Incorporating empirical knowledge into data-driven variable selection for quantitative analysis of coal ash content by laser-induced breakdown spectroscopy

Laser-induced breakdown spectroscopy(LIBS)has become a widely used atomic spectroscopic technique for rapid coal analysis.However,the vast amount of spectral information in LIBS contains signal uncertainty,which can affect its quantification performance.In this work,we propose a hybrid variable selection method to improve the performance of LIBS quantification.Important variables are first identified using Pearson's correlation coefficient,mutual information,least absolute shrinkage and selection operator(LASSO)and random forest,and then filtered and combined with empirical variables related to fingerprint elements of coal ash content.Subsequently,these variables are fed into a partial least squares regression(PLSR).Additionally,in some models,certain variables unrelated to ash content are removed manually to study the impact of variable deselection on model performance.The proposed hybrid strategy was tested on three LIBS datasets for quantitative analysis of coal ash content and compared with the corresponding data-driven baseline method.It is significantly better than the variable selection only method based on empirical knowledge and in most cases outperforms the baseline method.The results showed that on all three datasets the hybrid strategy for variable selection combining empirical knowledge and data-driven algorithms achieved the lowest root mean square error of prediction(RMSEP)values of 1.605,3.478 and 1.647,respectively,which were significantly lower than those obtained from multiple linear regression using only 12 empirical variables,which are 1.959,3.718 and 2.181,respectively.The LASSO-PLSR model with empirical support and 20 selected variables exhibited a significantly improved performance after variable deselection,with RMSEP values dropping from 1.635,3.962 and 1.647 to 1.483,3.086 and 1.567,respectively.Such results demonstrate that using empirical knowledge as a support for datadriven variable selection can be a viable approach to improve the accuracy and reliability of LIBS quantification.

Assessment of molecular markers and marker-assisted selection for drought tolerance in barley(Hordeum vulgare L.)

This review updates the present status of the field of molecular markers and marker-assisted selection(MAS),using the example of drought tolerance in barley.The accuracy of selected quantitative trait loci(QTLs),candidate genes and suggested markers was assessed in the barley genome cv.Morex.Six common strategies are described for molecular marker development,candidate gene identification and verification,and their possible applications in MAS to improve the grain yield and yield components in barley under drought stress.These strategies are based on the following five principles:(1)Molecular markers are designated as genomic‘tags’,and their‘prediction’is strongly dependent on their distance from a candidate gene on genetic or physical maps;(2)plants react differently under favourable and stressful conditions or depending on their stage of development;(3)each candidate gene must be verified by confirming its expression in the relevant conditions,e.g.,drought;(4)the molecular marker identified must be validated for MAS for tolerance to drought stress and improved grain yield;and(5)the small number of molecular markers realized for MAS in breeding,from among the many studies targeting candidate genes,can be explained by the complex nature of drought stress,and multiple stress-responsive genes in each barley genotype that are expressed differentially depending on many other factors.

Consequences of Insufficient Selectivity in Quantitative and Qualitative Chemical Analysis

A problem in chemical analysis in connection with measurements of a substance normally occurring in a sample, or identification of a substance which should not exist in a sample, is insufficient selectivity. In this article, we analyze this problem and propose remedies. We use a real doping case to illustrate how chemical noise causes a serious selectivity problem, probably causing a false positive outcome.

Selection of optimal window length using STFT for quantitative SNR analysis of LFM signal

An adaptive approach to select analysis window param- eters for linear frequency modulated （LFM） signals is proposed to obtain the optimal 3 dB signal-to-noise ratio （SNR） in the short- time Fourier transform （STFT） domain. After analyzing the instan- taneous frequency and instantaneous bandwidth to deduce the relation between the window length and deviation of the Gaus- sian window, high-order statistics is used to select the appropriate window length for STFT and get the optimal SNR with the right time-frequency resolution according to the signal characteristic under a fixed sampling rate. Computer simulations have verified the effectiveness of the new method.

A feature selection method combined with ridge regression and recursive feature elimination in quantitative analysis of laser induced breakdown spectroscopy

In the spectral analysis of laser-induced breakdown spectroscopy,abundant characteristic spectral lines and severe interference information exist simultaneously in the original spectral data.Here,a feature selection method called recursive feature elimination based on ridge regression(Ridge-RFE)for the original spectral data is recommended to make full use of the valid information of spectra.In the Ridge-RFE method,the absolute value of the ridge regression coefficient was used as a criterion to screen spectral characteristic,the feature with the absolute value of minimum weight in the input subset features was removed by recursive feature elimination(RFE),and the selected features were used as inputs of the partial least squares regression(PLS)model.The Ridge-RFE method based PLS model was used to measure the Fe,Si,Mg,Cu,Zn and Mn for 51 aluminum alloy samples,and the results showed that the root mean square error of prediction decreased greatly compared to the PLS model with full spectrum as input.The overall results demonstrate that the Ridge-RFE method is more efficient to extract the redundant features,make PLS model for better quantitative analysis results and improve model generalization ability.

