Tool health monitoring using airborne acoustic emission and a PSO-optimized neural network

Tool condition monitoring is in major focus nowadays in order to reduce production downtime due to breakdown maintenance, as timely detection of tool wear reduces the production cost. The paper provides an approach to monitor tool health for a CNC turning process using airborne acoustic emission and a PSO (Particle Swarm Optimization) optimized back-propagation neural network. Acoustic signals for good, average, and worn-out tools are recorded through a microphone. Back-propagation neural network are then trained and optimized using PSO algorithm to classify the tool health. PSO-optimized back-propagation neural network shows better performance for tool health classification as compared to simple back-propagation neural networks.