life of the buoy is effectively prolonged.

Description

Technical Field

[0001] The present application relates to the field of new material design technology, and in particular to an AI prediction-driven magnesium alloy dissolution behavior control method and system. Background Art

[0002] Magnesium-based alloys have been widely used in aviation, automobiles, electronic equipment and other fields due to their excellent lightweight properties and good biocompatibility. In particular, in the manufacture of disposable sonar buoys, magnesium-based alloys have become an ideal material choice due to their excellent solubility properties. However, traditional magnesium-based alloy production methods mainly focus on the setting of fixed components and lack precise adjustment and optimized design for specific application environments, resulting in their dissolution rate and corrosion performance in specific environments often failing to meet the requirements.

[0003] In the application of disposable sonar buoys, since the buoys need to work stably for a long time in a specific water environment, the dissolution rate and corrosion behavior of their magnesium-based alloys are crucial to the performance and service life of the buoys. Factors such as temperature, pH value, salinity, etc. in different water environments directly affect the corrosion rate and dissolution behavior of magnesium-based alloys. However, traditional alloy design methods often cannot accurately consider these environmental factors, resulting in limitations and uncertainties in composition design. Therefore, how to accurately design the alloy composition according to the specific use environment of the sonar buoy has become a technical problem that needs to be solved urgently. Summary of the invention

[0004] The purpose of this application is to provide an AI prediction-driven magnesium alloy dissolution behavior control method and system to solve the technical problem that traditional magnesium-based alloy production methods cannot accurately set the magnesium-based alloy composition parameters according to the sonar buoy usage environment.

[0005] In view of the above problems, the present application provides an AI prediction-driven magnesium alloy dissolution behavior control method and system.

[0006] In the first aspect, the present application provides an AI prediction-driven magnesium alloy dissolution behavior control method, which is implemented through an AI prediction-driven magnesium alloy dissolution behavior control system, including: collecting parameters of the water environment in which the sonar buoy is used to obtain an environmental data set, wherein the sonar buoy is a disposable sonar buoy; configuring alloy composition parameter thresholds for the magnesium-based alloy required for preparing the sonar buoy; using the environmental data set as a conditional constraint and the alloy composition parameter threshold as an optimization space, optimizing the alloy composition parameters based on a composition fitness evaluation function until convergence, and outputting the optimal composition parameters; preparing the sonar buoy according to the optimal composition parameters.

[0007] Optionally, the AI ​​prediction-driven magnesium alloy dissolution behavior control method also includes: the environmental data set includes the average water temperature, solution pH value and electrolyte concentration.

[0008] Optionally, the AI ​​prediction-driven magnesium alloy dissolution behavior control method also includes: configuring alloy composition parameter thresholds for the magnesium-based alloy required to prepare the sonar buoy, wherein the alloy composition parameter thresholds include alloy composition thresholds and addition ratio thresholds corresponding to each alloy component type, and the alloy component types include at least magnesium, zinc, aluminum and manganese.

[0009] Optionally, the AI ​​prediction-driven magnesium alloy dissolution behavior control method also includes: setting an environmental parameter tolerance range, and expanding the environmental data set according to the environmental parameter tolerance range to obtain an environmental data range set; using the environmental data range set as a constraint, guided by magnesium-based alloy preparation, performing preparation information retrieval based on big data, and collecting sample alloy composition parameter sets and sample performance parameter sets, wherein performance parameters include dissolution rate uniformity coefficient and corrosion resistance coefficient; using the sample alloy composition parameter set and the sample performance parameter set as training data, obtaining an alloy performance predictor based on machine learning training; utilizing the alloy performance predictor, using the alloy composition parameter threshold as the optimization space, optimizing the alloy composition parameters based on the composition fitness evaluation function, and outputting the optimal composition parameters.

[0010] Optionally, the AI ​​prediction-driven magnesium alloy dissolution behavior control method also includes: configuring Q prediction operators based on machine learning, wherein Q is an integer greater than or equal to 2, and the prediction operators include at least a BP neural network and a random decision forest; using the sample alloy composition parameter set and the sample performance parameter set as training data, and dividing the training data into Q equal parts to obtain Q training sets; using the Q training sets to perform supervised training on the Q prediction operators until convergence, outputting Q alloy performance prediction branches, and integrating to construct the alloy performance predictor, wherein the output of the alloy performance predictor is the mean of the output results of the Q alloy performance prediction branches.

