語系: 繁體中文
說明(常見問題)
圖資館首頁
登入
切換: 標籤 | MARC模式 | ISBD

整合均勻設計與克利金插值法於壓粉體燒結放電加工對表面鍍層參數最佳化設計 ...
~
國立高雄大學電機工程學系--工業技術整合產業研發碩士專班

 

  • 整合均勻設計與克利金插值法於壓粉體燒結放電加工對表面鍍層參數最佳化設計 = Optimal Parameter for Surface Modification by W-Powder Sinter Electrical Discharge Machining with Application of Uniform Design and Kriging Interpolation Methods
  • 紀錄類型: 書目-語言資料,印刷品 : 單行本
    並列題名: Optimal Parameter for Surface Modification by W-Powder Sinter Electrical Discharge Machining with Application of Uniform Design and Kriging Interpolation Methods
    作者: 朱瓊芳,
    其他團體作者: 國立高雄大學
    出版地: [高雄市]
    出版者: 撰者;
    出版年: 2012[民101]
    面頁冊數: 79面圖，表格 : 30公分;
    標題: 放電加工
    標題: Sinter
    電子資源: http://handle.ncl.edu.tw/11296/ndltd/71732358747186508162
    附註: 參考書目：面63-66
    摘要註: 本研究旨在探討以不同鎢鐵粉比例及壓粉體壓力製成之高溫燒結壓粉體電極，在不同放電持續時間和放電電流條件下，以CNC放電加工機對模具鋼進行放電改質後，對披覆分佈率及硬度之影響。並以披覆分佈率最大化及硬度最大化為目標，求解最佳的參數設定。 本文首先以U(_15^*)(〖15〗^7 )均勻表進行實驗設計，依照使用表將鎢鐵粉比例、壓粉體壓力、放電持續時間和放電電流製程參數安排於U(_15^*)(〖15〗^7 )均勻表之表頭上，以形成含15次實驗之均勻實驗。依照每次實驗的參數設定，製作高溫燒結壓粉體電極並實施放電加工，並以掃描式電子顯微鏡量測鍍層分佈率及洛氏硬度之數據。以克利金插值法，依實驗之輸入與輸出數據建立鍍層分佈率及洛氏硬度之克利金反應曲面模型。將克利金反應曲面視為目標函數，以非線性規劃求解鍍層分佈率最大之最佳參數設定及洛氏硬度最大之最佳參數設定。鍍層分佈率最大化之最佳參數設定為：壓粉體壓力785.8kg、鎢鐵粉比例5.6g：1.4g、放電持續時間4.35min、放電電流5.61A，此時鎢粉分佈率預測值為82.0658%。洛氏硬度最大化之最佳參數設定為：壓粉體壓力1236.9kg、鎢鐵粉比例5.63g：1.37g、放電持續時間5.11min、放電電流11.12A，此時洛氏硬度HRc預測值為11.0144。最後以驗證實驗評估克利金模型之最佳解是否可作為原問題之近似最佳解。驗證實驗顯示，鎢粉分佈率的真實值為83.33%，與預測值82.0658%相比，兩者只有1.2642%之誤差。洛氏硬度真實為HRc11.3，與預測值HRc11.0144相比，兩者只有2.52%之誤差，故以克利金模型之最佳解作為原問題之近似最佳解是可行的。本研究顯示將均勻實驗設計、克利金反應曲面模型及最佳化算法結合應用，可有效提昇壓粉體燒結放電加工對表面鍍層的品質特性，此套研究流程可作為業界面對最佳化製程參數問題時之參考。 In this thesis, we demonstrate the surface modification of SKD11 tool steel made from mixture powder of tungsten and iron in Wolfram Powder Sinter followed by Electrode Discharge Machining (EDM). Optimized process parameters were achieved by means of uniform experimental design(UED) and Kriging interpolation methods. 15 different experiments based on selected U(_15^*)(〖15〗^7 ) uniform design table of process parameters such as pressure of W-Fe electrode(X1), percentages ratio of W-Fe powder (X2), duration of discharge pulse (X3), and discharge pulse current(X4) have been conducted, and experiment results were verified by scanning electron microscopy (SEM) and Rockwell hardness testing. Process parameters in surface coating and hardness were optimized by Kriging interpolation methods using the input and output data of the uniform experimental design table base on Kriging response curve model as the target function. Optimized data of tungsten powder percentage of 82.0658% in surface coating process was derived using reference parameters in surface coating process as powder pressure 785.8kg, tungsten percentage ratio 5.6g: 1.4g,discharging duration time 4.35min,and discharging current 5.61A. Optimized data of HRc11.0144 in Rockwell hardness testing was derived using reference parameters in surface coating process as powder pressure 1,236.9kg, tungsten percentage ratio 5.63g: 1.37g, discharging duration time 5.11min, and discharging current 11.12A. The liability of applying Kriging model to approach the optimized parameters in process was tested and verified by experimental results. The results shows that only small deviation of 1.2642% between real value of tungsten powder percentage of 83.33% in stead of the predicted value of 82.0658%, and deviation of 2.52% between real value of HRc11.3 of Rockwell hardness testing in stead of the predicted value of HRc11.0144 by using Kriging method. As the result, we conclude that Kriging Surrogate Model is effective to achieve the optimized process parameters for surface modification of SKD11 tool steel.
館藏
  • 2 筆 • 頁數 1 •
 
310002292830 博碩士論文區(二樓) 不外借資料 學位論文 TH 008M/0019 542201 2514 2012 一般使用(Normal) 在架 0
310002292848 博碩士論文區(二樓) 不外借資料 學位論文 TH 008M/0019 542201 2514 2012 c.2 一般使用(Normal) 在架 0
  • 2 筆 • 頁數 1 •
評論
Export
取書館別
 
 
變更密碼
登入