Optimal sampling strategy for high-density polyethylene production processes

A large number of sampling points are required to ensure the accuracy of parameter estimation, which leads to high cost of sample analysis. An optimal sampling strategy was thus put forward based on Latin hypercube sampling to reduce the number of sampling points and ensure parameter estimation accuracy. By solving corresponding optimal sampling sub-problems with the candidate values of parameter space by Latin hypercube sampling, the selection frequency of each potential sampling point was constructed in the optimal sampling strategy. Sampling points were selected by virtue of heuristic strategy and iterative enhancement strategy, and parameter estimation was achieved. The numerical simulation results show that the sampling strategy proposed can greatly reduce the number of sampling points while ensuring parameter estimation accuracy. In addition, it is more stable than the optimal sampling method based on random sampling. © 2022 Zhejiang University. All rights reserved.