数控机床故障概率计算方法的研究

本文提出了计算数控机床各种常见故障发生概率的一般方法,以此来确定故障发生的各种可能性,从而为提高数控机床可靠性提供理论依据,并通过示例证明其有效性。     

基于SVM的人口预测

为了提高人口预测的精度,建立了基于支持向量机的预测模型,并采用此模型对实测数据进行了预测。与FNN组合预测的预测结果相比,建立的新型人口预测模型具有更好的预测精度。     

利用互联网大数据预测季度GDP增速的方法研究

大数据为季度GDP走势预测创新研究带来重要突破口。本文利用百度等网站的互联网大数据,基于代表性高维数据机器学习(和深度学习)模型,对我国2011-2018年季度GDP增速深入进行预测分析。研究发现,对模型中的随机干扰因素作出一定分布的统计假设,有助于降低预测误差,任由模型通过大量数据机械地学习和完善并不总是有利于模型预测能力的提升;采用对解释变量集添加惩罚约束的方法,可以有效地处理互联网大数据维度较高的棘手问题;预测季度GDP增速的最优大数据解释变量集的稳定性较高。     

Comparison of Support Vector Machines to Other Classifiers Using Gene Expression Data

The efficiency of independent controling variates in simulation of activity networks will be evaluated in the sense of mean square deviation of the estimator from the true value. The given result is asymptotic     

Measuring Time Series Predictability Using Support Vector Regression

Most studies involving statistical time series analysis rely on assumptions of linearity, which by its simplicity facilitates parameter interpretation and estimation. However, the linearity assumption may be too restrictive for many practical applications. The implementation of nonlinear models in time series analysis involves the estimation of a large set of parameters, frequently leading to overfitting problems. In this article, a predictability coefficient is estimated using a combination of nonlinear autoregressive models and the use of support vector regression in this model is explored. We illustrate the usefulness and interpretability of results by using electroencephalographic records of an epileptic patient.     

Online learning and forecast combination in unbalanced panels

This article evaluates the performance of a few newly proposed online forecast combination algorithms and compares them with some of the existing ones including the simple average and that of Bates and Granger (1969). We derive asymptotic results for the new algorithms that justify certain established approaches to forecast combination including trimming, clustering, weighting, and shrinkage. We also show that when implemented on unbalanced panels, different combination algorithms implicitly impute missing data differently, so that the performance of the resulting combined forecasts are not comparable. After explicitly imputing the missing observations in the U.S. Survey of Professional Forecasters (SPF) over 1968 IV-2013 I, we find that the equally weighted average continues to be hard to beat, but the new algorithms can potentially deliver superior performance at shorter horizons, especially during periods of volatility clustering and structural breaks.     

"机械原理"课堂教学妙语的运用

本文以笔者在"机械原理"课堂教学中积累的一些课堂教学妙语为例,说明了课堂教学妙语在活跃课堂气氛、剖析重难点、实现育人功能等方面的作用,指出课堂教学妙语的形成是教师广博的知识积累、长期的细心采撷、幽默的语言习惯和高超的语言驾御能力以及教学内容、师生互动和课堂情景触发的结果。     

Inductive,Evolutionary, and Neural Computing Techniques for Discrimination: A Comparative Study*

This paper provides a comparative study of machine learning techniques for two-group discrimination. Simulated data is used to examine how the different learning techniques perform with respect to certain data distribution characteristics. Both linear and nonlinear discrimination methods are considered. The data has been previously used in the comparative evaluation of a number of techniques and helps relate our findings across a range of discrimination techniques.     

Lot Splitting in Stochastic Flow Shop and Job Shop Environments*

In recent years many firms have been implementing small lot size production. Lot splitting breaks large orders into smaller transfer lots and offers the ability to move parts more quickly through the production process. This paper extends the deterministic studies by investigating various lot splitting policies in both stochastic job shop and stochastic flow shop settings using performance measures of mean flow time and the standard deviation of flow time. Using a computer simulation experiment, we found that in stochastic dynamic job shops, the number of lot splits is more important than the exact form of splitting. However, when optimal job sizes are determined for each scenario, we found a few circumstances where the implementation of a small initial split, called a “flag,” can provide measurable improvement in flow time performance. Interestingly, the vast majority of previous research indicates that methods other than equal lot splitting typically improves makespan performance. The earlier research, however, has been set in the static, deterministic flow shop environment. Thus, our results are of practical interest since they show that the specific method of lot splitting is important in only a small set of realistic environments while the choice of an appropriate number of splits is typically more important.     

An approachto the machine layout problem in a cellular manufacturing environment

The model of the machine layout problem MLP in a cellular manufacturing environment attains additional dimensions as it should satisfy the qualitative interconnections between the machines and the location restrictions of an existing factory environment. A new MLP model based on merging pre-emptive goal programming and simulated annealing has been developed for machine layout in cells. This model seeks to find feasible solutions by addressing practical issues of implementation as well as reducing the total travel distances for parts between machines. The new model can also be applied to facility layout problems FLP . The computational work is demonstrated by applying the model to problems of both quantitative and qualitative types, and has produced encouraging results. This model is particularly attractive for layout problems with realistic goals and constraints. To show the performance of the model in handling real-world problems, a practical example has been introduced and solved using the proposed model.