[go: up one dir, main page]

CN114373523B - Glass hardness prediction method based on squirrel optimization algorithm and machine learning algorithm - Google Patents

Glass hardness prediction method based on squirrel optimization algorithm and machine learning algorithm Download PDF

Info

Publication number
CN114373523B
CN114373523B CN202210281841.9A CN202210281841A CN114373523B CN 114373523 B CN114373523 B CN 114373523B CN 202210281841 A CN202210281841 A CN 202210281841A CN 114373523 B CN114373523 B CN 114373523B
Authority
CN
China
Prior art keywords
squirrel
glass
tree
algorithm
parameters
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202210281841.9A
Other languages
Chinese (zh)
Other versions
CN114373523A (en
Inventor
杨勇
翟华
韩江
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Hefei University of Technology
Original Assignee
Hefei University of Technology
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Hefei University of Technology filed Critical Hefei University of Technology
Priority to CN202210281841.9A priority Critical patent/CN114373523B/en
Publication of CN114373523A publication Critical patent/CN114373523A/en
Application granted granted Critical
Publication of CN114373523B publication Critical patent/CN114373523B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16C—COMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
    • G16C60/00—Computational materials science, i.e. ICT specially adapted for investigating the physical or chemical properties of materials or phenomena associated with their design, synthesis, processing, characterisation or utilisation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • G06N20/20—Ensemble learning
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/004—Artificial life, i.e. computing arrangements simulating life
    • G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16C—COMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
    • G16C20/00—Chemoinformatics, i.e. ICT specially adapted for the handling of physicochemical or structural data of chemical particles, elements, compounds or mixtures
    • G16C20/30—Prediction of properties of chemical compounds, compositions or mixtures
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16C—COMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
    • G16C20/00—Chemoinformatics, i.e. ICT specially adapted for the handling of physicochemical or structural data of chemical particles, elements, compounds or mixtures
    • G16C20/70—Machine learning, data mining or chemometrics
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
    • Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
    • Y02P90/30—Computing systems specially adapted for manufacturing

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Computing Systems (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Software Systems (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Artificial Intelligence (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Data Mining & Analysis (AREA)
  • Evolutionary Computation (AREA)
  • Physics & Mathematics (AREA)
  • Medical Informatics (AREA)
  • General Health & Medical Sciences (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Crystallography & Structural Chemistry (AREA)
  • Health & Medical Sciences (AREA)
  • Chemical & Material Sciences (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Computational Linguistics (AREA)
  • Biophysics (AREA)
  • Databases & Information Systems (AREA)
  • Biomedical Technology (AREA)
  • Molecular Biology (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention discloses a glass hardness prediction method based on a squirrel optimization algorithm and a machine learning algorithm, and relates to the field of electric digital data processing; the method comprises the following steps: constructing a hardness database of glass materials with different components; constructing a feature descriptor; providing a training set sample, and initializing parameters of a squirrel search algorithm; setting optimized parameters of a squirrel search algorithm, optimizing parameters of a Catboost algorithm by using the optimized parameters to obtain optimal parameters of the Catboost algorithm, and establishing a glass hardness prediction model; test sample data is input, and glass hardness is predicted. The method constructs a unique descriptor, uses the squirrel search optimization algorithm for optimizing the parameters of the Catboost algorithm, has a simple structure, improves the convergence speed and precision, can obviously improve the performance of the Catboost algorithm by the optimized optimal Catboost algorithm parameters, and has practical significance for improving the accuracy of predicting the glass hardness.

Description

基于松鼠优化算法和机器学习算法的玻璃硬度预测方法Glass hardness prediction method based on squirrel optimization algorithm and machine learning algorithm

技术领域technical field

本发明涉及电数字数据处理领域，尤其涉及一种基于松鼠优化算法和机器学习算法的玻璃硬度预测方法。The invention relates to the field of electrical digital data processing, in particular to a glass hardness prediction method based on a squirrel optimization algorithm and a machine learning algorithm.

背景技术Background technique

玻璃是一种非平衡、非晶的材料，它能自发地弛豫到过冷的液态。与晶体不同， 玻璃不需要满足严格的化学计量规则，可以被认为是化学元素的连续溶液。因此大量的元素可能成为构成玻璃材料的组分。80种化学元素以1 mol%的量变化可以产生

Figure 965089DEST_PATH_IMAGE001
种可能的玻璃成分。然而，报道的无机玻璃的数量只有
Figure 342981DEST_PATH_IMAGE002
种左右，这意味着探索具有特殊性能的新玻璃形成成分还有巨大的空间。Glass is a non-equilibrium, amorphous material that relaxes spontaneously to a supercooled liquid state. Unlike crystals, glasses do not need to meet strict stoichiometric rules and can be thought of as continuous solutions of chemical elements. Therefore, a large number of elements may become components constituting the glass material. 80 chemical elements can be produced by changing the amount of 1 mol%
Figure 965089DEST_PATH_IMAGE001
possible glass compositions. However, the number of reported inorganic glasses is only
Figure 342981DEST_PATH_IMAGE002
This means that there is enormous scope for exploring new glass-forming compositions with special properties.

硬度作为最重要的机械性能之一，反映了材料对抗永久变形的性能。玻璃材料的硬度不仅取决于化学组分还与其网络结构息息相关，一般而言，构建玻璃网络结构的网络形成体氧化物(如

Figure 678147DEST_PATH_IMAGE003
，
Figure 141489DEST_PATH_IMAGE004
等)越多，硬度值越大，而具有破网作用的网络修饰体氧化物(如碱金属和碱土金属氧化物等)越多，硬度值越小。影响硬度的主要因素为：玻璃的热历史、压力史和外界因素如测量时的温度、湿度以及施加载荷的大小等。现有玻璃材料设计方法中对硬度的预测主要基于组分的加权求和公式，这种预测方法存在预测准确度低的缺点。As one of the most important mechanical properties, hardness reflects the resistance of a material to permanent deformation. The hardness of glass materials depends not only on the chemical composition but also on its network structure. Generally speaking, the network former oxides (such as
Figure 678147DEST_PATH_IMAGE003
,
Figure 141489DEST_PATH_IMAGE004
etc.), the greater the hardness value, and the more network modifier oxides (such as alkali metal and alkaline earth metal oxides, etc.) with network breaking effect, the smaller the hardness value. The main factors that affect the hardness are: the thermal history of the glass, the pressure history and external factors such as temperature, humidity and the size of the applied load during measurement. The prediction of hardness in the existing glass material design methods is mainly based on the weighted summation formula of components, and this prediction method has the disadvantage of low prediction accuracy.

发明内容SUMMARY OF THE INVENTION

为了解决现有技术中无法准确且快速预测玻璃硬度的问题，本发明提出了一种基于松鼠搜索优化算法和机器学习算法的玻璃硬度预测方法。In order to solve the problem that the glass hardness cannot be accurately and quickly predicted in the prior art, the present invention proposes a glass hardness prediction method based on a squirrel search optimization algorithm and a machine learning algorithm.

