The volume 2005

Table of Contents

Teodor DUMITRIU: Development of a Simulink® toolbox for friction control design and compensation
Abstract  Full Text

Diana STEFANESCU, Emilia PECHEANU, Adina COCU: Pedagogical knowledge model based on conceptual graphs and ontology
Abstract  Full Text

Luminita DUMITRIU, Marian CRACIUN, Cristina SEGAL, Lucian GEORGESCU: Descriptive mining for the QSAR problem
Abstract  Full Text

Marian GAICEANU: Active power compensator of the current harmonics based on the instantaneous power theory
Abstract  Full Text

Adina COCU, Diana STEFANESCU: Uncertainty management using bayesian networks in student knowledge diagnosis
Abstract  Full Text

Cosmin Danut BOCANIALA, José Sá da COSTA and Severin BUMBARU: On the applicability of state-of-the-art fault diagnosis methodologies to simple and complex systems
Abstract  Full Text

Cornelia TUDORIE, Severin BUMBARU, Luminita DUMITRIU: Relative aggregation operator in database fuzzy querying
Abstract  Full Text

Cornelia TUDORIE, Cristian NEACSU, Ionel MANOLACHE: Fuzzy queries in romanian language an intelligent interface
Abstract  Full Text

Dan MUNTEANU, Severin BUMBARU: A survey of text clustering techniques used for web mining
Abstract  Full Text

Dan MUNTEANU, Severin BUMBARU: Classification process in a text document recommender system
Abstract  Full Text

Ion VONCILA, Maria ILIUTA, Mariana DUMITRESCU, Gelu GURGUIATU: The influence of electric transformers' specific parameters modification on power distributions and on technological appropriate consumption (tac) within the energetic system
Abstract  Full Text

Ion VONCILA, Nicolae BADEA, Dorel POPA: The distribution of the electricfield and of the current density in case of electrocorrosion countermeasures using thesacrifice anodes technique
Abstract  Full Text

Viorel DUGAN, Razvan SOLEA, Clara IONESCU and Daniela GHINITA: Three fuzzy reasoning models as a decision suport aid, to find an electrical energy tariff
Abstract  Full Text

 

 

Teodor DUMITRIU: Development of a Simulink® toolbox for friction control design and compensation


Abstract: This paper focuses on the development of a MATLAB/Simulink® library for servo-systems with friction as a part of a new simulation platform dedicated to model, analysis and control design of friction. It is well known that friction is a very important process for the control engineering both for high-precision servo – mechanisms and simple pneumatic and hydraulic systems. Highly nonlinear process, friction may result in steady state errors, limit cycles and poor performance. It is therefore important for control engineering to understand friction phenomena and to know how to deal with them. Moreover, a reliable library of friction models that captures the friction behavior provides an important tool in order to investigate by analysis and simulation the properties of friction that are relevant to control design.

Keywords: Stiction, Stribeck effect, KFM, pre-sliding displacement, stick-slip motion, Model-based friction compensation;

 

 

 

Diana STEFANESCU, Emilia PECHEANU, Adina COCU: Pedagogical knowledge model based on conceptual graphs and ontology


Abstract: Intelligent educational systems are knowledge-based systems (KBS) they can be developed by a generic knowledge-based system development methodology. In this paper, we present an ontology-based approach for formalizing different knowledge types. The formalism is based upon conceptual graphs. A priority concern to all research work in adaptive education is that of finding an appropriate representation for pedagogical knowledge. For implementation, we use the CoGITaNT environment (Conceptual Graphs Integrated Tools allowing Nested Typed graphs), a library of C++ classes (open-sources, developed by LIRMM CNRS, France) allowing the development of applications based on the CG knowledge representation scheme.

Keywords: Intelligent educational systems, Pedagogical knowledge modeling, Conceptual graphs.

 

 

 

Luminita DUMITRIU, Marian CRACIUN, Cristina SEGAL, Lucian GEORGESCU: Descriptive mining for the QSAR problem


Abstract: There are several approaches in trying to solve the Quantitative Structure-Activity (QSAR) problem. These approaches are based either on statistical methods or on predictive data mining using neural networks. Among the statistical methods, one should consider regression analysis, pattern recognition (such as cluster analysis, factor analysis and principal components analysis) or partial least squares. These approaches have a low explanatory capability or non at all. This paper attempts to establish a new approach in solving QSSAR problems using descriptive data mining. This way, the relationship between the chemical properties and the activity of a substance would be comprehensibly modeled.