[0011] Optionally, the AI ​​prediction-driven magnesium alloy dissolution behavior control method also includes: randomly selecting a number of initial composition parameters within the alloy composition parameter threshold; using the alloy performance predictor to perform performance prediction on the several initial composition parameters, respectively, and output a number of predicted rate uniformity coefficients and a number of predicted corrosion resistance coefficients; based on a composition fitness evaluation function, performing weighted calculation on the several predicted rate uniformity coefficients and the several predicted corrosion resistance coefficients, and outputting a number of composition fitnesses; optimizing the alloy composition parameters based on the composition fitness evaluation function and the several composition fitnesses, and outputting the optimal composition parameters.

[0012] Optionally, the AI ​​prediction-driven magnesium alloy dissolution behavior control method also includes: arranging a number of initial composition parameters from large to small according to composition fitness to generate an initial composition parameter sequence; randomly selecting multiple composition parameters at the alloy composition parameter threshold to replace the last 20% parameters of the initial composition parameter sequence to generate an updated initial composition parameter sequence; continuing to perform composition fitness evaluation and parameter iterative update based on the composition fitness evaluation function and the updated initial composition parameter sequence until a predetermined number of iterations is reached, and outputting the first parameter in the current initial composition parameter sequence as the optimal composition parameter.

[0013] In the second aspect, the present application also provides an AI prediction-driven magnesium alloy dissolution behavior control system for executing an AI prediction-driven magnesium alloy dissolution behavior control method as described in the first aspect, including: an environmental parameter acquisition module, used to collect parameters of the water environment of the sonar buoy and obtain an environmental data set, wherein the sonar buoy is a disposable sonar buoy; a composition parameter threshold configuration module, used to configure the alloy composition parameter threshold of the magnesium-based alloy required for preparing the sonar buoy; a composition parameter optimization module, used to use the environmental data set as a conditional constraint and the alloy composition parameter threshold as the optimization space, to optimize the alloy composition parameters based on the composition fitness evaluation function until convergence, and output the optimal composition parameters; a sonar buoy preparation module, used to prepare the sonar buoy according to the optimal composition parameters.

[0014] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0015] The environmental data set is obtained by collecting parameters of the water environment in which the sonar buoy is used; then, the alloy composition parameter threshold of the magnesium-based alloy required for preparing the sonar buoy is configured; then, the environmental data set is used as a conditional constraint, and the alloy composition parameter threshold is used as the optimization space, and the alloy composition parameter is optimized based on the composition fitness evaluation function until convergence, and the optimal composition parameters are output; finally, the sonar buoy is prepared according to the optimal composition parameters. That is to say, through environmental data drive and optimized design of alloy composition, the scientificity and accuracy of the magnesium-based alloy composition parameter setting can be improved, and the dissolution rate of the sonar buoy in a specific environment can be controlled and the corrosion performance can be stable, thereby effectively improving the performance and service life of the buoy, and reducing the experimental cost and time in the traditional research and development process.

[0016] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented according to the contents of the specification, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are specifically cited below. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easy to understand through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the present application or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0018] Figure 1 A schematic diagram of a flow chart of an AI prediction-driven magnesium alloy dissolution behavior control method for this application;

[0019] Figure 2 This is a schematic diagram of the structure of an AI prediction-driven magnesium alloy dissolution behavior control system for this application.

[0020] Description of reference numerals:

[0021] Environmental parameter collection module 11, component parameter threshold configuration module 12, component parameter optimization module 13, sonar buoy preparation module 14. DETAILED DESCRIPTION

[0022] This application solves the technical problem that the traditional magnesium-based alloy production method cannot accurately set the magnesium-based alloy composition parameters according to the sonar buoy use environment by providing an AI prediction-driven magnesium alloy dissolution behavior control method and system. Through environmental data drive and optimized design of alloy composition, the scientificity and accuracy of magnesium-based alloy composition parameter setting can be improved, and the dissolution rate of sonar buoys in specific environments can be controlled and the corrosion performance can be stable, thereby effectively improving the performance and service life of the buoy and reducing the experimental cost and time in the traditional research and development process.

[0023] Below, the technical solutions in the present application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments of the present application. It should be understood that the present application is not limited to the example embodiments described herein. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application. It should also be noted that, for the convenience of description, only the parts related to the present application are shown in the accompanying drawings, rather than all of them.

