臺灣博碩士論文加值系統

積體電路科技的進步，趨使電子產品朝向輕、薄、短、小發展，耗電量降低的同時也提高了內部構裝發熱密度。 個人電腦為了提升散熱效率，發展出眾多散熱機制如手動模式(Manual)、速度巡航模式(Speed Cruise Mode)、熱巡航模式(Thermal Cruise Mode) 及多點控制模式(Multi-point Control Mode)等四種模式， 但因散熱機制精確模型難以建構, 無法適時控制風扇轉速, 導致散熱效能不佳。
本論文應用自適應神經模糊推理系統 (Adaptive Neuro-Fuzzy Inference Systems, ANFIS)於個人電腦散熱系統, ANFIS 演算法結合模糊推論系統與類神經網路的特性， 充分發揮系統不確定性、 自我學習與組織能力的特性，本文透過收集散熱風扇的風扇轉速及實際散熱降溫能力特性資料作為演算法訓練數據，不需建構風扇電機的複雜函數，使演算法能自我學習進行風散轉速的最佳預測，達到最低消耗功率及散熱準確性之目標。
由實驗數據得知，本論文所提出的 ANFIS，與手動模式、速度巡航模式、熱巡航模式及多點控制模式應用於風散轉速控制進行比較分析， 當測試系統由高溫攝氏 80 度降至低溫 50 度， 風扇平均轉速可降低 7.67%，證明以 ANFIS 進行轉速控制能有效的降低運轉轉速，提高散熱系統效率。

With advances in integrated circuit technology, the demand for light-weighted, thin, short, small and reduced power consumer products increases day by day, also increased internal heat density. To improve the cooling efficiency of personal computers, many cooling methods are developed out, such as the manual mode, speed cruise mode, thermal cruise mode and multi-point fan control mode, and so on. But it is more difficult to achieve a mathematical model of the fan motor, so let the fan speed control is not good.
In this thesis, we propose an adaptive neuro-fuzzy inference system(ANFIS) approach to a PC cooling systems for reducing the time of searching the optimum speed of the cooling fan. ANFIS has combined fuzzy inference system and neural network two kinds of an intelligent algorithm, so it can be applied to uncertainty system fully. Even more, it has self- study and ability of the organization at the same time. In this thesis. Via the collection cooling fans of fan speed and cooling capacity data as a training algorithm data. The projected algorithm can be the best predictor of the cooling temperature of the motor speed to achieve the target minimum of power consumption and fan's speed accuracy.
The experiment results show that the proposed ANFIS is compared with manual mode, speed cruise mode, thermal cruise mode and multi-point fan control mode in fan speed control comparative analysis when the test system consists of a high temperature of 80 degrees Celsius down to cryogenic 50 degrees. Our experiment analysis shows that the fan speed can be reduced 7.67%, we can prove the effectiveness in reducing the fan speed, and improve the efficiency of the cooling system.

中文摘要--------------------------------------------------------------------------I
英文摘要--------------------------------------------------------------------------II
誌謝------------------------------------------------------------------------------III
目錄------------------------------------------------------------------------------IV
圖目錄----------------------------------------------------------------------------VI
表目錄----------------------------------------------------------------------------VIII

一、緒 論-------------------------------------------------------------------------1
1.1 前言----------------------------------------------------------------------1
1.2 研究動機------------------------------------------------------------------5
1.3 文獻回顧------------------------------------------------------------------6
1.4 論文架構------------------------------------------------------------------7

二、散熱系統架構------------------------------------------------------------------8
2.1 散熱系統概述--------------------------------------------------------------8
2.2 微處理機控制器（Microprocessor Control Unit）-----------------------------9
2.3 板翅式散熱器(Plate-Fin Heat Sinks)----------------------------------------12
2.4 PWM風扇架構---------------------------------------------------------------14
2.4.1 風扇馬達參數--------------------------------------------------------------19
2.5 電壓控制電流模塊架構------------------------------------------------------22
2.6 雙向電平移位器------------------------------------------------------------23
2.7 演算法概述----------------------------------------------------------------24
2.7.1 手動模式(Manual Mode)-----------------------------------------------------25
2.7.2 熱巡航模式(Thermal Cruise Mode)-------------------------------------------26
2.7.3 速度巡航模式(Speed Cruise Mode)-------------------------------------------28
2.7.4 多點控制模式(Multi-Point Control Mode)------------------------------------30

三、系統流程與智慧型演算法--------------------------------------------------------31
3.1 散熱系統概述--------------------------------------------------------------31
3.2 ANFIS應用於散熱系統-------------------------------------------------------33
3.3 自適應神經模糊推理系統的背景----------------------------------------------40
3.4 自適應神經模糊推理系統的法則----------------------------------------------41
3.5 類神經模糊系統(Neuro-Fuzzy system, NFS)-----------------------------------44
3.6 ANFIS 模式限制------------------------------------------------------------47
3.7 ANFIS 控制器設計----------------------------------------------------------48