本发明采用的技术方案是：The technical scheme adopted in the present invention is:

基于松鼠优化算法和机器学习算法的玻璃硬度预测方法，其特征在于，所述方法包括：The glass hardness prediction method based on squirrel optimization algorithm and machine learning algorithm, characterized in that the method includes:

步骤1：采集不同组分氧化物玻璃的硬度数据，构建玻璃硬度数据库，所述玻璃硬度数据库包括一一映射的玻璃组分和其对应的硬度；Step 1: collect the hardness data of oxide glasses with different components, and construct a glass hardness database, the glass hardness database includes one-to-one mapping of glass components and their corresponding hardnesses;

步骤2：基于化学特征，分别以元素摩尔含量

Figure 282621DEST_PATH_IMAGE005
、原子间库伦作用力
Figure 515019DEST_PATH_IMAGE006
、基于力场势的短程相互作用关系
Figure 21087DEST_PATH_IMAGE007
作为输入参数的描述符；Step 2: Based on the chemical characteristics, the molar content of the elements are respectively
Figure 282621DEST_PATH_IMAGE005
, Coulomb force between atoms
Figure 515019DEST_PATH_IMAGE006
, short-range interaction relationship based on force field potential
Figure 21087DEST_PATH_IMAGE007
Descriptor as input parameter;

步骤3：以步骤2构造的描述符为模型的输入，以步骤1构建的硬度数据库为模型的输出，构建训练集、测试集，建立Catboost模型；Step 3: Take the descriptor constructed in Step 2 as the input of the model, and take the hardness database constructed in Step 1 as the output of the model, construct a training set and a test set, and establish a Catboost model;

步骤4：引入松鼠搜索优化算法，优化选取的Catboost模型的参数；Step 4: Introduce the squirrel search optimization algorithm to optimize the parameters of the selected Catboost model;

步骤5：基于优化后的参数建立性能最优的Catboost模型；Step 5: Establish a Catboost model with the best performance based on the optimized parameters;

步骤6：针对待预测的玻璃组分，利用最优的Catboost模型预测该玻璃组分的玻璃硬度。Step 6: For the glass composition to be predicted, use the optimal Catboost model to predict the glass hardness of the glass composition.

进一步地，所述的基于松鼠优化算法和机器学习算法的玻璃硬度预测方法，其特征在于，步骤2包括如下步骤：Further, the described method for predicting glass hardness based on squirrel optimization algorithm and machine learning algorithm is characterized in that step 2 includes the following steps:

步骤2-1：以组成玻璃各个组分摩尔含量

Figure 34042DEST_PATH_IMAGE005
为一组描述符；Step 2-1: To make up the molar content of each component of the glass
Figure 34042DEST_PATH_IMAGE005
is a set of descriptors;

步骤2-2：以组成玻璃各个组分不同原子间库伦作用力构造描述符

Figure 916547DEST_PATH_IMAGE006
：Step 2-2: Construct descriptors based on Coulomb forces between different atoms of each component of the glass
Figure 916547DEST_PATH_IMAGE006
:

Figure 3452DEST_PATH_IMAGE008
Figure 3452DEST_PATH_IMAGE008

其中，

Figure 680421DEST_PATH_IMAGE009
和
Figure 665826DEST_PATH_IMAGE010
为离子
Figure 352022DEST_PATH_IMAGE011
和
Figure 293433DEST_PATH_IMAGE012
的有效离子电荷，
Figure 203620DEST_PATH_IMAGE013
和
Figure 128851DEST_PATH_IMAGE014
分别是组成元素
Figure 87580DEST_PATH_IMAGE011
和
Figure 414656DEST_PATH_IMAGE012
有效离子电荷
Figure 495744DEST_PATH_IMAGE009
和
Figure 642692DEST_PATH_IMAGE010
的摩尔分数；in,
Figure 680421DEST_PATH_IMAGE009
and
Figure 665826DEST_PATH_IMAGE010
for ions
Figure 352022DEST_PATH_IMAGE011
and
Figure 293433DEST_PATH_IMAGE012
The effective ionic charge of ,
Figure 203620DEST_PATH_IMAGE013
and
Figure 128851DEST_PATH_IMAGE014
the constituent elements
Figure 87580DEST_PATH_IMAGE011
and
Figure 414656DEST_PATH_IMAGE012
effective ionic charge
Figure 495744DEST_PATH_IMAGE009
and
Figure 642692DEST_PATH_IMAGE010
mole fraction of ;

步骤2-3：以组成玻璃各个组分不同原子间力场势的短程相互作用关系

Figure 405112DEST_PATH_IMAGE007
构造描述符
Figure 586694DEST_PATH_IMAGE007
：Step 2-3: Use the short-range interaction relationship between the force field potentials between different atoms of each component of the glass
Figure 405112DEST_PATH_IMAGE007
construct descriptor
Figure 586694DEST_PATH_IMAGE007
:

Figure 838684DEST_PATH_IMAGE015
Figure 838684DEST_PATH_IMAGE015

其中，S为组成元素合集，

Figure 472928DEST_PATH_IMAGE016
为组成元素有效离子电荷的摩尔分数，
Figure 39038DEST_PATH_IMAGE017
为Buckingham电势参数，p= -4, -3, -2, -1, 0, 1, 2, 3, 4。Among them, S is the set of constituent elements,
Figure 472928DEST_PATH_IMAGE016
is the mole fraction of the effective ionic charge of the constituent elements,
Figure 39038DEST_PATH_IMAGE017
are Buckingham potential parameters, p = -4, -3, -2, -1, 0, 1, 2, 3, 4.

进一步地，所述的基于松鼠优化算法和机器学习算法的玻璃硬度预测方法，其特征在于，步骤4包括如下步骤：Further, the described method for predicting glass hardness based on squirrel optimization algorithm and machine learning algorithm is characterized in that step 4 includes the following steps:

步骤4-1：选取Catboost算法需优化的参数，松鼠的位置坐标即为需要优化的参数，定义松鼠搜索优化算法的种群规模，维度，最大迭代次数，滑行距离参数和捕食者存在概率；Step 4-1: Select the parameters to be optimized by the Catboost algorithm, the position coordinates of the squirrel are the parameters to be optimized, and define the population size, dimension, maximum number of iterations, sliding distance parameters and predator existence probability of the squirrel search optimization algorithm;

步骤4-2：种群位置初始化，为

Figure 622597DEST_PATH_IMAGE018
只松鼠生成随机位置，第
Figure 248751DEST_PATH_IMAGE011
只松鼠的位置可以通过一个矢量来确定；所有松鼠的位置在边界范围内随机初始化，如下式：Step 4-2: Initialize the population position, which is
Figure 622597DEST_PATH_IMAGE018
A squirrel generates a random location, the first
Figure 248751DEST_PATH_IMAGE011
The position of a single squirrel can be determined by a vector; the positions of all squirrels are randomly initialized within the bounds as follows:

Figure 104711DEST_PATH_IMAGE019
Figure 104711DEST_PATH_IMAGE019

其中，

Figure 474513DEST_PATH_IMAGE020
代表第
Figure 427425DEST_PATH_IMAGE011
只松鼠第
Figure 958901DEST_PATH_IMAGE012
维的值，
Figure 567737DEST_PATH_IMAGE021
、
Figure 272388DEST_PATH_IMAGE022
分别为变量的上下边界，rand为[0, 1]之间的随机数；in,
Figure 474513DEST_PATH_IMAGE020
representative
Figure 427425DEST_PATH_IMAGE011
only squirrel
Figure 958901DEST_PATH_IMAGE012
dimension value,
Figure 567737DEST_PATH_IMAGE021
,
Figure 272388DEST_PATH_IMAGE022
are the upper and lower boundaries of the variable, and rand is a random number between [0, 1];

步骤4-3：根据步骤4-2每只松鼠的位置计算每只松鼠位置所代表参数的模型准确度；Step 4-3: Calculate the model accuracy of the parameters represented by the position of each squirrel according to the position of each squirrel in Step 4-2;

