数据中心内服务器能耗数据采集及特征分析

数据中心的高能耗低能效问题正受到广泛关注与研究，但目前没有公开的数据中心内服务器能耗数据集供研究人员使用，且现有过滤式特征选择并不能满足运维人员的需求。为此，提出了一套模拟数据中心内服务器运行状态的仿真环境架构，基于该架构采集了服务器运行各类任务时的多项性能指标和能耗数据。然后将基于因果关系的特征选择应用于能耗数据集的特征分析中，构建出可解释的特征子集和能耗预测结果。实验结果表明，因果特征子集大小约为过滤式特征子集大小的1/3到1/6，并且使用因果特征子集训练的模型在75%的情况下都取得了最优预测精度。     

Mixture ratio modeling of dynamic systems

Any knowledge extraction relies (possibly implicitly) on a hypothesis about the modelled-data dependence. The extracted knowledge ultimately serves to a decision-making (DM). DM always faces uncertainty and this makes probabilistic modelling adequate. The inspected black-box modeling deals with “universal” approximators of the relevant probabilistic model. Finite mixtures with components in the exponential family are often exploited. Their attractiveness stems from their flexibility, the cluster interpretability of components and the existence of algorithms for processing high-dimensional data streams. They are even used in dynamic cases with mutually dependent data records while regression and auto-regression mixture components serve to the dependence modeling. These dynamic models, however, mostly assume data-independent component weights, that is, memoryless transitions between dynamic mixture components. Such mixtures are not universal approximators of dynamic probabilistic models. Formally, this follows from the fact that the set of finite probabilistic mixtures is not closed with respect to the conditioning, which is the key estimation and predictive operation. The paper overcomes this drawback by using ratios of finite mixtures as universally approximating dynamic parametric models. The paper motivates them, elaborates their approximate Bayesian recursive estimation and reveals their application potential.     

基于C#的飞机疲劳试验实时预警系统设计

为满足飞机疲劳试验对试验数据实时监控的需求,采用C#语言开发了一套飞机疲劳试验实时预警系统。该系统以疲劳试验数据管理系统为基础,计算各测量点、各工况的平均值和标准差,通过多种方法设定预警阈值。该系统界面友好、功能完备,可实现疲劳试验数据的实时监控,能够及时发现异常试验数据,尽早发现结构损伤并采取有效措施,可以大幅降低维修成本、缩短维修周期。     

A machine learning‐based operational control framework for reducing energy consumption of an amine‐based gas sweetening process

This study proposes a data‐driven operational control framework using machine learning‐based predictive modeling with the aim of decreasing the energy consumption of a natural gas sweetening process. This multi‐stage framework is composed of the following steps: (a) a clustering algorithm based on Density‐Based Spatial Clustering of Applications with Noise methodology is implemented to characterize the sampling space of all possible states of the operation and to determine the operational modes of the gas sweetening unit, (b) the lowest steam consumption of each operational mode is selected as a reference for operational control of the gas sweetening process, and (c) a number of high‐accuracy regression models are developed using the Gradient Boosting Machines algorithm for predicting the controlled parameters and output variables. This framework presents an operational control strategy that provides actionable insights about the energy performance of the current operations of the unit and also suggests the potential of energy saving for gas treating plant operators. The ultimate goal is to leverage this data‐driven strategy in order to identify the achievable energy conservation opportunity in such plants. The dataset for this research study consists of 29 817 records that were sampled over the course of 3 years from a gas train in the South Pars Gas Complex. Furthermore, our offline analysis demonstrates that there is a potential of 8% energy saving, equivalent to 5 760 000 Nm³ of natural gas consumption reduction, which can be achieved by mapping the steam consumption states of the unit to the best energy performances predicted by the proposed framework.     

Driving behavior recognition and prediction based on Bayesian model

Since the existing intelligent driving systems are lack of efficiency and accuracy when processing huge number of driving data,a brand new approach of processing driving data was developed to identify and predicate human driving behavior based on Bayesian model.The approach was proposed to take two steps to deduce the specific driving behavior from driving data correspondingly without any supervision,the first step being using Bayesian model segmentation algorithm to divide driving data that inertial sensor collected into near-linear segments with the help of Bayesian model segmentation algorithm,and the second step being using extended LDA model to aggregate those linear segments into specific driving behavior (such as braking,turning,acceleration and coasting).Both offline and online experiments are conducted to verify this approach and it turns out that approach has higher efficiency and recognition accuracy when dealing with numerous driving data.     

