A weighted LS-SVM approach for the identification of a class of nonlinear inverse systems

In this paper, a weighted least square support vector machine algorithm for identification is proposed based on the T-S model. The method adopts fuzzy c-means clustering to identify the structure. Based on clustering, the original input/output space is divided into several subspaces and submodels are identified by least square support vector machine (LS-SVM). Then, a regression model is constructed by combining these submodels with a weighted mechanism. Furthermore we adopt the method to identify a class of inverse systems with immeasurable state variables. In the process of identification, an allied inverse system is constructed to obtain enough information for modeling. Simulation experiments show that the proposed method can identify the nonlinear allied inverse system effectively and provides satisfactory accuracy and good generalization. Supported by the National Natural Science Foundation of China (Grant No. 60874013) and the Doctoral Project of the Ministry of Education of China (Grant No. 20070286001)     

TOWARD A FUZZY MODEL OF POLISH INTERNET MORTGAGE MARKET

This article presents our continuing efforts to develop a model of the Internet-based mortgage market in Poland. It sums up modeling research done over the last 4 years, also showing the stages and the process of building and verifying the proposed model consisting of three submodels. The submodels are described for three market situations; that is, a stable market, financial crisis, and economic boom. The proposed model variables are analyzed statistically and the submodels are verified. The article presents the most current stage of model development and concludes with suggestions for further research in this area leading to the ultimate application of the proposed model in a real-life prognosis process.     

Mechanized material handling systems design and routing

Tinker AFB (TAFB) is the prime Air Force rework facility for jet engine parts that support standard engines. The wear and tear on the engine parts of the various aircraft currently produces about 10 million individual requirements annually for part inspection and subsequent discard or rework. Past practices have resulted in extremely long flow times for rework and a burgeoning requirement for conveyor capacity.

A new concept, called the Modular Repair Center (MRC), is being implemented to correct the problem. The MRC's have most of the processes necessary to inspect and repair components of a given engine assembly. They have the advantage of reducing the number of long moves, giving more accountability for quality and better tracking. A large network model of TAFB facility has been developed to determine the required work-in-process storage, conveyor capacity, shortest routes for parts to flow from one MRC to another and the maximum flow along each conveyor section. The model was analyzed using Floyd's shortest route algorithm. It resulted in reducing the “bottlenecks” on certain conveyor sections by rerouting some of the flow along low density traffic links and thus reducing the need for extra conveyor capacity. Simple formulas were designed to estimate the number of pallets generated by disassembly and each MRC to provide the demand requirements for conveyor capacity.     

流程工业能耗系统多维子模型及其开放性集成框架

流程工业组成因素多,运行结构复杂,针对单一模型无法合理、全面地描述其能源消耗系统各因素及其关联关系的问题,提出了流程工业能耗系统多维子模型及其集成化构建方法。首先,综合设备、能量、信息、人员等多方面因素,从静态结构描述、能源消耗的动态行为以及企业能耗系统运作的目的性3个层面建立流程工业能耗系统多维子模型;然后,分析了不同视角子模型关联关系;最后,详细研究了能耗系统多维模型的开放性集成框架,从而为流程工业能耗系统建立集信息流、能源流和物料流为一体,并同时反映内、外部关键因素的多维集成化模型提供了方法支持。这种内外关联、高度集成的流程工业能耗系统模型的构建将为流程工业综合性能效评估奠定基础。     

Reconfiguration of assembly systems: From conveyor assembly line to serus

Confronted with the dynamic and complex market environments, the traditional conveyor assembly line can no longer meet customers’ demands effectively. The way of reconfiguring conveyor assembly line to a more flexible manufacturing system has been attracting considerable attention both in the academics and production practices. Seru system, also called assembly cell system, is regarded as one of the most successful innovations of manufacturing system in reconfiguring conveyor assembly line. Such a manufacturing system merges considerable flexibility of job shops and high efficiency of conveyor assembly lines to some extent. In this paper, we investigate the problem of how to reconfigure conveyor assembly line to serus. A comprehensive mathematical model incorporating two issues of how many serus should be established and how many workers should be assigned to each seru is developed. Then the model is investigated by an industrial case and compared to Kaku's model with respect to the selected plan. The computation results validate that the proposed model is more suitable to analyze the reconfiguration problems from conveyor assembly line to serus.     