步骤4-4：根据飞行松鼠的准确度，按升序排列它们的位置；Steps 4-4: Arrange their positions in ascending order according to the accuracy of the flying squirrels;

步骤4-5：根据步骤4-4的排序将飞行松鼠按顺序分配到山核桃树、橡树和普通树，其中山核桃树代表全局最优解位置，橡树代表局部最优解位置；Step 4-5: According to the sorting of steps 4-4, the flying squirrels are assigned to the hickory tree, the oak tree and the common tree in order, wherein the hickory tree represents the global optimal solution position, and the oak tree represents the local optimal solution position;

步骤4-6：更新松鼠位置，如下式：Step 4-6: Update the squirrel position as follows:

（1）在橡树上的松鼠向山核桃树移动，(1) The squirrel on the oak tree moves toward the hickory tree,

Figure 283069DEST_PATH_IMAGE023
Figure 283069DEST_PATH_IMAGE023

其中，

Figure 985446DEST_PATH_IMAGE024
是随机滑行距离，
Figure 81578DEST_PATH_IMAGE025
是[0, 1]范围内的随机数，
Figure 589919DEST_PATH_IMAGE026
是山核桃树的位置，t表示当前迭代；滑动常数
Figure 720686DEST_PATH_IMAGE027
实现全局与局部搜索之间的平衡，经过大量分析论证，
Figure 593964DEST_PATH_IMAGE027
的值设为1.9；in,
Figure 985446DEST_PATH_IMAGE024
is the random glide distance,
Figure 81578DEST_PATH_IMAGE025
is a random number in the range [0, 1],
Figure 589919DEST_PATH_IMAGE026
is the position of the pecan tree, and t is the current iteration; the sliding constant
Figure 720686DEST_PATH_IMAGE027
To achieve a balance between global and local search, after a lot of analysis and demonstration,
Figure 593964DEST_PATH_IMAGE027
The value of is set to 1.9;

（2）在普通树上的松鼠向橡树移动，(2) A squirrel on a common tree moves towards an oak tree,

Figure 459283DEST_PATH_IMAGE028
Figure 459283DEST_PATH_IMAGE028

其中，

Figure 974578DEST_PATH_IMAGE029
是[0, 1]范围内的随机；in,
Figure 974578DEST_PATH_IMAGE029
is random in the range [0, 1];

（3）在普通树上的松鼠向山核桃树移动，(3) The squirrel on the common tree moves towards the hickory tree,

Figure 694273DEST_PATH_IMAGE030
Figure 694273DEST_PATH_IMAGE030

其中，

Figure 4031DEST_PATH_IMAGE031
是[0, 1]范围内的随机数；in,
Figure 4031DEST_PATH_IMAGE031
is a random number in the range [0, 1];

步骤4-7：根据步骤4-6更新后每只松鼠的位置计算每只松鼠位置所代表参数的模型准确度，升序排列位置，将飞行松鼠按顺序重新分配到山核桃树、橡子树和普通树；Step 4-7: Calculate the model accuracy of the parameters represented by the position of each squirrel according to the position of each squirrel after updating in steps 4-6, arrange the positions in ascending order, and reassign the flying squirrels to the hickory tree, acorn tree and common tree;

步骤4-8：判断季节变化条件是否满足，如满足更新普通树上松鼠位置，不满足维持原位置，如下式：Step 4-8: Determine whether the seasonal change conditions are satisfied, such as updating the position of the squirrel on the common tree, but not maintaining the original position, as follows:

（1）计算季节常量

Figure 605914DEST_PATH_IMAGE032
：(1) Calculate the seasonal constant
Figure 605914DEST_PATH_IMAGE032
:

Figure 659321DEST_PATH_IMAGE033
Figure 659321DEST_PATH_IMAGE033

（2）计算季节变化条件

Figure 233521DEST_PATH_IMAGE034
：(2) Calculate seasonal variation conditions
Figure 233521DEST_PATH_IMAGE034
:

Figure 714181DEST_PATH_IMAGE035
Figure 714181DEST_PATH_IMAGE035

其中，t和

Figure 68939DEST_PATH_IMAGE036
分别是当前和最大迭代值；where t and
Figure 68939DEST_PATH_IMAGE036
are the current and maximum iteration values, respectively;

（3）如果季节条件得到满足，则随机改变普通树上松鼠的位置：(3) If the seasonal conditions are satisfied, randomly change the position of the squirrel on the ordinary tree:

Figure 394878DEST_PATH_IMAGE037
Figure 394878DEST_PATH_IMAGE037

步骤4-9：重新计算各个松鼠位置上的准确度，升序排列位置，将飞行松鼠分配到山核桃树、橡子树和普通树；Steps 4-9: Recalculate the accuracy on each squirrel position, arrange the positions in ascending order, and assign flying squirrels to hickory, acorn, and common trees;

步骤4-10：重复步骤4-6到4-9，满足迭代条件或最大迭代次数时，结束优化过程；Step 4-10: Repeat steps 4-6 to 4-9, and end the optimization process when the iteration conditions or the maximum number of iterations are satisfied;

步骤4-11：输出山核桃树上松鼠的位置和准确度，山核桃树上松鼠的位置为Catboost参数值。Step 4-11: Output the position and accuracy of the squirrel on the hickory tree, the position of the squirrel on the hickory tree is the Catboost parameter value.

本发明的有益效果在于：The beneficial effects of the present invention are:

（1）本发明建立了基于化学特性的描述符，考虑了不同原子之间离子作用力和短程作用力的相互关系，更加符合实际玻璃的规律，预测结果更为准确；(1) The present invention establishes a descriptor based on chemical properties, and considers the relationship between ionic force and short-range force between different atoms, which is more in line with the laws of actual glass, and the prediction results are more accurate;

（2）本发明将松鼠搜索优化算法用于优化Catboost算法参数寻优，结构简单，提高收敛速度和精度，且寻优得到的最优Catboost算法参数可以较明显地提高Catboost算法的性能，对于提高预测玻璃硬度的准确性具有现实意义。(2) The present invention uses the squirrel search optimization algorithm to optimize the parameters of the Catboost algorithm, the structure is simple, the convergence speed and accuracy are improved, and the optimal Catboost algorithm parameters obtained by the optimization can significantly improve the performance of the Catboost algorithm. The accuracy of predicting glass hardness has practical significance.

附图说明Description of drawings

图1 本发明方法实施的流程图；Fig. 1 is a flow chart of the implementation of the method of the present invention;

图2 本发明描述符构造流程图 。Figure 2 is a flow chart of the construction of the descriptor of the present invention.

具体实施方式Detailed ways

下面将结合本发明实施例中的附图，对本发明实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是本发明一部分实施例，而不是全部的实施例。The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to 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, but not all of the embodiments.

