Research On Tool Fault Diagnosis Based On Fuzzy Support Vector Machine

With the development of economy and technology, NC machine tools has gradually become an important symbol of national level of modernization.However, the failure of NC machine tools has become a major bottleneck affecting productivity. NC machine tools at this stage in China’s small and medium enterprises do not have the use of the general state of the intelligent diagnostic function tool, mostly relying on experience to diagnose and analysis. Due to the limitations of human experience, efficiency and accuracy of the diagnosis of the presence of great uncertainty. For these reasons, these small and medium enterprises urgently need to use artificial intelligence methods for monitoring the state of wear of the tool in order to improve the accuracy of diagnosis.The main objective of the dissertation writing is to use scientific methods to collect data, and the use of artificial intelligence fusion technique for diagnosis tool for SME problems arise, and on this basis, the system is suitable for business and the realization of the overall design. Continue to conduct research and analysis about the issues raised above.Before the start of this dissertation and tool wear mechanism through fault diagnosis theory NC machine tools in-depth analysis, obtained fuzzy theory and support vector machines, and on this basis put forward the idea of fuzzy support vector machine. Establish relevant mathematical models and multi-fault classifier based on this idea. The simulation results by fuzzy support vector machine model can effectively solve the existing classification categories when the blind to improve the accuracy of classifiers, while the diagnostic process with ambiguity, and memory capacity, used diagnostic tool wear has high practical sex.Finally,based on the characteristics of the enterprise and tool failures. design and implement a system for small and medium enterprises as well as the database structure model.