In this Kdynote address, two types of information granules are considered : (ⅰ) one for set assignments of a concept descriptor and (ⅱ) the other for truthood assignment to the concept description verifier. The first is, the process which specifies the assignment of an object to a clump, a class, a group, etc., and hence defines the set membership with a relational constraint. the second is the assignment of the degree of truthood or the membership specification of the abstract concept of truthood which specifies the " veristic" constraint associated with the concept descriptor. The combination of these two distinct assignments let us generate four set and logic theories. This then leads to the concern of normal forms and their derivation from truth tables for each of these theories. In this regard, some of the fundamental issues arising in this context are discussed and certain preliminary answers are provided in order to highlight the consequences of these theories.

This paper presents a stabilization algorithm for a class of fuzzy systems with singleton consequect. To this aim, we introduce two canonical forms of an unforced fuzzy system and a stability theorem. A design example is shown to verify the stabilization algorithm.

In this paper, a fuzzy control technique using adjustable scale factors and Lyapunov Function for the precise position control of DC servo system is introduced. The suitable scale factors were selected and the stable control input using the stability theory of Lyapunov function cam be applied. Therefore, the controlled system have the robustness against disturbances and can be stabilized because of reinforced adaptivity. This proposed fuzzy controller is implemented on a 80586 micro-computer which have of fuzzy inference routine part, manipulating part of scale factors and DT-2801 data aquisition board.

In this paper, we analyze the effects of scaling factors on the performance of a fuzzy logic controller(FLC). The quantitative relation between input and output variables of FLC is obtained by using a qualsi-linear fuzzy model, and an approximate transfer function of FLC is dervied from the comparison of it with the conventional PID controller. Then we analyze in detail the effects of scaling factor using this approximate transfer function and root locus method. Also we suggest an on-line tuning method for scaling factors which employs an sample performance function and a variable reference for tuning index.

This paper presents an auto tuning method of PID controller based on the application of fuzzy logic. The proposed method combined the principles of PID control with fuzzy control, which cam considerably improve the performance index of PID controller. Simulation results show that higher performance and accuracy of overall system for desired value is achieved with our manner when compared to widely-used conventional tuning method.

This paper proposes fuzzy regression analysis with non-symmetric fuzzy coefficients. By assuming non-symmetric triangular fuzzy coefficients and applying the quadratic programming fomulation, the center of the obtained fuzzy regression model attains more central tendency compared to the one with symmetric triangular fuzzy coefficients. For a data set composed of crisp inputs-fuzzy outputs, two approximation models called an upper approximation model and a lower approximation model are considered as the regression models. Thus, we also propose an integrated quadratic programming problem by which the upper approximation model always includes the lower approximation model at any threshold level under the assumption of the same centers in the two approximation models. Sensitivities of Weight coefficients in the proposed quadratic programming approaches are investigated through real data.

In this paper, we formulate the DEA model with interval efficiency, there exist two phases of efficiency evaluation with respect to the upper limit and the lower limit. From these viewpoints, we can define two extreme points of efficiency. As a result, an interval efficiency for each DMU cam be obtained. We also formulate the interval cross-efficiency.

It is well known that traveling salesman problem (for short, TSP) is one of mot important problems for optimization, and almost all optimization problems result in TSP. This paper describes on an effective solution of TSP using genetic algorithm. The features of our method are summarized as follows : (1) By using division and unification method, a large problem is replaced with some small ones. (2) Smoothing method proposed in this paper enables us to obtain a fine approximate solution globally. Accordingly, demerits caused by division and unification method are decreased. (3) Parallel operation is available because all divided problems are independent of each other.

we introduce the concept of coverings, direct products, cascade products and wreath products of T-fuzzy finite state machines and investigate their algebraic structures.

In this paper, a fuzzy model based on the Polynomial Neural Network(PNN) structure is proposed to estimate the emission pattern for air pollutant in power plants. the new algorithm uses PNN algorithm based on Group Mehtod of Data Handling (GMDH) algorithm and fuzzy reasoning in order to identify the premise structure and parameter of fuzzy implications rules, and the least square method in order to identify the optimal consequence parameters. Both time series data for the gas furnace and data for the NOx emission process of gas turbine power plants are used for the purpose of evaluating the performance of the fuzzy model. The simulation results show that the proposed technique can produce the optimal fuzzy model with higher accuracy and feasibility than other works achieved previously.

We introduce the concept of a fuzzy c* -continuity. And we obtain some properties.

We introduce the concept of a fuzzy H-continuity and find some propertie.

A notion of 'fuzzy' convergence of filters on a set is introduced. We show that the collection of fuzzy L-limit spaces forms a cartesian closed topological category and obtain an interesting relationship between the notions of 'fuzzy' convergence structure and convergence approach spaces.

We introduce the concepts of extending a fuzzily constrained function and a fuzzy extending of a real function by using usual limit and illustrate them.

In this paper, we propose a new paradigm (NEW_BP) to be capable of overcoming limitations of the traditional backpropagation(OLD_BP). NEW_BP is based on the method of conjugate gradients with the normalized direction vectors and computes step size through the linear search which may be characterized by order statistics and golden section. Simulation results showed that NEW_BP was definitely superior to both the stochastic OLD_BP and the deterministic OLD_BP in terms of accuracy and rate of convergence and might sumount the problem of local minima. Furthermore, they confirmed us that stagnant phenomenon of training in OLD_BP resulted from the limitations of its algorithm in itself and that unessential approaches would never cured it of this phenomenon.

We propose a way to show the inherent learning complexity for the multilayer perceptron. We display the solution space and the error surfaces on the input space of a single neuron with two inputs. The evolution of its weights will follow one of the two error surfaces. We observe that when we use the back-propagation(BP) learning algorithm (1), the wight cam not jump to the lower error surface due to the implicit continuity constraint on the changes of weight. The self-relaxation approach is to explicity find out the best combination of all neurons' two error surfaces. The time complexity of training a multilayer perceptron by self-relaxationis exponential to the number of neurons.

In this paper, we propose another paradigm(QNBP) to be capable of overcoming Limitations of the traditional backpropagation(SDBP). QNBPis based on the method of Quasi -Newton(variable metric) with the nomalized direction vectors and computes step size through the linear search. Simulation results showed that QNBP was definitely superior to both the stochasitc SDBP and the deterministic SDBP in terms of accuracy and rate of convergence and might sumount the problem of local minima. and there was no different between DFP+SR1 and BFGS+SR1 combined algrothms in QNBP.

This paper suggests an integrated method for discovering knowledge from a large database. Our approach applies an attribute-oriented concept hierarchy ascension technique to extract generalized data from actural data in databases, induction of decision trees to measure the value of information, and knowledge reduction of rough set theory to remove dispensable attributes and attribute values. The integrated algorithm first reduce the size of database for the concept generalization, reduces the number of attributes by way of elimination condition attributes which have little influence on decision attribute, and finally induces simplified decision rules removing the dispensable attribute values by analyzing the dependency relationships among the attributes.

This paper aims to present some ideas for implementation of Uncertain Relational Databases (URD) which are extensions of classical relational databases. Our system firstly is based on possibility distribution and probability theory to represent and manipulate fuzzy and probabilistic information, secondly adopts flexible mechanisms that allow the management of uncertain data through the resources provided by both available relational database management systems and front-end interfaces, and lastly chooses dynamic SQL to enhance versatility and adjustability of systems.

Unification of Kohone SOM(Self-Organizing Maps) neural network with the branch-and-bound algorithm is presented for clustering large set of patterns. The branch-and-bound search technique is employed for designing coarse neural network learning paradaim. Those unification can be use for clustering or calssfication of large patterns. For classfication purposes further usefulness is possible, since only two clusters exists in the SOM neural network of each nodes. The result of experiments show the fast learning time, the fast recognition time and the compactness of clustering.