实施例1，如图1、2所示：Embodiment 1, as shown in Figures 1 and 2:

步骤1：集不同组分氧化物玻璃的硬度数据，构建玻璃硬度数据库，具体采集400组

Figure 354744DEST_PATH_IMAGE038
玻璃的硬度数据，构建玻璃硬度数据库；Step 1: Collect the hardness data of oxide glass with different components, build a glass hardness database, and collect 400 groups in detail
Figure 354744DEST_PATH_IMAGE038
The hardness data of glass, build glass hardness database;

步骤2：基于化学特征，分别以元素摩尔含量

Figure 803043DEST_PATH_IMAGE005
、原子间库伦作用力
Figure 582780DEST_PATH_IMAGE006
、基于力场势的短程相互作用关系
Figure 712410DEST_PATH_IMAGE007
作为输入参数的描述符；具体步骤为：Step 2: Based on the chemical characteristics, the molar content of the elements are respectively
Figure 803043DEST_PATH_IMAGE005
, Coulomb force between atoms
Figure 582780DEST_PATH_IMAGE006
, short-range interaction relationship based on force field potential
Figure 712410DEST_PATH_IMAGE007
Descriptor as input parameter; the specific steps are:

步骤2-1：以400组

Figure 526782DEST_PATH_IMAGE038
摩尔含量
Figure 893785DEST_PATH_IMAGE039
、
Figure 160819DEST_PATH_IMAGE040
和
Figure 94140DEST_PATH_IMAGE041
为三组描述符；Step 2-1: Take 400 Groups
Figure 526782DEST_PATH_IMAGE038
Molar content
Figure 893785DEST_PATH_IMAGE039
,
Figure 160819DEST_PATH_IMAGE040
and
Figure 94140DEST_PATH_IMAGE041
are three sets of descriptors;

步骤2-2：以Si、Na、Ca、O原子间库伦作用力构造描述符

Figure 763018DEST_PATH_IMAGE006
：Step 2-2: Construct descriptors with Coulomb forces between Si, Na, Ca, O atoms
Figure 763018DEST_PATH_IMAGE006
:

Figure 553120DEST_PATH_IMAGE008
Figure 553120DEST_PATH_IMAGE008

其中，

Figure 41870DEST_PATH_IMAGE009
和
Figure 44461DEST_PATH_IMAGE010
为离子
Figure 364584DEST_PATH_IMAGE011
和
Figure 528849DEST_PATH_IMAGE012
的有效离子电荷，
Figure 504895DEST_PATH_IMAGE013
和
Figure 45598DEST_PATH_IMAGE014
分别是组成元素
Figure 220227DEST_PATH_IMAGE011
和
Figure 555394DEST_PATH_IMAGE012
有效离子电荷
Figure 18736DEST_PATH_IMAGE009
和
Figure 910600DEST_PATH_IMAGE010
的摩尔分数；in,
Figure 41870DEST_PATH_IMAGE009
and
Figure 44461DEST_PATH_IMAGE010
for ions
Figure 364584DEST_PATH_IMAGE011
and
Figure 528849DEST_PATH_IMAGE012
The effective ionic charge of ,
Figure 504895DEST_PATH_IMAGE013
and
Figure 45598DEST_PATH_IMAGE014
the constituent elements
Figure 220227DEST_PATH_IMAGE011
and
Figure 555394DEST_PATH_IMAGE012
effective ionic charge
Figure 18736DEST_PATH_IMAGE009
and
Figure 910600DEST_PATH_IMAGE010
mole fraction of ;

步骤2-3：以Si、Na、Ca、O原子间力场势的短程相互作用关系

Figure 142998DEST_PATH_IMAGE007
构造描述符
Figure 649066DEST_PATH_IMAGE007
：Step 2-3: Based on the short-range interaction relationship between the Si, Na, Ca, and O atoms
Figure 142998DEST_PATH_IMAGE007
construct descriptor
Figure 649066DEST_PATH_IMAGE007
:

Figure 599704DEST_PATH_IMAGE015
Figure 599704DEST_PATH_IMAGE015

其中，S为组成元素合集，

Figure 278947DEST_PATH_IMAGE016
为组成元素有效离子电荷的摩尔分数，
Figure 631431DEST_PATH_IMAGE017
为Buckingham电势参数，p= -4, -3, -2, -1, 0, 1, 2, 3, 4；Among them, S is the set of constituent elements,
Figure 278947DEST_PATH_IMAGE016
is the mole fraction of the effective ionic charge of the constituent elements,
Figure 631431DEST_PATH_IMAGE017
is the Buckingham potential parameter, p = -4, -3, -2, -1, 0, 1, 2, 3, 4;

步骤3：以步骤2构造的描述符为模型的输入，以步骤1构建的硬度数据库为模型的输出，320组数据构建训练集、80组数据构建测试集，建立Catboost模型；Step 3: Take the descriptor constructed in Step 2 as the input of the model, take the hardness database constructed in Step 1 as the output of the model, construct a training set with 320 sets of data, and construct a test set with 80 sets of data, and establish a Catboost model;

步骤4-1：选取Catboost算法需优化的参数迭代数、学习率、深度，建立松鼠的位置，定义松鼠搜索优化算法的种群规模，维度，最大迭代次数；Step 4-1: Select the number of iterations, learning rate, and depth of parameters to be optimized by the Catboost algorithm, establish the position of the squirrel, and define the population size, dimension, and maximum number of iterations of the squirrel search optimization algorithm;

步骤4-2：种群位置初始化，为50只松鼠生成随机位置，第

Figure 308400DEST_PATH_IMAGE011
只松鼠的位置可以通过一个矢量来确定。所有松鼠的位置在边界范围内随机初始化，如下式：Step 4-2: Initialize the population location, generate random locations for 50 squirrels,
Figure 308400DEST_PATH_IMAGE011
The location of a squirrel can be determined by a vector. The positions of all squirrels are randomly initialized within the bounds as follows:

Figure 543073DEST_PATH_IMAGE019
Figure 543073DEST_PATH_IMAGE019

其中，

Figure 229269DEST_PATH_IMAGE020
代表第
Figure 170680DEST_PATH_IMAGE011
只松鼠第
Figure 284130DEST_PATH_IMAGE012
维的值，
Figure 6098DEST_PATH_IMAGE021
、
Figure 230406DEST_PATH_IMAGE022
分别为变量的上下边界，rand为[0, 1]之间的随机数；in,
Figure 229269DEST_PATH_IMAGE020
representative
Figure 170680DEST_PATH_IMAGE011
only squirrel
Figure 284130DEST_PATH_IMAGE012
dimension value,
Figure 6098DEST_PATH_IMAGE021
,
Figure 230406DEST_PATH_IMAGE022
are the upper and lower boundaries of the variable, and rand is a random number between [0, 1];

步骤4-3：根据步骤4-2每只松鼠的位置计算每只松鼠位置所代表参数的模型准确度；Step 4-3: Calculate the model accuracy of the parameters represented by the position of each squirrel according to the position of each squirrel in Step 4-2;

步骤4-4：根据飞行松鼠的准确度，按升序排列它们的位置；Steps 4-4: Arrange their positions in ascending order according to the accuracy of the flying squirrels;

步骤4-5：根据步骤4-4的排序将飞行松鼠按顺序分配到山核桃树、橡子树和普通树；Step 4-5: Allocate the flying squirrels to the hickory tree, acorn tree and common tree in order according to the order of step 4-4;

步骤4-6：更新松鼠位置，如下式：Step 4-6: Update the squirrel position as follows:

（1）在橡树上的松鼠向山核桃树移动，(1) The squirrel on the oak tree moves toward the hickory tree,

Figure 291903DEST_PATH_IMAGE023
Figure 291903DEST_PATH_IMAGE023

其中，

Figure 310674DEST_PATH_IMAGE024
是随机滑行距离，
Figure 270671DEST_PATH_IMAGE025
是[0, 1]范围内的随机数，
Figure 33091DEST_PATH_IMAGE026
是山核桃树的位置，t表示当前迭代。滑动常数
Figure 214673DEST_PATH_IMAGE027
实现全局与局部搜索之间的平衡，经过大量分析论证，
Figure 466663DEST_PATH_IMAGE027
的值设为1.9；in,
Figure 310674DEST_PATH_IMAGE024
is the random glide distance,
Figure 270671DEST_PATH_IMAGE025
is a random number in the range [0, 1],
Figure 33091DEST_PATH_IMAGE026
is the position of the hickory tree, and t represents the current iteration. sliding constant
Figure 214673DEST_PATH_IMAGE027
To achieve a balance between global and local search, after a lot of analysis and demonstration,
Figure 466663DEST_PATH_IMAGE027
The value of is set to 1.9;

（2）在普通树上的松鼠向橡树移动，(2) A squirrel on a common tree moves towards an oak tree,

Figure 100907DEST_PATH_IMAGE028
Figure 100907DEST_PATH_IMAGE028

其中，

Figure 401438DEST_PATH_IMAGE029
是[0, 1] 范围内的随机数。in,
Figure 401438DEST_PATH_IMAGE029
is a random number in the range [0, 1].

