Large scale variable fidelity surrogate modeling

Engineers widely use Gaussian process regression framework to construct surrogate models aimed to replace computationally expensive physical models while exploring design space. Thanks to Gaussian process properties we can use both samples generated by a high fidelity function (an expensive and accurate representation of a physical phenomenon) and a low fidelity function (a cheap and coarse approximation of the same physical phenomenon) while constructing a surrogate model. However, if samples sizes are more than few thousands of points, computational costs of the Gaussian process regression become prohibitive both in case of learning and in case of prediction calculation. We propose two approaches to circumvent this computational burden: one approach is based on the Nyström approximation of sample covariance matrices and another is based on an intelligent usage of a blackbox that can evaluate a low fidelity function on the fly at any point of a design space. We examine performance of the proposed approaches using a number of artificial and real problems, including engineering optimization of a rotating disk shape.     

Optical mapping and nanocoding approaches to whole-genome analysis

Recent application of microfluidic and nanofluidic approaches to genomics has enabled rapid analysis of whole human and other large genomes. Here, we present a review of optical mapping and nanocoding, two single-molecule whole-genome analysis systems, which are used for physical mapping of genomes. Optical mapping employs a microfluidic device, fabricated using soft lithography approaches, to achieve DNA presentation via DNA flow in microchannels. On the other hand, nanocoding leverages DNA confinement in nanoscale devices to achieve DNA presentation. In this review, we focus on the physical principles underlying DNA presentation in both of these systems and the application of these approaches towards better understanding of genomic structure and structural variation.     

Deploying learning designs across physical and web spaces: Making pervasive learning affordable for teachers

Pervasive computing devices and communication infrastructures enable learning situations that occur in both the physical and the virtual world. However, deploying these pervasive situations is still a challenge for teachers. This paper presents GLUEPS-AR, a system for deploying learning designs across physical and web spaces, using mainstream Virtual Learning Environments, Web 2.0 artifacts and Augmented Reality applications. GLUEPS-AR has been evaluated through a mixed methods study on the deployment of three authentic pervasive learning situations. Results highlight that GLUEPS-AR supports teachers in deploying their pedagogical ideas on pervasive learning environments, overcoming the main limitations of existing approaches.     

A cyber-physical experimentation environment for the security analysis of networked industrial control systems

Although many studies address the security of Networked Industrial Control Systems (NICSs), today we still lack an efficient way to conduct scientific experiments that measure the impact of attacks against both the physical and the cyber parts of these systems. This paper presents an innovative framework for an experimentation environment that can reproduce concurrently physical and cyber systems. The proposed approach uses an emulation testbed based on Emulab to recreate cyber components and a real-time simulator, based on Simulink, to recreate physical processes. The main novelty of the proposed framework is that it provides a set of experimental capabilities that are missing from other approaches, e.g. safe experimentation with real malware, flexibility to use different physical processes. The feasibility of the approach is confirmed by the development of a fully functional prototype, while its applicability is proven through two case studies of industrial systems from the electrical and chemical domain.     

One-class classification for oil spill detection

SAR oil spill classification is a challenging topic, which is tackled by semi-empirical ad hoc approaches supported by very qualified experts. In all cases, the feature space is empirically defined, and two-class classification approaches are used. Although this approach allows achieving acceptable operational results, there is still room for improving both the comprehension of the physical phenomenon and the performance of classification techniques. In this paper, we propose a novel approach to oil-spill classification based on the paradigm of one-class classification. A classifier is trained using only examples of oil spills, instead of using oil spills and look-alikes, as in two-class approaches. Further, since the feature space is empirically defined, we also propose an objective technique to select the most powerful one that is suited for the oil-spill detection task at hand. Results on two case study datasets are reported to validate the proposed approach.     

Example‐Driven Deformations Based on Discrete Shells

Despite the huge progress made in interactive physics‐based mesh deformation, manipulating a geometrically complex mesh or posing a detailed character is still a tedious and time‐consuming task. Example‐driven methods significantly simplify the modelling process by incorporating structural or anatomical knowledge learned from example poses. However, these approaches yield counter‐intuitive, non‐physical results as soon as the shape space spanned by the example poses is left. In this paper, we propose a modelling framework that is both example‐driven and physics‐based and thereby overcomes the limitations of both approaches. Based on an extension of the discrete shell energy we derive mesh deformation and mesh interpolation techniques that can be seamlessly combined into a simple and flexible mesh‐based inverse kinematics system.     