Stock Selection Based on a Hybrid Quantitative Method

Quantitative stock selection has become a research hotspot in the field of investment decision. As the data mining technology becomes mature, quantitative stock selection has made great progress. From the perspective of value investment, this paper selects top 200 stocks of A share in terms of market value. With the random forest (RF), financial characteristic variables with significant impact on SVR are screened out. At the same time with quantum genetic algorithm (QGA) superior to the traditional genetic algorithm (GA), SVR parameters are deeply and dynamically sought for, so as to build the RF-QGA-SVR model for year-to-year stock ranking. The quantitative stock selection model is built, and the empirical analysis of its stock selection performance is conducted. The conclusion is as follows: 1) Optimizing SVR with QGA has higher precision than the traditional genetic algorithm, and is more excellent than the traditional GA optimization;2) SVR after RF optimization of characteristic variables more significantly improves the accuracy of stock ranking and prediction;3) In the stock ranking obtained from the RF-QGA-SVR model, the yields of top stock portfolios are much higher than the market benchmark yield. At the same time, the yields of the top 10 stock portfolios are the highest, and the top 30 stock portfolios are the most stable. This study has positive reference significance on quantitative stock selection in the field of quantitative investment.

Variable selection in near infrared spectroscopy for quantitative models of homologous analogs of cephalosporins

Two universal spectral ranges(4550-4100 cm^(-1) and 6190-5510 cm^(-1))for construction of quantitative models of homologous analogs of cephalosporins were proposed by evaluating theperformance of five spectral ranges and their combinations,using three data sets of cephalos-porins for injection,ie.,cefuroxime sodium,cetriaxone sodium and cefoperazone sodium.Subsequently,the proposed ranges were validated by using eight calibration sets of otherhomologous analogs of cephalosporins for injection,namely cefmenoxime hydrochloride,ceftezole sodium,cefmetazole,cefoxitin sodium,cefotaxime sodium,cefradine,cephazolin sodium and ceftizoxime sodium.All the constructed quantitative models for the eight kinds of cephalosporinsusing these universal ranges could fulill the requirements for quick quantification.After that,competitive adaptive reweighted sampling(CARS)algorithm and infrared(IR)-near infrared(NIR)two-dimensional(2D)correlation spectral analysis were used to determine the scientific basis of these two spectral ranges as the universal regions for the construction of quantitativemodels of cephalosporins.The CAR.S algorithm demonstrated that the ranges of 4550-4100 cm^(-1) and 6190-5510 cm^(-1) included some key wavenumbers which could be attributed to content changes of cephalosporins.The IR-NIR 2D spectral analysis showed that certain wavenumbersin these two regions have strong correlations to the structures of those cephalosporins that wereeasy to degrade.

Quantitative Stock Selection Model Based on Long-Short Term Memory(LSTM)Neural Network

This article attempted to construct a multi-factor quantitative stock selection model,analyze the financial indicators and transaction data of listed companies in detail via the big data statistical test method,and to find out the alpha excess return relative to the market in the case of short stock index futures as a hedge in the Chinese market.

Quantitative Damage Detection for Planetary Gear Sets Based on Physical Models

Planetary gear set is the critical component in helicopter transmission train, and an important problem in condition monitoring and health management of planetary gear set is quantitative damage detection. In order to resolve this problem, an approach based on physical models is presented to detect damage quantitatively in planetary gear set. A particular emphasis is put on a feature generation and selection method, which is used for sun gear tooth breakage damage detection quantitatively in planetary gear box of helicopter transmission system. In this feature generation procedure, the pure torsional dynamical models of 2K-H planetary gear set is established for healthy case and sun gear tooth-breakage case. Then, a feature based on the spectrum of simulation signals of the dynamical models is generated. Aiming at selecting the best feature suitable for quantitative damage detection, a two-sample Z-test procedure is used to analyze the performance of features on damage evolution tracing. A feature named SR, which had better performance in tracking damage, is proposed to detect damage in planetary gear set. Meanwhile, the sun gear tooth-chipped seeded experiments with different severity are designed to validate the method above, and then the test vibration signal is picked up and used for damage detection. With the results of several experiments for quantitative damage detection, the feasibility and the effect of this approach are verified. The proposed method can supply an effective tool for degradation state identification in condition monitoring and health management of helicopter transmission system.

