Research on Intelligent Control and System Modeling to the Electromagnetic Induction Hyperthermia

摘要

The aim of this dissertation involved developing a temperature controller and system modeling for predicting the temperature response of tissues in electromagnetic induction hyperthermia (EIH) There are two critical challenges when applied to deep-seated tissue using EIH i e the temperature may not be measured accurately and the temperature control is susceptible to disturbance due to unpredictable plant parameter variations The finite element method (FEM) was suitable for coupled analysis with electromagnetic fields and heat transfer which could be used to predict temperature profiles in deep tissue implanted with magnetic materials To improve the accuracy of the FEM model an adaptive network fuzzy inference system (ANFIS) model was implemented on the basis of measured data and simulated data which were generated by the FEM model The ANFIS model can provide a large number of testing data to optimize the parameters in the FEM model and it can be used to expedite the optimization process Moreover system identification for EIH was analysed and selected with recurrent neural networks models to approximate various conditions of the temperature process The recurrent neural network was useful for establishing the temperature prediction model with the advantages of simple design and stable efficacy A closed-loop controller was applied to track a reference model to guarantee a desired temperature response The EIH system generates an alternating magnetic field to heat a high magnetic permeability material This wireless induction heating has few side effects when it is extensively applied to cancer treatment The effects of hyperthermia strongly depend on the precise control of temperature However during the treatment process the control performance is degraded due to severe perturbations and parameter variations In this study a modified self-learning fuzzy logic controller (SLFLC) with gain tuning mechanism was implemented to obtain high control performance in a wider range of treatment situations This implementation was performed by appropriately altering the output scaling factor of a fuzzy inverse model to adjust the control rules The proposed SLFLC was compared to the classical self-tuning fuzzy logic controller (STFLC) and fuzzy model reference learning control (FMRLC) Additionally the proposed SLFLC was verified by conducting in vitro experiments with porcine liver The experimental results indicate that the proposed controller shows greater robustness and excellent adaptability with respect to temperature control of the EIH system

獎項日期	2017 8月 14
原文	English
監督員	Cheng-Chi Tai (Supervisor)