Machines condition analysis based on output machining parameter and vibration signal monitoring

Abstract

This dissertation reports an investigation on the machine condition for a group of similar milling machines based on vibration signal and surface roughness quality. This study is motivated by an issue in maintenance field where an appropriate technique has to be applied to monitor the condition of a machine. In this work, the machining system studied are the group of milling machines within the laboratory in the School of Manufacturing Engineering, Universiti Malaysia Perlis. To further understand the subject of studies, the extensive literatures has been done related to condition monitoring, vibration analysis and also the effect to the surface roughness. The application of fishbone analysis in identifying the possible causes of the problem also has been included in the review. In this work, the vibration level have been measured at three different points of the machines (ram, spindle and vertical head). These vibration level is then analysed to determine the causes of inaccuracy in the machining quality. From the experimental results, it is found that surface quality is strongly affected by the vibration. Machine 2 is found in critical state due to the highest level of vibration as compared to the other machines. Machine 1 and 8 have demonstrated the unusual condition where the vibration level is moderate however the surface roughness value is high and does not synonym to vibration trend. In the present work, the root cause for vibration and poor surface quality are performed by using fishbone analysis for the affected machine. The findings are discussed and the conclusions reached in this project and recommendations for future work are enclosed