[0024] For example, please refer to the attached Figure 1 The present application provides an AI prediction-driven magnesium alloy dissolution behavior control method, which is applied to an AI prediction-driven magnesium alloy dissolution behavior control system, and specifically includes the following steps:

[0025] Step 1: Collect parameters of the water environment where the sonar buoy is used to obtain an environmental data set, wherein the sonar buoy is a disposable sonar buoy; Step 2: Configure the alloy composition parameter threshold of the magnesium-based alloy required for preparing the sonar buoy; Step 3: Take the environmental data set as a conditional constraint and the alloy composition parameter threshold as the optimization space, optimize the alloy composition parameters based on the composition fitness evaluation function until convergence, and output the optimal composition parameters; Step 4: Prepare the sonar buoy according to the optimal composition parameters.

[0026] Specifically, first, the parameters of the water environment in which the sonar buoy is used are collected to obtain an environmental data set, wherein the sonar buoy is a disposable sonar buoy, that is, the relevant data of the water environment to be used by the buoy is collected, such as temperature, pH value, salinity, dissolved oxygen concentration, solution composition, etc. These environmental parameters have an important influence on the corrosion rate, dissolution behavior and performance of the magnesium-based alloy. This information can be obtained through environmental monitoring equipment or existing water environment data recording systems (such as satellite data, laboratory environment simulation, etc.), and these data will become input conditions in the subsequent optimization process for adjusting the alloy composition design. Then, the alloy composition parameter threshold of the magnesium-based alloy required for the preparation of the sonar buoy is configured, that is, the range and restriction conditions of the magnesium-based alloy composition used for the sonar buoy are determined. For example, the selected alloy composition ratio (such as magnesium, aluminum, zinc, manganese, etc.) must meet certain minimum or maximum standards to ensure that the physical and chemical properties of the alloy meet the use requirements of the buoy.

[0027] Then, the environmental data set is used as a conditional constraint, the alloy component parameter threshold is used as the optimization space, and the alloy component parameter optimization is performed based on the component fitness evaluation function, that is, the optimization algorithm is used to search for the optimal solution in the set space of the alloy component, and the performance of different alloy component combinations is evaluated by setting an evaluation function (such as dissolution rate, corrosion current density, etc.), and the component combination that best meets the requirements is selected; the algorithm will continue to optimize the alloy composition until the best solution is found or the optimization criteria are met (that is, the result converges), and the optimal component parameters are output. Finally, the sonar buoy is prepared according to the optimal component parameters, that is, according to the optimized alloy composition, the corresponding preparation process (such as casting, extrusion, etc.) is adopted, and the alloy component ratio is accurately applied to the production process of the buoy to ensure that the prepared buoy has the expected performance (such as dissolution rate, corrosion resistance, etc.).

[0028] Through these steps, the overall process collects water environment data, combines it with the set magnesium-based alloy composition parameter thresholds, and uses the optimization algorithm to intelligently optimize the alloy composition. This process ensures that under specific environmental conditions, the magnesium-based alloy can achieve the expected dissolution rate, stable corrosion performance and other key properties, and ultimately produces a high-performance sonar buoy, which improves the reliability, service life and production efficiency of the buoy.

[0029] The AI ​​prediction-driven magnesium alloy dissolution behavior control method is applied to an AI prediction-driven magnesium alloy dissolution behavior control system, which can solve the technical problem that the traditional magnesium-based alloy production method cannot accurately set the magnesium-based alloy composition parameters according to the sonar buoy use environment. The environmental data set is acquired by collecting parameters of the water environment of the sonar buoy; then the alloy composition parameter threshold of the magnesium-based alloy required for the preparation of the sonar buoy is configured; then the environmental data set is used as a conditional constraint, and the alloy composition parameter threshold is used as the optimization space, and the alloy composition parameter is optimized based on the composition fitness evaluation function until convergence, and the optimal composition parameter is output; finally, the sonar buoy is prepared according to the optimal composition parameter. That is to say, through environmental data drive and alloy composition optimization design, the scientificity and accuracy of magnesium-based alloy composition parameter setting can be improved, and the sonar buoy can achieve controllable dissolution rate and stable corrosion performance in a specific environment, thereby effectively improving the performance and service life of the buoy, and reducing the experimental cost and time in the traditional research and development process.

[0030] Furthermore, the present application also includes:

[0031] The environmental data set includes mean water body temperature, solution pH value and electrolyte concentration.

