US 12,449,791 B2
Real-time prediction and regulation methods and systems of product quality based on process dynamic pattern
Xiaoli Luan, Wuxi (CN); Niannian Zheng, Wuxi (CN); Haiying Wan, Wuxi (CN); Shunyi Zhao, Wuxi (CN); Yuqing Ni, Wuxi (CN); and Fei Liu, Wuxi (CN)
Assigned to JIANGNAN UNIVERSITY, Wuxi (CN)
Filed by JIANGNAN UNIVERSITY, Wuxi (CN)
Filed on Feb. 6, 2023, as Appl. No. 18/164,822.
Claims priority of application No. 202210826869.6 (CN), filed on Jul. 14, 2022.
Prior Publication US 2024/0028016 A1, Jan. 25, 2024
Int. Cl. G05B 19/418 (2006.01)
CPC G05B 19/41875 (2013.01) 9 Claims
OG exemplary drawing
 
1. A real-time prediction method of product quality based on a process dynamic pattern, comprising:
S1: constructing a state space probability model of quality pattern dynamic motion equation;
S2: calculating a probability density function distribution of a quality pattern according to the state space probability model of quality pattern dynamic motion equation;
S3: performing optimized learning on parameters of the state space probability model of quality pattern dynamic motion equation according to the probability density function distribution, to obtain optimal model parameters and an optimized state space probability model of quality pattern dynamic motion equation; and
S4: performing online prediction on product quality indicators based on the optimized state space probability model of quality pattern dynamic motion equation,
wherein the performing optimized learning on parameters of the state space probability model of quality pattern dynamic motion equation, to obtain optimal model parameters and an optimized state space probability model of quality pattern dynamic motion equation specifically comprises steps of:
S31: calculating an expectation of a maximum likelihood function according to a probability density function distribution of the quality pattern in a Gaussian distribution;
S32: calculating the optimal model parameters by using an expectation maximum likelihood algorithm; and
S33: introducing the optimal model parameters into the constructed state space probability model of quality pattern dynamic motion equation, to obtain the optimized state space probability model of quality pattern dynamic motion equation,
wherein a method of the optimized learning in step S3 is selected from the group consisting of an expectation maximum likelihood algorithm, a genetic algorithm, a particle swarm optimization algorithm, a simulated annealing algorithm, a greedy algorithm, and a neighborhood search algorithm;
wherein a method for performing online prediction on product quality indicators in step S4 is selected from the group consisting of a Kalman filter method, an iterative least square method, a maximum a posteriori estimation method, a polynomial interpolation method, and a finite impulse response method; and
wherein the real-time prediction method further comprises a Bayesian network; and the real-time prediction method implements online prediction and performs online regulation on product quality indicators to ensure that product quality remains at an optimal level.