Missing values in data for fuzzy c-menas clustering is discussed. Two basic methods of fuzzy c-means, i.e., the standard fuzzy c-means and the entropy method are considered and three options of handling missing values are proposed, among which one is to define a new distance between data with missing values, second is to alter a weight in the new distance, and the third is to fill the missing values by an appropriate numbers. Experimental Results are shown.

It sis wisely stated that the most valuable knowledge that a person can acquire is the knowledge of how to learn. The human's learning is characterized by the ability to extract relationships between the different characters of a given situation . The ellipse is a first approach of comparison. We assimilate each character to a half axis of the ellipse and the result is a geometrical figure that varies according to values of the two characters. Thus, we take into account the two characters as an alone entity.

Gustafson and Kessel proposed a modified fuzzy c-Means algorithm based of the Mahalanobis distance. Though the algorithm appears more natural through the use of a fuzzy covariance matrix, it needs to calculate determinants and inverses of the c-fuzzy scatter matrices. This paper proposes a fuzzy clustering algorithm using pseudo mahalanobis distance, which is more easy to use and flexible than the Gustafson and Kessel's fuzzy c-Means.

It is wisely stated that the most valuable knowledge that a person cam acquire is the knowledge of how to learn. The human's learning is characterized by the ability to extract relationships between the different characters of a given situation. The ellipse is a first approach of comparison. We assimilate each character to a half axis of the ellipse and the result is a geometrical figure that varies according to values of the two characters. Thus, we take into account the two characters as an alone entity.

This paper is our first attempt to construct a information processing system such as the living creatures' brain based on artificial life technique. In this paper, we propose a method of constructing neural networks using bio-inspired emergent and evolutionary concept, Ontogeny of living things is realized by cellular automata model and Phylogeny that is living things adaptation ability themselves to given environment, are realized by evolutionary algorithms. Proposing evolving cellular automata neural systems are calledin a word ECANS. A basic component of ECANS is 'cell' which is modeled on chaotic neuron with complex characteristics, In our system, the states of cell are classified into eight by method of connection neighborhood cells. When a problem is given, ECANS adapt itself to the problem by evolutionary method. For fixed cells transition rule, the structure of neural network is adapted by change of initial cell' arrangement. This initial cell is to become a network b developmental process. The effectiveness and the capability of proposed scheme are verified by applying it to pattern classification and robot control problem.

Complex "lifelike" behaviors are composed of local interactions of individuals under fundamental rules of artificial life. In this paper, fundamental rules for cooperative group behaviors, "flocking" and "arrangement", of multiple autonomous mobile robots are represented by a small number of fuzzy rules. Fuzzy rules in Sugeno type and their related paramenters are automatically generated from clustering input-output data obtained from the algorithms the group behaviors. Simulations demonstrate the fuzzy rules successfully realize group intelligence of mobile robots.

Transporting a large table using multiple robotic agents requires at least two group behaviors of homing and herding which are to bo coordinated in a proper sequence. Existing GP methods for multi-agent learning are not practical enough to find an optimal solution in this domain. To evolve this kind of complex cooperative behavior we use a novel method called fitness switching. This method maintains a pool of basis fitness functions each of which corresponds to a primitive group behavior. The basis functions are then progressively combined into more complex fitness functions to co-evolve more complex behavior. The performance of the presented method is compared with that of two conventional methods. Experimental results show that coevolutionary fitness switching provides an effective mechanism for evolving complex emergent behavior which may not be solved by simple genetic programming.

We may gaze at some peculiar scenes of flocking of birds and fishes. This paper demonstrates that multiple agent mobile robots show complex behaviors from efficient and strategic rules. The simulated flock are realized by a distributed behavioral model and each mobile robot decides its own motion as an individual which moves constantly by sensing the dynamic environment.

CAM-Brain is a model to develop neural networks based in cellular automata by evolution, and finally aims at a model as and artificial brain,. In order to show the feasibility of evolutionary engineering to develop an artificial brain we have attempted to evolve a module of CAM-Brain for the problem to control a mobile robot, In this paper, we present some recent results obtained by analyzing the behaviors of the evolved neural module. Several experiments reveal a couple of problems that should be solved when CAM-Brain evolves to control a mobile robot. so that some modification of the original model is proposed to solve them. The modified CAM-Brain has evolved to behave well in a simulated environment, and a thorough analysis proves the power of evolution.

In this paper, we propose a feedback adaptive controller which need not adjustment of the scale factor. Numerical examples are included to illustrate the procedure of a adaptive control and to show the performance of the control system. We can observe that the output of control system, converges toward the reference of response.

A PID type fuzzy Controller is proposed based on a crisp type model in which the consequent parts of the fuzzy control rules are functional representation or real numbers. Using the conventional PID control theory, a new PID type fuzzy controller is developed, which retains the characteristics of the conventional PID controller. An advantage of this approach, is that it simplifies the complicated defuzzification algorithm which could be time consuming. Computer simulation results have shown that the proposed PID fuzzy controller has satisfactory tracking performance.

A novel fuzzy basis function vector- based adaptive control approach for Multi-input and Multi-output(MIMO) system is presented in this paper, in which the nonlinear plants is first linearised, the fuzzy basis function vector is then introduced to adaptively learn the upper bound of the system uncertainty vector, and its output is used as the paramenters of the compensator in the sense that both the robustness and the asymptotic error convergence can be obtained for the closed loop nonlinear control system.

This paper describe an intelligent cruise control system for automobile. With the remarkable numerical increase of automobiles on the road, the optimized traffic flow control using the cruise control is one of the very important traffic problems to overcome the limitation of an existing road capacity. Based on this idea that minimize the fuel cost and the air pollution, and accept a driver's needs for driving, we have developed an intelligent cruise control system for vehicle. This proposed intelligent fuzzy cruise controller was successfully implemented using the fuzzy algorithm, the i80c196 μ-controller board and the throttle valve actuator. The field test results on an linear road was introduced.

In this paper, it is shown how Fuzzy Pattern Matching can be applied to diagnosis of the most common faults of Rotating Machinery. The whole diagnosis process has been divided in three steps : Fault Detection, Fault Isolation and Fault Identification, whose possible results are described by linguistic patterns. Diagnosis will consist in obtaining a set of matching indexes that indexes that express the compatibility of the fuzzified features extracted from the measured vibration signals, with the knowledge contained in the corresponding patterns.

This paper suggests the method to recognize the various defect patterns of cold mill strip using binary decision tree constructed by genetic algorithm automatically. In case of classifying the complex the complex patterns with high similarity like the defect patterns of cold mill strip, the selection of the optimal feature set and the structure of recognizer is important for high recognition rate. In this paper genetic algorithm is used to select a subset of the suitable features at each node in binary decision tree. The feature subset of maximum fitness is chosen and the patterns are classified into two classes by linear decision function. After this process is repeated at each node until all the patterns are classified respectively into individual classes. In this way , binary decision tree classifier is constructed automatically. After construction binary decision tree, the final recognizer is accomplished by the learning process of neural network using a set of standard p tterns at each node. In this paper, binary decision tree classifier is applied to recognition of the defect patterns of cold mill strip and the experimental results are given to show the usefulness of the proposed scheme.

In this paper, we define and study another various generalizations of fuzzy continuous functions by using the concept of regular generalized fuzzy closed sets, A comparative study regarding the mutual interrelations among these functions along with those functions obtained in [3] is made. Finally, we have introduced and studied the notions of rgf-extremally disconnectedness and rgf-compactness

We introduce the notion of ${\gamma}$-fuzzy filter and ${\gamma}$-limit structure to L-fuzzy point. We show that the category ${\gamma}$Lim of ${\gamma}$-limit spaces is a cartesian closed topological construct containing the category LFTop of stratified L-fuzzy topological spaces as a bireflective subcategory.