（3）在普通树上的松鼠向山核桃树移动，(3) The squirrel on the common tree moves towards the hickory tree,

Figure 703107DEST_PATH_IMAGE030
Figure 703107DEST_PATH_IMAGE030

其中

Figure 125998DEST_PATH_IMAGE031
是[0, 1]范围内的随机数。in
Figure 125998DEST_PATH_IMAGE031
is a random number in the range [0, 1].

步骤4-7：根据步骤4-6更新后每只松鼠的位置计算每只松鼠位置所代表参数的模型准确度，升序排列位置，将飞行松鼠按顺序重新分配到山核桃树、橡子树和普通树；Step 4-7: Calculate the model accuracy of the parameters represented by the position of each squirrel according to the position of each squirrel after updating in steps 4-6, arrange the positions in ascending order, and reassign the flying squirrels to the hickory tree, acorn tree and common tree;

步骤4-8：判断季节变化条件是否满足，如满足更新普通树上松鼠位置，不满足维持原位置，如下式：Step 4-8: Determine whether the seasonal change conditions are satisfied, such as updating the position of the squirrel on the common tree, but not maintaining the original position, as follows:

（1）计算季节常量

Figure 247537DEST_PATH_IMAGE032
：(1) Calculate the seasonal constant
Figure 247537DEST_PATH_IMAGE032
:

Figure 351760DEST_PATH_IMAGE033
Figure 351760DEST_PATH_IMAGE033

（2）计算季节变化条件

Figure 304672DEST_PATH_IMAGE034
：(2) Calculate seasonal variation conditions
Figure 304672DEST_PATH_IMAGE034
:

Figure 101727DEST_PATH_IMAGE035
Figure 101727DEST_PATH_IMAGE035

其中，t和

Figure 444984DEST_PATH_IMAGE036
分别是当前和最大迭代值；where t and
Figure 444984DEST_PATH_IMAGE036
are the current and maximum iteration values, respectively;

（3）如果季节条件得到满足，则随机改变普通树上松鼠的位置：(3) If the seasonal conditions are satisfied, randomly change the position of the squirrel on the ordinary tree:

Figure 352897DEST_PATH_IMAGE037
Figure 352897DEST_PATH_IMAGE037

步骤4-9：重新计算各个松鼠位置上的准确度，升序排列位置，将飞行松鼠分配到山核桃树、橡子树和普通树；Steps 4-9: Recalculate the accuracy on each squirrel position, arrange the positions in ascending order, and assign flying squirrels to hickory, acorn, and common trees;

步骤4-10：重复步骤4-6到4-9，满足迭代条件或最大迭代次数时，结束优化过程；Step 4-10: Repeat steps 4-6 to 4-9, and end the optimization process when the iteration conditions or the maximum number of iterations are satisfied;

步骤4-11：输出山核桃树上松鼠的位置（Catboost参数值）和准确度。Step 4-11: Output the location (Catboost parameter value) and accuracy of the squirrel on the pecan tree.

步骤5：基于优化后的参数建立性能最优的Catboost模型；Step 5: Establish a Catboost model with the best performance based on the optimized parameters;

步骤6：针对待预测的玻璃组分，利用最优的Catboost模型预测该玻璃组分的玻璃硬度。Step 6: For the glass composition to be predicted, use the optimal Catboost model to predict the glass hardness of the glass composition.

实施例2，如图1、2所示：Embodiment 2, as shown in Figures 1 and 2:

步骤1：集不同组分氧化物玻璃的硬度数据，构建玻璃硬度数据库，具体采集800组

Figure 911048DEST_PATH_IMAGE038
玻璃的硬度数据，构建玻璃硬度数据库；Step 1: Collect the hardness data of oxide glass with different components, build a glass hardness database, and collect 800 groups
Figure 911048DEST_PATH_IMAGE038
The hardness data of glass, build glass hardness database;

步骤2：基于化学特征，分别以元素摩尔含量

Figure 879004DEST_PATH_IMAGE005
、原子间库伦作用力
Figure 709557DEST_PATH_IMAGE006
、基于力场势的短程相互作用关系
Figure 217899DEST_PATH_IMAGE007
作为输入参数的描述符；具体步骤为：Step 2: Based on the chemical characteristics, the molar content of the elements are respectively
Figure 879004DEST_PATH_IMAGE005
, Coulomb force between atoms
Figure 709557DEST_PATH_IMAGE006
, short-range interaction relationship based on force field potential
Figure 217899DEST_PATH_IMAGE007
Descriptor as input parameter; the specific steps are:

步骤2-1：以800组

Figure 83086DEST_PATH_IMAGE038
摩尔含量
Figure 956365DEST_PATH_IMAGE039
、
Figure 539793DEST_PATH_IMAGE040
和
Figure 851825DEST_PATH_IMAGE041
为三组描述符；Step 2-1: Take 800 Groups
Figure 83086DEST_PATH_IMAGE038
Molar content
Figure 956365DEST_PATH_IMAGE039
,
Figure 539793DEST_PATH_IMAGE040
and
Figure 851825DEST_PATH_IMAGE041
are three sets of descriptors;

步骤2-2：以Si、Na、Ca、O原子间库伦作用力构造描述符

Figure 571520DEST_PATH_IMAGE006
：Step 2-2: Construct descriptors based on Coulomb forces between Si, Na , Ca, O atoms
Figure 571520DEST_PATH_IMAGE006
:

Figure 615699DEST_PATH_IMAGE008
Figure 615699DEST_PATH_IMAGE008

其中，

Figure 483161DEST_PATH_IMAGE009
和
Figure 536567DEST_PATH_IMAGE010
为离子
Figure 110768DEST_PATH_IMAGE011
和
Figure 591428DEST_PATH_IMAGE012
的有效离子电荷，
Figure 705708DEST_PATH_IMAGE013
和
Figure 297226DEST_PATH_IMAGE014
分别是组成元素
Figure 725933DEST_PATH_IMAGE011
和
Figure 643074DEST_PATH_IMAGE012
有效离子电荷
Figure 219549DEST_PATH_IMAGE009
和
Figure 349179DEST_PATH_IMAGE010
的摩尔分数；in,
Figure 483161DEST_PATH_IMAGE009
and
Figure 536567DEST_PATH_IMAGE010
for ions
Figure 110768DEST_PATH_IMAGE011
and
Figure 591428DEST_PATH_IMAGE012
The effective ionic charge of ,
Figure 705708DEST_PATH_IMAGE013
and
Figure 297226DEST_PATH_IMAGE014
the constituent elements
Figure 725933DEST_PATH_IMAGE011
and
Figure 643074DEST_PATH_IMAGE012
effective ionic charge
Figure 219549DEST_PATH_IMAGE009
and
Figure 349179DEST_PATH_IMAGE010
mole fraction of ;

