| Summary: | 碩士 === 南亞技術學院 === 機械工程系機械與機電工程碩士班 === 107 === This study performed measurement experiments on three standard ring gauges of different sizes (25 mm, 85 mm, and 125 mm) using an L9 orthogonal array. The control factors included moving speed, scanning speed, rod length, and temperature, and the noise factors included pre-hit distance and retract distance. The quality objective was a nominal-the-best characteristic. Taguchi analysis revealed that the optimal parameter combination for the smallest standard ring gauge (25 mm) was A3B3C1D1 and produced a measurement value of 24.998108 mm. The optimal parameter combination of the original design in the L9 orthogonal array was A3B3C2D1 and produced a measurement value of 24.998115 mm, thereby presenting an improvement rate of 6.09%. For the medium standard ring gauge (85 mm), Taguchi analysis revealed that the optimal parameter combination was A1B1C1D1, which was identical to the optimal parameter combination of the original design in the L9 orthogonal array and thereby presented no improvement in the measurement quality of the coordinate measuring machine. For the largest standard ring gauge (125 mm), Taguchi analysis revealed that the optimal parameter combination was A1B1C1D1, which was identical to the optimal parameter combination of the original design in the L9 orthogonal array and thereby presented no improvement in the measurement quality of the coordinate measuring machine.
To improve the full-range measurement accuracy of a coordinate measuring machine, this study employed the Taguchi method to identify the single quality objective optimal parameter combinations for three standard ring gauges of different sizes (25 mm, 85 mm, and 125 mm). We then incorporated grey relational analysis to identify the multiple quality objective optimal parameter combination for the full range of the machine and increase its full-range measurement quality.
Keywords: coordinate measuring machine, standard ring gauge, Taguchi method, nominal-the-best characteristic, grey relational analysis
|
|---|
