This study presents (i) a cost-effective retrofit valve automation platform and (ii) an industrially validated, data-driven control strategy for galvanizing processes.
In aging industrial plants, automation upgrades are often constrained by high capit...
This study presents (i) a cost-effective retrofit valve automation platform and (ii) an industrially validated, data-driven control strategy for galvanizing processes.
In aging industrial plants, automation upgrades are often constrained by high capital costs and long procurement lead times, especially when manual valves must be replaced with dedicated control valves. To address this barrier, this study develops a retrofit-oriented automatic valve platform that integrates an encoder-based actuator with commercial manual valve bodies and provides digital connectivity via Modbus RTU and long-range wireless communication, enabling installation without re-piping or extended production downtime. The proposed actuator demonstrates practical responsiveness, achieving a full-stroke motion of a 1/2-inch ball valve from 0°to 90° with a settling time of 6 s.
In a separate but related effort, this research investigates a data-driven control methodology for zinc coating weight regulation in Continuous Galvanizing Lines (CGL), where strong nonlinearity and transport delay can limit conventional feedback control and where predictive model-based approaches can degrade under model mismatch in aging equipment. Operational robustness is improved through a Case-Based Reasoning (CBR)-based operating recipes from historical successful runs, combined with an adaptive feedback mechanism that updates controller gains according to line speed and process sensitivity. An asymmetric feedforward policy is additionally introduced to provide a safety margin and mitigate under-coating risks.
Field trials indicate that the coating-related control deviation remains within 10g/m² of the set value(SV) without falling below SV−10 g/m². supporting stable operation under practical Continuous Galvanizing Line(CGL) disturbances.