Interpreting the optical properties of oxide glasses with machine learning and Shapely additive explanations

Due to their excellent optical properties, glasses are used for various applications ranging from smartphone screens to telescopes. Developing compositions with tailored Abbe number (V_d) and refractive index at 587.6 nm (n_d), two crucial optical properties, is a major challenge. To this extent, machine learning (ML) approaches have been successfully used to develop composition–property models. However, these models are essentially black boxes in nature and suffer from the lack of interpretability. In this paper, we demonstrate the use of ML models to predict the composition-dependent variations of V_d and n_d. Further, using Shapely additive explanations (SHAP), we interpret the ML models to identify the contribution of each of the input components toward target prediction. We observe that glass formers such as SiO₂, B₂O₃, and P₂O₅ and intermediates such as TiO₂, PbO, and Bi₂O₃ play a significant role in controlling the optical properties. Interestingly, components contributing toward increasing the n_d are found to decrease the V_d and vice versa. Finally, we develop the Abbe diagram, using the ML models, allowing accelerated discovery of new glasses for optical properties beyond the experimental pareto front. Overall, employing explainable ML, we predict and interpret the compositional control on the optical properties of oxide glasses.     

计算机辅助玻璃组分设计

根据硅酸盐玻璃成分与物理性质的线性关系,利用VisualC 语言进行编程,实现玻璃组成成分的计算过程。通过对钨组、钼组以及铂组等多种电真空玻璃系统的性能计算与实际性能的比较分析可以看出,玻璃的密度、膨胀系数等性能参数的计算值和实测值具有非常好的一致性。采用VisualC 程序辅助玻璃成分设计,能够提高新型玻璃的研究效率。     

Prediction of thermodynamic properties of polyvinylpyrrolidone solutions by artificial neural networks coupled with genetic algorithm

In the present study, the artificial neural networks coupled with the genetic algorithm (ANN–GA) models were used to predict the thermodynamic properties of polyvinylpyrrolidone (PVP) solutions in water and ethanol at various temperatures, mass fractions, and molecular weights of polymer. The genetic algorithm (GA) was used to find the best weights and biases of the network and improve the performance of ANNs. The proposed model was composed of three input variables including the temperature of the solution, the mass fraction, and molecular weight of the polymer. Density, viscosity, and surface tension of PVP solutions with various molecular weights (10,000, 25,000, and 40,000) in water and ethanol have been measured in the temperature range 20–55°C and various mass fractions of polymer. The ANN–GA models were trained by the experimental datasets and the prediction of density, surface tension, and viscosity of PVP solutions was performed using these models. The predicted values were compared with the experimental ones and the mean absolute relative error was less than 0.5% for the density and surface tension and about 3% for the viscosity of solutions.     

Phosphate glasses with improved optical properties

A study was undertaken to see how some components can affect the physicochemical properties of PbO−BaO−Al₂O₃−P₂O₅ optical glasses. The possibility is demonstrated for the manufacture of phosphate-base flint glasses with a high refractive index and improved transmission in the visible and UV regions of the spectrum. Translated from Steklo i Keramika, No. 2, pp. 3–5, February, 1997.     

Predicting nepheline precipitation in waste glasses using ternary submixture model and machine learning

Nepheline precipitation in nuclear waste glasses during vitrification can be detrimental due to the negative effect on chemical durability often associated with its formation. Developing models to accurately predict nepheline precipitation from compositions is important for increasing waste loading since existing models can be overly conservative. In this study, an expanded dataset of 955 glasses, including 352 high-level waste glasses, was compiled from literature data. Previously developed submixture models were refitted using the new dataset, where a misclassification rate of 7.8% was achieved. In addition, nine machine learning (ML) algorithms (k-nearest neighbor, Gaussian process regression, artificial neural network, support vector machine, decision tree, etc.) were applied to evaluate their ability to predict nepheline precipitation from glass compositions. Model accuracy, precision, recall/sensitivity, and F1 scores were systemically compared between different ML algorithms and modeling protocols. Model prediction with an accuracy of ~0.9 (misclassification rate of ~10%) was observed for different algorithms under certain protocols. This study evaluated various ML models to predict nepheline precipitation in waste glasses, highlighting the importance of data preparation and modeling protocol, and their effect on model stability and reproducibility. The results provide insights into applying ML to predict glass properties and suggest areas for future research on modeling nepheline precipitation.     