[0032] Specifically, the environmental data set includes the average water temperature, solution pH and electrolyte concentration, which will serve as key conditions for optimizing the composition of magnesium-based alloys. During the optimization process, the input of these environmental factors can more accurately adjust the alloy composition to ensure that the dissolution rate, corrosion performance and stability of the buoy in a specific environment meet the expected requirements.

[0033] Furthermore, the present application also includes:

[0034] The alloy composition parameter thresholds of the magnesium-based alloy required for preparing the sonar buoy are configured, wherein the alloy composition parameter thresholds include alloy composition thresholds and addition ratio thresholds corresponding to each alloy composition type, and the alloy composition types include at least magnesium, zinc, aluminum and manganese.

[0035] Specifically, the alloy composition parameter thresholds of the magnesium-based alloy required for preparing the sonar buoy are configured, wherein the alloy composition parameter thresholds include alloy composition thresholds, alloy composition types refer to the basic elements that make up the alloy, such as magnesium, zinc, aluminum and manganese, and the addition ratio thresholds corresponding to each alloy composition type, that is, the specific addition ratio range defined for each alloy element (such as magnesium, zinc, aluminum, manganese, etc.), these ratio thresholds will affect the overall performance of the alloy and determine the corrosion rate, dissolution behavior, etc. of the alloy under different environmental conditions; for example, the content of magnesium may need to be between 80% and 90%, while the proportions of zinc, aluminum, manganese and other elements have their own ranges. For example, the zinc content should be between 5% and 8%, the aluminum content may be required to be between 3% and 6%, and the manganese content between 1% and 2%. The specific range depends on the required alloy properties.

[0036] Furthermore, the present application also includes:

[0037] An environmental parameter tolerance interval is set, and the environmental data set is expanded according to the environmental parameter tolerance interval to obtain an environmental data interval set; with the environmental data interval set as a constraint and magnesium-based alloy preparation as a guide, preparation information retrieval is performed based on big data, and a sample alloy composition parameter set and a sample performance parameter set are collected, wherein the performance parameters include a dissolution rate uniformity coefficient and a corrosion resistance coefficient; with the sample alloy composition parameter set and the sample performance parameter set as training data, an alloy performance predictor is obtained based on machine learning training; using the alloy performance predictor, with the alloy composition parameter threshold as the optimization space, alloy composition parameter optimization is performed based on a composition fitness evaluation function, and the optimal composition parameters are output.

[0038] Specifically, in order to cope with different changes in water environment, it is first necessary to set a tolerance interval for environmental parameters (such as temperature, pH value, electrolyte concentration, etc.), that is, to set the allowable fluctuation range of these parameters, which can reflect the uncertainty or change of the environment. Then, the environmental data set is expanded according to the environmental parameter tolerance interval, that is, by changing the numerical range of the environmental parameters, a set of data containing different environmental conditions is generated, which can ensure that the data set covers a wider range of actual conditions and obtain an environmental data interval set.

[0039] Next, with the environmental data interval set as a constraint and magnesium-based alloy preparation as a guide, preparation information retrieval is performed based on big data. Through big data technology, historical data, literature data, experimental results, etc. related to magnesium-based alloy preparation are retrieved. These data may include alloy composition, preparation process, environmental conditions, and final performance (such as dissolution rate, corrosion resistance, etc.), and sample alloy composition parameter sets and sample performance parameter sets are collected, wherein the performance parameters include dissolution rate uniformity coefficient and corrosion resistance coefficient. Then, the sample alloy composition parameter set and sample performance parameter set are used as training data, and the collected sample alloy composition and performance parameters are used as input data. The alloy performance prediction model is trained through a machine learning algorithm (such as regression analysis, neural network, etc.). Through training, the machine learning model can learn the complex relationship between alloy composition and performance, and generate an alloy performance predictor. This predictor can predict its performance indicators such as dissolution rate and corrosion resistance under given alloy composition parameters. Finally, the alloy performance predictor is used, and the alloy composition parameter threshold is used as the optimization space. The alloy composition parameter optimization is performed based on the composition fitness evaluation function, and the optimal composition parameters are output. By leveraging the advantages of big data and machine learning, we can extract patterns from large amounts of historical data and ensure the performance of sonar buoys in specific water environments through precise adjustments to the alloy composition, greatly improving design efficiency and accuracy.

