基于宽度学习的注塑产品质量预测方法

在注塑成型工业中,产品质量自动监测一直是注塑工业智能化发展的核心问题。高品质和大规模的产品质量数据采集成本高昂,导致数据样本量少、不同类别样本数据不平衡,为注塑产品质量预测提出了更高的挑战。为此,该文提出一种基于宽度学习方法的注塑产品质量预测模型,以产品的3维尺寸为预测目标,在普通的宽度学习系统(BLS)中加入最小p范数来改进得到模型p范数宽度学习系统(pN-BLS),解决小样本和不平衡数据的问题,提高模型对离群点的检测性能。在第4届工业大数据竞赛任务2《注塑成型工艺的虚拟量测和调机优化》数据集中,将192个参数特征与预测目标进行相关分析,提取相关性高的基础特征17个,衍生特征4个和调机参数2个作为模型的输入。将16600条数据平均分为训练集和测试集各8300条,与支持向量机 (SVM)、最近邻算法 (KNN)、多层感知机 (MLP)和BLS进行对比实验,实验结果显示pN-BLS具有更快速和更准确的预测效果。在实际缺陷检测应用中,pN-BLS能更准确地预测异常数据,具有更高的鲁棒性。     

Evolutionary model of the growth and size of firms

The key idea of this model is that firms are the result of an evolutionary process. Based on demand and supply considerations the evolutionary model presented here derives explicitly Gibrat’s law of proportionate effects as the result of the competition between products. Applying a preferential attachment mechanism for firms, the theory allows to establish the size distribution of products and firms. Also established are the growth rate and price distribution of consumer goods. Taking into account the characteristic property of human activities to occur in bursts, the model allows also an explanation of the size–variance relationship of the growth rate distribution of products and firms. Further the product life cycle, the learning (experience) curve and the market size in terms of the mean number of firms that can survive in a market are derived. The model also suggests the existence of an invariant of a market as the ratio of total profit to total revenue. The relationship between a neo-classic and an evolutionary view of a market is discussed. The comparison with empirical investigations suggests that the theory is able to describe the main stylized facts concerning the size and growth of firms.     

On a learning precedence graph concept for the automotive industry

Assembly line balancing problems (ALBP) consist in assigning the total workload for manufacturing a product to stations of an assembly line as typically applied in automotive industry. The assignment of tasks to stations is due to restrictions which can be expressed in a precedence graph. However, (automotive) manufacturers usually do not have sufficient information on their precedence graphs. As a consequence, the elaborate solution procedures for different versions of ALBP developed by more than 50 years of intensive research are often not applicable in practice.Unfortunately, the known approaches for precedence graph generation are not suitable for the conditions in the automotive industry. Therefore, we describe a new graph generation approach that is based on learning from past feasible production sequences and forms a sufficient precedence graph that guarantees feasible line balances. Computational experiments indicate that the proposed procedure is able to approximate the real precedence graph sufficiently well to detect optimal or nearly optimal solutions for a well-known benchmark data set. Even for additional large instances with up to 1,000 tasks, considerable improvements of line balances are possible. Thus, the new approach seems to be a major step to close the gap between theoretical line balancing research and practice of assembly line planning.     

Fractional commensurate, composition, and adding schemes: Learning trajectories of Jason and Laura: Grade 5

A case study of two 5th-Grade children, Jason and Laura, is presented who participated in the teaching experiment, Children’s Construction of the Rational Numbers of Arithmetic. The case study begins on the 29th of November of their 5th-Grade in school and ends on the 5th of April of the same school year. Two basic problems were of interest in the case study. The first was to provide an analysis of the concepts and operations that are involved in the construction of three fractional schemes: a commensurate fractional scheme, a fractional composition scheme, and a fractional adding scheme. The second was to provide an analysis of the contribution of interactive mathematical activity in the construction of these schemes. The phrase, “commensurate factional scheme” refers to the concepts and operations that are involved in transforming a given fraction into another fraction that are both measures of an identical quantity. Likewise, “fractional composition scheme” refers to the concepts and operations that are involved in finding how much, say, 1/3 of 1/4 of a quantity is of the whole quantity, and “fractional adding scheme” refers to the concepts and operations involved in finding how much, say, 1/3 of a quantity joined to 1/4 of a quantity is of the whole quantity. Critical protocols were abstracted from the teaching episodes with the two children that illustrate what is meant by the schemes, changes in the children’s concepts and operations, and the interactive mathematical activity that was involved. The body of the case study consists of an on-going analysis of the children’s interactive mathematical activity and changes in that activity. The last section of the case study consists of an analysis of the constitutive aspects of the children’s constructive activity, including the role of social interaction and nonverbal interactions of the children with each other and with the computer software we used in teaching the children.     

深度融合云环境自身架构的东西向攻击检测技术研究

随着云和容器等虚拟化技术的不断扩张,云、数据中心和企业网中的东西向流量呈快速增长趋势。如果不采集虚拟网络流量,用户80%的网络流量将呈现“黑盒”状态,无法对云平台内东西向量进行安全管控,在等保2.0云计算安全扩展中明确要求针对虚机之间、宿主机与虚机间的流量需要进行检测和异常告警。本文通过对业内现有东西向流量检测方案进行分析,研究并提出一种与云平台自身架构深度融合且非侵入式的流量获取方式,基于大数据分析和机器学习算法提升攻击检测能力的整体方案,此外还研究了融合过程中与云平台业务自动适配的问题。