Showing papers on "Neuro-fuzzy published in 1989"

Fuzzy control and fuzzy systems

Abstract: From the Publisher: Presents extensive and updated material concerned with the methodology and algorithms of fuzzy sets considered mainly in the context of control engineering and system modeling and analysis. Offers information on fuzzy sets and the concept of fuzzy control, reviewing selected applications and their origin. Discusses design aspects and theoretical developments in the design of fuzzy controllers. Includes comprehensive coverage of the paradigms and algorithms of fuzzy modeling.

Knowledge representation in fuzzy logic

Abstract: The author presents a summary of the basic concepts and techniques underlying the application of fuzzy logic to knowledge representation. He then describes a number of examples relating to its use as a computational system for dealing with uncertainty and imprecision in the context of knowledge, meaning, and inference. It is noted that one of the basic aims of fuzzy logic is to provide a computational framework for knowledge representation and inference in an environment of uncertainty and imprecision. In such environments, fuzzy logic is effective when the solutions need not be precise and/or it is acceptable for a conclusion to have a dispositional rather than categorical validity. The importance of fuzzy logic derives from the fact that there are many real-world applications which fit these conditions, especially in the realm of knowledge-based systems for decision-making and control. >

A description of the dynamic behavior of fuzzy systems

Abstract: An approach is presented for analyzing the global behavior of a fuzzy dynamical system that applies the concept and method of cell-to-cell mapping to obtain the evolving trend of the states of a fuzzy dynamical system. The behavior of the fuzzy system is characterized by equilibria, periodic motions, and their domain of attractions. Min-max operation accumulates the fuzziness of a fuzzy system in every step of iterations and makes the state evolution obscure. The proposed method transforms a given fuzzy mapping to at Z-to-Z mapping and does not accumulate fuzziness. Both the real and fuzzy initial state response analyses are discussed. An inverted pendulum controlled by a fuzzy controller is analyzed to illustrate the validity of the method. >

Qualitative Spatial Reasoning: A Semi-quantitative Approach Using Fuzzy Logic

Abstract: Qualitative reasoning is useful as it facilitates reasoning with incomplete and weak information and aids the subsequent application of more detailed quantitative theories. Adoption of qualitative techniques for spatial reasoning can be very useful in situations where it is difficult to obtain precise informationand where there are real constraints of memory, time and hostile threats. This paper formulates a computational model for obtaining all induced spatial constraints on a set of landmarks, given a set of approximate quantitative and qualitative constraints on them, which may be incomplete, and perhaps even conflicting.

Neural Network Models In Simulation: A Comparison With Traditional Modeling Approaches

Abstract: Neural models are enjoying a resurgence in systems research primarily due to a general interest in the connectionist approach to modeling in artificial intelligence and to the availability of faster and cheaper hardware on which neural net simulations can be executed. We have experimented with using a multi-layer neural network model as a simulation model for a basic ballistics model. In an effort to evaluate the efficiency of the neural net implementation for simulation modeling, we have compared its performance with traditional methods for geometric data fitting such as linear regression and surface response methods. Both of the latter approaches are standard features in many statistical software packages. We have found that the neural net model appears to be inadequate in most respects and we hypothesize that accuracy problems arise, primarily, because the neural network model does not capture the system structure characteristic of all physical models. We discuss the experimental procedure, issues and problems, and finally consider possible future research directions.

Distribution system planning with fuzzy models and techniques

Abstract: Fuzzy modelling of power systems is proposed, which takes into account the qualitative aspects and vagueness or uncertainty that do not have a random nature and therefore cannot be modelled by a probabilistic approach An operative procedure to derive fuzzy load diagrams associated with different types of energy consumption is presented The feasibility of the development of fuzzy linear programming models for power system planning is demonstrated DC and an AC fuzzy load flow formulations are developed and the quality and usefulness of the results obtained with them is shown using an example

Ship Auto-navigation Fuzzy Expert System (SAFES)

