(54.236.58.220) 您好!臺灣時間:2021/02/28 08:15
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果

詳目顯示:::

我願授權國圖
: 
twitterline
研究生:謝文祥
研究生(外文):SIE,WEN-XIANG
論文名稱:開發輪圈偏擺之偵測系統
論文名稱(外文):Development of a Prognosis System for Wheel Unbalance
指導教授:楊浩青楊浩青引用關係
指導教授(外文):YANG,HAW-CHING
口試委員:洪敏雄陳朝鈞許志華
口試委員(外文):HUNG,MIN-HSIUNGCHEN,CHAO-CHUNHSU,CHIN-HUA
口試日期:2020-07-22
學位類別:碩士
校院名稱:國立高雄科技大學
系所名稱:電機工程系
學門:工程學門
學類:電資工程學類
論文種類:學術論文
論文出版年:2020
畢業學年度:108
語文別:中文
論文頁數:90
中文關鍵詞:輪圈偏擺動平衡主軸振動小波包分解隨機森林
外文關鍵詞:Wheel unbalanceDynamic BalanceSpindle vibrationWavelet PacketRandom Forests
相關次數:
  • 被引用被引用:0
  • 點閱點閱:42
  • 評分評分:系統版面圖檔系統版面圖檔系統版面圖檔系統版面圖檔系統版面圖檔
  • 下載下載:0
  • 收藏至我的研究室書目清單書目收藏:0
以CNC輪圈設備進行加工時,工件品質與設備狀況會存在許多變異因素,例如,輪圈素材品質變異或機台不穩時,將造成輪圈工件夾持不正常,導致加工偏差或工件飛出,造成人員受傷與機台物品損壞。因此,如何在加工前判斷工件偏擺是否正常,實為有效加工的重要依據。
本研究開發一輪圈偏擺偵測系統。在偵測方法上,首先利用安裝在CNC機台最適當位置的加速規,並搭配不同主軸轉速來獲得主軸振動訊號。在分析方法上,擷取振動訊號的小波包特徵,並透過K近鄰演算法(KNN)與隨機森林(RF),可據以判斷工件偏擺狀態。最後,在顯示架構上,經由NODE-RED網頁介面,可進一步提供線上監測的能力。
研究結果顯示,經模擬輪圈與夾具承靠面之最大間距從0到1 mm,在主軸轉速分別為120、300、360 RPM等不同條件下,所開發系統可在最小間距0.3 mm於300 RPM時,基於Level 5的小波包與隨機森林模型,可於2秒內獲得90.3%  98.5%的判別率。因此,本研究所提出之輪圈偏擺偵測系統,將可應用實際機台的線上偵測所需。

During machining wheels by a CNC lathe, workpiece quality and the machine status may exist variation, such as wheel-clamping exception may cause from abnormal material or machine, could lead to quality or safety issues, and will result in operator or machine being damaged. Therefore, assessing runout in rotating of dynamic balance before production is crucial to effectively perform the wheel machining process.
This research develops a prognosis system for detetion of wheel runout. In detection, vibration signals at various speeds are collected from an accelerometer installed on an optimal location of CNC machine spindle. In analysis, rotating runout can be assessed by methods of KNN (K-nearest neighbor) and RF (Random Forests) served with vibration features extracted by wavelet package analysis. Finally, a web application based on Node-red is provided to online monitor and detect rotating runout in a CNC lathe.
The results indicate that the proposed system presents 90.3%98.5% accuracies of detecting minimum 0.3 mm (maximum spacing of bearing surface between wheel and fixture) at 300 RPM within two seconds based on the wavelet packets with 5 levels and RF model after simulating spacing varying from 0 to 1 mm and spindle speeds at 120, 300 and 360 RPM. Therefore, this system can fit needs of online diagnosis of dynamic balance of wheel manufcaturing.

目錄
摘要 i
Abstract ii
誌謝 iv
目錄 v
表目錄 viii
圖目錄 x
第1章 緒論 1
1.1 研究背景 1
1.2 研究目的 3
1.3 本文架構 5
第2章 理論方法 6
2.1 訊號來源 6
2.1.1 加速規 6
2.2 輪圈偏擺可能原因 8
2.3 特徵分析方法 8
2.3.1 快速傅立葉轉換 9
2.3.2 小波包分解 13
2.4 失效偵測分類 14
2.4.1 K-近鄰演算法(k nearest neighbor) 14
2.4.2 隨機森林(Random Forest) 17
第3章 系統開發 18
3.1 系統架構 18
3.1.1 系統概念 18
3.1.2 系統流程 19
3.2 振動資料收集 20
3.2.1 振動感測器安裝 20
3.2.2 實驗規劃 23
3.3 特徵計算 25
3.3.1 快速傅立葉轉換特偵判別 25
3.3.2 振動小波包分解 27
3.4 模型建立 28
3.4.1 K-近鄰演算法(k nearest neighbor)模型建立 28
3.4.2 隨機森林(Random Forest)模型建立 29
3.5 顯示介面 30
第4章 案例研究與分析 31
4.1 設備資料的擷取 31
4.1.1 機台規格與輪圈尺寸 31
4.1.2 安裝位置實驗設計 33
4.1.3 加速規軸向選擇 37
4.2 輪圈振動偏擺狀態特徵 40
4.2.1 偏擺特徵實驗 40
4.2.2 狀態標示 49
4.3模型建置 50
4.3.1基於WaveletPacket與KNN之失效偵測診斷 50
4.3.2基於WaveletPacket與RF之失效偵測診斷 52
4.4模型驗證 56
4.4.1學習模型資料比對 56
4.4.2學習模型資料比對結論 59
4.5加速規差異驗證 61
4.6振動資料特徵層數選擇 65
4.6.1層數實驗1 小波包Level 3 65
4.6.2層數實驗2 小波包Level 4 67
4.6.3層數實驗3 小波包Level 5 68
4.6.4層數實驗4 小波包Level 6 69
4.6.5振動資料特徵層數實驗總結 70
第5章 結論與未來展望 72
5.1.1 結論 72
5.1.2 未來展望 73
參考文獻 74