A hybrid data envelopment analysis and multi-attribute decision making approach to sustainability assessment

The concept of sustainability consists of three main dimensions: environmental, techno-economic, and social. Measuring the sustainability status of a system or technology is a significant challenge, especially when it needs to consider a large number of attributes in each dimension of sustainability. In this study, we first propose a hybrid approach, involving data envelopment analysis (DEA) and a multi-attribute decision making (MADM) methodologies, for computing an index for each dimension of sustainability, and then we define the overall sustainability index as the mean of the three measured indexes. Towards this end, we define new concepts of efficiency and cross-efficiency of order (p, q) where p and q are the number of inputs and outputs, respectively. For a given (p, q) , we address the problem of finding efficiency of order (p, q) by developing a novel DEA-based selecting method. Finally, we define the sustainability index as a weighted sum of all possible cross-efficiencies of order (p, q) . Form a computational viewpoint, the proposed selecting model significantly decreases the computational burden in comparison with the successive solving of traditional DEA models. A case study of the electricity-generation technologies in the United Kingdom is taken as a real-world example to illustrate the potential application of our method.     

惯性导航设备升沉数据测试系统研究

研制了一套基于STM32单片机与单轴摇摆台的测试系统,利用固定圆心、固定半径条件下的圆内三角形关系,通过拉线传感器采集单轴摇摆台复合运动中的升沉数据;同时,对惯性导航设备输出的升沉数据进行采集与处理。对测试系统输出的升沉数据进行了验证,实验结果表明:该测试系统的测试数据准确合理,可实现对惯性导航设备升沉参数的动态测试,弥补了现有同类测试系统的结构复杂、量程小等不足。     

Clustering adaptive canonical correlations for high-dimensional multi-modal data

Multi-modal canonical correlation analysis (MCCA) is an important joint dimension reduction method and has been widely applied to clustering tasks of multi-modal data. MCCA-based clustering is usually dimension reduction of high-dimensional data followed by clustering of low-dimensional data. However, the two-stage clustering is difficult to ensure the adaptability of dimension reduction and clustering, which will affect the final clustering performance. To solve the issue, we propose a novel clustering adaptive multi-modal canonical correlations (CAMCCs) method, which constructs a unified optimization model of multi-modal correlation learning and clustering. The method not only realizes discriminant learning of correlation projection directions under unsupervised cases, but also is able to directly obtain class labels of multi-modal data. Additionally, the method also realizes out-of-sample extension in class labels. Solutions of CAMCCs are optimized by an iterative way, and we analyze its convergence. Extensive experimental results on various datasets have demonstrated the effectiveness of the method.     

A new DEA common-weight multi-criteria decision-making approach for technology selection

This paper addresses an advanced manufacturing technology selection problem by proposing a new common-weight multi-criteria decision-making (MCDM) approach in the evaluation framework of data envelopment analysis (DEA). We improve existing technology selection models by giving a new mathematical formulation to simplify the calculation process and to ensure its use in more general situations with multiple inputs and multiple outputs. Further, an algorithm is provided to solve the proposed model based on mixed-integer linear programming and dichotomy. Compared with previous approaches for technology selection, our approach brings new contributions. First, it guarantees that only one decision-making unit (DMU) (referring to a technology) can be evaluated as efficient and selected as the best performer while maximising the minimum efficiency among all the DMUs. Second, the number of mixed-integer linear programs to solve is independent of the number of candidates. In addition, it guarantees the uniqueness of the final optimal set of common weights. Two benchmark instances are used to compare the proposed approach with existing ones. A computational experiment with randomly generated instances is further proceeded to show that the proposed approach is more suitable for situations with large datasets.     

Feature selection and classification using support vector machine and decision tree

Breast cancer is one of the human threats which cause morbidity and mortality worldwide. The death rate can be reduced by advanced diagnosis. The objective of this article is to select the reduced number of features the help in diagnosing breast cancer in Wisconsin Diagnostic Breast Cancer (WDBC). This proposed model depicts women who all have no cancer cells or in benign stage later develop into malignant (metastases). Due to the dynamic nature of the big data framework, the proposed method ensures high confidence and low execution time. Moreover, healthcare information growth chases an exponential pattern, and current database systems cannot adequately manage the massive amount of data. So, it is requisite to adopt the “big data” solution for healthcare information.