Infrared monitoring of aluminium milling processes for reduction of environmental impacts

In modern manufacturing contexts, process monitoring is an important tool aimed at ensuring quality standard fulfilment whilst maximising throughput. In this work, a monitoring system comprised of an infrared (IR) camera was employed for tool state identification and surface roughness assessment with the objective of reducing environmental impacts of a milling process. Two data processing techniques, based on statistical parameters and polynomial fitting, were applied to the temperature signal acquired from the IR camera during milling operations in order to extract significant features. These features were inputted to two different neural network based procedures: pattern recognition and fitting, for decision making support on tool condition and surface roughness evaluation respectively. These capabilities are discussed in terms of reducing waste products and energy consumption whilst further improving productivity.