In this paper, we introduce the concepts of intuitionistic fuzzy points and intuitionistic fuzzy neighborhoods. We investigate properties of continuous, open and closed maps in the intuitionistic fuzzy topological spaces, and show that the category of Chang's fuzzy topological spaces is a bireflective full subcategory of that of intuitionistic fuzzy topogical spaces.

We introduce the concept of a fuzzy irresolute mapping and a fuzzy semi-homeomorphism. And we find some properties of then.

We introduce the concept of a fuzzy vietories topology and we obtain its fundamental properties.

Fuzzy information retrieval which uses the concept of fuzzy relation is able to retrieve documents in the way based on not morphology but semantics, dissimilar to traditional information retrieval theories. Fuzzy information retrieval logically consists of three sets : the set of documents, the set of terms and the set of queries. It maintains a fuzzy relational matrix which describes the relationship between documents and terms and creates a thesaurus with fuzzy relational product. It also provides the user with documents which are relevant to his query. However, there are some problems on building a thesaurus with fuzzy relational product such that it has big time complexity and it uses fuzzy values to be processed with flating-point. Actually, fuzzy values have to be expressed and processed with floating-point. However, floating-point operations have complex logics and make the system be slow. If it is possible to exchange fuzzy values with binary values, we could expect sp eding up building the thesaurus. In addition, binary value expressions require just a bit of memory space, but floating -point expression needs couple of bytes. In this study, we suggest a new method of building a thesaurus, which accelerates the operation of the system by pre-applying an ${\alpha}$-cut. The experiments show the improvement of performance and reliability of the system.

The basic requirement of a distributed multimedia system are intramedia synchronization which asks the strict delay and jitter for the check period of media buffer and the scaling duration with periodic continuous media such as audio and video media, and intermedia synchronization that needs the constraint for relative time relations among them when several media are presented in parallel. In this paper, a distributed multimedia synchronization algorithm based on the fuzzy logic is presented and the performance is evaluated through simulation. Intramedia synchronisation algorithm uses the media scaling techniques and intermedia synchronization algorithm uses variable service rates on the basis of fuzzy logic to solve the multimedia synchronization problem.

A method of multimedia information data acquisition based on fuzzy rules is proposed, where the multimedia means the five senses of a human being. Observed information is characterized by VAGOT(visual, acoustic, gustatory, olfactory and tactile) time series data and the goal is to extract an appropriate subset of the VAGOT data based on a given instruction. Fuzzy rules based on visual and acoustic information are used to identify the appropriate time interval on the fireworks multimedia information.

Fuzzy rules are often obtained by experts who know an objective system well. fuzzy rules acquired by experts, however, do not express all input-output relations of the system. This paper proposes a method fuzzy rules are expressed in plain language so that the fuzzy rules are understood easily. The proposed method is applied to the control of the distance between cars and running through a crank-typed road, and the validity of the method is confirmed.

We are dealing with the preliminary diagnosis from the information of headache interview chart. We quantify the qualitative information based on the interview chart by dual scaling. Prototype of fuzzy diagnostic sets and the neural linear regression methods are established with these quantified data, These new methods can be used to classify new patient's tone of diseases with certain degrees of belief and its concerned symptoms. We call these procedures as neural Fuzzy Differential Diagnosis of Headache (NFDDH-1). Also we investigate three measures to medical diagnosis, where relations between symptoms and diseases are described by intutionistic fuzzy set (IFS) data. Two measures are described by nin-max and max-min IFS operators, respectively. Another measure is the similarity degree, i.e., IFS distance between patient's symptoms and prototypes of diseases. We consider some reasonable criteria for three measures in order to determine the label of headache, We will establish hree measures in NFDDH-2 and combine two packages as NFDDH

this paper presents the potential application of fuzzy logic to the automatic incident detection system. While the conventional incident detection algorithms are based on a binary decision process, the algorithm using fuzzy logic can incorporate ambiguity which occurs in determining incidents. Since collecting good amount of data to construct data base for incidents is pretty expensive, a traffic simulator called FRESIM is used to simulate traffic condition in a freeway. Incident data are obtained by changing input parameters of the simulator and the fuzzy algorithm generates fuzzy rule for determining normal and incident traffic conditions. In this paper, various steps are described to test the algorithm and its results are summarized.

If we want to compare the form of two objects, the human vision takes into account the parameter's width/length/height at the same time. however, the machine needs to compare width then lengths and finally height. In each comparison the machine considers only one character. The goal of this paper is to imitate the human manner of comparison and recognition by using two or three characters instead of one during the comparison. The ellipse is a first approach of comparison because it provides us a general and a simple relation that can link two parameters that are the half axis of the ellipse. Indeed, we assimilate each character to a half axis of the ellipse and the result is a geometrical figure that varies according to values of the two characters.

The interval approach to the linguistic expression coding nears us to the human idea. Thus, what seems "weak" for a person can appear very weak for another person or for the same person in others circumstances. However, the utilization of intervals is not restrained to the cases of linguistic expression coding. Indeed, the interval can facilitate the solution of several problems.

It is important to develop a method of assembling a set of sub pictures automatically into a mosaic picture , because a view through fiberscopes or microscopes with higher magnifying power is much larger than the field of view taken by a camera. This paper presents a method of assembling sub pictures, where roughly estimated junctions called approximate junctions are employed for matching triangles formed by selected junctions in sub pictures. To over come the difficulties in processing speed and noise corruption, fuzzy rules is applied to get fuzzy values for existence of approximate junctions and fuzzy similarity for congruent triangle matching. Some demonstration, exemplified by assembling microscopic metal matrix photographs, are given to show feasibility of this method.

This paper proposes a new data classification method based on the tolerant rough set that extends the existing equivalent rough set. Similarity measure between two data is described by a distance function of all constituent attributes and they are defined to be tolerant when their similarity measure exceeds a similarity threshold value. The determination of optimal similarity theshold value is very important for the accurate classification. So, we determine it optimally by using the genetic algorithm (GA), where the goal of evolution is to balance two requirements such that (1) some tolerant objects are required to be included in the same class as many as possible. After finding the optimal similarity threshold value, a tolerant set of each object is obtained and the data set is grounded into the lower and upper approximation set depending on the coincidence of their classes. We propose a two-stage classification method that all data are classified by using the lower approxi ation at the first stage and then the non-classified data at the first stage are classified again by using the rough membership functions obtained from the upper approximation set. We apply the proposed classification method to the handwritten numeral character classification. problem and compare its classification performance and learning time with those of the feed forward neural network's back propagation algorithm.

The Rough Set theory suggested by Pawlak in 1982 has been useful in AI, machine learning, knowledge acquisition, knowledge discovery from databases, expert system, inductive reasoning. etc. The main advantages of rough set are that it does not need any preliminary or additional information about data and reduce the superfluous informations. but it is a significant disadvantage in the real application that the inference result form is not the real control value but the divided disjoint interval attribute. In order to overcome this difficulty, we will propose approach in which Rough set theory and Neuro-fuzzy fusion are combined to obtain the optimal rule base from lots of input/output datum. These results are applied to the rule construction for infering the temperatures of refrigerator's specified points.

Software reuse is a well-known method to increase the productivity of software, nevertheless it is not employed well on real world. One of the important factors that this problem occurs is programers' distrust in the existing components. Therefore in this paper, to increase the reliability of reusability decision, we proposed a method which can analyze significance of the reusability decision metrics using Rough Set.

In this paper we will discuss a type of inductive learning called learning from examples, whose task is to induce general description of concepts from specific instances of these concepts. In many real life situations, however, new instances can be added to the set of instances. It is first proposed within the framework of rough set theory, for such cases, an algorithm to find minimal set of rules for decision tables without recalculation for overcall set of instances. The method of learning presented here is base don a rough set concept proposed by Pawlak[2][11]. It is shown an algorithm to find minimal set of rules using reduct change theorems giving criteria for minimum recalculation with an illustrative example. Finally, the proposed learning algorithm is applied to fuzzy system to learn sampled I/O data.