Mechanized material handling systems design and routing

Tinker AFB (TAFB) is the prime Air Force rework facility for jet engine parts that support standard engines. The wear and tear on the engine parts of the various aircraft currently produces about 10 million individual requirements annually for part inspection and subsequent discard or rework. Past practices have resulted in extremely long flow times for rework and a burgeoning requirement for conveyor capacity.A new concept, called the Modular Repair Center (MRC), is being implemented to correct the problem. The MRC's have most of the processes necessary to inspect and repair components of a given engine assembly. They have the advantage of reducing the number of long moves, giving more accountability for quality and better tracking. A large network model of TAFB facility has been developed to determine the required work-in-process storage, conveyor capacity, shortest routes for parts to flow from one MRC to another and the maximum flow along each conveyor section. The model was analyzed using Floyd's shortest route algorithm. It resulted in reducing the “bottlenecks” on certain conveyor sections by rerouting some of the flow along low density traffic links and thus reducing the need for extra converyor capacity. Simple formulas were designed to estimate the number of pallets generated by disassembly and each MRC to provide the demand requirements for conveyor capacity.     

基于AVR单片机的断链失速智能保护研究

阐述了输送机断链、失速保护装置的工作原理,详细介绍了以AVR ATmega16单片机为核心构成的保护装置的软硬件、控制算法和可靠性设计方案。该智能保护装置实现了对输送机断链、失速故障的智能可靠保护,有力保证了输送机安全、可靠运行。     

组织决策过程的随机有色Petr i网模型

随机有色Petri网可为组织决策过程提供描述框架和分析手段。基于SCPN的计算机图形建模仿真工具研究组织设计变量对组织决策效率的影响,这些变量包括决策人之间的协调结构、决策人偏好、决策方法、决策目标数、备选方案数等。仿真实例结果表明,决策人之间的协调结构对决策效率有明显的影响。     

Task response time for real-time distributed systems with resourcecontentions

An analytic model is proposed for estimating task response times in distributed systems with resource contentions. The model consists of two submodels. The first submodel is an extended queuing network model used for approximating module response times. This submodel is solved by a decomposition technique which reduces the computational complexity by two to three orders of magnitude when compared with a direct approach. The second submodel is a weighted control-flow graph model from which task response time can be obtained by aggregating module response time in accordance with the precedence relationships. Task response times estimated by the analytic model compare closely with simulation results. It is shown that resource contention delays depend on the availability of resources as well as on the invocation rates and response times of the modules that use the resources. The model can be used to study the tradeoffs among module assignments, scheduling policies, interprocessor communications, and resource contentions in distributed processing systems     

Paint batching problem on M-to-1 conveyor systems

An M-to-1 conveyor system consists of multiple upstream conveyors and a single downstream conveyor. In this paper, we investigate the paint batching problem on M-to-1 conveyor systems with the objective of minimizing setup costs. Our research is motivated by a vehicle re-sequencing problem at a major Korean automotive manufacturer. Setup costs are incurred when two consecutive jobs in the downstream conveyor do not share the same feature. Re-sequencing flexibility is limited by the precedence relationship among jobs in the upstream conveyors. First, we develop a mixed integer linear programming model and propose an efficient dynamic programming (DP) algorithm for a 2-to-1 conveyor system. However, because the suggested DP cannot guarantee optimality in general settings, we propose two efficient genetic algorithms (GAs) to find near optimal solutions. Specifically, we design the reordering operation for making offspring to satisfy the precedence condition. We show that the proposed GAs perform prominently with respect to optimality gap and computation time; thus, they are amenable to environments where solutions must be obtained within tight time constraints.     