Keywords: Quantitative Structure-Activity Relationship, data mining, association rules, classification.

 

 

 

Marian GAICEANU: Active power compensator of the current harmonics based on the instantaneous power theory


Abstract: The quality of the electrical current becomes a major concern. The proliferation of the power electronic converters, which are used extensively to control electrical apparatus in industrial and commercial applications (dc and ac variable speed motor drives, induction furnaces, power line conditioners, and industrial power supplies), is at the origin of the AC current distribution network pollution and the reactive power demand. These power electronic converters typically draw non-sinusoidal currents from the utility, causing interference with adjacent sensitive loads and limit the utilization of the available electrical supply. The quality of the electrical current thus becomes a significant concern for the distributors of energy and their customers. Recent progress as regards technology of the power electronics brings a capacity of compensation and correction of the harmonic distortion generated by the nonlinear loads. In this paper a parallel active filter prototype capable of reducing the total harmonic distortion in the supply for most current source or adjustable speed drive type loads is presented. A 33 kVA active power filter was developed for harmonic and reactive power compensation based on the instantaneous power theory. The active filter configuration requires the measurement of both the load and filter currents. Experimental results from a prototype active power filter confirm the suitability of the proposed approach. The actual 33kVA prototype converter has been built and tested in the SIEI S.p.A. (Italy) laboratory under the Marie Curie Post Doctoral research. The active power compensator is controlled by a high performance DSP platform, resulting in the following active filter features: source current reduction up to the 25th harmonic, 10% THD achievable for current source type loads, efficiency above 97%, does not cause resonance with other loads, operation in the presence of unbalanced loads, reactive power and harmonics compensation capability, load current and source current detection capability.

Keywords: Active Power Filter, instantaneous power theory, harmonics compensation

 

 

 

Adina COCU, Diana STEFANESCU: Uncertainty management using bayesian networks in student knowledge diagnosis


Abstract: In intelligent tutoring systems, student or user modeling implies dealing with imperfect and uncertain knowledge. One of the artificial intelligence techniques used for uncertainty management is that of Bayesian networks. This paradigm is recommended in the situation when exist dependencies between data and qualitative information about these data. In this work we present a student knowledge diagnosis model based on representation with Bayesian networks. The educational system incorporate a multimedia interface for accomplishes the testing tools. The results of testing sessions are represented and interpreted with probability theory in order to ensure an adapted support for the student. The aims of the computer assisted application that contains this diagnose module are to support the student in personalized learning process and errors explanation.

Keywords: numerical uncertainty management, intelligent computer based learning system, Bayesian network, probability theory, student diagnosis.

 

 

 

Cosmin Danut BOCANIALA, José Sá da COSTA and Severin BUMBARU: On the applicability of state-of-the-art fault diagnosis methodologies to simple and complex systems


Abstract: This paper performs an analysis on the applicability of state-of-the-art fault diagnosis methodologies to both simple and complex systems. Here, a complex system represents a system whose global behavior, which emerges from the interactions between its usually large number of basic components, is difficult to accurately describe via a model. First, the basic notions used in fault detection and isolation, are introduced. Then, short reviews are given for the main quantitative methods, qualitative reasoning based methods, and soft computing approaches. The next section is dedicated to recent distributed approaches to fault diagnosis of complex systems. Finally, some conclusions are given.

Keywords: Fault Diagnosis, Complex Systems, Quantitative Methods, Qualitative Reasoning, Soft Computing.

 

 

 

Cornelia TUDORIE, Severin BUMBARU, Luminita DUMITRIU: Relative aggregation operator in database fuzzy querying


Abstract: Fuzzy selection criteria querying relational databases include vague terms; they usually refer linguistic values form the attribute linguistic domains, defined as fuzzy sets. Generally, when a vague query is processed, the definitions of vague terms must already exist in a knowledge base. But there are also cases when vague terms must be dynamically defined, when a particular operation is used to aggregate simple criteria in a complex selection. The paper presents a new aggregation operator and the corresponding algorithm to evaluate the fuzzy query.