The Grey Analysis, Kriging and Selection Index of Flower Yield in Rugosa Rose

The analysis of grey system, kriging interpolation, and integration selection index were employed to investigate the relationships between the flower yield/plant （FY） and 15 other quantitative traits of 20 rugosa rose cultivars. The result showed that： The grey relational grade （GRG） of the number of flowers/plant （NF）, the number of branches/plant （NB）, the width of floral bud （WB）, and the weight/flower （WF） to the FY were larger （〉 0.5）; FY improved with the increase of NF and NB. Moreover, the indirect selection of either trait could not achieve improvement of FY. It is necessary to improve FY by multi-trait selection. The integration selection index （ISI） equation of FY was established with the characters NF, NB, WB, and WF： I= 0.3187x1 - 318.6x2 ＋ 670.1 x4 ＋ 6.3xa, index heritability = 0.8014, selective response of the integration breeding value = 245.8811. This will provide a theoretic basis for the genetic breeding of rugosa rose.

Quantitative Determination of Composition of Quaternary Cementitious Materials

Based on the principle of ENV 196-4 ＂Methods of testing cement - Part 4 Quantitative determination of constituents or Chinese Standard GB/12960-2007 Quantitative measurement of mineral admixtures in cement, methods were developed for quantitative determination of fly ash, slag and limestone powder in fresh cement pastes, mortars and concretes. Limestone powder was determined using thermal analysis method. The residue content of fly ash on an 80um sieve, and silt contents of aggregate were also considered during the quantitative determination of mineral composition of quaternary cementitious system. With the developed methods, the deviations between the measured and the actual mineral contents of the constituent in the eemantitious material in fresh cement paste, mortar and concrete, were within 3%.

Comparative Study of Variable Selection Using Genetic Algorithm with Various Types of Chromosomes

In this study,different methods of variable selection using the multilinear step-wise regression（MLR） and support vector regression（SVR） have been compared when the performance of genetic algorithms（GAs） using various types of chromosomes is used.The first method is a GA with binary chromosome（GA-BC） and the other is a GA with a fixed-length character chromosome（GA-FCC）.The overall prediction accuracy for the training set by means of 7-fold cross-validation was tested.All the regression models were evaluated by the test set.The poor prediction for the test set illustrates that the forward stepwise regression（FSR） model is easier to overfit for the training set.The results using SVR methods showed that the over-fitting could be overcome.Further,the over-fitting would be easier for the GA-BC-SVR method because too many variables fleetly induced into the model.The final optimal model was obtained with good predictive ability（R2 = 0.885,S = 0.469,Rcv2 = 0.700,Scv = 0.757,Rex2 = 0.692,Sex = 0.675） using GA-FCC-SVR method.Our investigation indicates the variable selection method using GA-FCC is the most appropriate for MLR and SVR methods.

Selection for growth performance of tank-reared Pacific white shrimp, Litopenaeus vannamei

Seven growth-related traits were measured to assess the selection response and genetic parameters of the growth of Pacific white shrimp, Litopenaeus vannamei, which had been domesticated in tanks for more than four generations. Phenotypic and genetic parameters were evaluated and fitted to an animal model. Realized response was measured from the difference between the mean growth rates of selected and control families. Realized heritability was determined from the ratio of the selection responses and selection differentials. The animal model heritability estimate over generations was 0.44±0.09 for body weight (BW), and ranged from 0.21±0.08 to 0.37±0.06 for size traits. Genetic correlations of phenotypic traits were more variable (0.51-0.97), although correlations among various traits were high (>0.83). Across generations, BW and size traits increased, while selection response and heritability gradually decreased. Selection responses were 12.28%-23.35% for harvest weight and 3.58%-13.53% for size traits. Heritability estimates ranged from 0.34±0.09 to 0.48±0.15 for harvest weight and 0.17±0.01-0.38±0.11 for size traits. All phenotypic and genetic parameters differed between various treatments. To conclude, the results demonstrated a potential for mass selection of growth traits in L. vannamei. A breeding scheme could use this information to integrate the effectiveness constituent traits into an index to achieve genetic progress.

“双碳”愿景下CO_(2)驱强化采油封存技术工程选址指标评价

在国家能源安全和“双碳”战略愿景下,CO_(2)驱强化采油封存技术(CO_(2)-EOR)因能助力油气行业转型发展,成为“低碳化”乃至“负碳化”的首选技术和最现实的选择。无论是实验、数值模拟还是现场实践,目前国内外学者对CO_(2)-EOR研究侧重于CO_(2)作为高效的驱油“催化剂”本身及油藏CO_(2)-EOR适应性认识,对于工程选址评价缺乏统一标准和系统研究。在充分调研国内外文献的基础上,结合中国CO_(2)-EOR应用进展和工程实践,明确了CO_(2)-EOR工程选址可行性评价所需的通用依据,指出了CO_(2)-EOR工程选址遵循“CO_(2)封存与驱油双统一”、安全性、经济性的专属性原则,并从CO_(2)-EOR工程选址的地质、工程、安全、经济4个要素开展了较详尽系统的研究,定性-定量构建了“4+8+27”CO_(2)-EOR工程选址三级指标评价体系(GESE),以期为油藏开展CO_(2)-EOR工程选址提供借鉴,助力中国碳减排技术的应用与发展。