In this paper, for efficient data classification in multispectral bands environment, a band feature extraction method using the Rough sets theroy is proposed. First, we make a look up table from training data, and analyze the properties of experimental multispectral image data, then select the efficient band usin indiscernibility relation of Rough sets theory from analysis results. Proposed method is applied to LAMDSAT TM data on 2, June, 1992. Among them, normal distributive data were experimented, mainly. From this, we show clustering trends that similar to traditional band selection results by wavelength properties, from this, we verify that can use the proposed method that centered on data properties to select the efficient bands, though data sensing environment change to hyperspectral band environments.

Underwater robotic vehicles (URVs) have been an important tool for various underwater tasks because they have greater speed, endurance, depth capability, and safety than human divers. As the use of such vehicles increases, the vehicle control system becomes one of the most critical subsytems to increase autonomy of the vehicle. The vehicle dynamics are nonlinear and their hydrodynamic coefficients are often difficult to estimate accurately. In this paper a new type of fuzzy model-based controller based on Takagi-Sugeno-Kang fuzzy model is designed and applied to the control of of an underwater robotic vehicle. The proposed fuzzy controller : 1) is a nonlinear controller, but a linear state feedback controller in the consequent of each local fuzzy control rule ; 2) can guarantee the stability of the closed-loop fuzzy system ; 3) is relatively easy to implement. Its good performance as well as its robustness to the change of parameters have been shown and compared with the re ults of conventional linear controller by simulation.

There have been several recent studies concerning the stability of fuzzy control systems and the synthesis of stabilizing fuzzy controller. This paper reports on a related study of the TS(Takagi-Sugeno) fuzzy systems, and it is shown that the controller synthesis problems for the nonlinear systems described by the TS fuzzy model can be reduced to convex problems involving LMIs(Linear matrix inequalities). After classifying the TS fuzzy systems into two families based on how diverse their input matrices are, different controller synthesis procedure is given for each of these families. A numerical example is presented to illustrate the synthesis procedures developed in this paper.

A systematic design method for PDC(Parallel Distributed Compensation)-type continuous time Takagi-Sugeno(T-S in short) fuzzy control systems which have inaccessible states is developed in this paper. Reduced-dimensional fuzzy state estimator is introduced from existing T-S fuzzy model using the PDC structure of Wang et al. [1] LMI(Linear Matrix Inequalities) problems which represent the stabililty of the reduced-dimensional fuzzy state estimator are derived. Pole placement constraints idea for each rules is adopted to determine the estimator gains and they are also revealed as LMI problems. these LMI problems are combined with Joh et al's [7][8] LMI problems for PDC -type continuous time T-S fuzzy controller design to yield a systematic design method for PDC -type continuous time T-S fuzzy control systems which have inaccessible states.

A new fuzzy control algorithm for the control of active magnetic bearing (AMB) systems is proposed in this paper. It combines PDC design of Joh et al. [8][9] and Namdani-gype control rules using fuzzy singletons to handle the nonlinear characteristics of AMB systems efficiently. They are named fine mode control and rough mode control , respectively. The rough mode control yields the fastest response for large deviation of the rotor and the fine mode control fives desired transient response for small deviation of the rotor. The proposed algorithm is applied a AMB systems to verify the performance of the method, The comparison of the proposed method to a linear controller using a linearized model about the equilibrium point and PDC algorithm in [7] show the superiority of the proposed algorithm.

This paper deals with the decision in vague knowledge, One method is a classic theory. That is to say, constraints and goals in the vague knowledge. Another method is the fuzzy catastrophe. If there exist two fuzzy variables, there may be a discontinuity which plays an important role in decision.

Dynamic programming is applicable to any situation where items from several groups must be combined to form an entity, such as a composite investment or a transportation route connecting several districts. The most desirable entity is constructed in stages by forming sub-entities that are candidates for inclusion in the most desirable entity are retained, and all other sub-entities are discarded. In the paper, the fuzzy dynamic programming is applied to the situation where each investment in the set has the following characteristics : the amount to be invested has several possible values, and the rte of return varies with the amount invested. Each sum that may be invested represents a distinct level of investment , and the investment therefore has multiple levels. A numeric example constructing a combination of multilevel investments is given in the paper.

Conventional marketing research generally focuses on a single layer's benefit. A notable example is the consumer layer providing managers with partial market information to evaluate relevant strategies. As generally known, marketing management encounters complex supply and demand behaviors, thereby necessitation that a successful marketing strategy adopt multi-layer considerations, such as the consumer layer, channel-retailer layer, and marketing planner layer. In light of above situation, this study applies fuzzy theory and the analytic hierarchy process(AHP) technique to analyze the performances of marketing strategies under multi-layer benefits, In addition, conventional marketing research has difficulty in efficiently allocating the limited budget so that each desired criterion can be significantly enhanced by a group of events. Therefore, a weighting structure among the goal, layers, criteria, and strategies(i.e. a group of events) is also developed herein to trace the influential process and assist marketing managers in efficiently allocating resources(i.e.budget).

An algorithm for representing the cubic spline interpolation of differentiable functions by a fuzzy system is presented in this paper. The cubic B-spline functions which form a basis for the interpolation function are used as the fuzzy sets for input fuzzification. The ordinal number of the coefficient cKL in the list of the coefficient cij's as sorted in increasing order, is taken to be the output fuzzy set number in the (k, l) th entry of the fuzzy rule table. Spike functions are used for the output fuzzy sets, with cij's as support boundaries after they are sorted. An algorithm to compute the support boundaries explicitly without solving the matrix equation involved is included, along with a few properties of the fuzzy rule matrix for the designed fuzzy system.

In this paper, we introduce the notions of fuzzy ${\gamma}$-regular open sets and fuzzy almost ${\gamma}$-continuous maps, and investigate some of their basic properties.

We introduce the concepts of fuzzy strongly ${\gamma}$-semiopen sets and fuzzy strongly ${\gamma}$-semicontinuous maps, and investigate some of their basic properties.

We introduce the concept of a 'fuzzy' map between sets by modifying the concept of the extension principle introduced by Dubois and Prade in [1] and study their properties. Using these we generalize Goguen's and Zadeh's extension principles in [2] and [3].

In this paper we introduce the concept of fuzzy integrals for set-valued mappings, which is an extension of fuzzy integrals for single-valued functions defined by Sugeno. And we give some properties including convergence theorems on fuzzy integrals for set-valued mappings.

In this paper, we suggested a neuro-fuzzy learning algorithm for tuning fuzzy rules, in which a fuzzy system model is of additive-type. Using the method, it is possible to reduce the computation size, since performing the fuzzy inference and tuning the fuzzy rules of each fuzzy subsystem model are independent. Moreover, the efficiency of suggested method is shown by means of a numerical example.

In this paper, we try to analyze two kinds of conventional neuro-fuzzy learning algorithms, which are widely used in recent fuzzy applications for tuning fuzzy rules, and give a summarization of their properties. Some of these properties show that uses of the conventional neuro-fuzzy learning algorithms are sometimes difficult or inconvenient for constructing an optimal fuzzy system model in practical fuzzy applications.

In this paper, a method for the improvement of learning speed and convergence rate was proposed applied it to physiological neural structure with the advantages of artificial neural networks and fuzzy theory to physiological neuron structure, To compare the proposed method with conventional the single layer perception algorithm, we applied these algorithms bit parity problem and pattern recognition containing noise. The simulation result indicated that our learning algorithm reduces the possibility of local minima more than the conventional single layer perception does. Furthermore we show that our learning algorithm guarantees the convergence.

