Online Cutting Temperature Prediction Using Ink-Jet Printed Sensors and Model Order Reduction Method

In metal cutting, how to measure the tool tip temperature is always an issue. The highest temperature occurs at the contact surface between the tool and the chip, which is difficult for non-contact measuring methods such as the infrared thermal imaging technique. For other measuring methods, such as thermocouples, an additional small hole is required to be drilled before the sensor is able to be placed at the designated position, which greatly increases the cost. This paper presented a cutting temperature measurement with an ink-jet printed thermistor array. The printed sensor had high thermal index β, which possessed high temperature sensitivity, while its miniature dimension contributed to a fast response time. The ink-jet printing sensors can be made in advance so the setup time is short. Also, the sensors can be easily installed at different locations on the tool or the workpiece. In order to estimate the tool tip temperature, the finite element method (FEM) was used with the measured temperatures as inputs, which was known as an inverse heat conduction problem (IHCP). In order to increase computation efficiency to meet the requirement of online monitoring, the model order reduction method (MOR) was applied. In both non-cutting and cutting experiments, the temperature history could be easily estimated. In this study, the tool tip temperature was updated in 0.72 s, while the errors were only about 10% in non-cutting tests. This made it possible for online monitoring of cutting temperatures, while complex tool geometry and boundary conditions were considered.