关键词： Rock   Elastic properties   Anisotropy   Machine learning  

摘要： This paper aims to demonstrate that the elastic stiffnesses and the anisotropic parameters of rocks can be accurately predicted from geophysical features such as the porosity, the density, the compression stress, the pore pressure and the burial depth using relevant machine learning methods. It also suggests that the extreme gradient boosting method is the best method for this purpose. It is more accurate, extremely faster to train and more robust than the artificial neural networks and the support vector machine methods. Very highR-squared scores was obtained for the predicted elastic stiffnesses of a relevant dataset that is available in the literature. This dataset contains different types of rocks, and the values of the features are in large ranges. An optimal set of parameters was obtained by considering an appropriate sensitivity analysis. The optimized model is very easy to implement in Python for practical applications.

关键词： ATP-binding cassette (ABC) transporters   fatty acid desaturase (FAD)   genetic map   glaucousness   large-scale sequence-based markers   Secale cereale L  

摘要： The standard approach to genetic mapping was supplemented by machine learning (ML) to establish the location of the rye gene associated with epicuticular wax formation (glaucous phenotype). Over 180 plants of the biparental F-2 population were genotyped with the DArTseq (sequencing-based diversity array technology). A maximum likelihood (MLH) algorithm (JoinMap 5.0) and three ML algorithms: logistic regression (LR), random forest and extreme gradient boosted trees (XGBoost), were used to select markers closely linked to the gene encoding wax layer. The allele conditioning the nonglaucous appearance of plants, derived from the cultivar Karlikovaja Zelenostebelnaja, was mapped at the chromosome 2R, which is the first report on this localization. The DNA sequence of DArT-Silico 3585843, closely linked to wax segregation detected by using ML methods, was indicated as one of the candidates controlling the studied trait. The putative gene encodes the ABCG11 transporter.

关键词： Image edge detection   Machine learning algorithms   Classification algorithms   Training   Machine learning   Visualization   Filtering algorithms   Machine learning   depth image information   edge detection   algorithmic design  

摘要： Machine learning algorithms have become a hot topic in current research due to their unique learning performance, and have achieved fruitful research and application results in various fields. In this paper, the idea of machine learning classification algorithm is applied to depth image edge detection, AdaBoost algorithm and decision tree are used for image edge detection. The algorithm is created from training set creation, depth image feature extraction and combination of AdaBoost and image depth information, creating image training sample sets, selecting image features, training algorithm classifiers, and simulating medical ultrasound image classifiers. Finally, the machine learning algorithm was simulated and tested. The experimental results show that the edge detection effect is good, the algorithm adaptability is strong, and no adjustment parameters are needed.

关键词： CO2 solubility   Ionic liquids   Machine learning   Group contribution   Artificial neural network   Support vector machine  

摘要： A comprehensive database containing 10,116 CO2 solubility data measured in various ionic liquids (ILs) at different temperatures and pressures is established. Based on this database, the relationship between CO2 solubility and IL structure, temperature and pressure is correlated using group contribution (GC) methods. Two different machine learning algorithms, namely artificial neural network (ANN) and support vector machine (SVM), are employed to develop the GC models. For the 2023 test-set data, the estimated MAE and R-2 are 0.0202 and 0.9836, respectively for the ANN-GC model and for the SVM-GC model they are 0.0240 and 0.9783, respectively. The distributions of prediction errors are plotted for both models to provide more comprehensive knowledge on the model performance. The results indicate that both of the models can give reliable predictions on the CO2 solubilities in ILs and the ANN-GC model performs slightly better than the SVM-based model. (C) 2020 Elsevier Ltd. All rights reserved.

关键词： Solar radiation   Machine learning   Climatic zones   Input combination   Meteorological factor  

摘要： The values of global solar radiation are important fundamental data for potential evapotranspiration estimation, solar energy utilization, climate change study, crop growth model, and etc. This research tried to explore the optimal combination of input meteorological factors and the machine learning methods for the estimation of daily solar radiation under different climatic conditions so as to improve the estimation accuracy. Based on the correlation between meteorological factors, different meteorological factor input combinations were established and the support vector machine method was used to estimate global solar radiation at 80 weather stations in four climatic regions of China mainland. The results showed that, the optimal combinations of input meteorological factors were different in the four different climatic zones in China mainland. Three meteorological factors of sunshine hours, extraterrestrial radiation, and air temperature had greater impacts on the solar radiation estimation. Adding the factor of precipitation could obviously improve the estimation accuracy in humid regions, but not remarkably in arid regions. Wind speed had very little influence on solar radiation estimation. The accuracies of machine learning methods were better than the Angstrom-Prescott formula and the multiple linear regression method. Among them, support vector machine and extreme learning machine were more appropriate. In some sites, the root mean square error of support vector machine method was even 20% less than that of the Angstrom-Prescott formula. In general, reasonable division of the areas and establishment of appropriate input combinations of meteorological factors according to the climatic conditions, combined with machine learning methods, can effectively improve the accuracy of solar radiation estimation.

