使用致冷晶片之模糊溫度控制器研製

本論文研製一種控制致冷晶片的模糊溫度控制器，經由使用者輸入溫度與保冷箱中溫度變化量，進行模糊化及模糊推論後輸出控制板，達到致冷晶片溫度控制的目的。使用致冷晶片而不使用冷媒的氟氯碳化物，係為改善溫室效應所帶來的氣溫上升。我們在Arduino微控制板撰寫模糊控制規則程式、脈波寬度調變模擬程式與溫度感測器偵測程式，由三種程式做結合後，進行偵測保冷箱內部溫度誤差及變化量，再藉由模糊化、模糊推論與重心法的解模糊功能，來達到使用者需求的溫度。這個致冷晶片的模糊溫度控制器，可控制的溫度為 9-12度，具有降低溫度及穩定溫度的效果，在未來將可適用於個人化冰箱與水族箱上溫度的控制。

關鍵字

溫室效應；冷媒；致冷晶片；模糊理論

並列摘要

This thesis presents a fuzzy temperature controller that can be used to regulate the cold-ended outputs of Thermo Electric Cooler (TEC) modules. Specifically, a fuzzy control system was employed in the developed controller for adjusting the temperature to a desirable range. Currently, thermoelectric cooling modules are used as a substitute of refrigerants to alleviate global warming problems. An Arduino microcontroller board was used to code a program comprising fuzzy logic control rules. A pulse-width modulation technique was applied to adjust and simulate the coded program. A defuzzification function involving fuzzification, inference, and center of gravity method was applied to detect the errors and variations in the temperature of a box. Through this method, the optimal output membership value was obtained and transmitted to the voltage output board for regulating the cooling modules and achieving a desirable temperature. The results revealed that the temperature of the system can be controlled to a range of 9–12 °C, verifying the refrigeration function of the proposed system. The proposed system can be applied to control the temperature of pet incubators or aquarium tanks.