Training neurofuzzy systems

A neurofuzzy system combines the positive attributes of a neural network and a fuzzy system by providing a transparent framework for representing linguistic rules with well defined modelling and learning characteristics. Unfortunately, their application is limited to problems involving a small number of input variables by the curse of dimensionality where the the size of the rule base and the training set increase as an exponential function of the input dimension. The curse can be alleviated by a number of approaches but one which has recently received much attention is the exploitation of redundancy. Many functions can be adequately approximated by an additive model whose output is a sum over several smaller dimensional subrnodels. This technique is called global partitioning and the aim of an algorithm designed to construct the approximation is to automatically determine the number of submodels and the subset of input variables for each submodel. The construction algorithm is an iterative process where each iteration must identify a set of candidate refinements and evaluate the associated candidate models. This leads naturally to the problem of how to train the candidate models and the approach taken depends on whether they contain one or multiple submodels.     

Identification of piecewise affine systems via mixed-integer programming

This paper addresses the problem of identification of hybrid dynamical systems, by focusing the attention on hinging hyperplanes and Wiener piecewise affine autoregressive exogenous models, in which the regressor space is partitioned into polyhedra with affine submodels for each polyhedron. In particular, we provide algorithms based on mixed-integer linear or quadratic programming which are guaranteed to converge to a global optimum. For the special case where the estimation data only seldom switches between the different submodels, we also suggest a way of trading off between optimality and complexity by using a change detection approach.     

Modeling and analysis of closed loop manufacturing systems using parameter coupling

Closed loop manufacturing systems (CLMSs) with recirculating material handling devices are extensively used in various industrial environments. The performance of such systems is impacted by many factors such as the total capacity of pallets, the actual number of pallets in the system, the machine reliability and processing time, the pallet index speed, and the positions of loading/unloading points. These factors make the accurate analysis and optimization of complex CLMSs very difficult and challenging. This paper presents a new parameter coupling technique to model and analyze a wide range of CLMSs. It is an enhancement based on the existing open production line analysis with unreliable assembly machines and finite buffers. Virtual assembly machines are introduced to represent the specific phenomena of CLMSs such as the recirculation of empty pallets and the sharing of conveyor space. Two types of parameter coupling patterns, the machine parameter coupling, and the buffer capacity coupling, are introduced to reflect the characteristics of the CLMSs. The parameter coupling technique is simple and effective for analyzing a broad range of CLMSs. Comparisons between this analytic method and simulation experiments demonstrate that the proposed parameter coupling technique is fast, accurate and robust.     

An inexact fuzzy-queue programming model for environmental systems planning

In this study, an inexact fuzzy-queue programming (IFQP) model is developed for municipal solid waste (MSW) management under uncertainty, where fuzzy queue (FQ) model is introduced into the interval-fuzzy linear programming (IFLP) framework. The proposed IFQP model can not only handle uncertainties that are presented in terms of fuzzy sets and discrete intervals, but also reflect the influence of FQ in decision-making problems. Moreover, it can help quantify the satisfaction degrees of the system cost and constraints under uncertainty. It improves upon the existing optimization model with advantages in data availability, uncertainty reflection and result analysis. In its solution process, the model is transformed into two deterministic submodels based on an interactive algorithm. Interval solutions which are stable in the decision space can be obtained by solving the two submodels sequentially. The developed model is applied to a case of long-term planning of MSW management systems to demonstrate its applicability. It can effectively reflect dynamic, interactive, and uncertain characteristics of municipal waste management systems. The decision variable solutions provide optimal schemes for waste flow allocation corresponding to lower and upper bounds of system cost. The results indicate that reasonable solutions have been generated. They are helpful for supporting: (i) decision making or justification of waste-allocation patterns of MSW management system, and (ii) in-depth analysis of tradeoffs among system cost, satisfaction degree, and environmental constraint under uncertainty.     

