Prediction and Diagnosis System for Malfunctions of Electric Scooters

  • 汪 聖翔

Student thesis: Master's Thesis


Information on structural vibration are usually utilized to judge its status Because vibration on different structural will cause different vibration modes even small differences Electric scooter used in the process the structural is likely to damage due to vibration fatigue So this research mainly focuses on designing a malfunction-predicting system for electric scooter in order to detect status of the electric scooter quickly and foresee a malfunction that may happen in the future An accelerometer and data acquisition module and LabVIEW are integrated to develop a measurement and diagnosis system in this thesis By utilizing vibration signals from the electric scooter to diagnose the possible malfunction before diagnosing features are needed for identifying information Frequency domain information frequency multiplication analysis and statistics are used to detect a certain status’s vibrational features This thesis aims to address the screw problem of the motor shaft and front axle shock absorber abnormalities etc The malfunction status is used to identify the fault based on the status frequency feature for locating the fault and identifying the information Furthermore it is evaluated whether the electric scooter has the malfunction signal feature to determine the type of fault in order to achieve a predictive diagnosis The experiment shows that the diagnostic accuracy of normal status is 93 33% a loose screw in the motor shaft has 95 56% a loose screw in the front axle has 90 56% and a shock absorber abnormality has 87 8% The results show that the frequency feature can effectively diagnose the status of an electric scooter
Date of Award2016 Sep 8
Original languageEnglish
SupervisorHeiu-Jou Shaw (Supervisor)

Cite this