The Establishment of Self-Tuning Fuzzy Temperature Control and Adaptive Network-Based Fuzzy Inference Temperature Prediction Model for Electromagnetic Thermotherapy System
碩士 === 國立成功大學 === 電機工程學系 === 103 === Electromagnetic thermotherapy uses a high frequency induction heating machine to generate heat for tumor tissues treating. The induction heating machine produces an electromagnetic field that induces currents on medical needles. The currents heat the medical need...
| Main Authors: | , |
|---|---|
| Other Authors: | |
| Format: | Others |
| Language: | zh-TW |
| Published: |
2015
|
| Online Access: | http://ndltd.ncl.edu.tw/handle/96181795011918964288 |
| Summary: | 碩士 === 國立成功大學 === 電機工程學系 === 103 === Electromagnetic thermotherapy uses a high frequency induction heating machine to generate heat for tumor tissues treating. The induction heating machine produces an electromagnetic field that induces currents on medical needles. The currents heat the medical needles to a specific temperature that is required for treating a tissue area. Accurate control of temperature is crucial to ensure that it is high enough to kill the tumor cells while not overheating the needles. This paper designed a self-tuning fuzzy logic controller (STFLC) that could improve the heating performance in medical applications under interferences and accurately control the temperature of the needles in order to perform safe treatments. This was designed by considering the dynamically environmental factors that will result in the uncertain heating model affecting temperature control. Furthermore, this paper built a temperature prediction model based on an adaptive network-based fuzzy inference system (ANFIS). It was trained by the heating database, which was generated by the finite element method (FEM) model to reconstruct the biomodel that was capable of accurate prediction of temperature in the treatment process. The ANFIS models could assist a temperature controller to effectively make real-time predictions and control temperature. The STFLC and the ANFIS models went through successful testing in experiments simulating the heating distance variation in the abdominal region during patient breathing. The maximum error of the treatment temperature was small and the system was shown to be capable of accurately predicting the change in real-time temperature.
|
|---|