應用時頻分析技術於車用齒輪箱振動測試台之故障特徵診斷與實驗驗證;Application of Time-Frequency Analysis Techniques for Fault Feature Diagnosis and Experimental Verification of Vehicle Gearboxes on a Vibration Test Rig

NCUIR > Engineering College > Graduate Institute of Mechanical Engineering > Electronic Thesis & Dissertation > Item 987654321/98751

Please use this identifier to cite or link to this item: https://ir.lib.ncu.edu.tw/handle/987654321/98751

Title:	應用時頻分析技術於車用齒輪箱振動測試台之故障特徵診斷與實驗驗證;Application of Time-Frequency Analysis Techniques for Fault Feature Diagnosis and Experimental Verification of Vehicle Gearboxes on a Vibration Test Rig
Authors:	蔡榮峻;TSAI, JUNG-CHUN
Contributors:	機械工程學系
Keywords:	電動車齒輪箱;實驗模態分析;振動診斷;包絡線分析;時頻分析;Electric Vehicle Gearbox;Experimental Modal Analysis;Vibration Diagnosis;Envelope Analysis;Time-frequency analysis
Date:	2025-08-20
Issue Date:	2025-10-17 13:15:11 (UTC+8)
Publisher:	國立中央大學
Abstract:	隨著全球電動車產業迅速發展，整車噪音特性由引擎轉為以電動馬達與齒輪箱為主，振動與異音問題成為影響乘坐品質的關鍵因素。為掌握齒輪箱在實際運轉下的動態行為，需建立高穩定性之振動檢測平台與故障診斷技術。本研究以車用齒輪箱為對象，發展一套整合多體動力學(Multi-Body Dynamics, MBD)、有限元素分析(Finite Element Analysis, FEA)與時頻訊號分析之系統性診斷方法。首先依據幾何模型與傳動參數，建構齒輪箱動態模型，模擬齒輪運轉受力與負載分佈。對振動測試台結構進行模態與應力分析，並透過實驗模態法(Experimental Modal Analysis, EMA)驗證數值模型準確性。根據模態響應與高應力區域結果，優化測試台結構設計，以提升自然頻率並降低與齒輪嚙合頻率的共振風險。本研究亦於變轉速條件下，於時頻分析觀察結構系統的頻率響應行為，並針對齒輪撞傷與軸承外環局部損傷等故障情形，應用RMS、頻譜分析及包絡線分析技術進行診斷，建立對應之頻率指標與辨識準則。研究結果證實所建模擬與分析方法具良好準確性與應用潛力，可為未來傳動系統及振動測試平台之品質監測與預診斷基礎。;With the rapid development of the global electric vehicle industry, the primary sources of vehicle noise have shifted from internal combustion engines to electric motors and gearboxes. Vibration and abnormal noise have become key factors affecting ride comfort. To understand the dynamic behavior of gearboxes under actual operating conditions, it is essential to establish a highly stable vibration testing platform and effective fault diagnosis techniques. This study focuses on vehicle gearboxes and develops a systematic diagnostic method that integrates Multi-Body Dynamics (MBD), Finite Element Analysis (FEA), and time-frequency signal analysis. Based on the geometric model and transmission parameters, a dynamic model of the gearbox is constructed to simulate gear operation forces and load distribution. Modal and stress analyses are conducted on the vibration test rig structure, and the accuracy of the numerical model is validated through Experimental Modal Analysis (EMA). Using modal responses and high-stress region results, the structure of the test rig is optimized to raise natural frequencies and reduce the risk of resonance with gear mesh frequencies. Under variable speed conditions, time-frequency analysis is employed to observe the system’s frequency response behavior. For fault cases such as gear tooth impact and localized damage to bearing outer races, diagnostic techniques including RMS, spectrum analysis, and envelope analysis are applied to establish corresponding frequency indicators and identification criteria. The results confirm that the proposed modeling and analysis methods offer high accuracy and practical applicability, providing a solid foundation for future quality monitoring and predictive diagnostics of transmission systems and vibration testing platforms.
Appears in Collections:	[Graduate Institute of Mechanical Engineering] Electronic Thesis & Dissertation

Files in This Item:

File	Description	Size	Format
index.html		0Kb	HTML	28	View/Open

社群 sharing

Loading...