Abstract: The growing demand for automation in ship navigation and for various assessments of harbors and waterways lends increasing importance to so-called fast-time simulation techniques and to ship simulator experiments that include man-in-the-loop simulation. The authors present an automatic navigation system known as SAFES that they have developed combining fuzzy theory and an expert-system, and that is applicable to the automation of ship navigation and to the modeling of the navigator in fast-time simulation. SAFES can treat various ship operations in both open seas and restricted waterways, as well as scenarios involving multiship encounters. It is pointed out that the expert system approach is essential for dealing with the complexities of the navigator's decision-making process, which cannot easily be handled in a sequential programming style such as one using FORTRAN. Each function of the navigator's orders and the helmsman's behavior can, however, also be represented by a non-fuzzy description. But once a general-purpose fuzzy controller/reasoner is prepared that passes one fuzzy output from two fuzzy inputs, almost all functions can be easily handled by fuzzy control/reasoning because most functions of human operation are originally characterized by only fuzzy descriptions.

Fuzzy systems identification

Abstract: The paper deals with fuzzy system identification. A general identification approach for multi-input/single-output fuzzy systems is presented, which includes structure identification, parameter (fuzzy relation) estimation, and the associated self-learning algorithm. Zadeh's possibility distribution plays an important role in identification and the use of fuzzy models thus constructed. Numerical examples are provided which show the advantages of the proposed identification algorithm and the effectiveness of the self-learning algorithm. Comparison shows that the proposed method can produce the fuzzy model with higher accuracy than previously achieved in other work. In the application example, the proposed identification approach has been used to construct fuzzy models for a fluidised catalytic cracking unit in a big refinery. The resultant fuzzy models are accurate enough for industrial application purpose.

Fuzzy connectionist expert systems

Abstract: Summary form only given, as follows. A fuzzy connectionist expert system with learning capabilities is described. The system uses a recruitment-of-cells learning algorithm for knowledge acquisition, and also allows the translation of rules into an equivalent connectionist network. The system contains conventional expert system features such as explanation and consultation facilities. Tests have been conducted to see its capabilities of learning knowledge bases of classification domains from examples, and in all cases it has been compared favorably against existing conventional systems. >

On the use of neural networks and fuzzy logic in speech recognition

Abstract: A rule-based phoneme recognition method is proposed. This method uses neural networks for acoustic feature detection and fuzzy logic for the decision procedure. Rules for phoneme recognition are prepared for each pair of phonemes (pair-discrimination rules). Recognition experiments were performed using Japanese city names uttered by two male speakers. About 80% of the errors occurring in conventional template matching, which the discrimination rules were designed to recover, were in fact recovered (an improvement in recognition rate of 4.0 to 8.0%). This confirms the effectiveness of the proposed method. >

On training of artificial neural networks

Abstract: A theory and methodology are presented for training artificial neural networks in a general setting. Starting with defining general concepts, and analyzing associated properties of artificial neural networks, the authors formalize, categorize, and characterize artificial neural networks from a system point of view. They focus on the analysis aspect of artificial neural nets to address and investigate trainability and representability; on the synthesis aspect of artificial neural nets to provide design principles to the systems; and on the algorithmic aspect of the artificial neural nets to develop an effective and efficient learning paradigm. >

A fuzzy expert database system

Abstract: Fuzzy concepts always exist in much of human reasoning as well as decision making. This paper presents a fuzzy expert database system which is an integration of a fuzzy expert system building tool called SYSTEM Z-II and a database management system called Rdb/VMS. This system is able to extract fuzzy data and terms stored in a database and used in the fuzzy reasoning in an expert system. It can also retrieve information by fuzzy database-queries which are generated by the expert system automatically. Many expert systems in different domain areas such as decision making can be constructed. Sample applications are described to demonstrate the flexibility and power of this system. The fuzzy query language defined and used in the system can also be used independently as a fuzzy enquiry tool in database applications.

Two approaches for information retrieval through fuzzy associations

Abstract: Two approaches for the formulation of information through fuzzy associations are presented. A fuzzy association is introduced as a fuzzy relation defined on a set of indices to a database. One approach is the extension of fuzzy indices to a database using fuzzy associations. A fuzzy association is a generalization of the concept of fuzzy thesauri. An algorithm for fuzzy information retrieval based on this approach is developed. The other approach represents the retrieval process as a block diagram. Maximum and minimum operations are used instead of the ordinary sum and product operations on the diagram. Studies of advanced indexing, such as the clustering of articles, are represented as feedback on the diagram. Properties of fuzzy information retrieval, such as level fuzzy sets and set operations on responses of the retrieval system, are discussed using the diagram representation. >

