Yokogawa Electric Corporation and JSR Corporation announce the successful conclusion of a field test in which AI was used to autonomously run a chemical plant for 35 days, a world first. This test confirmed that reinforcement learning AI can be safely applied in an actual plant, and demonstrated that this technology can control operations that have been beyond the capabilities of existing control methods (PID control/APC) and have up to now necessitated the manual operation of control valves based on the judgments of plant personnel. In this field test, the AI solution successfully dealt with the complex conditions needed to ensure product quality and maintain liquids in the distillation column at an appropriate level while making the maximum possible use of waste heat as a heat source. In so doing it stabilized quality, achieved high yield, and saved energy.
While rain, snow, and other weather conditions were significant factors that could disrupt the control state by causing sudden changes in the atmospheric temperature, the products that were produced met rigorous standards and have since been shipped. Furthermore, as only good quality products were created, fuel, labor, time, and other losses that occur when off-spec products are produced were all eliminated. Safe operations were ensured through a three-step process.
The AI used in this control experiment, the Factorial Kernel Dynamic Policy Programming (FKDPP) protocol, was jointly developed by Yokogawa and the Nara Institute of Science and Technology (NAIST) in 2018 and was recognized at an IEEE International Conference on Automation Science and Engineering as being the first reinforcement learning-based AI in the world that can be utilized in plant management.
