关键词： artificial neural network   mechanical properties   neuro-fuzzy system   rubberized concrete   waste rubber  

摘要： The use of waste materials as a binder or aggregate in the concrete mixture is a great step towards sustainability in the construction industry. Waste rubber (WR) can be used as coarse and fine aggregates in concrete and improves the crack resistance, impact resistance, and fatigue life of the produced concrete. However, the mechanical properties of rubberized concrete degrade significantly by replacing the natural aggregate with WR. To have accurate estimations of the mechanical properties of rubberized concrete, two machine learning methods consisting of artificial neural network (ANN) and neuro-fuzzy system (NFS) were served in this study. To do this, a comprehensive dataset was collected from reliable literature, and two scenarios were addressed for the selection of input variables. In the first scenario, the critical ratios of the rubberized concrete and the concrete age were considered as the input variables. In contrast, the mechanical properties of concrete without WR and the percentage of aggregate volume replaced by WR were assumed as the input variables in the second scenario. The results show that the first scenario models outperform the models proposed by the second scenario. Moreover, the developed ANN models are more reliable than the proposed NFS models in most cases.

关键词： Desiccant wheel   Multiple linear regression   Multilayer Perceptron   Decision Tree   Modelling  

摘要： In this paper, several alternative models were developed with multiple linear regression and machine learning algorithms to determine the output states of silica gel desiccant wheels for balanced flow. The decision tree method was used for this purpose for the first time in open literature. All the models developed include six input parameters and a wider range than those available in the literature. Predictions from the models were compared with the master dataset used to derive the models, each of the five sub-datasets that make up the master dataset and with data available in the literature. It was determined that the most suitable models are those coded as multiple linear regression-IV (MLR-IV), multilayer perceptron regressor-III (MLPR-III) and decision tree-VII (DTVII), and DT-VII is the best among them. The determination coefficient and root mean square error for temperature were found to be 0.9894 and 0.8743 degrees C for MLR-IV, 0.9817 and 1.1526 degrees C for MLPR-III, 0.9986 and 0.3295 degrees C for DT-VII, respectively. The corresponding values for humidity ratio were 0.9912 and 0.3701 g kg(-1) for MLR-IV, 0.9885 and 0.4227 g kg(-1) for MLPR-III, 0.9994 and 0.0995 g kg(-1) for DT-VII, respectively. The results obtained revealed that the proposed models can be used safely in preliminary design, simulation and dynamic energy analysis of systems with desiccant wheels.

关键词： Hard constraint   Theory guided   Physics informed   Sparse observation   Projection   Constraint patch  

摘要： Machine learning models have been successfully used in many scientific and engineering fields. However, it remains difficult for a model to simultaneously utilize domain knowledge and experimental observation data. The application of knowledge-based symbolic artificial intelligence (AI) represented by expert systems is limited by the expressive ability of the model, and data-driven connectionism AI represented by neural networks is prone to produce predictions that might violate physical principles. In order to fully integrate domain knowledge with observations and make full use of the strong fitting ability of neural networks, this study proposes theory-guided hard constraint projection (HCP). This deep learning model converts physical constraints, such as governing equations, into a form that is easy to handle through discretization, and then implements hard constraint optimization through projection in a patch. Based on rigorous mathematical proofs, theory-guided HCP can ensure that model predictions strictly conform to physical mechanisms in the constraint patch. The training process of theory-guided HCP only needs a small amount of labeled data (sparse observation), and it can supervise the model by combining the coordinates (label-free data) with domain knowledge. The performance of the theory-guided HCP is verified by experiments based on a published heterogeneous subsurface flow problem. The experiments show that theory-guided HCP requires fewer data, and achieves higher prediction accuracy and stronger robustness to noisy observations, than the fully connected neural networks and soft constraint models. Furthermore, due to the application of domain knowledge, theory-guided HCP possesses the ability to extrapolate and can accurately predict points outside of the range of the training dataset. (C) 2021 The Author(s). Published by Elsevier Inc.