[0040] Furthermore, the present application also includes:

[0041] Q prediction operators are configured based on machine learning, wherein Q is an integer greater than or equal to 2, and the prediction operators include at least a BP neural network and a random decision forest; the sample alloy composition parameter set and the sample performance parameter set are used as training data, and the training data is equally divided into Q parts to obtain Q training sets; the Q prediction operators are supervised and trained using the Q training sets until convergence, and Q alloy performance prediction branches are output, and the alloy performance predictor is integrated to construct, wherein the output of the alloy performance predictor is the mean of the output results of the Q alloy performance prediction branches.

[0042] Specifically, first, Q prediction operators are configured based on machine learning, where Q is an integer greater than or equal to 2, and the prediction operators include at least BP neural network and random decision forest, where Q represents the number of prediction models used, and Q is an integer greater than or equal to 2. By configuring multiple prediction operators (such as BP neural network and random decision forest), the advantages of multiple models can be integrated, the overfitting problem that may occur in a single model can be reduced, and the accuracy of the prediction can be improved. BP neural network is a classic deep learning algorithm, which can capture the complex nonlinear relationship between alloy composition and performance by training the neural network model through the back propagation algorithm; random decision forest is an integrated learning method, which improves the stability and accuracy of prediction through the combination of multiple decision tree models, and is particularly suitable for complex classification and regression tasks.

[0043] The previously collected sample alloy composition and sample performance parameters are used as training data, and these data are equally divided into Q parts. Each training data will be used to train each prediction operator separately, so as to ensure that each operator can be trained on different data sets, improve the generalization ability of the model, and obtain Q training sets; further use the Q training sets to supervise the Q prediction operators, each prediction operator will use one of the training sets for training, and adjust the model parameters through the supervised learning method until the training process converges. The goal of the training is to enable each prediction operator to accurately predict the alloy properties (such as dissolution rate, corrosion resistance, etc.), the model loss function value during the training process no longer decreases significantly, and the model parameters are stable, that is, the training reaches the optimal state. The Q trained prediction operators are combined into an alloy performance predictor, and the final output of the alloy performance predictor is the mean of the prediction results of the Q prediction operators. In this way, the error of a single model can be reduced and the reliability of the overall prediction results can be improved. For example, BP neural network and random decision forest may give different prediction results, but their mean can represent a more robust and accurate prediction.

[0044] Furthermore, the present application also includes:

[0045] A number of initial composition parameters are randomly selected within the alloy composition parameter threshold; the alloy performance predictor is used to perform performance prediction on the several initial composition parameters respectively, and a number of predicted rate uniformity coefficients and a number of predicted corrosion resistance coefficients are output; based on a composition fitness evaluation function, a weighted calculation is performed on the several predicted rate uniformity coefficients and the several predicted corrosion resistance coefficients, and a number of composition fitnesses are output; based on the composition fitness evaluation function and the several composition fitnesses, alloy composition parameters are optimized, and the optimal composition parameters are output.

[0046] Specifically, first, several initial component parameters are randomly selected at the alloy component parameter threshold. These initial components will be used as the starting point of the optimization algorithm for subsequent performance prediction and optimization processes. Then, the previously trained alloy performance predictor (such as a machine learning model) is used to predict the performance of these initially selected alloy component parameters, and several predicted rate uniformity coefficients and several predicted corrosion resistance coefficients are output. The rate uniformity coefficient describes the uniformity of the dissolution rate in different regions. The larger the coefficient, the more uniform the dissolution process; the corrosion resistance coefficient describes the ability of the alloy to resist corrosion in a specific environment. The higher the corrosion resistance coefficient, the better the stability of the alloy in the environment. Then, based on the component fitness evaluation function, the several predicted rate uniformity coefficients and several predicted corrosion resistance coefficients are weightedly calculated. The purpose of the weighted calculation is to evaluate the fitness of each initial component parameter, that is, their performance under a specific environment. According to actual application requirements, after weighted calculation, several component fitness values ​​can be output. Each component fitness value corresponds to an initial alloy component parameter, indicating the comprehensive performance of the component under current environmental conditions. Finally, the alloy component parameters are optimized based on the component fitness evaluation function and several component fitnesses, and the optimal component parameters are output.

[0047] Furthermore, the present application also includes:

[0048] Arrange a number of initial composition parameters from large to small according to the composition fitness to generate an initial composition parameter sequence; randomly select a plurality of composition parameters at the alloy composition parameter threshold to replace the last 20% parameters of the initial composition parameter sequence to generate an updated initial composition parameter sequence; continue to perform composition fitness evaluation and parameter iterative update based on the composition fitness evaluation function and the updated initial composition parameter sequence until a predetermined number of iterations is reached, and output the first parameter in the current initial composition parameter sequence as the optimal composition parameter.