In this paper, we proposed a new fuzzy supervised learning algorithm. We construct, and train, a new type fuzzy neural net to model the linear activation function. Properties of our fuzzy neural net include : (1) a proposed linear activation function ; and (2) a modified delta rule for learning algorithm. We applied this proposed learning algorithm to exclusive OR,3 bit parity using benchmark in neural network and pattern recognition problems, a kind of image recognition.

In this paper, we propose a learning algorithm of fuzzy neural networks with trapezoidal fuzzy weights. This fuzzy neural networks can use fuzzy numbers as well as real numbers, and represent linguistic information better than standard neural networks. We construct trapezodal fuzzy weights by the composition of two triangles, and devise a learning algorithm using the two triangular membership functions, The results of computer simulations on numerical data show that the fuzzy neural networks have high fitting ability for target output.

Subsymbolic Knowledge processing is said to be changed states of networks constructed from small elements. subsymbolic systems also make it possible to use connectionist models for knowledge processing. Connectionist realization such modulus are modulus linked together for solving a given problem. We study using neural networks as distinct actions. The output vectors produced by the neural networks are consider as a new facts. These new facts are then processed to activate another networks or used in the current production rule, The production rule is applying knowledge stored in the knowledge base to make inference. After neural networks knowledge base is constructed and trained. We present a running sample of incorporating neural network knowledge base. We implement using rochester connectionist simulator. We suggest that incorporating neural network knowledge base. Therefore incorporated neural network knowledge base ensures a cleaner solution which results in better perfor s.

In this paper we first show that uniform PR systems and half independent PR systems have same dynamics, and then an important property of this two kinds of systems is derived. The most important property of uniform PR systems is that they have the ability of classifying m-dimensional problabilistic vector into in classes. The significance of studying the dynamics of uniform PR systems are tried from the beginning with a uniform PR system.

This paper presents a method of object classification from dynamic image based on fuzzy inference algorithm which is suitable for low speed such as, conveyor, uninhabited transportation. At first, by using feature parameters of moving object, fuzzy if - then rule that can be able to adapt the wide variety of surroundings is developed. Secondly, implication function for fuzzy inference are compared with respect the proposed algorithm. Simulation results are presented to testify the performance and applicability of the proposed system.

An optical system which can autonomously form and display an impression of a picture made up by many figures has been developed. This system consists of optical fuzzy-neurons which calculate the correlation between the input picture and the reference image by incoherent optics. The calculated signal is applied to an amplifier whereby the output signal increases, then decreases according to increase of the input signal . These outputs are synthesized, and are used for changing the position where the system gaze on a part of the input picture by light beam. In this system, the light intensity used for gazing changes chaotically, The attractor drawn from the change of light intensity corresponds to the impression of the picture. This paper shows the results that are calculated by the numerical simulation. The system has been simulated to express the impression for a picture formed by 4figures.

We propose a fuzzy rule based method for structure preserving dimensionality reduction. This method selects a small representative sample and applies Sammon's method to project it. The input data points are then augmented by the corresponding projected(output) data points. The augmented data set thus obtained is clustered with the fuzzy c-means(FCM) clustering algorithm. Each cluster is then translated into a fuzzy rule for projection. Our rule based system is computationally very efficient compared to Sammon's method and is quite effective to project new points, i.e., it has good predictability.

Power systems have uncertain dynamics due to a variety of effects such as lightning, severe storms and equipment failures. The variation of the effective reactance of a transmission line due to a fault is an example of uncertainty in power system dynamics. Hence, a robust controller to cope with these uncertainties is needed. Recently fuzzy controllers have become quite popular for robust control due to its capability of dealing with unstructured uncertainty. Thus in this paper we design an adaptive fuzzy controller using an input-output linearization approach for the transient stabilization and voltage regulation of a power system under a sudden fault. Simulation results show that satisfactory performance is achieved by the proposed controller.

The helicopter system is non-linear and complex. Futhermore, because of absence of an accurate mathematical model, it is difficult accurately to control its attitude. But we can control the non-modeled system with the uncertainty and unstructure using the fuzzy control algorithm. Therefore, we apply optimized fuzzy controllers for the control of its elevation angle and azimuth one using expert's intuitions and knowledges. The simulation and experimental results of the hellicopter simulator CE150 with MATLAB shall be introduced.

As the strip caster system that produces a regular steel plate can be considered as a complicate nonlinear multi-variable system, it is not easy to obtain an effective control system. One way to overcome the difficulties is to apply the intelligent neuro-fuzzy fusion approach in developing the control scheme. The neuro-fuzzy control scheme possesses several distinct advantages, including the fact that it doesn't need the exact mathematical modeling of controlled plant and can provided some robustness in the control scheme. In this paper, an intelligent neuro-fuzzy for the stripe caster system will be proposed. The effectiveness of the proposed scheme will be demonstrated by computer simulation.

In this paper, we implemented the variable fuzzy speed controller of an IM(induction motor) using the fuzzy control algorithms. Specially, we proposed a self-tuning technique of scale factors which could make easily the fuzzy speed controller optimize. Comparing with the conventional PI speed controller, the dynamic performances of a proposed fuzzy controller such as the reaching time, the maximum overshoot and the robustness against load disturbance were substantially improved.

In this paper, a new approach to modeling of nonlinear systems using fuzzy theory is presented. So as to handle a variety of nonlinearity and reflect the degree of confidence in the informations about system, we combine multiple model method with hierarchical prioritized structure. The mountain clustering technique is used in partition of system, and TSK rule structure is adopted to form the fuzzy rules. Back propagation algorithm is used for learning parameters in the rules. Computer simulations are performed to verify the effectiveness of the proposed method. It is useful for the treatment fo the nonlinear system of which the quantitative math-approach is difficult.

In this paper, we propose a stable fuzzy logic controller architecture for inverted pendulum,. In the design procedure, we represent the fuzzy system as a Takagi-Sugeno fuzzy model and construct a global fuzzy logic controller by considering each local state feedback controller and a supervisory controller, Unlike usual parallel distributed controller, one can design a global stable fuzzy controller without finding a common Lyapunov function by the proposed method. A simulation is performed to control the inverted pendulum to show the effectiveness and feasibility of the proposed fuzzy controller.

In this paper, merits of both fuzzy and PID controllers are combined. The combined controller is designed such that the tuning of the PID controller is achieved by the basic fuzzy controller via its rule base. The proposed scheme avoids the tuning of PID parameters which is always a time consuming task, difficult to carry out and often poorly done. Computer simulations are made to demonstrate the satisfactory tracking performance of the combined fuzzy-PID controller.

This paper proposes a method which can resolve the problem of exisiting fuzzy PI controller using optimal scaling gains obtained by genetic algorithm. The new method adapt a fuzzy logic controller as a high level controller to perform scaling gain algorithm between two pre-determined sets.

As technology in a computer hardware and software advances, efficient information retrieval from multimedia database gets highly demanded. Recently, it has been actively exploited to retrieve information based on the stored contents. However, most of the methods emphasize on the points which are far from human intuition or emotion. In order to overcome this shortcoming , this paper attempts to apply interactive genetic algorithm to content-based image retrieval. A preliminary result with subjective test shows the usefulness of this approach.

Elementary school pupils frequently have difficult acquiring creative writing skills, i.e. how to develop the idea to be expressed, how to compose the materials of the outcome or contents. In this paper, we focus on the problems of how to support creative writing work to arrange materials in order to formulate ideas as the stories develop. the stories develop. The basic ideas behind the method are that(1) a basic story is automatically generated by GA-based operations and shown to a user as sequences of pictures, (2) IGA(Interactive Genetic Algorithm) is used to evaluate and select a preferred story, (3) the results are combined with previously stored stories using case-based techniques. Based on these ideas, we have developed a computer supported environment for this purpose and conducted related experiments.

