Modeling Of BJT With Fuzzy System And ANFIS

Modeling Of BJT With Fuzzy System And ANFIS

Ms. Sonal A. Kale

Sipna college of Engg. And Tech. Amravati.

Prof. Nilesh Kasat

Sipna college of Engg. And Tech. Amravati.

Department of electronics and telecommunication

Abstract

The ability of ANFIS will be follow various curves of the bipolar transistor and comparing with self defined Fuzzy Systems. The results show worse diagnostic than self-defined fuzzy if the input data has higher dispersion. Another achievement show two things:

1. Self defined fuzzy modeling is more powerful while we dont want to involve with more rules 2-ANIFIS just in the condition which very higher number of rules and in the same time higher number of training data suggests more accurate model considering that the reason of being less of our training data is because of

   that we aware of relations between data.

   Introduction

   A bipolar junction transistor (BJT) is a three-terminal electronic device constructed of doped semiconductor material and may be used in amplifying or switching applications. Bipolar transistors are so named because their operation involves both electrons

   and holes. Charge flow in a BJT is due to bidirectional diffusion of charge carriers across a junction between two regions of different charge concentrations. By design, most of the BJT collector current is due to the flow of charges injected from a high-concentration emitter into the base where they are minority carriers that diffuse toward the collector, and so BJTs are classified as minority- carrier devices.

   A BJT consists of three differently doped semiconductor regions, the emitter region, the

   base region and the collector region.These regions are,

   respectively, p type, n type and p type in a PNP, and n type, p type and n type in a NPN transistor. Each semiconductor region is connected to a terminal, appropriately labeled: emitter (E), base (B) and collector (C).The base is physically located between the emitter and the collector and is made from lightly doped, high resistivity material. The collector surrounds the emitter region, making it almost impossible for the electrons injected into the base region to escape being collected, thus making the resulting value of very close to unity, and so, giving the transistor a large . A cross section view of a BJT indicates that the collectorbase junction has a much larger area than the emitterbase junction. The transistors, is usually not a symmetrical device. This means that

   interchanging the

   collector and the emitter makes the transistor leave the forward active mode and start to operate in reverse mode. Because the transistor's internal structure is usually optimized for forward- mode operation, interchanging the collector and the emitter makes the values of and in reverse operation much smaller than those in forward operation; often the of the reverse mode is lower than 0.5. The lack of symmetry is primarily due to the doping ratios of the emitter and the collector. The emitter is heavily doped, while the collector is lightly doped, allowing a large reverse bias voltage to be applied before the collectorbase junction breaks down. The collectorbase junction is reverse biased in normal operation. The reason the emitter is heavily doped is to increase the emitter injection efficiency: the ratio of carriers injected by the emitter to those

   injected by the base. For high current gain, most of the carriers injected into the emitterbase junction must come from the emitter.

   Literature work and review

   The problem of Bipolar Transistor Modeling with Fuzzy Systems and ANFIS in this BJTs are classified as minority- carrier devices . Semiconductor transistor terminal appropriately labeled emitter (E), base (B) and collector (C) . The emitterbase junction the collectorbase junction breaks down the carriers injected into the emitterbase junction must come from the emitter various parts were doped to make them into semiconductors, etc . how the device responds to changes in the applied voltages and currents. Base-Emitter junction this

   voltage/current characteristic curve has an exponential-like shape similar to that of

   a normal PN Junction diode a bit from device to device and with the temperature quickly draw over to the Collector any free electrons which enter the Base region from the Emitter . Bipolar Transistors as it contains quite a lot of detailed information change either the base current or the applied Collector potential ; the Base and Emitter it eventually stops drawing any electrons out of the device and the Collector current falls towards zero . This system contains two inputs namely x and y and an output or Z which is associated with the following rules

   .This layer is the last layer of the network and is composed of one node and adds up all inputs of the node.