[0049] Specifically, first, according to the aforementioned component fitness evaluation function, the multiple alloy components initially selected are evaluated, and these components are sorted according to the size of the fitness. The sorting order is from high to low, which means that the components with high fitness are placed in front and the components with low fitness are placed in the back; through sorting, a component parameter sequence is formed, in which the alloy components in the front represent the selection with better performance, and the components in the back are relatively poor. Then, in the component parameter sequence, the last 20% of the parameters (i.e., the components with lower fitness) are selected, and these poor parameters are replaced by randomly selecting new component parameters within the alloy component parameter threshold range. This operation is to introduce a new search space to avoid the algorithm from falling into a local optimal solution; after the replacement operation, a new initial component parameter sequence is generated, which contains part of the original components and part of the new components. The updated initial composition parameter sequence is further used to continue the composition fitness evaluation, and the fitness of the alloy composition is updated according to the evaluation results, thereby optimizing the composition parameters. This process will be iterated multiple times. In each iteration, the alloy composition parameters will be adjusted according to the fitness evaluation and update rules until a predetermined number of iterations is reached. The predetermined number of iterations can be set according to demand, usually determined by experiments or experience. After multiple iterative optimizations, the first one in the current initial composition parameter sequence (i.e., the component with the best fitness) will be selected as the final optimal composition parameter as the optimal formula for the magnesium-based alloy.

[0050] This process increases the ability of exploration and optimization in the optimization of alloy composition parameters by introducing random replacement, sorting and iterative updating. Through multiple iterations, the adaptability of the alloy composition is gradually improved, and finally the optimal combination of alloy composition is found. This process effectively avoids local optimal solutions and improves the accuracy and adaptability of the alloy composition, which can meet the performance requirements in specific application environments.

[0051] In summary, the AI ​​prediction-driven magnesium alloy dissolution behavior control method provided in this application has the following technical effects:

[0052] The environmental data set is obtained by collecting parameters of the water environment in which the sonar buoy is used; then, the alloy composition parameter threshold of the magnesium-based alloy required for preparing the sonar buoy is configured; then, the environmental data set is used as a conditional constraint, and the alloy composition parameter threshold is used as the optimization space, and the alloy composition parameter is optimized based on the composition fitness evaluation function until convergence, and the optimal composition parameters are output; finally, the sonar buoy is prepared according to the optimal composition parameters. That is to say, through environmental data drive and optimized design of alloy composition, the scientificity and accuracy of the magnesium-based alloy composition parameter setting can be improved, and the dissolution rate of the sonar buoy in a specific environment can be controlled and the corrosion performance can be stable, thereby effectively improving the performance and service life of the buoy, and reducing the experimental cost and time in the traditional research and development process.

[0053] Embodiment 2: Based on the same invention concept as the AI ​​prediction-driven magnesium alloy dissolution behavior control method in the aforementioned embodiment, the present application also provides an AI prediction-driven magnesium alloy dissolution behavior control system, see Attachment 2. Figure 2 ,include:

[0054] An environmental parameter acquisition module 11 is used to collect parameters of the water environment in which the sonar buoy is used and obtain an environmental data set, wherein the sonar buoy is a disposable sonar buoy; a composition parameter threshold configuration module 12 is used to configure the alloy composition parameter threshold of the magnesium-based alloy required for preparing the sonar buoy; a composition parameter optimization module 13 is used to use the environmental data set as a conditional constraint and the alloy composition parameter threshold as an optimization space to optimize the alloy composition parameters based on a composition fitness evaluation function until convergence and output the optimal composition parameters; a sonar buoy preparation module 14 is used to prepare the sonar buoy according to the optimal composition parameters.

[0055] Furthermore, the AI ​​prediction-driven magnesium alloy dissolution behavior control system is also used for: the environmental data set includes the average water temperature, solution pH value and electrolyte concentration.

[0056] Furthermore, the AI ​​prediction-driven magnesium alloy dissolution behavior control system is also used to: configure the alloy composition parameter thresholds of the magnesium-based alloy required for preparing sonar buoys, wherein the alloy composition parameter thresholds include alloy composition thresholds and addition ratio thresholds corresponding to each alloy component type, and the alloy component types include at least magnesium, zinc, aluminum and manganese.

