標題: 使用基因演算法之壓電式平面喇叭最佳化設計
Optimization of Piezoelectric Radiator Using Genetic Algorithms
作者: 黃敬宏
Chinghong Huang
白明憲
Mingsian Bai
機械工程學系
關鍵字: 基因演算法;壓電材料;最佳化;genetic algorithms;piezoelectric material;optimization
公開日期: 2001
摘要: 通訊電子產品的微小化趨勢為目前產業之主流,隨著機體和法規環境的要求,例如駕駛時禁打行動電話之免持聽筒規定,行動電話及PDA之重要元件揚聲器亦有微小化的必要。本研究的目的是希望利用基因演算法(Genetic Algorithms)的優勢,建立出一套設計壓電式平面叭方法。在以往的平面喇叭皆是以電磁式的致動器為驅動原件,其缺點為無法突破厚度的限制,而壓電材料具有形狀可塑度高、高響應速率、出力大、省電等種整優點,以其為驅動原件的新式平面喇叭將可克服電磁式平面喇叭在厚度上的限制。在本文中建立出壓電式平面喇叭之動態方程式,以基因演算法最佳化參數,使其性能提昇。並利用程式模擬與實驗的程序來驗證其結果,而根據結果進行討論。
The miniaturization of communication electronic manufactures is the master stream in industrial circles. Recently, the electric elements become smaller and the loudspeaker has to change the structure in order to increase its practicability, efficiency and performance. In this paper, the PZT-excited panel speaker is modeled by using energy method, and assumed-modes method. The system is transformed into the modal space, and then the results are obtained in the physical space by a transformation matrix. The genetic algorithms (GA) is used to optimize the design. The results are used in the simulation and experiment. Performance indices including frequency response, sound pressure level (SPL), on-axis response, total harmonic distortion (THD), directional response, sensitivity and efficiency are evaluated. It is found in the experimental results that the PZT-excited panel speaker has poor response at low frequency. In order to improve the performance, we chose suitable material and boundary condition.
URI: http://140.113.39.130/cdrfb3/record/nctu/#NT900489011
http://hdl.handle.net/11536/69126
顯示於類別:畢業論文