This paper describes a design of graphical user interface for a simulated breeding tool with multifield. The term field is used here as a population of visualized individuals that are candidates of selection. Multi-field interface enables the user to breed his/her favorite phenotypes by selection independently in each field, and he/she can copy arbitrary individual into another field. As known on genetic algorithms, a small population likely leads to premature convergence trapped by a local optimum, and migration among plural populations is useful to escape from local optimum. The multi-field user interface provides easy implementation of migration and wider diversity. We show the usefulness of multi-field user interface through an example of a breeding system of 2D CG images.

This paper introduces our three approaches to reduce the burden of human interactive EC operators: (1) improvement of the interface of presenting individuals, (2)improvement of the interface of inputting fitness values, and (3) fast EC convergence. We propose methods to display individuals in order of predicted fitness values by neural networks or Euclidean distance measure for (1), to input quantized fitness values for (2), and to make a new elite by approximating the EC search space with a quadratic function for (3). They are evaluated through simulations and subjective testes, and their effects have shown.

In this paper, an LP problem with convex polyhedral objective coefficients is treated. In the problem, the interactivities of the uncertain objective coefficients are represented by a bounded convex polyhedron (a convex polytope). We develop a computation algorithm of a maxmin achievement rate solution. To solve the problem, first, we introduce the relaxation procedure. In the algorithm, a sub-problem, a bilevel programing problem, should be solved. To solve the sub-problem, we develop a solution method based on a branch and bound method. As a result, it is shown that the problem can be solved by the repetitional use of the simplex method.

In this paper we consider multiobjective linear programming problems with coefficients of the objective functions specified by possibility distributions. Possibly and necessarily efficient solution sets are defined as funny solution sets whose membership grades represent possibility and necessity degrees to which a feasible solution is efficient. Considering efficiency condition and its dual condition in ordinary multiobjective linear programming problem, we propose efficiency test methods based on an extreme ray generation method. Since the proposed methods can be put in the part of a bi-section method, we can develop calculation and methods of the degree of possible and necessary efficiency for feasible solutions.

Fuzzy mathematical programming (FMP) can be treated an uncertainty condition using fuzzy concept. Further, it can be extended to the multiple objective (or goal) programming problem, naturally. But we feel that FMP have some shortcomings such as the fuzzy number in FMP is the one dimesional possibility set, so it can not be represented the relationship between them, and, in spite of FMP includes some (uncertainty) fuzzy paramenters, many alogrithms are only obtained a crisp solution.In this study, we propose a method of FMS. Our method use the scenario approach (or fuzzy random variables) to represent the relationship between fuzzy numbers, and can obtain the fuzzy solution.

DEA(data envelopment analysis) is a non-parametric technique for measuring and evaluating the relative efficiencies of a set of entities with common crisp inputs and outputs. In fact, in a real evaluation problem input and output data of entities often flucturate. These fluctuating data can be represented as linguistic variables characterized by fuzzy numbers. Based on a fundamental CCR model, a fuzzy DEA model is proposed to deal with fuzzy input and output data, Furthermore, a model that extends a fuzzy DEA to a more general case is also proposed with considering the relation between DEA and RA (regression analysis) . the crisp efficiency in CCR modelis extended to an L-R fuzzy number in fuzzy DEA problems to reflect some uncertainty in real evaluation problems.

Although each de Morgan algebra has not always fixed points(centers), it has always fixed cores, the natural extention of fixed points. Fixed cores, of they do not degenerate to fixed points, are Boolean algebras, It is also shown the necessary and sufficient condition a algebra to be a Kleene algebra(fuzzy algebra) is that it has just one fixed core.

In this note by considering the notions of F-polygroups, the product of F-polygroups, F-subpolygroups and (weak) normal F-subpolygroups two questions are given. Then by an example it is shown that the answer of one of the questions (posed in the paper [12]) is in general negative. In other words, the product of two normal F-subpolygroups need not be normal.

It is well known that a Boolean algebra is one of the most important algebra for engineering. A fuzzy algebra, which is referred to also as a Kleen algebra, is obtained from a Boolean algebra by replacing the complementary law in the axioms of a Bloolean algebra with the Kleen's law, where the Kleen's law is a weaker condition than the complementary law. Removal of the Kleen's law from a Kleen algebra gives a De Morgan algebra. In this paper, we generate lattice structures of the above related algebraic systems having finite elements by using a computer. From the result, we could find out a hypothesis that the structure excepting for each element name between a Kleene algebra and a De Morgan algebra is the same from the lattice standpoint.

Logic functions such as fuzzy switching functions and multiple-valued Kleenean functions, that are models of Kleene algebra have been studied as foundation of fuzzy logic. This paper deals with a new kinds of functions-fuzzy switching functions with constants-which have features of both the above two kinds of functions . In this paper, we propose new canonical forms for enumerating them. They are much useful to estimate simply the number of fuzzy switching functions with constants.

An image of a set that produces a multiset from an ordinary set and its extension to fuzzy multisets is considered. For each input element, its image is added to the output regardless whether or not there already exists the same image in the output. theoretical properties such as commutativity of the image with $\alpha$-cut or multiset addition are proved. Generalization to the image by multivariable functions is moreover defined.

For a fuzzy system modeled by a fuzzy hypergraph, two fuzzy similarity measures are proposed : one for the fuzzy similarity between fuzzy sets and the other between elements in fuzzy sets. The propose measures can represent the realistic similarities which can not be given by the existing measures. With and example, it is shown that it can be used in the behavior analysis in an organization.

In dealing with representing knowledge under uncertainty there is a sustain tendency to increase flexibility in order to avoid problems of inconsistency in the knowledge. Many knowledge systems(information retrieval systems, expert system) include hybrid representation models. Funny retrieval systems appear as a complement or as an enrichment of this models. In this paper, we describe dynamic rule modification through situation assessment for uncertainty management.

This paper deals with our proposed fuzzy inference method, in which the fuzzy relation is represented by the membership functions of the antecedent and consequent parts, it is not used any fuzzy composition. The strong point of this method is that the membership function of an inferred conclusion has a simple shape and thus its meaning can be interpreted easily. Firstly, the proposed method is explained, and then it is applied to fuzzy modeling of distributed data.

This paper examined the relations among the multidimensional linear interpolation(MDI) and fuzzy reasoning , and neural networks, and showed that an showed that an MDI is a special form of Tsukamoto's fuzzy reasoning and regularization networks in the perspective of fuzzy reasoning and neural networks, respectively. For this purposes, we proposed a special Tsukamoto's membership (STM) systemand triangular basis function (TBF) networks, Also we verified the condition when our proposed TBF becomes a well-known radial basis function (RBF).

This paper defines Fuzzy Logic Units(FLUs) which are piece wise finite elements in multidimension Euclidean space, and redefines triangular membership functions which are different from those defined in traditional literature. By analyzing FLUs, this paper gives a constructive method of fuzzy rules in fuzzy logic systems based on finite element method. The simulation results of single machine to infinite bus system show the effectiveness of the proposed method in this paper.

In this paper, we propose a new technique of codification. The purpose of this method is to take in consideration the natural language nuances and the fuzziness that characterizes the human reasoning. So, we warranted a means of more flexible encoding that translates as well the linguistic descriptions. Its principle is simple and intuitive. It consists simply in replacing in ambiguous cases, a unique number by an interval. The introduction of the new codification necessitates the elaboration of metric or similarity in order to compare two intervals. This comparison must take in consideration the difference of their size, the remoteness of their center and the width of their intersection. In consequence, we defined three new fuzzy concepts : "fuzzy inclusion degree", "fuzzy resemblance degree," and " fuzzy curve".

In this work, we present a method to deal with the interval valued decision making systems. First, we propose a new type of equality measure based on the Ordered Weighted Averaging (OWA) operator. The proposed equality measure has a structure to render the extreme values of the measure by choosing a suitable weighting vector of the OWA operator. From this property, we derive a bidirectional fuzzy inference network which can be applied for the decisionmaking systems requiring the inverval valued decisions.