四、實驗結果與分析----------------------------------------------------------------55
4.1 實驗條件------------------------------------------------------------------55
4.2 實驗結果------------------------------------------------------------------55
4.2.1 自適應神經模糊推理系統實驗------------------------------------------------56
4.2.2 手動模式實驗--------------------------------------------------------------57
4.2.3 熱巡航模式實驗------------------------------------------------------------58
4.2.4 速度巡航模式實驗----------------------------------------------------------59
4.2.5 多點控制模式實驗----------------------------------------------------------60
4.3 實驗結果比較--------------------------------------------------------------61

五、結論及未來展望----------------------------------------------------------------64
5.1 結論----------------------------------------------------------------------64
5.2 未來展望------------------------------------------------------------------65

參考文獻 ------------------------------------------------------------------66
發表論文 ------------------------------------------------------------------68
自 述 ------------------------------------------------------------------69

[1]Qualcomm Technologies, “Snapdragon S4 Processor: Coolest Kid on the Block” , US, Jun, 2012.
[2]吳培立 , 《筆記型電腦散熱系統之研究》 , 國立成功大學, 碩士論文, 台南, 2004.
[3]Chassis Plans, “white paper - cooling and noise” , US, May, 2008.
[4]陳廷彰, 《風扇噪音分析與有孔平板受風扇噪音激振的聲場分析》, 國立中央大學,碩士論文,桃園縣, 2008.
[5]賴紳洧, 《模糊邏輯理論在攜帶型電腦電源管理和散熱控制之應用》, 國立中央大學,碩士論文,桃園縣, 2007.
[6]Intel Corporation, “4-Wire Pulse Width Modulation (PWM) Controlled Fans Specification” , US, July 2004.
[7]Winbond Corporation “winbond lpc i/o” , US, Jan 2006.
[8]Alex Doboli, Edward H. Currie, “Introduction to Mixed-Signal, Embedded Design” , New York, USA, 2011.
[9]A. Hughes, “Electric Motors and Drives. Fundamentals, Types and Applications”, New York: Elsevier, Third edition,2006.
[10]Jyh-Shing R . Jang , “Fuzzy Modeling Using Generalized Neural Networks and Kalman Filter Algorithm. ” Proceedings of the 9th National Conference on Artificial Intelligence, Anaheim, CA, USA, July 14–19 2. pp. 762–767.
[11]J. S. R. Jang , “ANFIS: Adaptive-network-based fuzzy inference systems,” IEEE Trans. Systems, Man, and Cybernetics, Vol. 23, No.3, pp. 665-685, 1993.
[12]李允中, 王小璠, 蘇小春, 《模糊理論與其應用. 》 初版, 全華科技圖書公司, 2001.
[13]J. S. R. Jang, C.T. Sun and E. Mizutani, “Neuro-fuzzy and soft computing”, prentice Hall, International, Inc. 1997.
[14]W. J. Wang, S. F. Lee, and T. C. Chian “Fuzzy control for the intersection’s traffic light near the exit ramp of a freeway,” Automatic Control Conference, 2001, Da-Shing, Taiwan, ROC.
[15]王文俊, 《認識Fuzzy》. 第三版, 全華科技圖書公司, 2008.
[16]張鼎, 《32位元單晶片機C語言編程：基於PIC32》, 人民郵電出版社, 2009.
[17]Tan, R.H.G. , Goh, Y.H. , Wong, Y.Q. , Mok, V.H. “Energy Efficient Cooling Fan for PC Chassis” Innovative Technologies in Intelligent Systems and Industrial Applications, 2009.
[18]秉昱科技譯，《模糊邏輯與類神經模糊 實例說明》，儒林圖書有限公司， 2000.
[19]羅強華，《類神經網路-MATLAB的應用》，高立圖書有限公司， 2005。
[20]葉怡成，《類神經網路模式應用與實作》，儒林圖書有限公司， 2000。
[21]Intel Corporation, Microsoft Corporation, Advanced Power Management (APM) BIOS Interface Specification., Rev 1.2, February 1996
[22]Hewlett-Packard Corporation, Intel Corporation, Microsoft Corporation, Phoenix Technologies Ltd., Toshiba Corporation, Advanced Configuration and Power Interface Specification., Rev 3.0, September 2004.
[23]维基百科， “Computer cooling”， https://en.wikipedia.org/wiki/Computer_cool ing ， May 2015
[24]Winbond，“w83627ehf/ef、w83627ehg/eg、winbond lpc i/o”， Jan 2006