关键词： Type-2 diabetes   Emulation   Computational modelling   Risk prediction   Self-assessment  

摘要： Background The aim of the present paper is to construct an emulator of a complex biological system simulator using a machine learning approach. More specifically, the simulator is a patient-specific model that integrates metabolic, nutritional, and lifestyle data to predict the metabolic and inflammatory processes underlying the development of type-2 diabetes in absence of familiarity. Given the very high incidence of type-2 diabetes, the implementation of this predictive model on mobile devices could provide a useful instrument to assess the risk of the disease for aware individuals. The high computational cost of the developed model, being a mixture of agent-based and ordinary differential equations and providing a dynamic multivariate output, makes the simulator executable only on powerful workstations but not on mobile devices. Hence the need to implement an emulator with a reduced computational cost that can be executed on mobile devices to provide real-time self-monitoring. Results Similarly to our previous work, we propose an emulator based on a machine learning algorithm but here we consider a different approach which turn out to have better performances, indeed in terms of root mean square error we have an improvement of two order magnitude. We tested the proposed emulator on samples containing different number of simulated trajectories, and it turned out that the fitted trajectories are able to predict with high accuracy the entire dynamics of the simulator output variables. We apply the emulator to control the level of inflammation while leveraging on the nutritional input. Conclusion The proposed emulator can be implemented and executed on mobile health devices to perform quick-and-easy self-monitoring assessments.

关键词： Parameter Importance Assessment (PIA)   Avalanche susceptibility modeling   Machine learning (ML)   Boruta algorithm   Support Vector Machine (SVM)  

摘要： Due to ongoing climate change, water mass redistribution and related hazards are getting stronger and frequent. Therefore, predicting extreme hydrological events and related hazards is one of the highest priorities in geosciences. Machine Learning (ML) methods have shown promising prospects in this venture. Every ML method requires training where we know both the output (extreme event) and input (relevant physical parameters and variables). This step is critical to the efficacy of the ML method. The usual approach is to include a wide variety of hydro-meteorological observations and physical parameters, but recent advances in ML indicate that the efficacy of ML may not improve by increasing the number of input parameters. In fact, including unimportant parameters decreases the efficacy of ML algorithms. Therefore, it is imperative that the most relevant parameters are identified prior to training. In this study, we demonstrate this concept by predicting avalanche susceptibility in Leh-Manali highway (one of the most severely affected regions in India) with and without Parameter Importance Assessment (PIA). The avalanche locations were randomly divided into two groups: 70% for training and 30% for testing. Then, based on temporal and spatial sensor data, eleven avalanche influencing parameters were considered. The Boruta algorithm, an extension of Random Forest (RF) ML method that utilizes the importance measure to rank predictors, was used and it found nine out of eleven parameters to be important. Support Vector Machine (SVM) based ML technique is used for avalanche prediction, and to be comprehensive, four different kernel functions were employed (linear, polynomial, sigmoid, and radial basis function (RBF)). The prediction accuracy for linear, polynomial, sigmoid, and RBF kernels, with all the eleven parameters were found to be 80.4%, 81.7%, 39.2%, and 85.7%, respectively. While, when using selected parameters, the prediction accuracy for linear,

关键词： codon usage bias   metabolic modelling   metabolic engineering   quantitative proteomics   systems biology  

摘要： Proteins are responsible for most physiological processes, and their abundance provides crucial information for systems biology research. However, absolute protein quantification, as determined by mass spectrometry, still has limitations in capturing the protein pool. Protein abundance is impacted by translation kinetics, which rely on features of codons. In this study, we evaluated the effect of codon usage bias of genes on protein abundance. Notably, we observed differences regarding codon usage patterns between genes coding for highly abundant proteins and genes coding for less abundant proteins. Analysis of synonymous codon usage and evolutionary selection showed a clear split between the two groups. Our machine learning models predicted protein abundances from codon usage metrics with remarkable accuracy, achieving strong correlation with experimental data. Upon integration of the predicted protein abundance in enzyme-constrained genome-scale metabolic models, the simulated phenotypes closely matched experimental data, which demonstrates that our predictive models are valuable tools for systems metabolic engineering approaches. (C) 2021 Elsevier Ltd. All rights reserved.

关键词： 重症医学   机器学习   可视化分析   CiteSpace软件  

摘要： 目的基于知识可视化分析探讨2010~2020年机器学习在重症医学领域的应用情况。方法收集2010~2020年中国知网、万方数据和Web of Science(WOS)收录的机器学习方法在重症医学领域应用的相关文献。采用Cite Space软件对作者、国家、研究机构、共被引文献和关键词进行提取,并对该领域的合作关系以及热点、前沿话题进行探讨。结果2010~2020年期间共检索出4889篇中文文献和8036篇英文文献,发文量呈逐年上升趋势。在英文文献的作者、国家和机构研究主体分析中,发文量前十的机构及发文量最大、H指数最高的作者均来自美国,而中国在此领域尚有明显差距。"危险因素(risk factor)"和"病死率(mortality)"分别是中英文文献中频次最高的关键词。同时"危重症(critical illness)"在WOS中的突增年份从2018年开始一直持续至今,并且该关键词在中文文献中突增年份持续最长,表明机器学习应用已成为当前重症医学界持续关注的研究热点。在对WOS核心数据库的共被引共现网络聚类中,"脓毒症"和"机器学习"是最活跃的两个聚类。被引次数最高的2篇文献来自于"脓毒症"聚类,均与脓毒症的定义有关。在"机器学习"的聚类中,高被引文献集中在公开重症监护数据库-Ⅲ(medical information mark for intensive care-Ⅲ,MIMIC-Ⅲ)的分析,可见数据对机器学习应用于重症医学的重要性。结论机器学习是重症医学领域当前的研究热点之一,重点关键词为危重病病死率、危险因素和结局判断,脓毒症、谵妄、急性肾损伤和COVID-19等是最受关注的危重疾病。未来应对这些热点问题进行深入研究,增加国内外学者的合作与交流。

关键词： 抽油机井   示功图   机器学习   故障诊断  

摘要： 数字化油田建设使抽油机井实时生产数据呈现爆发式增长,为油井运行工况的实时诊断和故障分析提供了重要数据支持。利用大数据和智能算法对油井工况进行快速诊断分析,是油丼生产决策和智慧油田建设的重要内容。通过示功图判断工况是油井故障诊断的主要方法,示功图识别准确率影响着油丼故障判断的结果。本项目提出了一种基于卷积神经网络的示功图识别模型,通过参数优化和测试,形成了具有7层网络结构的CNN模型,实现了五种油井工况的准确分类。经训练和测试,该CNN模型收敛性好,模型训练准确率达到了99%,测试准确率达到了97%以上,达到了预期效果,对油井故障智能诊断具有重要的研究和应用价值。

关键词： 气溶胶   定量遥感   反演   机器学习   深度学习  

摘要： 机器学习方法近年来取得突破进展,其遥感应用从目标识别和地物分类领域,发展到定量化反演的多个领域。气溶胶定量遥感因其机理复杂,反演参数的种类和精度受到限制,机器学习为气溶胶遥感带来了新的研究和应用技术手段。本文汇总现有研究进展将气溶胶机器学习方法归纳为卫星遥感反演气溶胶光学厚度AOD(Aerosol Optical Depth)、卫星遥感反演其他气溶胶参数、卫星遥感反演颗粒物浓度(PMx)、地基气溶胶遥感4类。结合作者研究工作,通过分析讨论,归纳机器学习用于气溶胶定量遥感的条件为:(1)物理模型无法使用;(2)已有模型卫星产品精度低;(3)已有模型精度高但计算速度低。从应用的角度来说,可以借助于更多的具有相关性的输入信息,发挥机器学习在反演产品种类、反演精度、计算效率等方面的优势;而对定量遥感来说,应该同时重视挖掘遥感数据本身的信息来提高反演能力,并通过误差分析等手段反馈对遥感机理的理解,使机器学习与遥感机理研究相互促进。

关键词： 锂离子电池   机器学习   荷电状态   健康状态   剩余寿命  

摘要： 本文旨在综述机器学习方法在锂离子电池状态(包括荷电状态、健康状态和剩余可用寿命)估计领域的研究进展。首先,阐述机器学习方法在电池状态估计中的应用现状。然后,归纳电池状态估计机器学习方法的5个具体实施环节,即数据准备、模型选择与评价、超参数确定、数据预处理和模型训练,并提出了融合精度、实施成本和鲁棒性的学习算法评价方法。最后指出超参数确定方法仍存在的场景适应性问题,并建议建立多区域、跨季节、多模式和长时间的车用电池工况数据库,促进电池状态估计机器学习算法的实用性和普适性等方面的研究。