The purpose of this paper is to develop a neurofuzzy modeling & inference system which can determine principle dimensions and hull factors in an initial ship design. Neurofuzzy modeling & inference for a hull form design (NeFHull) applies the given input-output data to the fuzzy theory. NeFHull also deals the fuzzificated values with neural networks. NeFHull redefines normalized input-output data as membership functions and executes the fuzzficated information with backporpagation-neural -networks. A hybrid learning algorithms utilized in the training of neural networks and examining the usefulness of suggested method through mathematical and mechanical examples.

This paper proposes a new concept of optimal shortest path algorithm which reduce average vehicle wating time and improve average vehicle speed, Electro sensitive traffic system can extend the traffic cycle when three are many vehicles on the road or it can reduce the traffic cycle when there are small vehicles on the road. But electro sensitive traffic light system doesn't control that kind of function when the average vehicle speed is 10km -20km. Therefore, in this paper to reduce vehicle waiting time we developed design of traffic cycle software tool that can arrive destinination as soon as possible using optimal shortest pass algorithm. Computer simulation result proved 10%-32% reducing average vehicle wating time and average vehicle speed which can select shortest route using built in G.P.S. vehicle is better than not being able to select shortest route function.

Nowadays, with increasing many vehicles on restricted roads, the conventional traffic light creates prove startup-delaytime and end-lag-time. The conventional traffic light loses the function of optimal cycle. And so, 30∼45% of conventional traffic cycle is not matched to the present traffic cycle. In this paper proposes electrosensitive traffic light using fuzzy look up table method which will reduce the average vehicle waiting time and improve average vehicle speed. Computer simulation results prove that reducing the average vehicle waiting time which proposed considering passing vehicle length for optimal traffic cycle is better than fixed signal method which doesn't consider vehicle length.

In this paper, we propose a new design method of an optimal fuzzy logic controller using co-evolutionary concept. In general, it is very difficult to find optimal fuzzy rules by experience when the input and/or output variables are going to increase. Futhermore proper fuzzy partitioning is not deterministic ad there is no unique solution. So we propose a co-evolutionary method finding optimal fuzzy rules and proper fuzzy membership functions at the same time. Predator-Prey co-evolution and symbiotic co-evolution algorithms, typical approaching methods to co-evolution, are reviewed, and dynamic fitness landscape associated with co-evolution is explained. Our algorithm is that after constructing two population groups made up of rule base and membership function, by co-evolving these two populations, we find optimal fuzzy logic controller. By applying the propose method to a path planning problem of autonomous mobile robots when moving objects applying the proposed method to a pa h planning problem of autonomous mobile robots when moving objects exist, we show the validity of the proposed method.

In this paper, a controller using fuzzy-genetic algorithms is proposed for pat-tracking of WMR. A fuzzy controller is implemented so as to adjust appropriate crossover rate and mutation rate. A genetic algorithms is also implemented to have adaptive adjustment of control gain during optimizing process. To check effectiveness of this algorithms, computer simulation is applied.

The success of a fuzzy logic control system solving any given problem critically depends on the architecture of th network. Various attempts have been made in optimizing its structure its structure using genetic algorithm automated designs. In a regular genetic algorithm , a difficulty exists which lies in the encoding of the problem by highly fit gene combinations of a fixed-length. This paper presents a new approach to structurally optimized designs of a fuzzy model. We use a messy genetic algorithm, whose main characteristics is the variable length of chromosomes. A messy genetic algorithms used to obtain structurally optimized fuzzy models. Structural optimization is regarded important before neural network based learning is switched into. We have applied the method to the exampled of a cart-pole balancing.

This paper presents a new clustering algorithm called FCM algorithm for the design of fuzzy controller. FCM is an extended version of FCM(Fuzzy c-Means) algorithm and can estimate the number of clusters automatically and give membership grades $u_{ik}$ suitable for making fuzzy control rules. This paper also shows an example of its application to the line pursuit control of a car.

This paper presents an application of genetic algorithms(GAs) for optimal configuration of distribution network. Three problems have been used to show how genetic algorithms are modified and applied. Solutions to the problems are found by minimizing the cost function which is directly related with balancing the loads. Simulation results show that genetic algorithms are technically feasible if they are tailored to meet the needs of real problems.

Proposed here is a parallel genetic algorithm accompanied with intermittent migration among subpopulations. It is intended to maintain diversity in the population for a long period . This method was applied to finding out the global maximum of some multimodal functions for which no other methods seem to be useful . Preferable results and their detailed analysis are also presented.

A methodology based on the concept of variable string length GA(VGA) is developed for determining automatically the number of hyperplanes and their appropriate arrangement for modeling the class boundaries of a given training data set in RN. The genetic operators and fitness functionare newly defined to take care of the variability in chromosome length. Experimental results on different artificial and real life data sets are provided.

A Sequencing problem is one to find an ordered sequence of some entities which maximizes (or minimize) some objective function. This paper introduces an new type of sequencing problems, named a Sequencing problem with fuzzy preference relation is previded for the evaluation of the quality of sequences, It presents how such a problem can be formulated in the point of objective function. In addition, it proposes a genetic algorithm applicable to such a sequencing problem.

Fuzzy control has been used successfully in many practical applications. In traditional methods, experience and control knowledge of human experts are needed to design fuzzy controllers. However, it takes much time and cost. In this paper, an automatic design method for fuzzy controllers using genetic algorithms is proposed. In the method, we proposed an effective encoding scheme and new genetic operators. The maximum number of linguistic terms is restricted to reduce the number of combinatorial fuzzy rules in the research space. The proposed genetic operators maintain the correspondency between membership functions and control rules. The proposed method is applied to a cart centering problem. The result of the experiment has been satisfactory compared with other design methods using genetic algorithms.

We introduce the concept of a bi-population scheme for real-coded GAs consisting of an explorer sub-Ga and an exploiter sub-GA. The explorer sub-GA mainly performs global exploration of the search space, and incorporates a restart mechanism to help avoid being trapped at local optima. The exploiter sub-GA performs exploitation of fit local areas of the search space around the neighborhood of the best-so-far solution. Thus the search function of the algorithm is divided. the proposed technique exhibits performance significantly superior to standard GAs on two complex highly multimodal problems.

In this paper, lower and upper possibility distributions are identified to reflect two extreme opinions in portfolio selection problems based on upper and lower possibility distributions are formalized as quadratic programming problems. Portfolios for compromising two extreme opinions from upper and lower possibility distributions and balancing the opinions of a group of experts can be obtained by quadratic optimization problems, respectively.

This paper investigates a facility location problem where there are possible demand points with demand occuring probabilites and actual distances between these points and the facility site to be determined are ambiguous, Further we define the fuzzy goal with respect to the maximum value among the actual distances between demand points and the facility. We determine the site of facility maximizing the minimal satisficing degree under the chance constraint. We propose the geometric algorithm to find this optimal site.

In this paper, we consider a bi-objective vehicle routing problem to minimize total distance traveled and maximize minimum integrated satisfaction level of selecting desirable routes in an fuzzy graph. The fuzzy graph reflects a real delivery situation in which there are a depot, some demand points, paths linking them, and distance and integrated satisfaction level are associated with each route. For solving the vi-objective problem we introduce a concept of routing vector and define non-dominated solution for comparing vectors. An efficient algorithm involving a selection method of non-dominated solutions based on DEA is proposed for the vehicle routing problem with rigid distance and integrated satisfaction level.

This paper makes a study of the Shapley value in cooperative fuzzy games, games with fuzzy coalitions, which enable the representation of players' participation degree to each coalition. The Shapley value has so far been introduced only in an class of fuzzy games where a coalition value is not monotone with respect to each player's participation degree. We consider a more natural class of fuzzy games such that a coalition value is monotone with regard to each player's participation degree. The properties of fuzzy games in this class are investigated. Four axioms of Shapley functions are described and a Shapley function of a fuzzy fame in the class is given.