[0057] Furthermore, the AI ​​prediction-driven magnesium alloy dissolution behavior control system is also used to: set an environmental parameter tolerance range, and expand the environmental data set according to the environmental parameter tolerance range to obtain an environmental data range set; use the environmental data range set as a constraint, use magnesium-based alloy preparation as a guide, perform preparation information retrieval based on big data, collect sample alloy composition parameter sets and sample performance parameter sets, wherein performance parameters include dissolution rate uniformity coefficient and corrosion resistance coefficient; use the sample alloy composition parameter set and sample performance parameter set as training data, and obtain an alloy performance predictor based on machine learning training; use the alloy performance predictor, use the alloy composition parameter threshold as the optimization space, optimize the alloy composition parameters based on the composition fitness evaluation function, and output the optimal composition parameters.

[0058] Furthermore, the AI ​​prediction-driven magnesium alloy dissolution behavior control system is also used to: configure Q prediction operators based on machine learning, wherein Q is an integer greater than or equal to 2, and the prediction operators include at least a BP neural network and a random decision forest; use the sample alloy composition parameter set and the sample performance parameter set as training data, and divide the training data into Q equal parts to obtain Q training sets; use the Q training sets to perform supervised training on the Q prediction operators until convergence, output Q alloy performance prediction branches, and integrate to construct the alloy performance predictor, wherein the output of the alloy performance predictor is the mean of the output results of the Q alloy performance prediction branches.

[0059] Furthermore, the AI ​​prediction-driven magnesium alloy dissolution behavior control system is also used to: randomly select a number of initial composition parameters within the alloy composition parameter threshold; use the alloy performance predictor to perform performance prediction on the several initial composition parameters respectively, and output a number of predicted rate uniformity coefficients and a number of predicted corrosion resistance coefficients; based on a composition fitness evaluation function, perform weighted calculation on the several predicted rate uniformity coefficients and the several predicted corrosion resistance coefficients, and output a number of composition fitnesses; optimize the alloy composition parameters based on the composition fitness evaluation function and the several composition fitnesses, and output the optimal composition parameters.

[0060] Furthermore, the AI ​​prediction-driven magnesium alloy dissolution behavior control system is also used to: arrange a number of initial composition parameters from large to small according to composition fitness to generate an initial composition parameter sequence; randomly select multiple composition parameters at the alloy composition parameter threshold to replace the last 20% parameters of the initial composition parameter sequence to generate an updated initial composition parameter sequence; continue to perform composition fitness evaluation and parameter iterative update based on the composition fitness evaluation function and the updated initial composition parameter sequence until a predetermined number of iterations is reached, and output the first parameter in the current initial composition parameter sequence as the optimal composition parameter.

[0061] Each embodiment in this specification is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The AI ​​prediction-driven magnesium alloy dissolution behavior control method and specific examples in the aforementioned embodiment 1 are also applicable to the AI ​​prediction-driven magnesium alloy dissolution behavior control system of this embodiment. Through the aforementioned detailed description of the AI ​​prediction-driven magnesium alloy dissolution behavior control method, those skilled in the art can clearly understand the AI ​​prediction-driven magnesium alloy dissolution behavior control system of this embodiment, so for the sake of brevity of the specification, it will not be described in detail here. For the system disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part description.

[0062] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

[0063] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application belong to the scope of the present application and its equivalent technology, the present application is also intended to include these modifications and variations.

Claims

1. AI prediction-driven magnesium alloy dissolution behavior control method, characterized in that the method include: Collecting parameters of the water environment where the sonar buoy is used to obtain an environmental data set, wherein the sonar buoy is a disposable sonar buoy; Configuring alloy composition parameter thresholds for magnesium-based alloys required for preparing sonar buoys; Taking the environmental data set as a conditional constraint and the alloy composition parameter threshold as an optimization space, optimizing the alloy composition parameters based on the composition fitness evaluation function until convergence, and outputting the optimal composition parameters; The sonar buoy is prepared according to the optimal composition parameters.

2. The AI ​​prediction-driven magnesium alloy dissolution behavior control method according to claim 1 is characterized in that: The environmental data set includes mean water body temperature, solution pH value and electrolyte concentration.

