The proceedings contains 142 papers. Topics discussed include risk and reliability, uncertainty analysis, control, fuzzy data analysis, civil engineering applications in Japan, structural engineering applications, pat...
详细信息
The proceedings contains 142 papers. Topics discussed include risk and reliability, uncertainty analysis, control, fuzzy data analysis, civil engineering applications in Japan, structural engineering applications, patterns and fuzzy reasoning, uncertainty in engineering designs, fuzzy decision analysis, neural networks, genetic methods, approximate reasoning, fuzzy informationprocessing, optimization, uncertainty in logic, transportation and scheduling, fuzzy sets and systems in signal processing, numerical methods, uncertainty in databases, uncertainty in information, knowledge bases, and complexity.
A nonlinear 5-layer artificial neural autoencoder network for image data compression is constructed and trained using the back propagation algorithm and medical CT-images. The influence of linear and nonlinear pre/pos...
详细信息
A nonlinear 5-layer artificial neural autoencoder network for image data compression is constructed and trained using the back propagation algorithm and medical CT-images. The influence of linear and nonlinear pre/postprocessing operations is studied as well as an alternative compression scheme. Important implementational issues of neural networks are addressed as well as autoencoder issues. One of the results of this work is a compression/decompression tool that provides maximum flexibility and can be used independent from the training environment.
In this paper, we attempt to develop a practical decision support system for the damage assessment of structural corrosion. This system aims to aid inexperienced inspectors to judge whether a certain bridge should be ...
详细信息
In this paper, we attempt to develop a practical decision support system for the damage assessment of structural corrosion. This system aims to aid inexperienced inspectors to judge whether a certain bridge should be repaired or not. For this purpose, it is attempted to apply the neural network technique for the damage assessment. The learning ability of the neural network is useful to save the working time and load necessary in the inspection and analysis.
In this paper, a first stage for a fuzzy neural active controller is developed. Neural network technology is utilized in the real-time identification of the mode of vibration of any structural system under seismic loa...
详细信息
In this paper, a first stage for a fuzzy neural active controller is developed. Neural network technology is utilized in the real-time identification of the mode of vibration of any structural system under seismic loading. The parallel processing nature of neural networks fit the nature of multi-degree-of-freedom systems. Such a property would result in considering the whole structure and eliminating the need for problem reduction. The proposed mode identifier is a multi-layer neural network. The pattern identification process is considered a state evaluation stage that is required as a first step in the development of a fuzzy neural active controller. Such a controller would select a suitable fuzzy control strategy based on the identified displacement pattern.
Functional equivalence between multilayered neural networks (MNN's) and fuzzy systems with a singleton fuzzifier, a product inference, a centroid defuzzifier, and a sinusoidal membership function is discussed in t...
详细信息
Functional equivalence between multilayered neural networks (MNN's) and fuzzy systems with a singleton fuzzifier, a product inference, a centroid defuzzifier, and a sinusoidal membership function is discussed in this paper. First, a normalized structure of MNN's is given in terms of input-output equations of MNN's. Fuzzy basis function network (FBFN) expansions of multi-input and single output (MISO) fuzzy systems are then given in order to describe the input-output relationships of fuzzy systems. Sinusoidal membership functions are introduced for fuzzy systems with a graded value over [0, 1]. Functional equivalence between the two systems is analytically shown. Finally, universal approximation capability of FBFN's is briefly discussed.
For automatic rule extraction from a set of input-output data examples, decision tree generating methods such as ID3 and Fuzzy ID3 play a major role. These methods, however, are difficult to apply when there is a tend...
详细信息
For automatic rule extraction from a set of input-output data examples, decision tree generating methods such as ID3 and Fuzzy ID3 play a major role. These methods, however, are difficult to apply when there is a tendency for the examples to change dynamically. This paper presents a new method for adaptive rule extraction with the Fuzzy Self-Organizing Map and the results of the simulations to present the effectiveness by a comparison with other methods such as RBF and GA. We got the result that our method is superior to other methods for automatic and adaptive rule extraction.
For planning to construct dam and management of constructing, it is so important to quantitatively evaluate grouting works on improvement of the watertightness. This paper presents that evaluation system can be modele...
详细信息
For planning to construct dam and management of constructing, it is so important to quantitatively evaluate grouting works on improvement of the watertightness. This paper presents that evaluation system can be modeled based on neural network using resistivity data measured in site. Sensitivity analysis is carried out using this system. From above analysis, an experimental process and sense of engineer can be described by the model of the relationship between the variation of resistivity and the states of grouting put into rock mass.
In this paper we introduce a two-level modular neuro-fuzzy network based on incentive games where the modules are organized as autonomous local optimizers in a leader-follower game hierarchy. Incentive-reaction pairs ...
详细信息
In this paper we introduce a two-level modular neuro-fuzzy network based on incentive games where the modules are organized as autonomous local optimizers in a leader-follower game hierarchy. Incentive-reaction pairs are used as a measure for the capacity and responsiveness assessment of each follower module. Learning within the follower modules is performed in a traditional error-based manner (e.g., backpropagation) The allocation of targets and incentives to each follower module, on the other hand is independent of connection weights;incentive games are used for that purpose. Two important advantages of the new architecture are its physically significant follower module outputs and the context-based enhancement it makes to backpropagation.
A number of recently proposed methods of computational intelligence, specifically fuzzy-neural networks, fuzzy classifiers generated by fuzzy clustering, and fuzzy data analysis methods based on evolutionary algorithm...
详细信息
A number of recently proposed methods of computational intelligence, specifically fuzzy-neural networks, fuzzy classifiers generated by fuzzy clustering, and fuzzy data analysis methods based on evolutionary algorithms are outlined giving emphasis to the specific characteristic of methods which differentiate them from others.
暂无评论