   Propose work ANFIS uses two neural network and fuzzy logic approaches. When these two systems are combined, they may qualitatively and

   quantitatively achieve an appropriate result that will include either fuzzy intellect or calculative abilities of neural network. As other fuzzy systems the ANFIS rules. We may recognize five distinct layers in the structure of ANFIS network which makes it as a multi- layer network. A kind of this network, which is a Sugeno type fuzzy system with two inputs and one output, is indicated in Figure1 . As shown in Figure1, this system contains two inputs namely x and y and an output or Z which is associated with the following rules .

   Figure 1

   Rule 1 If (x is A1) and (y is B1) then Z1=p Rule 2 If (x is A2) and (y is B2) then Z2=p1

   In this system, Ai, Bi and Zi2x+qx+q12y+ry+r12 are fuzzy sets and systems output respectively. pi , qi and r are designing parameters which are obtained during the learning process. Then we may explain the various layers functions of this network as follows:

   Layer 1: In this layer, each node is equal to a fuzzy set and output of a node in the respective fuzzy set is equal to the input variable membership grade. The parameters of each node determine the membership function form in the fuzzy set of that node.

   Layer 2: In this layer the input signals values into each node are multiplied by each other and a rule firing strength is calculated.

   Layer 3: These layer nodes calculate rules relative weight.

   Layer 4: This layer is named rules layer which is obtained from multiplication of normalized firing strength (has been resulted in the previous layer) by first order of Sugeno fuzzy rule.

   Layer 5: This layer is the last layer of the network and is composed of one node and adds up all inputs of the node.

   According to figure 2 the first layer in ANFIS structure will performs fuzzy formation and second layer will be performs fuzzy and fuzzy rules. The third layer will be performs the normalization of the membership functions and the fourth layer will be the conclusive part of fuzzy rules and finally, the last layer will calculates the network output. According to these, it is obvious that the first and fourth layers in ANFIS structure are adaptive layers in which C in layer 1 are known as premise parameters that are related to membership function of

   fuzzy input. We will instructed ANFIS network by 23 percent of empirical data. 23 percent of primary data which had been considered for testing the appropriate of the modeling were entered into ANFIS model.

   Figure 2

   2Results obtained of self-defined method were compared with Anfis. Considering the results, it is obvious that proposed modeling b ANFIS with few numbers of rules and self-defined fuzzy modeling are efficient

   and valid and it can also be promoted to more general states. In a closed loop current compression cycle, a small portion of the current of bipolar transistor circulates through the cycle components while most of the current stays inside the loop. The worst scenario of current circulation is when large amounts of current become logged in the system. In this paper, an Adaptive Neuro Fuzzy Inference System (ANFIS) and a simple self-defined fuzzy model will be used for modeling the character of important parameters of bipolar transistor. In this way, we may considered the model with two inputs and one output. The input parameters are voltage of collector emitter and current of collector. The output parameter is current of base of transistor. For training Anfis, we prepared data according the transistor characteristics. Then, we will randomly

   divided empirical data into train and test sections in order to accomplish modeling.

   Conclusion

   In a closed loop current compression cycle, a small portion of the current of bipolar transistor circulates through the cycle components while most of the current stays inside the loop. The worst scenario of current circulation is when large amounts of current become logged in the system. In this paper, an Adaptive Neuro Fuzzy Inference System (ANFIS) and a simple self-defined fuzzy model are used for modeling the character of important parameters of bipolar transistor. In this way, we considered the model with two inputs and one output. The input parameters are voltage of collector emitter and current of collector. The output parameter is current of base of transistor. For training ANFIS, we

   prepared data according the transistor characteristics. Then, we randomly divided empirical data into train and test sections in order to accomplish modeling. We

   instructed ANFIS network by 23 percent of empirical data. 23 percent of primary data which had been considered for testing the appropriate of the modeling were entered into ANFIS model. Results obtained of self- defined method were compared with ANFIS. Considering the results, it is obvious that proposed modeling by ANFIS with few numbers of rules and self-defined fuzzy modeling are efficient and valid and it can also be promoted to more general states.