基于改进ELM的递归最小二乘时序差分强化学习算法及其应用

针对值函数逼近算法对精度及计算时间等要求,提出了一种基于改进极限学习机的递归最小二乘时序差分强化学习算法。首先,将递推方法引入到最小二乘时序差分强化学习算法中消去最小二乘中的矩阵求逆过程,形成递推最小二乘时序差分强化学习算法,减少算法的复杂度及其计算量。其次,考虑到LSTD（0）算法收敛速度慢,加入资格迹增加样本利用率提高收敛速度的算法,形成LSTD（λ）算法,以保证在经历过相同数量的轨迹后能收敛于真实值。同时,考虑到大部分强化学习问题的值函数是单调的,而传统ELM方法通常运用具有双侧抑制特性的Sigmoid激活函数,增大了计算成本,提出采用具有单侧抑制特性的Softplus激活函数代替传统Sigmoid函数,以减少计算量提高运算速度,使得该算法在提高精度的同时提高了计算速度。通过与传统基于径向基函数的最小二乘强化学习算法和基于极限学习机的最小二乘TD算法在广义Hop-world问题的对比实验,比较结果证明了所提出算法在满足精度的条件下有效提高了计算速度,甚至某些条件下精度比其他两种算法更高。     

Study of the iron-containing sodium borosilicate two-phase glasses and porous glasses by optical spectroscopy

The results of studying the iron oxide-containing two-phase sodium borosilicate glass and porous glass based on it using the method of optical spectroscopy in the visible and infrared regions are presented. Decoding the transmittance spectra of this glass in these spectral ranges using the literary data and their comparative analysis with two-phase sodium borosilicate glass that does not contain iron were carried out.     

Modeling and optimization of photocatalytic/photoassisted-electro-Fenton like degradation of phenol using a neural network coupled with genetic algorithm

Oxidation of phenol in aqueous media using supported TiO₂ nanoparticles coupled with photoelectro-Fenton-like process using Mn²⁺ cations as catalyst of electro-Fenton reaction was studied. The influence of the basic operational parameters such as initial pH of the solution, applied current, initial Mn²⁺ concentration, initial phenol concentration and kind of ultraviolet (UV) light on the oxidizing efficiency of phenol was studied. An artificial neural network (ANN) model was coupled with genetic algorithm to predict phenol oxidizing efficiency and to find an optimal condition for maximum phenol removal. The findings indicated that ANN provided reasonable predictive performance (R² = 0.949).     

Nonlinear model predictive control based on support vector machine and genetic algorithm

This paper presents a nonlinear model predictive control (NMPC) approach based on support vector machine (SVM) and genetic algorithm (GA) for multiple-input multiple-output (MIMO) nonlinear systems. Individual SVM is used to approximate each output of the controlled plant. Then the model is used in MPC control scheme to predict the outputs of the controlled plant. The optimal control sequence is calculated using GA with elite preserve strategy. Simulation results of a typical MIMO nonlinear system show that this method has a good ability of set points tracking and disturbance rejection.     

Surface nitrogen doping of optical glasses (a review)

The need is demonstrated for improvement in the service properties of the surfaces of optical components fabricated from chemically unstable glasses. A number of solid and gaseous reagents are proposed for surface nitrogen doping of optical glasses and the requirements they should meet are defined. The process variables involved in the nitrogen doping of glass at temperatures belowt _g are examined. The service properties of lead-containing, phosphate and borate glasses with a nitrogen-doped surface are listed. Translated from Steklo i Keramika, No. 1, pp. 6–8, January, 1997.     

