In this work we propose an immune approach for learning neurofuzzy systems, namely NEFDIAG (NEuro Fuzzy DIAGnosis). NEFDIAG is a software devoted primarily to creation, training and test of a classification Neuro-Fuzzy system of industrial process failures. But in case of great number of input variables NEFDIAG structure grows essentially and the dimensionality of learning task becomes a problem. Existing methods of NEFDIAG learning allow only identifying parameters of NEFDIAG without modifying its structure. We propose an immune Artificial learning approach for NEFDIAG learning based on clonal selection and immune network theories. It allows not only to identify NEFDIAG parameters but also to reduce number of neurons in hidden layers (rules layer) of NEFDIAG.

Dans cet article, nous présentons AntTreeStoch un nouvel algorithme de classification hiérarchique et non supervisée. Cet algorithme est basé sur l'auto-organisation observée chez les fourmis réelles. L'émergence des déplacements et les assemblages des fourmis, basés sur la similarité entre les individus permet d'identifier les états de fonctionnements d'un système industriel dynamique et complexe. L'algorithme offre aussi la possibilité de créer de nouvelles classes pour les données non identifiées. En effet le choix d'utiliser AntTreeStoch dans notre système de diagnostic a été motivé par la possibilité d'utiliser à la fois des données numériques et symboliques. Les expériences effectuées sur la base de données Iris, montrent que AntTreeStoch fournit de très bons résultats.

As a result from the demanding of process safety, reliability and environmental constraints, a called of fault detection and diagnosis system become more and more important. In this article some basic aspects of TSK (Takigi Sugeno Kang) neuro-fuzzy techniques for the prognosis and diagnosis of manufacturing systems are presented. In particular, a neuro-fuzzy model that can be used for the identification and the simulation of faults prognosis models is described. The presented model is motivated by a cooperative neuro-fuzzy approach based on a vectorized recurrent neural net-work architecture. The neuro-fuzzy architecture maps the residuals into two classes: a one of fixed direction residuals and another one of faults belonging to rotary kiln.

This paper proposes a novel hybrid algorithm for fault diagnosis of rotary kiln based on a binary ant colony (BACO) and support vector machine (SVM). The algorithm can find a subset selection which is attained through the elimination of the features that produce noise or are strictly correlated with other already selected features. The BACO algorithm can improve classification accuracy with an appropriate feature subset and optimal parameters of SVM. The proposed algorithm is easily implemented and because of use of a simple filter in that, its computational complexity is very low. The performance of the proposed algorithm is evaluated through two real Rotary Cement kiln datasets. The results show that our algorithm outperforms existing algorithms.

The continuing evolution of technology and human behavior puts the company in an uncertain and evolving environment. The company must be responsive and even proactive; therefore, control performance becomes increasingly difficult. Choosing the best method of ensuring control by the management policy of the company and its strategy is also a decision problem. The aim of this paper is the comparative study of three methods: the Balanced Scorecard, GIMSI and SKANDIAs NAVIGATOR for choosing the best method for ensuring the orderly following the policy of the company while maintaining its durability. Our work is divided into three parts. We firstly proposed original structural and kinetic metamodels for the three methods that allow an overall view of a method. Secondly, based on the three metamodels, we have drawn a generic comparison to analyze completeness of the method. Thirdly, we performed a restrictive comparison based on a restrictive set of criteria related to the same aspect example organizational learning, which is one of the bricks of knowledge management for a reconciliation to a proactive organization in an environment disturbed and uncertain, and the urgent needs. We note that we applied the three methods are applied in our precedent works.