步骤2-3：以Si、Na、Ca、O原子间力场势的短程相互作用关系

Figure 163551DEST_PATH_IMAGE007
构造描述符
Figure 782751DEST_PATH_IMAGE007
：Step 2-3: Based on the short-range interaction relationship between the Si, Na , Ca, and O atoms
Figure 163551DEST_PATH_IMAGE007
construct descriptor
Figure 782751DEST_PATH_IMAGE007
:

Figure 49784DEST_PATH_IMAGE015
Figure 49784DEST_PATH_IMAGE015

其中，S为组成元素合集，

Figure 983105DEST_PATH_IMAGE016
为组成元素有效离子电荷的摩尔分数，
Figure 651984DEST_PATH_IMAGE017
为Buckingham电势参数，p= -4, -3, -2, -1, 0, 1, 2, 3, 4；Among them, S is the set of constituent elements,
Figure 983105DEST_PATH_IMAGE016
is the mole fraction of the effective ionic charge of the constituent elements,
Figure 651984DEST_PATH_IMAGE017
is the Buckingham potential parameter, p = -4, -3, -2, -1, 0, 1, 2, 3, 4;

步骤3：以步骤2构造的描述符为模型的输入，以步骤1构建的硬度数据库为模型的输出，640组数据构建训练集、160组数据构建测试集，建立Catboost模型；Step 3: Take the descriptor constructed in Step 2 as the input of the model, take the hardness database constructed in Step 1 as the output of the model, construct a training set with 640 sets of data, and construct a test set with 160 sets of data, and establish a Catboost model;

步骤4-1：选取Catboost算法需优化的参数迭代数、学习率、深度，建立松鼠的位置，定义松鼠搜索优化算法的种群规模，维度，最大迭代次数；Step 4-1: Select the number of iterations, learning rate, and depth of parameters to be optimized by the Catboost algorithm, establish the position of the squirrel, and define the population size, dimension, and maximum number of iterations of the squirrel search optimization algorithm;

步骤4-2：种群位置初始化，为50只松鼠生成随机位置，第i只松鼠的位置可以通过一个矢量来确定。所有松鼠的位置在边界范围内随机初始化，如下式：Step 4-2: Initialize the population position, generate random positions for 50 squirrels, and the position of the i -th squirrel can be determined by a vector. The positions of all squirrels are randomly initialized within the bounds as follows:

Figure 442085DEST_PATH_IMAGE019
Figure 442085DEST_PATH_IMAGE019

其中，

Figure 196415DEST_PATH_IMAGE020
代表第
Figure 667847DEST_PATH_IMAGE011
只松鼠第
Figure 738703DEST_PATH_IMAGE012
维的值，
Figure 168547DEST_PATH_IMAGE021
、
Figure 144593DEST_PATH_IMAGE022
分别为变量的上下边界，rand为[0, 1]之间的随机数；in,
Figure 196415DEST_PATH_IMAGE020
representative
Figure 667847DEST_PATH_IMAGE011
only squirrel
Figure 738703DEST_PATH_IMAGE012
dimension value,
Figure 168547DEST_PATH_IMAGE021
,
Figure 144593DEST_PATH_IMAGE022
are the upper and lower boundaries of the variable, and rand is a random number between [0, 1];

步骤4-3：根据步骤4-2每只松鼠的位置计算每只松鼠位置所代表参数的模型准确度；Step 4-3: Calculate the model accuracy of the parameters represented by the position of each squirrel according to the position of each squirrel in Step 4-2;

步骤4-4：根据飞行松鼠的准确度，按升序排列它们的位置；Steps 4-4: Arrange their positions in ascending order according to the accuracy of the flying squirrels;

步骤4-5：根据步骤4-4的排序将飞行松鼠按顺序分配到山核桃树、橡子树和普通树；Step 4-5: Allocate the flying squirrels to the hickory tree, acorn tree and common tree in order according to the order of step 4-4;

步骤4-6：更新松鼠位置，如下式：Step 4-6: Update the squirrel position as follows:

（1）在橡树上的松鼠向山核桃树移动，(1) The squirrel on the oak tree moves toward the hickory tree,

Figure 419717DEST_PATH_IMAGE023
Figure 419717DEST_PATH_IMAGE023

其中，

Figure 859925DEST_PATH_IMAGE024
是随机滑行距离，
Figure 195092DEST_PATH_IMAGE025
是[0, 1]范围内的随机数，
Figure 658434DEST_PATH_IMAGE026
是山核桃树的位置，t表示当前迭代。滑动常数
Figure 737249DEST_PATH_IMAGE027
实现全局与局部搜索之间的平衡，经过大量分析论证，
Figure 31964DEST_PATH_IMAGE027
的值设为1.9；in,
Figure 859925DEST_PATH_IMAGE024
is the random glide distance,
Figure 195092DEST_PATH_IMAGE025
is a random number in the range [0, 1],
Figure 658434DEST_PATH_IMAGE026
is the position of the hickory tree, and t represents the current iteration. sliding constant
Figure 737249DEST_PATH_IMAGE027
To achieve a balance between global and local search, after a lot of analysis and demonstration,
Figure 31964DEST_PATH_IMAGE027
The value of is set to 1.9;

（2）在普通树上的松鼠向橡树移动，(2) A squirrel on a common tree moves towards an oak tree,

Figure 538031DEST_PATH_IMAGE028
Figure 538031DEST_PATH_IMAGE028

其中，

Figure 488670DEST_PATH_IMAGE029
是[0, 1]范围内的随机数。in,
Figure 488670DEST_PATH_IMAGE029
is a random number in the range [0, 1].

（3）在普通树上的松鼠向山核桃树移动，(3) The squirrel on the common tree moves towards the hickory tree,

Figure 167913DEST_PATH_IMAGE030
Figure 167913DEST_PATH_IMAGE030

其中

Figure 520397DEST_PATH_IMAGE031
是[0, 1]范围内的随机数。in
Figure 520397DEST_PATH_IMAGE031
is a random number in the range [0, 1].

步骤4-7：根据步骤4-6更新后每只松鼠的位置计算每只松鼠位置所代表参数的模型准确度，升序排列位置，将飞行松鼠按顺序重新分配到山核桃树、橡子树和普通树；Step 4-7: Calculate the model accuracy of the parameters represented by the position of each squirrel according to the position of each squirrel after updating in steps 4-6, arrange the positions in ascending order, and reassign the flying squirrels to the hickory tree, acorn tree and common tree;

步骤4-8：判断季节变化条件是否满足，如满足更新普通树上松鼠位置，不满足维持原位置，如下式：Step 4-8: Determine whether the seasonal change conditions are satisfied, such as updating the position of the squirrel on the common tree, but not maintaining the original position, as follows:

（1）计算季节常量

Figure 197366DEST_PATH_IMAGE032
:(1) Calculate the seasonal constant
Figure 197366DEST_PATH_IMAGE032
:

Figure 635300DEST_PATH_IMAGE033
Figure 635300DEST_PATH_IMAGE033

（2）计算季节变化条件

Figure 603388DEST_PATH_IMAGE034
：(2) Calculate seasonal variation conditions
Figure 603388DEST_PATH_IMAGE034
:

Figure 75957DEST_PATH_IMAGE035
Figure 75957DEST_PATH_IMAGE035

其中，t和

Figure 923828DEST_PATH_IMAGE036
分别是当前和最大迭代值。where t and
Figure 923828DEST_PATH_IMAGE036
are the current and maximum iteration values, respectively.

