Cold bulk metal forming has made large-scale production of small complex solid parts economically feasible. Tooling used in metal forming poses many uncertainties in the preliminary cost estimation and production process. Continual tool replacement and maintenance will dramatically reduce productivity and raise manufacturing cost. An on-line tool condition monitoring system applying artificial neural networks (ANN) to integrate information from multiple sensors for cold bulk metal forming has been proposed. The information used in the ANN to monitor the condition of the tool will be forces, acoustic emission signals and some forming process conditions (tool temperature, knock rates and surface lubrication of in feed material). Preliminary work using both strain g9uges and piezo-electric sensors has attempted to assess force signatures for process monitoring purposes in an industrial situation. It was apparent from this preliminary work that for a bulk deformation process in a production situation, a multi-sensor approach will be necessary for effective monitoring of tool condition.
展开▼
