 Open Access
 Total Downloads : 3203
 Authors : Akhil V. Gite, Raksha M. Bodade, Bhagyashri M. Raut
 Paper ID : IJERTV2IS2027
 Volume & Issue : Volume 02, Issue 02 (February 2013)
 Published (First Online): 28022013
 ISSN (Online) : 22780181
 Publisher Name : IJERT
 License: This work is licensed under a Creative Commons Attribution 4.0 International License
ANFIS Controller and Its Application
Akhil V. Gite, Raksha M. Bodade, Bhagyashri M. Raut
Department of Electronics & Telecommunication Jawaharlal Darda Institute of Engineering and Technology Yavatmal
Maharashtra State, INDIA
Abstract
The concept of fuzzy logic and artificial neural network for control problem has been grown into a popular research topic in recent years. The reason is that the classical control theory usually requires a mathematical model for designing the controller. The inaccuracy of mathematical modeling of the plants usually degrades the performance of the controller, especially for nonlinear and complex control problems. The advent of the fuzzy logic controllers (FLC) and the neural controllers based on multilayered neural networks has inspired new resources for the possible realization of better and more efficient control. In recent years, the integration between fuzzy logic and neural network namely fuzzy neural network (FNN) has been proposed and developed; generally the combination of fuzzy logic and neural network is called as ANFIS (Adaptive Neuro Fuzzy Inference System). Neural system has many input and also has multiple outputs but the fuzzy logic has multiple inputs and single output, so the combination of this two is know as ANFIS which is used for nonlinear applications.[1]
Keywords: ANFIS, Artificial neural network, Temperature control, PID.

INTRODUCTION
Process control systems are often nonlinear and difficult to control accurately. Their dynamic models are more difficult to derive than those used in aerospace or robotic control, and they tend to change in an unpredictable way. The conventional PID controllers, in various combinations have been widely used for industrial processes due to their simplicity and effectiveness for linear systems, especially for first and second order systems. It has been well known that Proportional Integral Derivative (PID) controllers can be effectively used for linear systems, but usually cannot be used for higher order and nonlinear systems [3]. This paper addresses an application that involves the temperature control system also it address to the application of water bath temperature controller. Other controller such as PID was not accurate for
nonlinear system but this ANFIS controller is accurate as compared to other controllers also it is useful for nonlinear system. This is the biggest advantage of this controller. So by training the neural and combining with will gate the ANFIS Controller which used for temperature water bath control, intelligent temperature water bath control

LITERATURE SURVEY
Fuzzy systems have the ability to represent comprehensive linguistic knowledge: given for example by a human expert and perform reasoning by means of rules. However, fuzzy systems do not provide a mechanism to automatically acquire[2] or tune those rules. On the other hand neuralnetworks are adaptive systems that can be trained and tuned from a set of samples. Once they are trained, neuralnetworks can deal with new input data by generalizing the acquired knowledge. Nevertheless, it is very difficult to extract and understand that knowledge. In other words, fuzzy systems and neuralnetworks are complementary paradigms [4]. The fuzzy logic controllers and the neural networks (both static and dynamic) are two modern system analyses which had been applied successfully in many practical applications .These two techniques are very useful when the system under study is partially unknown and previously assumed to be nonlinear. The combination between the two methods (Neurofuzzy control systems) is a powerful identification and control technique [5]. In recent years, Fuzzy Inference Systems (FISs) and Artificial Neural Networks (ANNs) have attracted considerable attention as candidates for novel computational systems because of the variety of the advantages that they offer over conventional computational systems. Unlike other classical control methods, Fuzzy Logic Control (FLC) and ANNs are more model free controllers, i.e. they do not require exact mathematical model of the system [6]. For nonlinear modeling, neural networks and neurofuzzy modeling approaches have received a great deal of attention. The drawbacks are the complexity and the darkness of their structures, especially, in modeling complex nonlinear systems. It has been shown that many types of nonlinearities in industry processes are effectively modeled with
hybrid models, like heating and cooling processes, fermentation, solid drying processes, continuous stirred tank reactor (CSTR) [7]. The temperature system has non linearity, long delay time, and large time constant and undetermined system. At present Industries uses the PID technique for temperature control. PID control is a crisp control, the self tuning of the P, I, D parameters are quite difficult and the resultant control is with overshoot and with large time constants [4]. Fuzzy logic used to express uncertainty in an expert system. The problem of Fuzzy controller is reduced to acquisition of a correct set of IFTHEN rules that can be obtained by human expert. ANFIS controller is the combination of fuzzy logic and ANN and capable to generate expert systems by itself. [8].The fixed gain feedback controllers (PID) are insufficient to compensate for parameter variations in the plant as well as to adopt to changes in the process environment. The Advent of Fuzzy Logic Control (FLC) has inspired new resources for more efficient control. It does not require a priori model of the process for implementation. In the traditional FLC the optimal membership values are found by trial and error which is laborious. Also the conventional method of transferring the range of input variables into corresponding universe of discourse is time consuming. The transfer of the range of input variables into corresponding universe of discourse can be carried out using Artificial Neural Networks [9]. The applications of neural networks to control systems have become increasingly important. The backpropagation neural network based on the
generalized delta learning rule, a gradient descent search technique, has been widely used. The backpropagation algorithm has several disadvantages, among which is lack of guaranteed convergence, but its simple yet powerful mathematical algorithm has made it the mainstay of neurocomputing. Before the neural network can be used as a controller, it first must learn the model of the plant. There are several learning architectures proposed whereby the neural network may be trained [10].

