China 
Africa 
Explore chapters and articles related to this topic
Design Aspects of Fuzzy Systems and Fuzzy Logic Applications
Published in K. Sundareswaran, A Leaner’s Guide to Fuzzy Logic Systems, 2019
With a limited information base, however, it is possible to construct a fuzzy logic system. The information need not be exact. Existing knowledge can be readily expressed in linguistic form and a fuzzy logic system can be constructed. The fuzzy logic system can then be activated for numerous inputs and corresponding outputs can be determined. This simple input-output pattern can now be used to train a neural network. A neural network, trained in this way, is termed a “neuro-fuzzy system”. Thus, the symbolic representation in fuzzy logic and the numerical processing of neural networks are integrated to form the powerful tool of neuro-fuzzy systems. In such a coupling, common characteristics of both schemes inherently exist. Neuro-fuzzy approaches have been successfully used in many applications, such as parameter estimation, process control and forecasting.
Fuzzy logic systems
Published in A. W. Jayawardena, Environmental and Hydrological Systems Modelling, 2013
Neuro-fuzzy (or fuzzy-neuro) refers to the combination of artificial neural networks and fuzzy logic. Hybrid artificial intelligence, first proposed by Jang (1993), synergizes the human-like reasoning style of fuzzy systems with the connectionist structure of artificial neural networks. Neuro-fuzzy systems are universal approximators with the ability to incorporate ‘IF–THEN’ rules. Fuzzy models cannot learn from data but are interpretable, whereas neural networks can learn from data and are accurate but not easily interpretable. The neuro-fuzzy modelling research field can be divided into linguistic fuzzy modelling that focuses on interpretability and precise fuzzy modelling that focuses on accuracy. These two types are respectively identified by the Mamdani model (Mamdani and Assilian, 1975) and the TSK model (Takagi and Sugeno, 1974, 1985).
Dynamic soil modelling for a settlement-driven TBM control system
Published in J. Saveur, (RE)Claiming the Underground Space, 2003
D.J.M. Ngan-Tillard, A.S. Elkadi, G. Swinnen
In the Netherlands also, Obladen (Obladen et al., 2001) reported on the development of a system in competition with the IBCS, the I2DOS, Intelligent Integrated Drilling Operating System, patented by Ballast Nedam (Obladen, 2001). This neural network based system is said to be able to predict deformation at 5 to 10 m in front of the TBM with a coefficient of correlation of 0.95. Unfortunately, the I2DOS input and output parameters are not described and no results are shown in the paper. Also, the training and validation data sets are too limited to be confident in the performance of such a system in a wide range of configurations. The I2DOS suffers from being a complete black box that makes total abstraction of the ground surrounding the TBM. It does not have the flexibility and reliance of a neuro-fuzzy based system. The fuzzy character of a neuro-fuzzy system allows integrating expert knowledge, dealing with uncertainty and imprecision, performing inference and making decisions.
Image noise removal using optimal deep learning-based noisy pixel identification and image enhancement
Published in The Imaging Science Journal, 2021
S.P. Premnath, J. Arokia Renjith, J. P. Ananth
A fuzzy system that uses a learning algorithm inspired by or created from neural network theory to establish its parameters (fuzzy sets and fuzzy rules) through the processing of data samples is referred to as a neuro-fuzzy system. The neuro-fuzzy model [28] integrates the learning power of ANN and explicit knowledge demonstration of fuzzy inference systems. Here, the noisy pixels are considered as input. ANFIS's model is divided of premise and consequence components. To identify the qualities based on the optimisation strategy, ANFIS is trained. The novel image matrix is produced using the noisy pixel's nearest neighbour and is provided as, For addressing uncertainty problems, the produced new values of the pixel are fed to neuro fuzzy model. where symbolise neuro-fuzzy system. After producing neuro fuzzy group, the new pixel is adapted as an input and a new image is produced with a noise pixel-based location. The output generated with neuro fuzzy system is
A new prediction model of solar radiation based on the neuro-fuzzy model
Published in International Journal of Ambient Energy, 2020
Mohammad Reza Parsaei, Hossein Mollashahi, Ayda Darvishan, Mahdi Mir, Rolando Simoes
Between groups of different alliances of transfer methods, logsig and tansig were established to be the best grouping features. Logsig and tansig were used for the hidden and output layers, respectively. Additionally, the number of neurons was selected to bid less amount of RMSE (Root Mean Square Error) as well as MAPE (Mean Absolute Percentage of Error) as well as a large cost of correlation coefficient (r). The RMSE was applied to clarify the model's fit and the change among real and forecasted data. MAPE was evaluated to find the absolute average error among the real and forecasted amounts (Alam, Kaushik, and Garg 2009; Ghadimi, Afkousi-Paqaleh, and Nouri 2013). Finally, r showed how strong the relation among the metered and forecasted value was. where N is the number of forecast samples; PV fi is the predicted PV power, is the real PV power and its average is presented by . The neuro-fuzzy model establishes an intelligent strategy that chains fuzzy logic with neural networks in order to have enhanced results. Hence, the ANFIS model can be labelled as a fuzzy scheme armed with a training method. It is fairly rapid and has appropriate training results which is able to associate with the best neural networks. In this work, the weights of the proposed forecast engine are optimised with an intelligent algorithm as presented in the following section:
Integrated Fuzzy DEA-ANFIS to Measure the Success Effect of Human Resource Spirituality
Published in Cybernetics and Systems, 2018
Mohammad Reza Taghizadeh Yazdi, Mohammad Mahdi Mozaffari, Salman Nazari-Shirkouhi, Seyed Mohammad Asadzadeh
Neuro-fuzzy modeling (Jang 1993; Brown and Harris 1994) refers to the way of applying various learning techniques developed in the neural network literature to fuzzy modeling or a fuzzy inference system. Neuro-fuzzy system, which combines neural networks and fuzzy logic, has recently gained considerable interest in research and application. The neuro-fuzzy approach added the advantage of reduced training time due not only to its smaller dimensions but also to the fact that the network can be initialized with parameters relating to the problem domain. Such results emphasize the benefits of the fusion of fuzzy and neural network technologies as it facilitates an accurate initialization of the network in terms of the parameters of the fuzzy reasoning system.