Preference Learning for Cognitive Modeling: A Case Study on Entertainment Preferences

Learning from preferences, which provide means for expressing a subject's desires, constitutes an important topic in machine learning research. This paper presents a comparative study of four alternative instance preference learning algorithms (both linear and nonlinear). The case study investigated is to learn to predict the expressed entertainment preferences of children when playing physical games built on their personalized playing features ( entertainment modeling). Two of the approaches are derived from the literature-the large-margin algorithm (LMA) and preference learning with Gaussian processes-while the remaining two are custom-designed approaches for the problem under investigation: meta-LMA and neuroevolution. Preference learning techniques are combined with feature set selection methods permitting the construction of effective preference models, given suitable individual playing features. The underlying preference model that best reflects children preferences is obtained through neuroevolution: 82.22% of cross-validation accuracy in predicting reported entertainment in the main set of game survey experimentation. The model is able to correctly match expressed preferences in 66.66% of cases on previously unseen data (p -value = 0.0136) of a second physical activity control experiment. Results indicate the benefit of the use of neuroevolution and sequential forward selection for the investigated complex case study of cognitive modeling in physical games.     

Perception and biomechanics data in a manual handling task: a comparative study

This paper explores the use of subjective perception tasks and its correlations with biomechanical data in the evaluation of manual material handling. Three main dimensions were considered for perception: physical regroups sensations issued from a specific body area; operative regroups feelings related to the execution of the task; and performance regroups feelings that involve a judgement on the execution or reflect overall sensations. The following questions were then explored. To what extent are perception data related to biomechanics data? Do both approaches lead to similar conclusions or interpretations when effect of practice, format and off-centre were tested? How can they complement one another? The task consisted of transferring 50 series of three 15 kg loads in order to verify the impact of free practice, format (box/cylinder) and load centre of gravity position. Eleven subjects rated perception on a CR-10 scale (Borg 1982) after each series. The session was completed with an interview on perception. The net resulting moment was systematically found to be the best correlated with data perception. While all physical and performance items corresponded in various ways to biomechanics data, perceptions associated with operative dimension appeared to be less related with biomechanical data. As regards the impact of practice, format and off-centre, both approaches would lead to the same conclusions, except for the effect of the off-centre. Verbal data add rational information about how or why perception can or cannot be reflected in biomechanics data. How both approaches can be matched more closely in manual handling is discussed.     

INTRODUCING META-LEVELS TO QUALITATIVE REASONING

Current approaches to qualitative reasoning are largely based on a fixed framework for modeling the physical world and concentrate on the reasoning methods that support qualitative reasoning. This paper argues that we need several levels of abstraction and different viewpoints on how to model the physical world, in order to create systems that reason about the physical world in a flexible way. We present a framework that integrates the three basic approaches to qualitative reasoning and show how this framework can be used as a basis for a flexible qualitative reasoning system.     

PCVM.ARIMA: predictive consolidation of virtual machines applying ARIMA method

Cloud computing adopts virtualization technology, including migration and consolidation of virtual machines, to overcome resource utilization problems and minimize energy consumption. Most of the approaches have focused on minimizing the number of physical machines and rarely have devoted attention to minimizing the number of migrations. They also decide based on the current resources utilization without considering the demand for resources in the future. Some approaches minimize the number of active physical machines and Service Level Agreement (SLA) violations with the number of unnecessary migrations. They consider the current resource utilization of physical machines and neglect from demands for future resource requirements. As a result, as time passes, the number of unnecessary migrations, and subsequently, the rate of SLA violations in data centers increases. Alternatively, several approaches only focus on a hardware level and reduce the physical machine’s dynamic power consumption. The lack of control over the overload of physical machines increases the amount of violation. In this paper, a framework called PCVM.ARIMA is presented that focuses on the dynamic consolidation of virtual machines over the minimum number of physical machines, minimize the number of unnecessary migrations, detect the physical machine overloading, and SLA based on the ARIMA prediction model. Moreover, the Dynamic Voltage and Frequency Scaling (DVFS) technique is used to apply the optimal frequency to heterogeneous physical machines. The experimental results show that the presented framework significantly reduces energy consumption while it improves the QoS factors in comparison to some baseline methods.

