In recent years soft computing paradigms (neuro fuzzy, feedforward neural network(FFNN), fuzzy models, wavelet networks(WN), etc.) have been used in a large variety of science such as physics, mathematics, engineering and etc. by combination of fuzzy logic, neural network and wavelet theory, fuzzy wavelet neural network(FWNN) is constructed. These networks are one of the important and popular paradigms of soft computing, have been used in many researches. In FWNN, input-output mapping is provided by dilations and translations of a mother wavelet. The consequent part of each fuzzy rules is a linear combination of finite wavelet functions. In this research, ability of FWNN to improve the performance accuracy and increasing the speed of convergence, have made the motivation of designing the proposed FW model. Generally, learning method is one of the most important factor in determining the ability of neural networks. Many different learning algorithms have been proposed for training FW. Back propagation (BP) is the most frequently used learning algorithms. Although high performance accuracy of BP learning algorithm, it has some bottlenecks included slow convergence, presence of local minima, over-trained, updating parameters iteratively. Many different researches have been conducted to improve learning methods, but all these methods are iterative-based and need to update parameters. Extreme learning machine (ELM) is a learning method for single hidden layer feed forward neural network (SLFN). Unlike to traditional justify; MARGIN: 0cm 0cm 0pt; unicode-bidi: embed; DIRECTION: ltr; mso-layout-grid-align: none" Keywords :Fuzzy logic, Fuzzy wavelet neural network, Extreme Learning Machine algorithm, Single hidden Layer Feedforward Neural network