（3）如果季节条件得到满足，则随机改变普通树上松鼠的位置：(3) If the seasonal conditions are satisfied, randomly change the position of the squirrel on the ordinary tree:

Figure 645796DEST_PATH_IMAGE037
Figure 645796DEST_PATH_IMAGE037

步骤4-9：重新计算各个松鼠位置上的准确度，升序排列位置，将飞行松鼠分配到山核桃树、橡子树和普通树；Steps 4-9: Recalculate the accuracy on each squirrel position, arrange the positions in ascending order, and assign flying squirrels to hickory, acorn, and common trees;

步骤4-10：重复步骤4-6到4-9，满足迭代条件或最大迭代次数时，结束优化过程；Step 4-10: Repeat steps 4-6 to 4-9, and end the optimization process when the iteration conditions or the maximum number of iterations are satisfied;

步骤4-11：输出山核桃树上松鼠的位置（Catboost参数值）和准确度。Step 4-11: Output the location (Catboost parameter value) and accuracy of the squirrel on the pecan tree.

步骤5：基于优化后的参数建立性能最优的Catboost模型；Step 5: Establish a Catboost model with the best performance based on the optimized parameters;

步骤6：针对待预测的玻璃组分，利用最优的Catboost模型预测该玻璃组分的玻璃硬度。Step 6: For the glass composition to be predicted, use the optimal Catboost model to predict the glass hardness of the glass composition.

实施例模型性能如下表：The performance of the embodiment model is as follows:

表一：实施例1、2模型性能表Table 1: Model performance table of Examples 1 and 2

Figure 870104DEST_PATH_IMAGE042
Figure 870104DEST_PATH_IMAGE042

以上所述，仅为本发明较佳的具体实施方式，但本发明的保护范围并不局限于此，任何熟悉本技术领域的技术人员在本发明揭露的技术范围内，根据本发明的技术方案及其发明构思加以等同替换或改变，都应涵盖在本发明的保护范围之内。The above description is only a preferred embodiment of the present invention, but the protection scope of the present invention is not limited to this. The equivalent replacement or change of the inventive concept thereof shall be included within the protection scope of the present invention.

Claims (2)

1. The glass hardness prediction method based on the squirrel optimization algorithm and the machine learning algorithm is characterized by comprising the following steps of:
step 1: acquiring hardness data of oxide glass with different components, and constructing a glass hardness database, wherein the glass hardness database comprises glass components mapped one by one and hardness corresponding to the glass components;
step 2: based on chemical characteristics, in terms of element molar content
Figure RE-DEST_PATH_IMAGE001
Coulomb force between atoms
Figure RE-DEST_PATH_IMAGE002
Short-range interaction relation based on force field potential
Figure RE-DEST_PATH_IMAGE003
A descriptor as an input parameter;
and step 3: taking the descriptor constructed in the step 2 as the input of the model, taking the hardness database constructed in the step 1 as the output of the model, constructing a training set and a test set, and establishing a Catboost model;
and 4, step 4: introducing a squirrel search optimization algorithm to optimize the parameters of the selected Catboost model;
and 5: establishing a Catboost model with optimal performance based on the optimized parameters;
and 6: predicting the glass hardness of the glass component by utilizing an optimal Catboost model aiming at the glass component to be predicted;
the step 2 comprises the following steps:
step 2-1: in terms of the molar content of each component constituting the glass
Figure 568274DEST_PATH_IMAGE001
Is a set of descriptors;
step 2-2: construction of descriptors by coulomb forces between different atoms of the components of the constituent glasses
Figure 649362DEST_PATH_IMAGE002
：
Figure RE-DEST_PATH_IMAGE004
Wherein,
Figure RE-DEST_PATH_IMAGE005
and
Figure RE-DEST_PATH_IMAGE006
is an ion
Figure RE-DEST_PATH_IMAGE007
And
Figure RE-DEST_PATH_IMAGE008
the effective ionic charge of (a) is,
Figure RE-DEST_PATH_IMAGE009
and
Figure RE-DEST_PATH_IMAGE010
are respectively a constituent element
Figure 874938DEST_PATH_IMAGE007
And
Figure 637358DEST_PATH_IMAGE008
effective ionic charge
Figure 818941DEST_PATH_IMAGE005
And
Figure 805351DEST_PATH_IMAGE006
the mole fraction of (c);
step 2-3: by short-range interaction of different interatomic force field potentials of various components of glass
Figure 439595DEST_PATH_IMAGE003
Construction descriptor
Figure 5705DEST_PATH_IMAGE003
：
Figure RE-DEST_PATH_IMAGE011
Wherein S is a collection of constituent elements,
Figure RE-DEST_PATH_IMAGE012
is the mole fraction of the effective ionic charge of the constituent elements,
Figure RE-DEST_PATH_IMAGE013
p = -4, -3, -2, -1, 0, 1, 2, 3, 4 for Buckingham potential parameters.
2. The squirrel optimization algorithm and machine learning algorithm-based glass hardness prediction method according to claim 1, wherein the step 4 comprises the steps of:
step 4-1: selecting parameters needing to be optimized by a Catboost algorithm, wherein the position coordinates of the squirrel are the parameters needing to be optimized, and defining the population scale, the dimensionality, the maximum iteration times, the sliding distance parameters and the predator existence probability of the squirrel search optimization algorithm;
step 4-2: the position of the population is initialized to
Figure RE-DEST_PATH_IMAGE014
Random positions were generated by squirrels alone, second
Figure 383073DEST_PATH_IMAGE007
The position of only squirrel can be determined by a vector; the positions of all squirrels were randomly initialized within the bounds, as follows:
Figure RE-DEST_PATH_IMAGE015
wherein,
Figure RE-DEST_PATH_IMAGE016
represents the first
Figure 274805DEST_PATH_IMAGE007
Only squirrel (a Chinese character of pine)
Figure 661924DEST_PATH_IMAGE008
The value of the dimension(s) is,
Figure RE-DEST_PATH_IMAGE017
、
Figure RE-DEST_PATH_IMAGE018
respectively the upper and lower boundaries of the variable, and rand is [0, 1 ]]A random number in between;
step 4-3: calculating the model accuracy of the parameters represented by the position of each squirrel according to the position of each squirrel in the step 4-2;
step 4-4: arranging their positions in ascending order according to the accuracy of flying squirrels;
and 4-5: according to the sequence of the step 4-4, sequentially distributing the flying squirrels to a hickory nut, an oak tree and a common tree, wherein the hickory nut represents the global optimal solution position, and the oak tree represents the local optimal solution position;
and 4-6: updating squirrel position as follows:
(1) the squirrel on the oak moves towards the pecan tree,
Figure RE-DEST_PATH_IMAGE019
wherein,
Figure RE-DEST_PATH_IMAGE020
is the random sliding distance of the sliding block,
Figure RE-DEST_PATH_IMAGE021
is [0, 1 ]]A random number within the range of the random number,
Figure RE-DEST_PATH_IMAGE022
is the position of the hickory tree, t represents the current iteration; sliding constant
Figure RE-DEST_PATH_IMAGE023
Realizes the balance between the global search and the local search, and through a large amount of analysis and demonstration,
Figure 828463DEST_PATH_IMAGE023
the value of (d) is set to 1.9;
(2) squirrels on ordinary trees move toward the oak,
Figure RE-DEST_PATH_IMAGE024
wherein,
Figure RE-DEST_PATH_IMAGE025
is [0, 1 ]]A random number within a range;
(3) the squirrel on the common tree moves toward the hickory tree,
Figure RE-DEST_PATH_IMAGE026
wherein,
Figure RE-DEST_PATH_IMAGE027
is [0, 1 ]]A random number within a range;
and 4-7: calculating the model accuracy of the parameters represented by the position of each squirrel according to the updated position of each squirrel in the steps 4-6, arranging the positions in an ascending order, and redistributing the flying squirrel to the pecan tree, the oak tree and the common tree in sequence;
and 4-8: judging whether the seasonal variation condition is satisfied, if the updating of the position of the squirrel on the common tree is satisfied, the original position is not satisfied, and the following formula is shown:
(1) calculating seasonal constants
Figure RE-DEST_PATH_IMAGE028
：
Figure RE-DEST_PATH_IMAGE029
(2) Calculating seasonal variation conditions
Figure RE-DEST_PATH_IMAGE030
：
Figure RE-DEST_PATH_IMAGE031
Wherein,t and
Figure RE-DEST_PATH_IMAGE032
current and maximum iteration values, respectively;
(3) randomly changing the position of squirrels on the common tree if seasonal conditions are met:
Figure RE-DEST_PATH_IMAGE033
and 4-9: recalculating the accuracy of each squirrel position, arranging the positions in an ascending order, and distributing the flying squirrels to the pecan trees, the oak trees and the common trees;
step 4-10: repeating the steps 4-6 to 4-9, and finishing the optimization process when the iteration condition or the maximum iteration times is met;
and 4-11: and outputting the position and the accuracy of the squirrel on the hickory nut tree, wherein the position of the squirrel on the hickory nut tree is a Catboost parameter value.
CN202210281841.9A 2022-03-22 2022-03-22 Glass hardness prediction method based on squirrel optimization algorithm and machine learning algorithm Active CN114373523B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210281841.9A CN114373523B (en) 2022-03-22 2022-03-22 Glass hardness prediction method based on squirrel optimization algorithm and machine learning algorithm

