In-Situ Monitoring and Closed-Loop Control in Vat Photopolymerisation: Sensors, State Estimation, Digital Twins and Adaptive Exposure
Segurola, Juan
2026-08-27 · Informe · Versión 1.1
Los títulos y las descripciones bibliográficas se conservan en el idioma del registro original. Los enlaces a los PDF indican los idiomas disponibles.
Vat photopolymerisation can expose a complete layer in seconds, but the material state evolves through oxygen consumption, gelation, conversion, heat generation, vitrification, shrinkage and separation from the vat film. Conventional working curves and test coupons characterize this response before production and therefore remain open-loop calibration. Recent research has introduced ultrasonic sensing, interferometry, thermal imaging, Raman spectroscopy, optical monitoring, force sensing, machine learning, digital twins and visual defect repair. The evidence, however, spans different levels. Ultrasonic and interferometric systems estimate layer thickness, degree of cure or modulus in real time; thermal-chemical models infer inaccessible interface temperature and conversion; digital twins reconstruct geometry, stiffness and defects; and visual- guided or hybrid model-data systems modify scraper motion, repair exposure, filling time or other process parameters during the same build. An adjacent hydrogel study demonstrates true feedback control of storage modulus using a piezoelectric sensor and bang-bang or fuzzy logic controllers, but it is not a VPP printing demonstration. This review separates measurement, state estimation, prediction, feedforward compensation, alarm/interlock, in-situ repair and feedback control. It defines sensor calibration, ground-truth, latency, blind- zone, uncertainty and intervention-efficacy requirements; proposes reporting aids for deadline compliance, estimator error, observability coverage and correction efficacy; and provides a qualification matrix and failure-mode table. No universal controller, acceptance threshold or VPP-specific standard is proposed. The central principle is: monitoring observes the process; estimation interprets the observation; prediction projects the future; and closed-loop control changes the process and proves that the change improved the declared outcome.