   References

   1. Manipulator Based on Neural Networks with Adaptive Learning Rate, Asian Journal of Information Technology 4 (10): 927-934. 2005

   2. Min-Yuan Cheng, Hsing-Chih Tsai, Erick Sudjono, Evolutionary fuzzy hybrid neural network

      for project cash flow control Engineering Applications of Artificial Intelligence, Volume 23,

      Issue 4, Pages 604-61,

      2010.

   3. Rong-Jong Wai, Hybrid fuzzy neural- network control for nonlinear motor-toggle servo

      mechanism IEEE Transactions on Control Systems Technology, IEEE Transactions on Volume[1] N. Guersi, M. Djeghaba, D. Lefebvre, F. Druaux, E. Leclercq, Tracking Control for Robot: 10, Page(s): 519

      532, 2002.

   4. M. Kumngern, K. Dejhan, Versatile Dual- Mode Class-AB Four- Quadrant Analog Multiplier International Journal of Signal Processing Volume 2, Number 4, 2005 ISSN 1304-4494, 2005

   5. Keshmiri, R, and Mohamad Shahri, A.,( 2007), Intelligent ABS Fuzzy Controller for Diverse

      Road Surfaces, World Academy of Science, Engineering and Technology 29.

   6. Tiem,N.V, Hoan,V.H., (2009) jpan, Synthesis of Adaptive Control for anti- lock braking systems based on Fuzzy Logic and Neural Network Control, Hanoi University of Transport andCommunications, Vietnam.

Intelligent Systems.

1. Andon ,V. Topalov, Oniz,Y, Kayacan,E and Kaynak,O. ,May (2011), Neuro-fuzzy control of antilock braking system using sliding mode incremental learning algorithm, Neurocomputing,

   Volume 74, Issue 11,

   Pages 1883-1893.2009 abs jadval-2

2. S. Sánchez Solano, A. Barriga, C. J. Jiménez, J.

   L. Huertas(1997) DESIGN AND APPLICATION OF DIGITAL FUZZY CONTROLLERS Sixth IEEE International Conference on Fuzzy Systems (FUZZ-IEEE97), Barcelona – Spain, July 1- 5, Vol. 2, pp. 869-

   874.

3. S. Aminifar, A. Khoei, Kh. Haidi, Gh. Yosefi ,( 2006 )A digital CMOS fuzzy logic controller chip using new fuzzifier and max circuit AEU – International Journal of Electronics and Communications, Volume 60, Issue 8, 1 September, pp. 557-566.

4. Bryk, M.; Wielgus, A,( 2010) Digital implementation of a programmable type-2 fuzzy logic controller ,17th IEEE Conferences MIXDES June 24-26, Wroclaw, Poland ,pp. 255

   – 258 .

5. Bouras, S.; Kotronakis, M.; Suyama, K.; Tsividis, Y.( 1998)

   Mixed analog-digital fuzzy logic controller with continuous-

   amplitude fuzzy inferences and defuzzification Fuzzy

   Systems,IEEE Transactions on Volume: 6, Issue: 2 pp. 205 215.

6. Saavedra, P.; Zrilic, J.; Ramirez-Angulo, J.( 1999) A compact mixed- mode VLSI block

   fuzzifier 42nd IEEE Midwest Symposium on circuits and Systems vol. 1

   , pp. 493 – 495 .

7. Gh. Yosefi, S. Aminifar, Sh. Neda, M.A. Daneshwar(2011) Design

   of a mixed-signal digital CMOS fuzzy logic controller (FLC) chip using new current mode circuits AEU International Journal of Electronics and Communications, Volume 65, Issue 3, March ,pp.

   173-181.

8. Gianluca Giustolisi, Gaetano Palumbo, and Salvatore Pennisi,(2002) Current-Mode A/D

   Fuzzy Converter, IEEE Transactions on Fuzzy Systems, Vol. 10, August

   ,No. 4, pp. 533-540.

9. I.Baturone, S. Sánchez-Solano, J. L. Huertas,(2000) Cmos Design of A Current-Mode Multiplier/Divider Circuit with Applications to Fuzzy Controllers, Analog Integrated Circuits and Signal Processing, Kluwer Acade publishersV. 23, Jun, N. 3, pp. 199-210.