遗传-分布估计算法求解化工生产中一类带多工序的异构并行机调度问题

针对化工生产中广泛存在的一类带多工序的异构并行机调度问题,即部分产品需多工序加工,同时不同产品间带序相关设置时间的异构并行机调度问题（heterogeneous parallel machine scheduling problem with multiple operations and sequence-dependent setup times, HPMSP_MOSST）,提出了一种遗传-分布估计算法（genetic algorithm-estimation of distribution algorithm, GA-EDA）,用于优化最早完工时间（makespan）。首先,提出了一种基于GA的概率模型训练机制,用来提高概率模型在算法进化初期的信息积累量,进而提高搜索的效率;其次,设计了一种有效的GA与EDA混合策略,使得算法的全局探索和局部开发能力得到合理平衡。计算机模拟验证了GA-EDA的有效性和鲁棒性。     

Automated docking of peptides and proteins by using a genetic algorithm combined with a tabu search

A genetic algorithm (GA) combined with a tabu search (TA) hasbeen applied as a minimization method to rake the appropriateassociated sites for some biomolecular systems. In our dockingprocedure, surface complementarity and energetic complementarityof a ligand with its receptor have been considered separatelyin a two-stage docking method. The first stage was to find aset of potential associated sites mainly based on surface complementarityusing a genetic algorithm combined with a tabu search. Thisstep corresponds with the process of finding the potential bindingsites where pharmacophores will bind. In the second stage, severalhundreds of GA minimization steps were performed for each associatedsite derived from the first stage mainly based on the energeticcomplementarity. After calculations for both of the two stages,we can offer several solutions of associated sites for everycomplex. In this paper, seven biomolecular systems, includingfive bound complexes and two unbound complexes, were chosenfrom the Protein Data Bank (PDB) to test our method. The calculatedresults were very encouraging—the hybrid minimizationalgorithm successfully reaches the correct solutions near thebest binded modes for these protein complexes. The docking resultsnot only predict the bound complexes very well, but also geta relatively accurate complexed conformation for unbound systems.For the five bound complexes, the results show that surfacecomplementarity is enough to find the precise binding modes,the top solution from the tabu list generally corresponds tothe correct binding mode. For the two unbound complexes, dueto the conformational changes upon binding, it seems more difficultto get their correct binding conformations. The predicted resultsshow that the correct binding mode also corresponds to a relativelylarge surface complementarity score. In these two test cases,the correct solution can be found in the top several solutionsfrom the tabu list. For unbound complexes, the interaction energyfrom energetic complementarity is very important, it can beused to filter these solutions from the surface complementarity.After the evaluation of the energetic complementarity, the conformationsand orientations close to the crystallographically determinedstructures are resolved. In most cases, the smallest root meansquare distance (r.m.s.d.) from the GA combined with TA solutionsis in a relatively small region. Our program of automatic dockingis really a universal one among the procedures used for thetheoretical study of molecular recognition.     

Performance modification of third-order optical nonlinearity of chalcogenide glasses by nanocrystallization

We demonstrate the modification of the third-order optical nonlinearity (TONL) of chalcogenide glasses (within the GeS₂–In₂S₃–CsCl ternary system) by nanocrystallization, i.e., by controlling the precipitation of nanocrystals (in pure In₂S₃ phase) within the amorphous background. Compared with the parent glass, the resultant chalcogenide glass ceramics (ChGCs) have unchanged infrared transmittance but modifiable optical bandgap energy with treatment duration. Both nonlinear refraction coefficient and nonlinear absorption coefficient of the ChGCs are increased due to the appearance of In₂S₃ nanocrystals. ChGCs subjected to heat treatment for 1.3–2 h are found to exhibit the optimum TONL performance.     

Modeling of a reaction network and its optimization by genetic algorithm

Continuous endeavors are going on in many research works to find out the strategy to mathematically model and optimize complex reaction networks in order to maximize the main product and at the same time keeping the reactor dimensions within some acceptable limits. The aim of this work is to provide with a strategy for efficient modeling and optimization of reaction networks for reaction controlled processes. Genetic algorithm (GA) has been used for optimizing complex search spaces with multiple optima. Formation of styrene monomer from the ethylbenzene dehydrogenation, with several by-products in a fixed bed reactor, is taken as an example for this study. Two activation energies are found to be the best in term of maximizing styrene productivity.     