     

Assessment of physical strain in younger and older subjects using heart rate and scalings of perceived exertion

Physical strain is usually evaluated physiologically by using the heart rate (HR) or psychologically using scaled perceived exertion (PE). In order to explore the relationship between these approaches, we compared the series of both measurements taken simultaneously during work on a bicycle ergometer. A total of 29 participants aged between 27 and 41, 42 and 56, and 57 and 71 years took part in 28 minutes of cycling with systematically increased and decreased load as well as in 7 hours of continuous cycling with low to medium exertion, interrupted by brief peak loads at high to very high exertion levels. Results revealed that, while both measurements are suitable to capture physical strain, HR is not as specific as PE. The older participants further perceived comparable physical loads as more demanding as the younger participants. With an aging workforce, results have potential implications for the organisation of work places.     

Computer-aided physical multi-domain modelling: Some experiences from education and industrial applications

For the purposes of this paper, computer-aided physical modelling means a type of modelling in which a computer-aided approach is used, with the basic aim being to maintain the physical structure of a real system or its topology as much as possible in the model. Bond graphs represent a very efficient and traditional approach. However, new, object-oriented and multi-domain tools based on the Modelica language are more appropriate for industrial staff or for the people who do not have a deep insight into modelling and simulation. In this paper we describe several educational and industrial application projects in the Dymola–Modelica environment: a process-systems library, two mechanical systems (an inverted pendulum and a laboratory helicopter), a model of thermal and radiation flows in buildings and two models of processes in mineral-wool production, i.e., a pendulum system and a recuperator system. We describe some experiences from these projects, but also from a more general use of the Matlab–Simulink and Dymola–Modelica environments over many years. One simple conclusion is that we need to educate with two approaches: a more physical and advanced acausal Modelica-like approach, but also a more traditional causal or block-oriented approach according to the historical CSSL standard. The important advantages and disadvantages of both approaches are described. The Modelica-based approach enables true ‘physical’ modelling with fully reusable components. However, there is a particular danger, i.e., users occasionally forget some basic modelling principles when using sophisticated libraries. The result is a very complex modelling structure that is relatively inefficient for the simulation and sometimes has many numerical problems. It is usually very difficult to detect the real reasons for that.     

Parallelization methods for implementation of discharge simulation along resin insulator surfaces

In this paper, we will investigate the implementation of the parallelization approaches used in the program of discharge simulation along resin insulator surfaces in SF₆/N₂ gas mixture which initially consumes a great deal of computational time. In a general way, this simulation program spent 10 days of execution to achieve satisfactory research results. For this reason, the goal of our paper is to reduce the execution time by parallelizing this program. Three parallelization approaches were used in our simulation: (i) splitting by different types of the charged particles using a distributed-memory approach, (ii) splitting by physical domain using a distributed-memory approach, and (iii) splitting by both domain and charged particles using multi-level distributed and shared memory approach. At last, the three approaches are tested on a Linux cluster composed of six dual-core PCs, and the experimental results show that all the parallelization approaches achieve the goal of reducing the execution time to a certain extent. In addition, among these approaches, the multi-level approach offers the most effective parallelization method for implementing this simulation on symmetrical multi-processing (SMP) clusters.     

Supervised machine learning in multimodal learning analytics for estimating success in project‐based learning

