Algorithm for sheet thickness measurement and its application to thermoformed PMMA sheet

One of the evaluative criteria utilized to ascertain the quality of a thermoformed product is the ultimate thickness of the sheet achieved through the processing. In determining the visual performance of optical products such as aircraft canopy, windscreens, etc., thickness distribution is crucial. Consequently, precise thickness measurement is the most essential aspect of quality control. To measure thickness, a variety of mechanical and optical measurement systems, in addition to several manual systems, are available. The manual intervention restricts the measurement system to lesser measurements and simplifies the system to simple geometries. Furthermore, post-processing of the image or point data is necessary to obtain thickness distribution using an optical measurement system. However, manual intervention is exceptionally time-consuming and may result in inaccurate outcomes. As a result, the current investigation put forth an algorithm based on machine learning to measure the precise thickness distributions from point data obtained through the measuring system. The algorithm's functionality is illustrated through the thermoforming of a PMMA hemispherical dome at various forming pressures. Point data for thickness measurements of the hemispherical domes were acquired using the Rapid-I system of measurement. Utilizing the proposed algorithm, the thickness distribution of the hemispherical domes was measured accurately and efficiently.