Hybridized frame inference and fuzzy reasoning system and method

Abstract: The invention is directed to an inference mechanism to execute human logical and fuzzy thought including intuitive thought. The mechanism consists of two independent engines. One engine is a frame inference engine based on human logical thought. The other engine is a fuzzy reasoning engine base on human fuzzy thought. The mechanism utilizes a common domain control to connect the two engines. After reading in the domain by the fuzzy reasoning engine, the engine draws inferences and writes the fuzzy reasoning result. The conventional inference engine operates in accordance with the results in the domain. For intuitive thought, for example, a procedure (model) is introduced for realizing an operation or a situation of an object. The frame engine commences execution of the procedure (model). The procedure (model) predicts objectives states under a present condition and outputs a predicted performance value requiring fuzzy reasoning on the domain.

Fuzzy data communication system

Abstract: A fuzzy data communication system includes a first fuzzy computer storing fuzzy functions and rules, a transmitter compiling the fuzzy functions and rules stored in the first computer into a message to transmit the message, a receiver receiving the transmitted message to decompile fuzzy functions and rules from the received message, and a second fuzzy computer implementing a fuzzy inference based on the decompiled fuzzy functions and rules.

A new method for approximate reasoning

Abstract: A method based on the revision principle for approximate reasoning in fuzzy-valued logic and fuzzy-linguistic-valued logic is introduced. Two kinds of fuzzy truth as well as the two types of fuzzy linguistic truth, which are constructed by the combination of the former, are distinguished. The revision principle for approximate reasoning in fuzzy-valued logic is introduced. The method is extended to approximate reasoning in fuzzy-linguistic-valued logic. With the unification of universe of discourse, the method can be used for fuzzy linguistic concept inference. >

Handling knowledge in high order neural networks: the combinatorial neural model

Abstract: Summary form only given, as follows. A description is given of the combinatorial neural model, a high-order neural network suitable for classification tasks. The model is based on fuzzy set theory, neural sciences studies, and expert knowledge analysis results. It presents interesting properties such as modularity, explanation capacity, knowledge and data representation, high speed of training, incremental learning, generalization capacity, processing of uncertain and incomplete data, and ability to reason nonmonotonically when representing only relevant evidence, and graceful decay. >

A fuzzy expert system shell using both exact and inexact reasoning

Abstract: This paper presents a comprehensive expert system shell which can deal with both exact and inexact reasoning. A prototype of this proposed shell, code named as SYSTEM Z-IIe, has been implemented successfully. It is a rule-based system which employs fuzzy logic and numbers for its reasoning. Two basic inexact concepts, fuzziness and uncertainty, are both used and distinct from each other clearly in the system. Moreover, these two concepts have been built into two levels for inexact reasoning, i.e. the level of the rules and facts, and the level of the values of the objects of these rules and facts. Other features of Z-IIe include multiple fuzzy propositions in rules and dual fact input mechanisms. It also allows any combinations of fuzzy and normal terms and uncertainties. Fuzzy numeric comparison logic control is also available for the rules and facts. Its natural language interface which uses English with restricted syntax improves the efficiency of knowledge engineering. Z-IIe is also coupled to a Database Management System for supplying facts from existing databases if appropriate. All these features can be combined to build very powerful expert systems and are illustrated by an example.

Fuzzy sets and their applications to artificial intelligence

Abstract: Publisher Summary This chapter discusses some of the basic theories of fuzzy sets. It demonstrates that various significant problems in the realm of uncertainty can be dealt with the use of fuzzy sets theory. This theory is aimed at the development of a body of concepts and techniques for dealing with sources of uncertainty or imprecision that are nonstatistical in nature. A similarity exists between the concepts of fuzziness and probability. The problems in which they are used are similar and even identical. These are problems in which indeterminacy is encountered because of random factors, inexact knowledge, or the theoretical impossibility of obtaining exact solutions. The chapter also discusses the relation of fuzziness and fuzzy statistics in the form of typicality theory, which is essential to the applicability of fuzzy set theory in artificial intelligence and expert systems. Using fuzzy logic, the expert system accepts vague data, compares it to all the rules in its memory simultaneously, and assigns each rule a weight.