Multimodal learning analytics provides researchers new tools and techniques to capture different types of data from complex learning activities in dynamic learning environments. This paper investigates the use of diverse sensors, including computer vision, user‐generated content, and data from the learning objects (physical computing components), to record high‐fidelity synchronised multimodal recordings of small groups of learners interacting. We processed and extracted different aspects of the students' interactions to answer the following question: Which features of student group work are good predictors of team success in open‐ended tasks with physical computing? To answer this question, we have explored different supervised machine learning approaches (traditional and deep learning techniques) to analyse the data coming from multiple sources. The results illustrate that state‐of‐the‐art computational techniques can be used to generate insights into the "black box" of learning in students' project‐based activities. The features identified from the analysis show that distance between learners' hands and faces is a strong predictor of students' artefact quality, which can indicate the value of student collaboration. Our research shows that new and promising approaches such as neural networks, and more traditional regression approaches can both be used to classify multimodal learning analytics data, and both have advantages and disadvantages depending on the research questions and contexts being investigated. The work presented here is a significant contribution towards developing techniques to automatically identify the key aspects of students success in project‐based learning environments, and to ultimately help teachers provide appropriate and timely support to students in these fundamental aspects.     

Field modeling with sampled distances

Traditional mesh-based approaches to the modeling and analysis of physical fields within geometric models require some form of topological reconstruction and conversion in the mesh generation process. Such manipulations tend to be tedious and error-prone manual processes that are not easily automated. We show that most field problems may be solved directly by using approximate distance fields computed from designed or sampled geometric data, thus avoiding many of the difficult reconstruction and meshing problems. With distances we can model fields that satisfy boundary conditions while approximating the governing differential equations to arbitrary precision. Because the method is based on sampling, it provides natural control for multi-resolution both in geometric detail of the domain and in accuracy of the computed physical field. We demonstrate the field modeling capability with several heat transfer applications, including a typical transient problem and a ‘scan and solve’ approach to the simulation of a physical field in a real-world artifact.     

分布式网络系统的任意预设时间分组一致

研究任意预设时间控制下的多智能体网络系统分组一致性问题.设计非零分组投影参数下任意预设时间控制协议,使得分布式网络系统在物理允许范围内的任意预设时间内迅速实现分组一致,该预设时间与系统参数和初始值都无关系.基于代数图论、李雅普诺夫稳定性和矩阵理论等,分别讨论无向和有向拓扑网络情形下,多智能体系统实现预设时间分组一致的充分条件.独轮车的多智能体系统仿真实验验证了所提方法的有效性.     

Bond graph modeling from an object oriented modeling point of view

Along with an ever increasing model complexity, a so-called object oriented approach to physical systems modeling has become more and more popular throughout the last few years. Frequently used keywords are multi-domain modeling, model reuse, and non-causal equations. On the other hand the physical systems modeling methodology based on bond graphs has been in use worldwide since Paynter devised bond graphs more than 35 years ago. It seems that due to different roots and a different terminology, aspects of one of the two approaches are not fully appreciated by those who adhere to the other modeling paradigm. By relating features of object-oriented modeling (OOM) to corresponding ones of the older bond graph methodology, it is pointed out what both modeling approaches have in common and what is different. As a working modeling language, Modelica is used since it seems that this object oriented modeling language is going to receive an increasing attention as a neutral exchange format between proprietary modeling tools. As an application example that combines the electrical, the hydraulic and the mechanical energy domain in a single system, a hydraulic drive with a controlled displacement pump is considered.     

Providing personalized converged services based on flexible network reconfiguration

Recent developments show a progressing convergence of networks, services, and applications, which results in the necessity for new approaches to provide a ubiquitous experience on converged networks for specialized high data-rate services. A fundamental diversifying attribute for the future physical network convergence paradigm is pressing. In this paper, we make researches on converged resource reconfiguration model of substrate physical network based on hardware virtualization and provide a serv...     

Does Ergonomics Improve Product Quality and Reduce Costs? A Review Article

Competition is an ongoing challenge confronting industrial corporations, particularly automobile manufacturing. Striving to improve product quality and productivity, automotive industries have used different quality management approaches, such as reduced variability, total quality management, and lean management, over recent years. Furthermore, incorporating proactive ergonomics such as physical and organizational ergonomics and psychosocial factors into the structure of a company is considered to be a support for productivity and quality. Several studies have shown the effects of ergonomics on better quality. Application of both quality management approaches and ergonomics in an integrated manner in the manufacturing production system is emphasized because they are similar concepts with the same objectives, that is, to improve efficiency. In this study, a comprehensive review was undertaken and 25 studies were reviewed in order to define how integration of an ergonomic approach in the manufacturing production system can reduce defects and improve quality in the production process.