基于机器学习方法的空气质量预测与影响因素识别

空气质量指数(AQI)的精准预测及影响因素识别,对空气污染防护和治理具有重要现实意义。选取北京市2014年第一季度至2022年第二季度AQI作为研究对象,探究六大污染物、五个气象因子和十四个经济变量对空气质量影响。选用DT,RF,GBDT和XGBoost模型对AQI进行预测,并使用稳定性选择方法定量分析各个变量对AQI的贡献。结果表明:四种模型方法均有良好的预测效果,其中XGBoost和RF的预测效果最优;六大污染物中PM2.5,PM10浓度和气象因素中的风速和气压对AQI影响较大;十四个经济变量对AQI的影响差异较大,其中城镇居民人均可支配收入、第三产业GDP和规模以上工业总产值等对AQI影响较大,而第一产业GDP和公路货物运输量等影响较小。

基于深度卷积神经网络的股票交易模型研究

为探讨卷积神经网络模型(CNN模型)的有效性,构建基于卷积神经网络模型的沪深300选股策略。首先通过分层法和IC测试法对CNN模型预测得到的上涨因子进行有效性测试;其次基于年化收益率、最大回撤等风险与收益指标,判断上涨因子选股策略的具体表现,进一步验证CNN选股模型的有效性。随后构建了基于宽基指数的择时策略,结果表明CNN模型在上证50指数上预测表现最佳。卷积神经网络的量化选股和择时模型的研究结论证实,卷积神经网络不仅能在沪深300中选出表现更好的股票,而且在量化择时方面也同样有效。

拉曼光谱结合光谱特征区间筛选算法快速定量鉴别植物调和油品质

本研究提出了一种基于拉曼光谱与光谱特征区间筛选算法实现植物调和油中高价值植物油含量快速定量检测的方法。首先,将粒子群优化(particle swarm optimization,PSO)算法与灰狼优化(grey wolf optimization,GWO)算法融合构建混合智能优化算法,即PSOGWO算法。其次,将PSOGWO与组合移动窗口(combined moving window,CMW)策略结合构建新型的拉曼光谱特征区间筛选算法,即PSOGWO-CMW算法。然后,将玉米油(corn oil,CO)和特级初榨橄榄油(extra virgin olive oil,EVOO)以不同比例配制为CO-EVOO植物调和油,并采集其拉曼光谱。将拉曼光谱输入偏最小二乘回归、PSO-CMW、GWO-CMW和PSOGWO-CMW模型预测EVOO含量,并比较建模效果。结果表明,PSOGWO-CMW模型具有最佳的预测性能。采用本方法与气相色谱-质谱法分别检测真实的CO-EVOO植物调和油样本中EVOO含量,结果表明两者的检测性能无显著差异。本方法快速、准确,亦可用于其他植物调和油中高价值植物油含量的快速定量检测。

糯玉米茎秆穿刺强度QTL分析与基因组选择

茎秆穿刺强度是衡量玉米茎秆机械强度和抗倒伏能力的重要指标之一,本研究以衍生于糯玉米自交系衡白522和通系5的198个重组自交系为试验材料,对茎秆穿刺强度进行数量性状位点(QTL)分析和基因组选择研究。单个环境QTL分析共检测到4个控制糯玉米茎秆穿刺强度的QTL,每个QTL的表型变异贡献率均小于10.00%,且仅在单个环境中被检测到;多个环境QTL分析共检测到8个QTL与环境互作,其加性效应总共可解释24.64%的表型变异,加性效应与环境互作贡献率为17.51%;上位性QTL分析共检测到4对QTL与QTL互作,可解释8.25%的表型变异。基因组选择中,当训练群体占群体总数的80%,随机选择500个标记即可获得较高的预测准确性;但是根据单个环境QTL分析结果,选择机率常用对数值排名前200的标记,即可大幅度提高基因组选择预测准确性。

多维度深水浅层建井方式优选方法研究

深水钻井作业具有海洋环境多变、地质条件复杂、作业风险大和日费率高等特点,深水浅层建井的方式难以选择。针对海底土质强度、建井质量、作业时效、经济性、风险控制等因素,根据现场作业工程数据,分别对不同建井工艺的适应性进行了分析,建立了单维度的适应性分级机制,并借助雷达图进行定量可视化,形成了多维度深水浅层建井方式优选方法。研究结果表明,水深500~1500 m的南海北部海域,单井作业时,喷射法一般为浅层建井最优方法,作业效率相比钻入法可提高50%以上。该方法已在中国南海几十口深水井取得良好应用效果,可为深水复杂地层建井方式选择提供定量评价方法。