In this paepr, we proposed a new way to make a pre-pruned searching tree for GO game moves from macroscopic strategy described in linguistic expression. The strategy was a consequence of macroscopic recognition of GO game situations. The definitions of fuzzy macroscopic strategy, fuzzy tactics and tactical sequences using fuzzy set are shown and its family, so called "multi level fuzzy set". Some examples are also shown.

This paper presents an efficient method which estimates the systems with unmodeled dynamics using D-L networks. This method is applied for estimating the system with unmodeled dynamics from only input-output information , so it can exclude additional procedure for system description and reduce the computational burden required for real-time estimation. Higher convergence speed is achieved in this manner in comparison with widely-used conventional methods.

This paper presents an explanation regarding on-line identification of a fuzzy system. The fuzzy system to be identified is assumed to be in the type of singletion consequent parts and be represented by a linear combination of fuzzy basis function (FBF's). For on-line identification, squared-cosine (SCOS) fuzzy basis function is introduced to reduce the number of parameters to be identified and make the system consistent and differentiable. Then the parameters of the fuzzy system are identified on-line by the gradient search method. Finally, a computer simulation is performed to illustrate the validity of the suggested algorithms.

Most of systems has nonlinearity . And also accurate modelings of these uncertain nonlinear systems are very difficult. In this paper, a fuzzy modeling technique for the stabilization control of an IP(inverted pendulum) system with nonlinearity was proposed. The fuzzy modeling was acquired on the basis of ANFIS(Adaptive Neuro Fuzzy Infernce System) which could learn using a series of input-output data pairs. Simulation results showed its superiority to the PID controller. We believe that its applicability can be extended to the other nonlinear systems.

Fuzzy random variables with piecewise linear membership functions are introduced from a practical viewpoint. The estimation of the expected values of these fuzzy random variables is also discussed and statistical application is denonstratied by using a real data set.

A design method of rule -based fuzzy modeling is presented for the model identification of complex and nonlinear systems. The proposed rule-based fuzzy modeling implements system structure and parameter identification in the efficient form of " IF..., THEN,," statements. using the theories of optimization and linguistic fuzzy implication rules. The improved complex method, which is a powerful auto-tuning algorithm, is used for tuning of parameters of the premise membership functions in consideration of the overall structure of fuzzy rules. The optimized objective function, including the weighting factors, is auto-tuned for better performance of fuzzy model using training data and testing data. According to the adjustment of each weighting factor of training and testing data, we can construct the optimal fuzzy model from the objective function. The least square method is utilized for the identification of optimum consequence parameters. Gas furance and a sewage treatment proce s are used to evaluate the performance of the proposed rule-based fuzzy modeling.

This paper presents several softcomputing techniques such as neural networks, fuzzy logic and genetic algorithms : Neural networks as brain metaphor provide fundamental structure, fuzzy logic gives a possibility to utilize top-down knowledge from designer, and genetic algorithms as evolution metaphor determine several system parameters with the process of bottom up development. With these techniques, we develop a pattern recognizer which consists of multiple neural networks aggregated by fuzzy integral in which genetic algorithms determine the fuzzy density values. The experimental results with the problem of recognizing totally unconstrained handwritten numeral show that the performance of the proposed method is superior to that of conventional methods.

The paper is part of an investigation by the authors on development of a knowledge based frame work for multimodal medical image in collaboration with the All India Institute of Medical Science, new Delhi. After presenting the key aspects of the Dempster-Shafer Evidence theory we have presented implementation of registration and fusion of T₁and T₂ weighted MR images and CT images of the brain of an Alzheimer's patient for minimising the uncertainty and increasing the reliability for dianostics and therapeutic planning.

This paper presents a systematic developement of a formal approach to inference in approximate reasoning. We introduce some measures of similarity and discuss their properties. Using the concept of similarity index we formulate two methods for inferring from vague knowledge. In order to illustrate the effectiveness of the proposed technique we use it to develop a vowel recognition system.

Recently we extended the fuzzy model for rule based systems incorporating an importance factor for each rule. The model permits for both unrestricted as well as non-negative importance factors. We use this extended model to design a fuzzy rule based classifier system which uses both the firing strength of the rule and the importance factor to decide the class label. The effectiveness of the scheme is established using several data sets.

The probabilistic relaxation(PR) scheme based on the conditional probability and probability space partition has the important property that when its compatibility coefficient matrix (CCM) has uniform components it can classify m-dimensional probabilistic distribution vectors into different classes. When consistency or inconsistency measures have been defined, the properties of PRs are completely determined by the compatibility coefficients among labels of labeled objects and influence weight among labeled objects. In this paper we study the properties of PR in which both compatibility coefficients and influence weights are uniform, and then a learning rule for such PR system is derived. Experiments have been performed to verify the effectiveness of the learning rule.

This paper proposes compression of image data using neural networks based on conjugate gradient method and dynamic tunneling system. The conjugate gradient method is applied for high speed optimization .The dynamic tunneling algorithms, which is the deterministic method with tunneling phenomenon, is applied for global optimization. Converging to the local minima by using the conjugate gradient method, the new initial point for escaping the local minima is estimated by dynamic tunneling system. The proposed method has been applied the image data compression of 12 ${\times}$12 pixels. The simulation results shows the proposed networks has better learning performance , in comparison with that using the conventional BP as learning algorithm.

Mathematical morphology (MM) has been introduced as a powerful tool for studying the geometrical properties of images, MM is a good approach to digital image processing , which is based on the shape feature. The MM operators such as dilation, erosion, closing and opening have been applied successfully to image noise reduction. The MM filters can easily filter the noise when the noise factors are known. However it is very difficult to reduce the noise when images are ambiguous, because the boundary between the noise and object is vague. In this paper, we propose a new method to reduce noise from ambiguous images by using Fuzzy Mathematical Morphology (FMM) operators. Performance evaluation via simulations show that the FMM filters efficiently reduce the image noise. Furthermore, the FMM filters show a good performance compared with the conventional filters.

Due to the temporal spatial correlation of the image sequence, the motion vector of a block is highly related to the motion vectors of its adjacent blocks in the same image frame. If we can obtain useful and enough information from the adjacent motion vectors, the total number of search points used to find the motion vector of the block may be reduced significantly. Using that idea, an efficient fuzzy prediction search (FPS) algorithm for block motion estimation is proposed in this paper. Based on the fuzzy inference process, the FPS can determine the motion vectors of image blocks quickly and correctly.

This paper describes an in-line lossless data compression method using adaptive binary arithmetic coding. To achieve better compression efficiency , we employ an adaptive fuzzy -tuning modeler, which uses fuzzy inference to deal with the problem of conditional probability estimation. The design is simple, fast and suitable for VLSI implementation because we adopt the table -look-up approach. As compared with the out-comes of other lossless coding schemes, our results are good and satisfactory for various types of source data.

Mechanical valve is one of the most widely used implantable artificial organs, Since its failure (mechanical failures and thrombosis to name two representative example) means the death of patient, its reliability is very important and early noninvasive detection is essential requirement . This paper will explain the method to detect the thrombosis formation by spectral analysis and neural network. In order quantitatively to distinguish peak of a normal valve from that of a thrombotic valve, a 3 layer backpropagation neural network, which contains 7,000 input nodes, 20 hidden layer and 1output , was employed. The trained neural network can distinguish normal and thrombotic valve with a probability that is higher than 90% . In conclusion, the acoustical spectrum analysis coupled with a neural network algorithm lent itself to the noninvasive monitoring of implanted mechanical valves. This method will be applied to be applied to the performance evaluation of other implantable rtificial organs.