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210281841.9A CN114373523B (en) 2022-03-22 2022-03-22 Glass hardness prediction method based on squirrel optimization algorithm and machine learning algorithm

Publications (2)

Publication Number Publication Date
CN114373523A CN114373523A (en) 2022-04-19
CN114373523B true CN114373523B (en) 2022-06-03

Family

ID=81146137

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210281841.9A Active CN114373523B (en) 2022-03-22 2022-03-22 Glass hardness prediction method based on squirrel optimization algorithm and machine learning algorithm

Country Status (1)

Country Link
CN (1) CN114373523B (en)

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116187167A (en) * 2022-12-30 2023-05-30 中建材玻璃新材料研究院集团有限公司 Machine learning-based method and system for predicting depth of stress layer after glass tempering
CN117497087B (en) * 2023-12-20 2024-04-26 浙江大学 Oxide glass performance prediction method based on interpretable high-dimensional spatial prediction model
CN117976093A (en) * 2024-01-11 2024-05-03 中建材玻璃新材料研究院集团有限公司 Glass transition temperature prediction method based on zebra optimization algorithm
CN118969144A (en) * 2024-07-19 2024-11-15 中建材玻璃新材料研究院集团有限公司 Glass elastic modulus prediction method based on spider bee optimization algorithm and machine learning algorithm
CN119400287B (en) * 2024-12-31 2025-04-29 成都光明光电股份有限公司 Optical glass chemical strengthening process prediction method, system, equipment and medium

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104062422A (en) * 2014-06-10 2014-09-24 华东理工大学 Predicating method for transformation temperature and hardness of glass
CN106444625A (en) * 2016-09-18 2017-02-22 合肥工业大学 Cutter head servo control method and cutter head servo control device for glass cutting machine
CN110683749A (en) * 2019-10-31 2020-01-14 江苏瑞特钢化玻璃制品有限公司 Production process of high-strength toughened glass
CN110728401A (en) * 2019-10-10 2020-01-24 郑州轻工业学院 Short-term power load prediction method of neural network based on squirrel and weed hybrid algorithm

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR102523472B1 (en) * 2016-08-01 2023-04-18 삼성전자주식회사 Method and apparatus for searching new material
CN110850477B (en) * 2019-11-07 2024-11-08 中建四局第一建设有限公司 A resistivity cave identification method based on squirrel search algorithm

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104062422A (en) * 2014-06-10 2014-09-24 华东理工大学 Predicating method for transformation temperature and hardness of glass
CN106444625A (en) * 2016-09-18 2017-02-22 合肥工业大学 Cutter head servo control method and cutter head servo control device for glass cutting machine
CN110728401A (en) * 2019-10-10 2020-01-24 郑州轻工业学院 Short-term power load prediction method of neural network based on squirrel and weed hybrid algorithm
CN110683749A (en) * 2019-10-31 2020-01-14 江苏瑞特钢化玻璃制品有限公司 Production process of high-strength toughened glass

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
Explainable machine learning algorithms for predicting galss transition temperatures;Edesio Alcobaca,et al.;《Acta Materialia》;20200415;92-100 *
触控屏用盖板玻璃的硬度分析;刘亚茹 等;《玻璃搪瓷与眼镜》;20211231;7-13 *

Also Published As

Publication number Publication date
CN114373523A (en) 2022-04-19

Similar Documents

Publication Publication Date Title
CN114373523B (en) Glass hardness prediction method based on squirrel optimization algorithm and machine learning algorithm
Bhattoo et al. Understanding the compositional control on electrical, mechanical, optical, and physical properties of inorganic glasses with interpretable machine learning
CN109948865A (en) A Path Planning Method for TSP Problems
CN106527381B (en) A kind of fast evaluation method towards parallel batch processing machine dynamic dispatching
WO2024139684A1 (en) Machine-learning-based method and system for predicting stress layer depth after glass tempering
CN114595873B (en) Gray correlation-based DA-LSTM short-term power load prediction method
CN110322020B (en) Adaptive learning rate scheduling for distributed random gradient descent
CN111260032A (en) Neural network training method, image processing method and device
CN108320059B (en) Workflow scheduling evolution optimization method and terminal equipment
CN108204944A (en) The Buried Pipeline rate prediction method of LSSVM based on APSO optimizations
CN111832101A (en) Construction method of a cement strength prediction model and cement strength prediction method
CN106599936A (en) Characteristic selection method based on binary ant colony algorithm and system thereof
CN114021470B (en) Relay storage life prediction method based on AMFO algorithm and SVM algorithm
CN106909560A (en) POI sorting method
CN109146196B (en) A method for predicting water consumption in residential communities
CN107590538B (en) A Dangerous Source Identification Method Based on Online Sequence Learning Machine
WO2025148464A1 (en) Glass transition temperature prediction method based on zebra optimization algorithm
CN118969144A (en) Glass elastic modulus prediction method based on spider bee optimization algorithm and machine learning algorithm
CN114912331A (en) Optimization method, device, equipment and medium for cabin stiffener
CN118438445A (en) Fruit sorting robot track planning method based on multi-objective optimization
Tausch Researchgate, RG-Scores, or a true Research Gate to Global Research? On the limits of the RG factor and some scientometric evidence on how the current RG score system discriminates against economic and social sciences and against the developing countries
CN114613426B (en) System development tree construction method based on dynamic multi-objective optimization
CN113868916B (en) LFVPSO-BPNN-based multi-loop groove cabling temperature rise prediction method
Cao et al. Multi agent collaborative search algorithm with adaptive weights
CN115965082A (en) Phylogenetic tree construction method and system based on deep learning and clustering search

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant