Integration of machine learning and first principles models

Model building and parameter estimation are traditional concepts widely used in chemical, biological, metallurgical, and manufacturing industries. Early modeling methodologies focused on mathematically capturing the process knowledge and domain expertise of the modeler. The models thus developed are termed first principles models (or white-box models). Over time, computational power became cheaper, and massive amounts of data became available for modeling. This led to the development of cutting edge machine learning models (black-box models) and artificial intelligence (AI) techniques. Hybrid models (gray-box models) are a combination of first principles and machine learning models. The development of hybrid models has captured the attention of researchers as this combines the best of both modeling paradigms. Recent attention to this field stems from the interest in explainable AI (XAI), a critical requirement as AI systems become more pervasive. This work aims at identifying and categorizing various hybrid models available in the literature that integrate machine-learning models with different forms of domain knowledge. Benefits such as enhanced predictive power, extrapolation capabilities, and other advantages of combining the two approaches are summarized. The goal of this article is to consolidate the published corpus in the area of hybrid modeling and develop a comprehensive framework to understand the various techniques presented. This framework can further be used as the foundation to explore rational associations between several models.     

川西彭州地区雷口坡组潮坪相薄储层辨识机理研究

川西彭州地区三叠系雷口坡组雷四上亚段潮坪相薄储层识别难度极大。围绕如何从复合地震强反射中区分并识别上、下两套储层面临的地球物理难题,采用先“分”后“合”的研究思路,基于实际地层结构及不同储层叠加样式建立正演模型,利用全波场波动方程正演模拟技术,剖析了不同主频条件下薄储层的地震响应特征,通过波形差异化分析,从复合地震响应中“剥离”出了两套储层所引起的地震响应特征及变化规律,明确了两套储层在不同频带下的地震识别标志和识别方法,为该区强反射界面干扰下两套薄互层储层辨识机理分析及精准预测奠定了基础。基于不同频带下薄储层辨识机理的分析结果,定性预测了薄储层平面展布,提出了深层潮坪相薄储层识别和预测难题的解决方案,为该区地震资料品质评价、面向薄储层的地震采集技术设计、地震资料处理及薄储层预测提供了依据和指导。     

Tribological behavior of uncoated and DLC-coated CoCr and Ti-alloys in contact with UHMWPE and PEEK counterbodies

There are very few studies of PEEK wearing against DLC. In this study, DLC and Zr-DLC coatings were deposited by magnetron sputtering onto Ti alloy substrates. Coated and un-coated Ti alloy samples were tribologically tested in multidirectional pin-on-disk equipment against UHMWPE and PEEK. CoCr alloy was also tested as control. Each of the materials was tested for 2 million cycles (Mc) in bovine serum lubricant (37 °C). Gravimetrical weight loss and changes in surface profile were similar when tested against UHMWPE. However, Zr-DLC samples delaminated after 1.2 Mc. When tested against PEEK, both coated samples delaminated, despite that the coatings maintained their structural integrity. The synergetic effect of stress-induced corrosion through biological fluid and high cyclic shear stress may have caused interface fatigue and subsequent delamination of both coatings.     

Integrating social and cognitive aspects of belief dynamics: towards a unifying framework

Belief change and spread have been studied in many disciplines—from psychology, sociology, economics and philosophy, to biology, computer science and statistical physics—but we still do not have a firm grasp on why some beliefs change more easily and spread faster than others. To fully capture the complex social-cognitive system that gives rise to belief dynamics, we first review insights about structural components and processes of belief dynamics studied within different disciplines. We then outline a unifying quantitative framework that enables theoretical and empirical comparisons of different belief dynamic models. This framework uses a statistical physics formalism, grounded in cognitive and social theory, as well as empirical observations. We show how this framework can be used to integrate extant knowledge and develop a more comprehensive understanding of belief dynamics.     

Physics‐based model of lithium‐ion batteries running on a circuit simulator

The authors developed a physics‐based equivalent circuit model of a lithium‐ion battery (LIB) whose parameters are continually updated, reflecting the theoretical calculation results of the Butler‐Volmer equation, diffusion equations of the lithium‐ion and lithium, and Nernst equations of the liquid and solid phases. The developed model was applied to the charge/discharge simulations of an LIB, and the experimental and simulated results of constant current discharges and pulsed‐charge/discharge were found to be in excellent agreement. In particular, using the developed model, analyzing transient responses of the LIB derived from the transition of the electric double layer charging to the electrode reaction is possible. These results demonstrate that the electrochemical performance of an LIB can be calculated on a circuit simulator using the developed model.     

物理教学与创造能力培养

针对目前我国高等教育改革中素质教育强调的学生创造能力的培养问题 ,我们以创造学基本理论为基础 ,探讨了大学物理教学过程中创新教育的策略和方法 ,提出了基于事实的因果关系进行创造性思想教育 ,培养学生创造能力为目标的教学新理念     

SASW测试相位差不确定性分析

在SASW测试中，受高阶振型波及道距等的影响，从折叠相位差曲线中无法通过折叠展开方法确定各离散频率点的绝对相位差．绝对相位差与折叠相位差相差2面弧度)的整数倍，整数取值不同，由一条折叠相位差曲线计算的相速度．频率曲线也不同，它们并不完全与表面波各阶振型频散曲线对应．若不同振型表面波影响在频率域是分离的，可以得到相应频率范围该振型波的频散曲线。在这些曲线中，总有一条相速度．频率曲线与常规折叠展开方法得到的相速度．频率曲线对应．将不同道距下得到的这些相速度-频率曲线堆叠在一起，并用灰度来表示各频散点出现的概率，由此可以研究基阶振型以及高阶振型波的频散．     

敲击-回波法中传感器影响的LTI系统理论分析

利用一维波动方程对理想自由桩在桩顶受冲击力作用的动态过程进行数学仿真，用有限元法程序计算出桩顶处不考虑传感器的速度响应理论时程曲线，并由LTI系统卷积积分及傅立叶变换得出冲击力和速度响应的频域表示；根据LTI系统理论对考虑传感器特性的动态过程进行分析，得出速度响应的频域表示；将典型的传感器频率特性与不考虑传感器的理想自由桩的速度响应在频域进行数字合成，经傅立叶逆交换还原成含有传感器特性的速度响应时程曲线：调节冲击力脉宽，计算与同一传感器特性对应的速度响应，得出冲击力与传感器对速度响应的联合影响规律并举出实例进行验证；最后，提出一种新的变换域抑制法，使速度响应中的传感器特性影响有可能通过计算机程序来抑制。     

基于虚拟技术的物理仿真实验系统的设计

随着信息技术的高速发展,虚拟现实技术在众多领域得到了广泛应用,本文基于此,采用B/S三层架构,结合Microsoft Visual C++2010及3D技术,构建了基于虚拟技术的物理仿真实验系统。     

引信起爆控制电路冗余设计技术研究

引信起爆控制电路通过对目标探测信号的处理和决策,给出发火控制信号,其可靠性直接决定了其发火可靠度,工程上大多采用直接双套并联方式进行冗余设计。针对引信起爆控制电路如何进行冗余设计问题,设计了并联冗余和非冗余两种电路,采用数理统计的方法计算了其全寿命周期可靠度,结果表明非冗余电路可靠度略高于并联冗余电路。采用故障物理方法,分别通过加速寿命试验和故障物理仿真,得出了起爆控制电路在引信服役期内的温度应力、力学环境应力、电应力下的失效寿命远大于寿命期的结论。综合数理统计和故障物理研究结果,给出了引信起爆控制电路可采用非冗余设计的建议。