This work appears in pattern recognition in the agronomic domain, especially for the identification of the leaves of plants, while using the adaptive technique of neuronal networks. In this article, we will expose our tool; which is intended for two categories of specialists, the first consisting of researchers in the field of botany, as the second, so all scientists, who may use this work in their own applications. We will expose also, the capacities of generalization of the neuronal networks and their implementation to our problem

Fault diagnosis and failure prognosis are essential techniques in improving the safety of many manufacturing systems. Therefore, on-line fault detection and isolation is one of the most important tasks in safety-critical and intelligent control systems. Computational intelligence techniques are being investigated as extension of the traditional fault diagnosis methods. This paper discusses the Temporal Neuro-Fuzzy Systems (TNFS) fault diagnosis within an application study of a manufacturing system. The key issues of finding a suitable structure for detecting and isolating ten realistic actuator faults are described. Within this framework, data-processing interactive software of simulation baptized NEFDIAG (NEuro Fuzzy DIAGnosis) version 1.0 is developed. This software devoted primarily to creation, training and test of a classification Neuro-Fuzzy system of industrial process failures. NEFDIAG can be represented like a special type of fuzzy perceptron, with three layers used to classify patterns and failures. The system selected is the workshop of SCIMAT clinker, cement factory in Algeria.

In this work we propose an immune approach for learning neurofuzzy systems, namely NEFDIAG (NEuro Fuzzy DIAGnosis). NEFDIAG is a software devoted primarily to creation, training and test of a classification Neuro-Fuzzy system of industrial process failures. But in case of great number of input variables NEFDIAG structure grows essentially and the dimensionality of learning task becomes a problem. Existing methods of NEFDIAG learning allow only identifying parameters of NEFDIAG without modifying its structure. We propose an immune Artificial learning approach for NEFDIAG learning based on clonal selection and immune network theories. It allows not only to identify NEFDIAG parameters but also to reduce number of neurons in hidden layers (rules layer) of NEFDIAG.

Dans cet article, nous présentons AntTreeStoch un nouvel algorithme de classification hiérarchique et non supervisée. Cet algorithme est basé sur l'auto-organisation observée chez les fourmis réelles. L'émergence des déplacements et les assemblages des fourmis, basés sur la similarité entre les individus permet d'identifier les états de fonctionnements d'un système industriel dynamique et complexe. L'algorithme offre aussi la possibilité de créer de nouvelles classes pour les données non identifiées. En effet le choix d'utiliser AntTreeStoch dans notre système de diagnostic a été motivé par la possibilité d'utiliser à la fois des données numériques et symboliques. Les expériences effectuées sur la base de données Iris, montrent que AntTreeStoch fournit de très bons résultats.

The basic idea of this work was to study the application of expert systems and fuzzy logic in the field of diagnostic and industrial maintenance. For this, a fuzzy expert system designed, developed and simulated in Ain Touta cement society in Batna in the East of Algeria. Dedicated to control cement mill. The application of fuzzy logic and expert systems to control the difference shown in the control system using fuzzy regulators for the operation of the grinding without unnecessary stops, it also helps the operator to know the maintenance task to perform. In addition, regulators decentralization allows the availability of fuzzy control, even if one of the regulators is absent, it does not prevent the other to complete its task to control fineness, mill's temperature and feed.

As a result from the demanding of process safety, reliability and environmental constraints, a called of fault detection and diagnosis system become more and more important. In this article some basic aspects of TSK (Takigi Sugeno Kang) neuro-fuzzy techniques for the prognosis and diagnosis of manufacturing systems are presented. In particular, a neuro-fuzzy model that can be used for the identification and the simulation of faults prognosis models is described. The presented model is motivated by a cooperative neuro-fuzzy approach based on a vectorized recurrent neural network architecture. The neuro-fuzzy architecture maps the residuals into two classes: a one of fixed direction residuals and another one of faults belonging to rotary kiln.

This paper proposes a novel hybrid algorithm for fault diagnosis of rotary kiln based on a binary ant colony (BACO) and support vector machine (SVM). The algorithm can find a subset selection which is attained through the elimination of the features that produce noise or are strictly correlated with other already selected features. The BACO algorithm can improve classification accuracy with an appropriate feature subset and optimal parameters of SVM. The proposed algorithm is easily implemented and because of use of a simple filter in that, its computational complexity is very low. The performance of the proposed algorithm is evaluated through two real Rotary Cement kiln datasets. The results show that our algorithm outperforms existing algorithms.