基于Gath-Geva算法和核极限学习机的多阶段间歇过程软测量

间歇过程具有较强的非线性,多阶段、慢时变及批次间存在变化,采用单一预测模型不能反映间歇过程的多阶段特性及阶段间过渡特性。提出一种基于Gath-Geva聚类和核极限学习机（kernel extreme learning machine,KELM）的多模型软测量方法。首先采用主成分分析（principal component analysis,PCA）对输入做特征提取,然后利用Gath-Geva算法对间歇过程进行多阶段工况划分,根据生产工况特性划分为不同的操作阶段后,分别建立局部KELM模型。对任一待预测样本,分别计算其对应各局部模型的预测值,最后采用贝叶斯集成,将其隶属于各局部模型的模糊隶属度作为权重和预测值融合得到最终预测值。以青霉素发酵数据进行实验测试,结果表明所提多模型算法相较于单一模型,具有更高的预测精度。     

拉曼光谱结合机器学习识别肥皂

通过拉曼光谱和机器学习对不同品牌、种类的肥皂进行准确和快速识别。通过采集不同厂家、不同品牌生产的56种肥皂的拉曼光谱数据,将拉曼光谱进行预处理后,进行系统聚类对拉曼光谱分类,结合光谱特征峰的比对,把样本分成5类。利用机器学习的SVM、KNN、贝叶斯方法,建立不同方法的判别分析,通过检验三种判别方法的准确度比对,综合比较,得到KNN的模型分类效果最好,分类正确率为96.4%。通过这种方法可以对肥皂样本进行快速准确的检验,帮助勘查人员进行肥皂检材识别。     

Generation of operating procedures for a mixing tank with a micro genetic algorithm

This paper explores the use of a micro genetic algorithm that uses variable-length chromosomes and a seeding scheme based on tabu search. The problem is to find the sequence of actions that have to be executed in the shortest time possible, but also in a way that minimizes the possibility of situations that may endanger the plant personnel and plant facilities. The proposed approach was tested on the generation of the optimum sequences for startup and shutdown of a mixing vessel similar to the equipment used in the synthesis of acrylic acid. The results show that the proposed method outperforms the traditional GA algorithm both in terms of the quality of the solution and computational effort.     

Development of compositions of uviol glasses with elevated radiation and optical resistance

The processes occurring in the glass shells of bactericidal lamps are considered. It is shown that it is expedient to develop new compositions of uviol glasses with elevated radiation and optical resistance (ROR). It was found that an increase in the total amount of alkaline-earth oxides and BaO individually at the expense of CaO +MgO decreases solarization of glasses of the SiO₂-MgO-CaO-BaO-Na₂O-K₂O system. The introduction of Fe₂O₃ in the composition with a decrease in the content of MgO for a certain concentration of Fe²⁺ and Fe³⁺ improves the anti-reflection properties of uviol glasses in the ultraviolet spectral region and increases the ROR to short-wave ultraviolet radiation. Addition of metallic aluminum increases the Fe²⁺Fe³⁺ ratio. The physicochemical properties of the synthesized compositions are presented.     

Fluorophosphate glasses doped with PbSe quantum dots and their nonlinear optical characteristics

A new fluorophosphate glass matrix containing PbSe quantum dots characterized by a narrow size distribution (ΔR/R ∼ 5–7%, where R is the size of a quantum dot) is prepared under heat treatment. It is demonstrated that fluorophosphate glasses can be doped by introducing (ZnSe + PbO) or PbSe into the composition of the glass. The optical absorption spectra of the quantum dots corresponding to the strong quantum confinement regime are studied over a wide range of quantum-dot sizes (4–18 nm). The nonlinear optical absorption is investigated at a wavelength of 1.54 μm. It is revealed that the introduction of (ZnSe + PbO) rather than of PbSe into the composition of the glass improves the bleaching contrast owing to a more uniform distribution of the quantum dots over the volume of the glass sample.