关键词： Cross-section calculation   Engineering simulator   HTGR   Machine learning methods  

摘要： To give better reactor physics results, the detailed 3-D diffusion module has been implemented in HTGR engineering simulator system (ESS), which needs more efficient cross-section calculations to meet real-time requirement of ESS. To calculated the HTGR's cross-sections, more reactor variables are needed with larger scope and more complex non-linear relationships, which makes the polynomial method used well in LWR simulators hard to meet full range accuracy. In this paper, a new cross-section calculation method based on machine learning is proposed. The proposed method uses deep neuron network to consider the complex non-linear relationships between reactor variables and uses tree regression to achieve accurate cross-sections in full range, which are trained from the scattered database generated by V.S.O.P.. With the numerical tests, the proposed cross-section calculation method is proved to achieve much more accurate cross-section results than present polynomial method and also fast enough to be used in the HTGR ESS. (C) 2020 Elsevier Ltd. All rights reserved.

关键词： Motorcycle   motorcycle crash   injury severity   data mining   classification techniques   road safety  

摘要： This study uses classification algorithms to establish models to predict the severity of crash injuries when motorcycle crashes occur. In this study, the power of multi-layer perceptron (MLP), rule induction (PART) and classification and regression trees (SimpleCart) models for predicting the severity of motorcycle crash was evaluated by comparing their results. To accomplish this objective, motorcycle crash data set extracted from the National Road Traffic Crash Database at the Building and Road Research Institute in Ghana. The data set was classified into four injury severity categories: fatal, hospitalised, injured and damage. The data collected from this database will provide means to directly compare and rank the data mining models, while also allowing for the identification of variables that are significantly influencing the severity of motorcycle crash. The results showed that among the tested classification algorithms, the SimpleCart model with an average accuracy of 73.81% outperformed the PART model (73.45%) and the MLP (72.16%) model based on a 10-fold cross-validation approach. The results revealed that the most significant factors associated with motorcycle crash injury severity were location type, settlement type, time of the crash, collision type and collision partner.

关键词： Association rule mining   Bayesian network   Defects   Machine learning   Swiss cheese model  

摘要： Defects pose considerable risks to construction projects in terms of both cost and quality, and identifying defects thus is crucial to effective construction quality management. In this study, data for 45 cases were obtained from the Public Construction Management Information System (PCMIS) of Taiwan. A combined machine learning method comprising association rule mining and a Bayesian network was employed to identify the relationships between defects as well as their occurrence probabilities. A total of 33 association rules and 11 high-risk defects were detected. The Swiss cheese model (SCM) was used to formulate four defensive layers and analyze the high-probability, strong-correlation, and multipath characteristics of high-risk defects. Specifically, the defect quality control inspection not implemented had the highest risk values. In the correlation analysis, more high-risk defects meant reduced inspection scores and construction quality;risk values and inspection scores had a strong negative correlation (r=-0.85). This study proposes innovative hybrid machine learning to evaluate the risks of defects, and the SCM was implemented to establish the risk factors and hierarchical relationships of the defects to determine their priority order in management. Future studies should analyze the time series of defects and employ sequential data to forecast their order of occurrence and relationships at different times, thereby increasing the understanding of dynamic construction projects. (C) 2020 American Society of Civil Engineers.

关键词： 电厂   垃圾分类   电费补贴   激励   主成分  

摘要： 针对现有居民垃圾分类效果不理想以及垃圾发电厂的处理速度日渐难以满足城市正常运转的需求的难题,本文深入研究了当前各大城市广泛采用的分类宣传、社区督导、行为激励等手段,提出了一种基于电网电厂数据交互、协议互动的电费补贴激励居民垃圾分类的方法,构建了基于工业互联网技术的多源异构电网电厂数据交互平台,并采用了k-近邻决策算法实现居民家庭门牌与电力缴费户号关联,提出一种发电厂数据奖惩指数分级算法实现垃圾分类对电厂效益提升的区间评估。为预测该套激励方法的效果,本文分别采用PSM模型算法预测分析该措施与宣传、督导等其他措施相比的实施效果,以及SPSS主成分分析法预测分析该措施与费用减免、积分兑换等其他激励措施相比的实施效果,最终证明该方法可有效提升全社会垃圾分类行为主动性。

关键词： 计算机视觉   人工智能   检测准确率   人体姿态识别   人体行为   机器学习方法   姿态检测系统   Open  

摘要： 人工智能与计算机视觉的结合是划时代的产物,计算机视觉智能化引起了新一轮的人工智能潮流,上至太空望远,下至海底探秘,计算机视觉技术已经有着丰富的实际应用。作为计算机视觉技术的延伸,人体姿态识别技术主要用于检测人体姿态或预测人体行为,在智能家居、安防等领域具有重要现实意义和研究价值。本文针对目前采用的浅层机器学习方法无法消除背景干扰,检测准确率不高等缺点,采用改进的神经网络模型对人体姿态进行检测,实验结果表明本文采用的方法能准确识别人体基本姿态,具有较好的准确率和可靠性。