Research Of Tool Wear Condition Monitoring Based On Cutting Sound Signal And Image Processing

The key to ensuring smooth production is to monitor tool wear condition in the large-scale automation production. This paper researches on the tool wear condition monitoring by cutting sound signal and workpiece surface texture based on analysis of the relative situation.The first chapter addresses the research background, summarizes the research status quo of the tool condition monitoring, and analyses the characteristics and the advantage. Therefore, the research problems are presented.The second chapter summarizes the tool wear mechanism, wear form and the blunt tool criteria, and analyzes the factors in affecting cutting sounds and workpiece surface texture during the tool wear. Furthermore, the experiment project is presented.The third chapter analyzes the signal from two different point of view (time field and frequency field) b ased o n t he t heory o f s ignal a nalysis. In t he t ime field, t ime series is statistically analyzed; while in the frequency field, all kinds of frequency chart are estimated again; then one method is compared with another one. Finally, we obtain feasible appraisal criteria for the tool wear status.The fourth chapter analyzes the images for workpiece surface texture using characteristic parameters for the two main theories of the gray co-occurrence matrix and gray run length based on the theory of image texture analysis. Accordingly, the ideal parameters are obtained that can reflect the tool wear status.The fifth chapter analyzes independently the advantages and disadvantages of the sound and images based on the information fusion using artificial nerve network, thus getting the constringent nerve network model.The sixth chapter gives the summary and prospects on this research.