TSK fuzzy modeling for tool wear condition in turning processes: An experimental study

This paper presents an experimental study for turning process in machining by using Takagi-Sugeno-Kang (TSK) fuzzy modeling to accomplish the integration of multi-sensor information and tool wear information. It generates fuzzy rules directly from the input-output data acquired from sensors, and provides high accuracy and high reliability of the tool wear prediction over a wide range of cutting conditions. The experimental results show its effectiveness and satisfactory comparisons relative to other artificial intelligence methods.     

A Fast Algorithm for Adaptive Prefix Coding

In this paper we present a new algorithm for adaptive prefix coding. Our algorithm encodes a text S of m symbols in O(m) time, i.e., in O(1) amortized time per symbol. The length of the encoded string is bounded above by (H+1)m+O(nlog ² m) bits where n is the alphabet size and H is the entropy. This is the first algorithm that adaptively encodes a text in O(m) time and achieves an almost optimal bound on the encoding length in the worst case. Besides that, our algorithm does not depend on an explicit code tree traversal. A preliminary version of this paper appeared in the Proceedings of the 2006 IEEE International Symposium on Information Theory (ISIT 2006). M. Karpinski’s work partially supported by a DFG grant, Max-Planck Research Prize, and IST grant 14036 (RAND-APX). Y. Nekrich’s work partially supported by IST grant 14036 (RAND-APX).     

Integrating working and enrichment approach learning: a document

Integrating working and learning is seen as a desirable alternative to traditional training regimes. An integrated approach to working and learning is more gradual, contextual, problem oriented and cost effective. Knowledge technology aims to catalyze workplace learning, but requires the right organisational culture and brings additional costs regarding the articulation, representation and transmission of knowledge. Our approach reduces these costs by making articulation a natural part of collaborative working, designing for both informal and formal knowledge, and facilitating the transition of socially situated knowledge through enriched documents. To be successful, our approach has certain prequesites concerning organisational culture, and the nature of shared documents, organisational knowledge and work activities.     

Building ensemble classifiers using belief functions and OWA operators

A pervasive task in many forms of human activity is classification. Recent interest in the classification process has focused on ensemble classifier systems. These types of systems are based on a paradigm of combining the outputs of a number of individual classifiers. In this paper we propose a new approach for obtaining the final output of ensemble classifiers. The method presented here uses the Dempster–Shafer concept of belief functions to represent the confidence in the outputs of the individual classifiers. The combing of the outputs of the individual classifiers is based on an aggregation process which can be seen as a fusion of the Dempster rule of combination with a generalized form of OWA operator. The use of the OWA operator provides an added degree of flexibility in expressing the way the aggregation of the individual classifiers is performed.     

First-order syntactic characterizations of minimal entailment,domain-minimal entailment,and Herbrand entailment

This paper investigates a consistent versioncwa _s of Reiter's closed-world assumptioncwa. It provides a syntactic characterization of minimal entailment ? _min : for every ∨-sentence ? and for every ∨-theory Σ, $$\varphi \varepsilon cwa_S \left( \sum \right)iff\sum \vdash _{min} \varphi .$$ A version of this characterization remains valid if not all relations are subject to minimization. These two characterizations do not use the domain-closure axiom nor the unique-names assumptionuna, although they may be easily modified to ones that incorporateuna. A similar result for Herbrand entailment ? _Her , by means of generalized domain-closure axiomdca _s , is provided: for every ∨-sentence ? and every ∨-theory Σ, $$\varphi \varepsilon dca_S \left( \sum \right)iff\sum \vdash _{Her} \varphi .$$ Finally, a syntactic characterization of domain-minimal entailment ? _dom in terms of a versionmda _s of minimal-domain assumption is shown: for every ∨-sentence ? and for every ∨-theory Σ, $$\varphi \varepsilon mda_S \left( \sum \right)iff\sum \vdash _{dom} \varphi .$$ The proving power of these entailments is then evaluated. In particular, it is shown that (1) neither ? _min nor its versions are strong enough to derive positive sentences from Σ unless they are first-order provable from Σ however, a double application of ? _min has enough power to derive such positive sentences; (2) ? _Her has the strength of infinitary rule of inference but cannot derive existential nor quantifier-free sentences from Σ unless they are first-order provable from Σ (3) ? _Her and ? _dom can derive from Σ certain positive facts about = which are otherwise unprovable from Σ and (4) ? _dom cannot derive from Σ sentences without positive occurrences of = unless they are first-order provable from Σ. Moreover, the paper relatescwa _s to Reiter'scwa and to Minker's generalized closed-world assumptionGCWA and its extension.     

Combining classifiers under probabilistic models: experimental comparative analysis of methods

This work will present a review of the concept of classifier combination based on the combined discriminant function. We will present a Bayesian approach, in which the discriminant function assumes the role of the posterior probability. We will propose a probabilistic interpretation of expert rules and conditions of knowledge consistency for expert rules and learning sets. We will suggest how to measure the quality of learning materials and we will use the measure mentioned above for an algorithm that eliminates contradictions in the rule set. In this work several recognition algorithms will be described, based on either: (i) pure rules, or; (ii) rules together with learning sets. Furthermore, the original concept of information unification, which enables the formation of rules on the basis of learning set or learning set on the basis of rules will be proposed. The obtained conclusions will serve as a spring‐board for the formulation of new project guidelines for this type of decision‐making system. At the end, experimental results of the proposed algorithms will be presented, both from computer generated data and for a real problem from the medical diagnostics field.     

Crack and flaw identification in elastodynamics using Kalman filter techniques

Filter-driven optimization based on the extended Kalman filter concept is used here for the numerical solution of crack and flaw identification problems in elastodynamics. The mechanical modeling of the studied two-dimensional problem, which includes the effect of unilateral contact along the sides of the crack, is done with the help of the boundary element method. The effect of various dynamical test loads and the applicability of this method for crack and defect identification in disks are investigated. The work has been supported by the German Research Foundation (DFG). Partial support has been provided by a Greek-German Research Cooperation Grant (IKYDA2001). This support is greatfully acknowledged.     

Defect identification in 3-D elastostatics using a genetic algorithm

An inverse problem in engineering mechanics is considered where the position and the geometry of three-dimensional, ellipsoidal defects are identified by using measurements of the mechanical response under static loading on the external surface of the structure. The problem is solved by appropriate combination of genetic optimization (GO) and boundary element method (BEM) and following previously published two-dimensional problems. The three-dimensional case presents some additional difficulties. Furthermore, the function of several genetic operators and the effect of the parameters of genetic optimization on the efficiency of the solution has been numerically examined.