[1]A. Khadersab and S. Shivakumar, “Parametric vibration Analysis of Rotating Machinery,” Materials Today: Proceedings, vol. 5, no. 11, Part 3, pp. 25688-25696, 2018/01/01/ 2018.
[2]J. Piotrowski, Shaft Alignment Handbook, CRC Press, 2006.
[3]T. H. Patel and A. K. Darpe, “Vibration response of misaligned rotors,” Journal of Sound and Vibration, vol. 325, no. 3, pp. 609-628, 2009.
[4]L. A. D. S. Pedotti, R. M. Zago, and F. Fruett, “Fault diagnostics in rotary machines through spectral vibration analysis using low-cost MEMS devices,” IEEE Instrumentation & Measurement Magazine, vol. 20, no. 6, pp. 39-44, 2017, doi: 10.1109/MIM.2017.8121950.
[5]I. Atoui, H. Meradi, R. Boulkroune, R. Saidi, and A. Grid, “Fault detection and diagnosis in rotating machinery by vibration monitoring using FFT and Wavelet techniques,” in 2013 8th International Workshop on Systems, Signal Processing and their Applications (WoSSPA), 12-15 May 2013 2013, pp. 401-406, doi: 10.1109/WoSSPA.2013.6602399.
[6]劉博榕, 主動式旋轉自動平衡儀之設計與實作, 碩士, 機械與機電工程學系, 國立臺灣海洋大學, 基隆市, 2013.
[7]E. García Plaza, P. J. Núñez López, and E. M. Beamud González, “Efficiency of vibration signal feature extraction for surface finish monitoring in CNC machining,” Journal of Manufacturing Processes, vol. 44, pp. 145-157, 2019.
[8]J. Herwan, S. Kano, O. Ryabov, H. Sawada, and M. Watanabe, “Comparing Vibration Sensor Positions in CNC Turning for a Feasible Application in Smart Manufacturing System,” International Journal of Automation Technology, vol. 12, pp. 282-289, 2018.
[9]B. Chen, X. Chen, B. Li, Z. He, H. Cao, and G. Cai, “Reliability estimation for cutting tools based on logistic regression model using vibration signals,"”Mechanical Systems and Signal Processing, vol. 25, no. 7, pp. 2526-2537, 2011.
[10]F.Smarra, G. D. Di Girolamo, V. De Iuliis, A. Jain, R. Mangharam, and A. D’Innocenzo, “Data-driven switching modeling for MPC using Regression Trees and Random Forests,” Nonlinear Analysis: Hybrid Systems, vol. 36, p. 100882, 2020.

[11]M. Gohari and A. M. Eydi, “Modelling of shaft unbalance: Modelling a multi discs rotor using K-Nearest Neighbor and Decision Tree Algorithms,” Measurement, vol.151,p.107253,2020.
[12]S. Sabut, A. Subudhi, and M. Dash, “Automated segmentation and classification of brain stroke using expectation maximization and random forest classifier,” Biocybernetics and Biomedical Engineering, 04/24 2019.
[13]C. Li, R.-V. Sanchez, G. Zurita, M. Cerrada, D. Cabrera, and R. E. Vásquez, “Gearbox fault diagnosis based on deep random forest fusion of acoustic and vibratory signals,” Mechanical Systems and Signal Processing, vol. 76-77, pp. 283-293, 2016/08/01/ 2016.
[14]R. Couronne, P. Probst, and A.-L. Boulesteix, “Random forest versus logistic regression: a large-scale benchmark experiment,” BMC bioinformatics, vol. 19, no. 1, p. 270, 2018.
[15]文淵閣工作室,2019,Python 機器學習與深度學習特訓班,碁峯資訊股份有限公司。
[16]高揚、衛崢、尹會生,2016,白話大數據與機器學習,機械工業出版社。R. C. Gonzalez, R. E. Woods, Digital Image Processing, 2nd 2002, ISBN
0-20- 118075-8。
[17]M. Stéphane, A Wavelet Tour of Signal Processing, 3rd 2009, ISBN
978-0-12-374370-1。
[18]DFT(離散傅立葉變換)與FFT(快速傅立葉變換)初識https://codertw.com/%E4%BA%BA%E5%B7%A5%E6%99%BA%E6%85%A7/131224/
[19]https://www.cnblogs.com/by-dream/p/10088976.html,決策數(Decision Tree),博客園。
[20]http://www.taroballz.com/2018/07/14/ML_decision_tree/,決策數(Decision Tree),Taroballz。
[21]東台精機股份有限公司https://www.tongtai.com.tw/tw/
[22]KEYENCE 台灣基恩斯https://www.keyence.com.tw/ss/products/measure/select/application/runout_vibration.jsp
[23]Wes McKinney、唐學韜等譯,2014,利用 Python 進行數據分析,機械工業出版社。
[24]麥好,2014,機器學習實踐指南案例應用解析,機械工業出版社。

電子全文 電子全文(網際網路公開日期:20250722)
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top
系統版面圖檔 系統版面圖檔