Deep reinforcement learning (RL) has attracted research interest in the manufacturing area in recent years, but real implemented applications are rarely found. This is because agents have to explore the given environments many times until they learn how to maximize the rewards for actions, which they provide to the environments. While training, random actions or exploration from agents may be disastrous in many real-world applications, and thus, people usually use computer generated simulation environments to train agents. In this paper, we present a RL experiment applied to temperature control of a chamber for ultra-precision machines. The RL agent was built in Python and PyTorch framework using a Deep Q-Network (DQN) algorithm and its action commands were sent to National Instruments (NI) hardware, which ran C codes with a sampling rate of 1 Hz. For communication between the agent and the NI data acquisition unit, a data pipeline was constructed from the subprocess module and Popen class. The agent was forced to learn temperature control while reducing the energy consumption through a reward function, which considers both temperature bounds and energy savings. Effectiveness of the RL approach to a multi-objective temperature control problem was demonstrated in this research.

Recently, interest in astronomy has increased internationally, and the technological development of lenses for large space telescopes is progressing. The multi-order diffractive engineered (MODE) lenses can make a large space telescope light and thin. However, because glass lenses are difficult to machine, we have adopted a method of molding at high temperature and high pressure. The STAVAX is commercially available chrome alloy stainless steel, and it is applied as various mold materials. The ultrasonic vibration cutting was adopted for ultra-precision machining because the tool wear is severe when cutting the STAVAX with a diamond tool. To achieve a flat surface for smooth ultrasonic vibration cutting, we performed a precise shape cutting using a CBN tool and confirmed and observed changes in the surface roughness and hardness depending on the cutting conditions. The ultrasonic vibration cutting was performed on the surface of the machine using a CBN tool, and the surface roughness was observed. It was confirmed that the surface roughness was impacted by the surface hardness. The specimens with low surface hardness showed the highest surface roughness at approximately 3 nm.