3. The AI ​​prediction-driven magnesium alloy dissolution behavior control method according to claim 2 is characterized in that: The alloy composition parameter thresholds of the magnesium-based alloy required for preparing the sonar buoy are configured, wherein the alloy composition parameter thresholds include alloy composition thresholds and addition ratio thresholds corresponding to each alloy composition type, and the alloy composition types include at least magnesium, zinc, aluminum and manganese.

4. The AI ​​prediction-driven magnesium alloy dissolution behavior control method according to claim 3 is characterized in that: Taking the environmental data set as a conditional constraint, taking the alloy composition parameter threshold as an optimization space, and performing alloy composition parameter optimization based on a composition fitness evaluation function, the method includes: Setting an environmental parameter tolerance interval, and expanding the environmental data set according to the environmental parameter tolerance interval to obtain an environmental data interval set; Taking the environmental data interval set as a constraint and magnesium-based alloy preparation as a guide, preparation information retrieval is performed based on big data to collect sample alloy composition parameter sets and sample performance parameter sets, wherein the performance parameters include a dissolution rate uniformity coefficient and a corrosion resistance coefficient; Using the sample alloy composition parameter set and the sample performance parameter set as training data, an alloy performance predictor is obtained based on machine learning training; The alloy property predictor is used to optimize the alloy composition parameters based on the alloy composition parameter threshold as the optimization space, and the alloy composition parameters are optimized based on the composition fitness evaluation function to output the optimal composition parameters.

5. The AI ​​prediction-driven magnesium alloy dissolution behavior control method according to claim 4 is characterized in that: Alloy property predictors are obtained based on machine learning training, including: Configure Q prediction operators based on machine learning, where Q is an integer greater than or equal to 2, and the prediction operators include at least a BP neural network and a random decision forest; The sample alloy composition parameter set and the sample performance parameter set are used as training data, and the training data is equally divided into Q parts to obtain Q training sets; The Q prediction operators are supervisedly trained using the Q training sets until convergence, and Q alloy property prediction branches are outputted to integrate and construct the alloy property predictor, wherein the output of the alloy property predictor is the mean of the output results of the Q alloy property prediction branches.

6. The AI ​​prediction-driven magnesium alloy dissolution behavior control method according to claim 4, characterized in that: Utilizing the alloy property predictor, taking the alloy composition parameter threshold as the optimization space, optimizing the alloy composition parameters based on the composition fitness evaluation function, and outputting the optimal composition parameters, including: Randomly selecting a number of initial composition parameters within the alloy composition parameter threshold; Using the alloy performance predictor, respectively predict the performance of the several initial composition parameters, and output several predicted rate uniformity coefficients and several predicted corrosion resistance coefficients; Based on the component fitness evaluation function, weighted calculation is performed on the plurality of predicted rate uniformity coefficients and the plurality of predicted corrosion resistance coefficients, and a plurality of component fitnesses are output; Based on the composition fitness evaluation function and a plurality of composition fitnesses, alloy composition parameters are optimized and the optimal composition parameters are output.

7. The AI ​​prediction-driven magnesium alloy dissolution behavior control method according to claim 6, characterized in that: Optimizing alloy composition parameters based on the composition fitness evaluation function and a plurality of composition fitnesses, and outputting the optimal composition parameters, includes: Arrange several initial component parameters from large to small according to component fitness to generate an initial component parameter sequence; Randomly select multiple composition parameters at the alloy composition parameter threshold to replace the last 20% parameters of the initial composition parameter sequence to generate an updated initial composition parameter sequence; Based on the component fitness evaluation function and the updated initial component parameter sequence, component fitness evaluation and parameter iterative updating are continued until a predetermined number of iterations is reached, and the first parameter in the current initial component parameter sequence is output as the optimal component parameter.

8. AI prediction-driven magnesium alloy dissolution behavior control system, characterized in that: The steps for implementing the AI ​​prediction-driven magnesium alloy dissolution behavior control method according to any one of claims 1 to 7 include: An environmental parameter collection module is used to collect parameters of the water environment of the sonar buoy and obtain an environmental data set, wherein the sonar buoy is a disposable sonar buoy; A composition parameter threshold configuration module is used to configure the alloy composition parameter threshold of the magnesium-based alloy required for preparing the sonar buoy; A composition parameter optimization module is used to optimize the alloy composition parameters based on the composition fitness evaluation function with the environmental data set as a conditional constraint and the alloy composition parameter threshold as an optimization space until convergence, and output the optimal composition parameters; The sonar buoy preparation module is used to prepare the sonar buoy according to the optimal composition parameters.

Images