US 12,449,794 B2
Production system
Hideki Sato, Yokohama (JP)
Assigned to CHIYODA CORPORATION, Kanagawa (JP)
Appl. No. 18/997,888
Filed by CHIYODA CORPORATION, Yokohama (JP)
PCT Filed Aug. 5, 2022, PCT No. PCT/JP2022/030089
§ 371(c)(1), (2) Date Jan. 23, 2025,
PCT Pub. No. WO2024/029069, PCT Pub. Date Feb. 8, 2024.
Prior Publication US 2025/0264871 A1, Aug. 21, 2025
Int. Cl. G05B 19/418 (2006.01)
CPC G05B 19/41885 (2013.01) [G05B 2219/32359 (2013.01)] 10 Claims
OG exemplary drawing
 
1. A production system that performs a first process and a second process continuously, comprising:
a first device;
a second device;
at least one first sensor; and
a processor, wherein
the first device obtains a first processed product from a raw material by performing the first process, which is any one of a chemical reaction, a separation, and a drying,
the second device obtains a second processed product from the first processed product by performing the second process, which is any one of a chemical reaction, a separation, and a drying,
the first sensor detects a first device information that is a state of the first device,
the processor controls the first device based on a preset first condition list including a plurality of control parameters of the first device, and modifies a preset second condition list including a plurality of control parameters of the second device, and controls the second device based on the modified second condition list,
the first device information includes a flow rate of the raw material and at least one of an inlet temperature of the first device, an outlet temperature of the first device, an internal temperature of the first device, an internal pressure of the first device, and a flow rate of the raw material in the first device,
the preset first condition list includes at least one of a target internal temperature of the first device, a target internal pressure of the first device, and a target flow rate of the raw material in the first device,
the preset second condition list includes at least one of a target internal temperature of the second device, and a target internal pressure of the second device,
the processor acquires a yield of a first processed product based on a first model that outputs the yield of the first processed product in response to an input including a raw material information related to a property of the raw material and the first device information,
the processor acquires a modified value of the preset second condition list based on a second model that outputs a modified value of the preset second condition list in response to an input including the yield of the first processed product,
the processor acquires a modified second condition list based on the preset second condition list and the modified value of the preset second condition list,
the processor acquires an estimated first device information based on a third model that outputs the estimated first device information, which is information related to a state of the first device and is not detected by the first sensor, in response to an input including the raw material information, the first device information, and the yield of the first processed product,
the processor acquires a modified value of the preset first condition list based on a fourth model that outputs a modified value of the preset first condition list in response to an input including at least one of the estimated first device information and the yield of the first processed product,
the processor acquires a modified first condition list based on the preset first condition list and the modified value of the preset first condition list,
the processor controls the first device based on the modified first condition list instead of the preset first condition list, and
the first model and the second model include at least one of a machine learning model and a mathematical model.