Intelligent Temperature Control of a Stretch Blow Molding Machine Using Deep Reinforcement Learning

Stretch blow molding serves as the primary technique employed in production of polyethylene terephthalate (PET) bottles. Typically, a stretch machine consists various components, including preform infeed system, transfer heating bottle discharge etc. Of particular significance is temperature control within which significantly influences quality PET bottles, especially when confronted with environmental changes between morning and evening during certain seasons. The on-site operators often need to adjust infrared lamps system several times. adjustment process heavily relies on personnel’s experience, causing challenge for manufacturers. Therefore, this paper takes object uses deep reinforcement learning method develop an intelligent approach adjusting parameters. proposed aims address issues such interference aging variation lamps. Experimental results demonstrate that achieves automatic parameters process, effectively mitigating influence ensuring stable surface ±2 ℃ target temperature.