Prediction of temperature distribution in human BEL exposed to 900MHz mobile phone radiation using ANFIS

• Published in: Applied soft computing Vol. 37; pp. 1029 - 1036
• Main Authors: Develi, Ibrahim, Sorgucu, Ugur
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01.12.2015
• Subjects: 900 MHz mobile phone radiation; Adaptive neuro-fuzzy inference system (ANFIS); Modeling the temperature effect of electromagnetic fields; Soft computing; Temperature effect of electromagnetic fields; Thermal effects of electromagnetic fields; 900MHz mobile phone radiation; Modeling the temperature effect of electromagnetic fields; Temperature effect of electromagnetic fields; Soft computing; Thermal effects of electromagnetic fields; Adaptive neuro-fuzzy inference system (ANFIS)
• ISSN: 1568-4946; 1872-9681
• DOI: 10.1016/j.asoc.2015.04.055

•An experimental setup is designed to investigate the effect of 900MHz electromagnetic field.•EM field strength of RF generator, antenna distances applied, exposure time and the measured depths were systematically changed, while establishing the varying conditions.•A soft computing method is firstly used in the literature to determine the temperature distribution of tissue equivalent liquid exposed to electromagnetic field. In this paper, an application of adaptive neuro-fuzzy inference system (ANFIS) for the prediction of temperature distribution in a human brain equivalent liquid (BEL), resulting from exposure to electromagnetic radiation, is presented. In the first phase of the study, temperature distributions, resulting from different exposure conditions of electromagnetic (EM) fields, are experimentally determined. Since experimental determination of temperature distribution in each point of the BEL is time-consuming and hard to implement, ANFIS is employed to determine the distribution without new measurements. The use of ANFIS is found to be very useful in terms of the prediction of temperature distribution within the BEL after exposure to radiation. Numerical results show that the proposed method can be employed for the prediction of temperature distribution in a human BEL with high accuracy without requiring any experimental measurements.