Keywords: Artificial Intelligence, Database, Fuzzy Logic, Fuzzy Queries

 

 

 

Cornelia TUDORIE, Cristian NEACSU, Ionel MANOLACHE: Fuzzy queries in romanian language an intelligent interface


Abstract: The most accessible interfaces querying databases must be so intelligent, able to understand natural language expressions and including vague terms in selection criteria. The paper proposes a general architecture for a flexible database interface and also a real implementation of such a system. It is general enough, so it can be connected to any database, after a specific knowledge base description. The natural language processing is mainly based on lexico-syntagmatical analyse; the vague criteria interpreting and evaluating are based on the fuzzy logic.

Keywords: Artificial Intelligence, Database, Fuzzy Logic, Fuzzy Queries, Natural Language

 

 

 

Dan MUNTEANU, Severin BUMBARU: A survey of text clustering techniques used for web mining


Abstract: This paper contains an overview of basic formulations and approaches to clustering. Then it presents two important clustering paradigms: a bottom-up agglomerative technique, which collects similar documents into larger and larger groups, and a top-down partitioning technique, which divides a corpus into topic-oriented partitions.

Keywords: information retrieval, algorithms, machine learning, web mining, text clustering

 

 

 

Dan MUNTEANU, Severin BUMBARU: Classification process in a text document recommender system


Abstract: This paper presents the classification process in a recommender system used for textual documents taken especially from web. The system uses in the classification process a combination of content filters, event filters and collaborative filters and it uses implicit and explicit feedback for evaluating documents.

Keywords: information systems, information analysis, algorithms, machine learning, agents

 

 

 

Ion VONCILA, Maria ILIUTA, Mariana DUMITRESCU, Gelu GURGUIATU: The influence of electric transformers' specific parameters modification on power distributions and on technological appropriate consumption (tac) within the energetic system


Abstract: The paper deals with a comparative analysis – using DINIS(E) programming environment – of the way electric transformers' specific variables (Iron losses, the parameters of the equivalent scheme, the level of primary power supply, etc.) influences power distributions, active and reactive, from a well-defined contour within the energetic system. Also, the paper deals with the way interconnection of a great number of transformation units, with different functional characteristics may influence the level of losses within the global system – result as a consequence of multiple connections – level ultimately emphasised by the technological appropriate consumption (TAC) .

Keywords: electric transformers, electric power supply networks

 

 

 

Ion VONCILA, Nicolae BADEA, Dorel POPA: The distribution of the electricfield and of the current density in case of electrocorrosion countermeasures using thesacrifice anodes technique


Abstract: An electric field oriented from sea water towards the ship, can develop an extraction work. The corrosion process was thus diminished, by a change in the ship’s hull mass, due to a continuous tore of the material from its bulkheads. The issue that arises is an attenuation of the extraction electric forces and a reduction in the process of mass changing. This problem can be solved by using an impressed electric field, opposite to the one developed during the corrosion process, field which can minimize the extraction work generated by this process.

Keywords: electrocorrosion, sacrifice anodes technique

 

 

 

Viorel DUGAN, Razvan SOLEA, Clara IONESCU and Daniela GHINITA: Three fuzzy reasoning models as a decision suport aid, to find an electrical energy tariff


Abstract: This contribution is a laboratory-work developed as an example of approximate (fuzzy) reasoning for students, possible to be used as a decision – support to estimate an electrical energy (EE) price for consumers. The three fuzzy tariff estimation models that are developed, integrate not only the S.C Electrica S.A.-single-supplier rate position, but and some (social) constraints/ compulsions of National Authority of Settlements from Energy (NASE) beginning with 1999, in this transition period from Romania. Although is possible, the paper not refer to a partial-price concrete case (internal tariff used in certain year, production price, transport price, distribution price, spot price, or an external price to be sold electrical energy, etc). This “laboratory-work-paper” shows how, by changing the parameters of S.C Electrica S.A. and NASE, it is possible to can perform sensitivity tests on the tariff function model, until can obtain an acceptable and true price. In this aim, the three fuzzy models use different rules for pricing: conservative, aggressive, and different order of words concerning the rules respectively, finally doing a comparation among prices and models. The paper not finished all fuzzy possibilities (rules) which can influences the expected value of a some EE tariff but, with certitude, can create a discussion base, about the way of approximate/ fuzzy reasoning, as a modality to find and to refine an EE price.

Keywords: fuzzy reasoning decision-support, fuzzy different models, EE tariff estimation