A general model for the heavy metal pollution of aquatic ecosystems: Model development

It is demonstrated how it is possible to set up a heavy metal model for an aquatic ecosystem in general. It does not imply that the same model can be applied for every case study. It is - as always - necessary to make modifications from case to case. However, by use of 5 submodels the variations from case to case are minor in the submodels, while the major question becomes which submodels to select from case to case. The 5 submodels are: a distribution model, a model of metal speciation, a model for release of metal from sediment, a model related to the metal concentrations in biota and an effect model. It is discussed which components should comprise each submodel and which modifications it should be expected to make in different cases.     

一种随机着色Petri网及模型的性能分析

针对随机Petri网（SPN）在系统性能分析时,其状态空间随着系统规模增大而指数性增长,造成求解稳定状态概率的复杂性的不足,提出了一种随机着色Petri网（SCPN）。分析了它的有界性和可达性,证明了它同构于一个一维连续时间的马尔可夫链;同时,也分析了随机着色Petri网用于建模和系统性能定量分析的方法。     

基于Flink的分布式在线集成学习框架研究

在大数据环境背景下,传统机器学习算法多采用单机离线训练的方式,显然已经无法适应持续增长的大规模流式数据的变化。针对该问题,提出一种基于Flink平台的分布式在线集成学习算法。该方法基于Flink分布式计算框架,首先通过数据并行的方式对在线学习算法进行分布式在线训练;然后将训练出的多个子模型通过随机梯度下降算法进行模型的动态权重分配,实现对多个子模型的结果聚合;与此同时,对于训练效果不好的模型利用其样本进行在线更新;最后通过单机与集群环境在不同数据集上做实验对比分析。实验结果表明,在线学习算法结合Flink框架的分布式集成训练,能达到集中训练方式下的性能,同时大大提高了训练的时间效率。     

Blocking in closed-loop conveyor systems connected in series with discrete and deterministic material flow

We consider m closed-loop conveyor systems each having a single loading and a single unloading station. The output flow of products (units) from the i^th conveyor system is the input flow to i+l^th conveyor system (i=1,2,..,m-1). Each conveyor system moves continuously but the speed rates are not necessarily the same. Products to be transported or temporarily stored on the conveyor systems are identical and they are placed on the conveyor along their length. Interarrival times of products to the first conveyor system as well as service times of the servers at the unloading stations are all known and deterministic. An arriving unit at the loading station of any conveyor system is blocked and lost to the system when it interferes with the recirculating products of that conveyor system.

An algorithm is developed to measure the effects of recirculation times of products, and other parameters of the system such as the ratios between product length^* and conveyor lengths, and arrival/service rates on the performance of the system.     

The reliability of life-critical computer systems

Summary In order to aid the designers of life-critical, fault-tolerant computing systems, accurate and efficient methods for reliability prediction are needed. The accuracy requirement implies the need to model the system in great detail, and hence the need to address the problems of large state space, non-exponential distributions, and error analysis. The efficiency requirement implies the need for new model solution techniques, in particular the use of decomposition/aggregation in the context of a hybrid model. We describe a model for reliability prediction which meets both requirements. Specifically, our model is partitioned into fault occurrence and fault/error handling submodels, which are represented by non-homogeneous Markov processes and extended stochastic Petri nets, respectively. The overall aggregated model is a stochastic process that is solved by numerical techniques. Methods to analyze the effects of variations in input parameters on the resulting reliability predictions are also provided.This work was supported in part by NASA grant NAG 1-70 and by an equipment grant from the Concurrent Computer Corp     

成套机电设备培根制作机的PLC电气控制系统的应用设计研究

针对培根制作工艺落后的问题,为了适应新式培根制作工艺的要求,设计制作了一套培根制作机。培根制作机是一种自动化程度要求较高的成套机电设备,由传送带、穿刺机、翻身机和滚肉机四大部分组成,用PLC技术设计培根制作机的控制电路系统,实现了对培根制作机新工序的自动控制。研究结果表明,该培根制作机适应了自动化控制的要求,在实际使用中性能稳定,操作方便,生产效率高,取得了很好的经济效益。