Neural fuzzy system

Lin, C.-T., and Lee, G. (1996), Neural Fuzzy Systems A Neuro-Fuzzy Synergism to Intelligent Systems, Prentice HaU, Upper Saddle River, NJ. 

Analyses of the data generated by the molecular dynamics method is an important part of the overall simulation. We have developed a number of very useful tools for the examination of the results such as to extract the maximum information. These include high resolution spectral estimators and the application of neural/fuzzy systems. A more detailed description of the methods and examples of application can be found in a recent review article [11]. 

Neural networks and adaptive fuzzy systems are model free estimators [12]. This quality makes it possible to find relationships between a series of examples and the desired results [13]. For example, one might be interested in predicting the long-time behavior of a polymer chain, say to the millisecond. By providing a number of examples of the dynamics on a short time scale, say picoseconds, a neural network is able to map out the important relationships and thus make interpolations and extrapolations to times that were not given as examj. Furthermore, in the same manner, the size discrepancy between most emulations and their experimental counterparts, can be dosed by using the predictive power of neural/fuzzy systems. 

Many types of combinations between fuzzy systems and neural networks have been proposed and studied. In what follows, we use the definitions and classification proposed by Detlef Nauck, from the Department of Computer Science, Technical University of Braunschweig, GermanyA neuro-fuzzy or neural fuzzy system is a combination of neural networks and fuzzy systems in such a way that neural networks are used to determine parameters of fuzzy systems. The main intention of a neuro-fuzzy approach is to create or improve a fuzzy system by means of neural network methods. The system should always be interpretable in terms of fuzzy if-then rules. A fuzzy... 

Sometimes fuzzy logic controllers are combined with pattern recognition software such as artificial neural networks (Kosko, Neural Networks and Fuzzy Systems, Prentice Hall, Englewood Cliffs, New Jersey, 1992). 

The architecture of an ANFIS model is shown in Figure 14.4. As can be seen, the proposed neuro-fuzzy model in ANFIS is a multilayer neural network-based fuzzy system, which has a total of five layers. The input (layer 1) and output (layer 5) nodes represent the descriptors and the response, respectively. Layer 2 is the fuzzification layer in which each node represents a membership. In the hidden layers, there are nodes functioning as membership functions (MFs) and rules. This eliminates the disadvantage of a normal NN, which is difficult for an observer to understand or to modify. The detailed description of ANFIS architecture is given elsewhere (31). 

Here, the premise is described by a membership function for the linguistic variable high and the function for the detection limit is the sum of the blank signal, y, and three times the standard deviation of the blank signal, Sg, (cf. Eq. (4.3)). Optimization of the parameters in the premise part of the rules is adaptively done by combining the fuzzy rule-based system with a neural network. Consider an adaptive neuro-fuzzy system with two inputs, and... 

Narendra, K.S. 1992. Adaptive control of dynamical systems using neural networks. In Handbook of Intelligent Control Neural, Fuzzy and Adaptive Approaches. D.A. White and D. A. Sofge, Eds. pp. 141-184. Van Nostrand Reinhold, New York. 

Dengyou Xia. 2007. Fire Risk Evaluation Model of High-Rise Buildings Based on Multilevel BP Neural Network. Fuzzy Systems and Knowledge Discovery 24-27. 

Artifical Neural Networks Computer Algorithms Databases Fourier Series Functional Analysis Fuzzy Sets, Fuzzy Logic, and Fuzzy Systems Statistics, Bayesian Wavelets... 

Martin del Brio B, Sanz Molina A (2007) Artificial neural networks and fuzzy systems (in Spanish), 3rd edn. Alfaomega Editor Group, Mexico, pp 289-292... 

When applying intelligent methodologies such as fuzzy systems, neural networks, ANFIS, etc., one of the main difficulties is the tuning of their parameters, which are the key to the success of these methodologies. Their parameters vary depending on the complexity of the problem and the method used to find the solution and in some cases, they stem from our own ability to conceptualize the problem itself, taking into account, the inputs of the system and the expected output values. 

Kosko B. Neural Networks and Fuzzy Systems, Englewood Cliffs Prentice-HaU, Inc. 1992. 

Neural Networks and Fuzzy Systems Neuron cell... 

Recurrent Neural Networks Fuzzy Systems Design Example Genetic Algorithms... 

The performance of the fuzzy control system can be further enhanced by fine-tunning the parameters of the fuzzy system (membership functions, rules, etc) by using a neural network or a global optimization scheme. First results in this direction, using the Particle Swarm Optimization method, have been presented in Marinakis et al. 2008. Furthermore, from the practical point of view a big deficiency of fuzzy controllers are... 

---