The Electromagnetic Ehermotherapy System Based on The Fuzzy Temperature Controller with the Combination of Neural Network Learning Mechanism and Reference Temperature Curve Model

• Main Authors: Yuan-RueiSyu, 許元瑞
• Other Authors: Cheng-Chi Tai
• Format: Others
• Language: zh-TW
• Published: 2016
• Online Access: http://ndltd.ncl.edu.tw/handle/35711702412870528532

碩士 === 國立成功大學 === 電機工程學系 === 104 === Electromagnetic thermotherapy refers to high-frequency current passing through an induction coil to generate an alternating magnetic field outside human body so that a magnetic metal needle appears eddy current internally to further heat the needle and achieve the specific temperature with therapeutic effect in the tissue. To ensure secure burning of tumor cells and avoid the necrosis of normal cells caused by high temperature, the control of burning temperature is extremely strict. In terms of uncertain therapeutic environment and system, a fuzzy temperature controller with the combination of reference temperature curve model and neural network learning is proposed in this study, aiming to overcome various interferences in the therapeutic environment, accurately control the temperature response curve for heating metal needle, and precede stable and secure thermal treatment. The experiment verifies that using the fuzzy temperature controller with the combination of reference temperature curve model and neural network learning under different or same therapeutic distance could effectively control it on the set therapeutic temperature curve. Besides, the simulation of changing heating distance resulted from the chest and abdomen of a patient when breathing shows the maximum therapeutic temperature error within 2%.