SYSTEM OVERVIEW
Takagi and Sugeno proposed the TS fuzzy model in1985. Students called it as Sugeno fuzzy model after. The Sugeno fuzzy model is a nonlinear model. It can aptly express the dynamic characteristic of complex systems. Furthermore, it is the fuzzy inference model that is in the most common use. A typical fuzzy rule in a Sugeno fuzzy model has the format:
If x is A and y is B, Then Z=f(x,y)
Where A & B are fuzzy set antecedent, Z=f(x,y) is a crisp function in the Consequent.
Let us assume the fuzzy inference system with two inputs x and y and one output. Suppose that the rule base contains two fuzzy ifthen rules of Takagi and Sugeno's type as given below:

ANFIS structure for 2 input variables for TSK Model
Layer 1: Every node i in this layer is a square node with a node function
where x is the input to node 0, and A is the linguistic label associated with this node function.
In other words, is the membership function of A, and itspecifies the degree to which the given x satisfies the quantifier Ai. Gaussian Membership function is chosen with maximum equal to 1 and minimum equal to 0. Parameters in this layer are referred to as the premise parameters. Membership functions are used for each of the input in this layer. [1]
Layer 2: Every node in this layer is a circle node labeled H, which multiplies the incoming signals and sends the product out. For instance,
Each node output represents the firing strength of a rule.[1]
Layer 3: Every node in this layer is a circle node labeled N. The ith node calculates the ratio of the ith rule's firing strength to the sum of all rules' firing strengths:
For convenience, outputs of this layer will be called normalized firing strengths. [1]
Layer 4: Every node 1 in this layer is a square node with a node function
Where , is the output of layer 3 and {pi, qi, ri} is the parameter set. Parameters in this layer will be referred to as consequent parameters.[1]
Layer 5: The single node in this layer is a circle node labeled E that computes the overall output as the summation of all incoming, signals, i.e.

ANFIS Model
This is the block diagram of ANFIS Controller, ANFIS controller is the combination of neural network and Fuzzy Logic. Many inputs are applied to the neural network depending upon the inputs the neural network has some standard output, so depending upon the input and the output the neural network is trained, after training the neural network the output is applied to the fuzzy logic which generates the IF THEN rules and membership functions, This is done in MATLAB Below is the block diagram of ANFIS controller

APPLICATION
This ANFIS controller is widely used for controlling the nonlinear system. As this is the best controller as compared to conventional PID controller, and other controller. This controller is used in Temperature water bath controller. Also
this controller is used in planes to controller them now a days research is going on for Intelligent Planes which learn by themselves and do take take off and landing so these are the applications

CONCLUSION
In this paper a ANFIS controller for temperature control of a water bath system is proposed. ANFIS based NFC is suitable for adaptive temperature control of a water bath system. As ANFIS is the combination of Neural Network and Fuzzy Logic, and it gives accuracy to nonlinear systems Hence ANFIS is the good controller as compared to other controller, and it is widely being used.
ACKNOWLEDGEMENT
The authors would like to thank firstly, our GOD, and all friends who gave us any help related to this work. Finally, the most thank is to our families and to our country INDIA which born us.
REFRENCES

Jafar Tavoosi, Majid Alaei, BehrouzJahani, Mohammad Amin Daneshwar, A Novel Intelligent Control System Design for Water BathTemperature Control ISSN 19918178

T.P.Mote, Dr.S.D.Lokhande, Temperature Control System Using ANFIS ISSN: 22312307,
Volume2, Issue1, March 2012

Ajay B Patil Adaptive Neuro Fuzzy Controller for Process Control System, 978142442806 9/08/2008, IEEE.

Kaijun Xu, Xiaoping Qiu ,Xiaobing ,Li Yang Xu A dynamic neurofuzzy controller for gasFired water heater, 33049/08 /2008 IEEE.

Valdez D., Ortiz V., Cabrera A. and Chairez I Extended Kalman FilterWeights Adjustment For Neonatal Incubator Neurofuzzy Identification, 0 780394895/06/$20.00/Â©2006 IEEE.

Otman M. Ahtiwash and Mohd Zaki Abdulmuin,
V.N. Alexandrov An Adaptive NeuroFuzzy Approach for Modeling and Control of Nonlinear Systems, ICCS 2001, LNCS 2074, 198207, 2001. SpringerVerlag Berlin Heidelberg.
[7]J. A. Vieira, F. Morgado Dias and A. M. Mota Hybrid NeuroFuzzy NetworkPriori Knowledge Model in Temperature Control of a Gas Water Heater System, 0769524575/0 /2005, IEEE. [8]S.Ravi P .A.Balakrishnan, Modeling And Control of an ANFIS Temperature Controller For Plastic Extrusion Process, 978142447770 8/10/2010 IEEE.
Advanced Control Schemes for Temperature Regulation of Air Heat Plant 078035406 0/99/1999, IEEE.

Marzuki Khalid and Sigeru Omatu A Neural Network Controller for Temperature Control System, 0272 1708/92/1992IEEE.