Research On Tool Fault Diagnosis Based On Membrane Computing

Tools usually be worn in the machining production process, If we can not stop to detect or replacethem in time, which maybe result in processing discontinuity, causing the workpiece scrapped or evendamaging the machine tool or making the flexible manufacturing system paralysis,and therefore it may leadto great conomic losses.So it is essential to monitor and diagnosis the tools’ working state online.A new method, based on empirical mode decomposition (EMD) and membrane computing (P system)for tools fault monitoring and diagnosis, is introduced to monitor and judge tools’ processing status in thispaper. Firstly, the cutting force signal was selected as research object through comparative analysis.A largenumber of experimental data was obtained with the cutting force measuring system developed by KistlerCompany. Secondly, using the intrinsic mode functions (IMF) based on the empirical mode decompositionmethod, we can get the energy value of each IMF component though Hilbert transform and set them as thefeature vectors. Finally, based on membrane computing, the hierarchical structure of the membrane, theobjects, multiple sets and evolutionary rules were studied to construct a new adaptive spiking neuron Psystem (ASN P system). Based on the ASNP system the fault diagnosis model was set up for tools’ cuttingforce signal.A new adaptive spiking P system identification model has been researched in this paperby using the membrane computing theory. The simulation results show that the proposed model hasgood performance in tools fault diagnosis.