Abstract
MILP (Mixed Integer Linear Programming) method for simultaneous gross error detection and data reconciliation has been proved to be an efficient way to adjust process data with material and other balance constraints. But the efficiency will decrease significantly when the MILP method is applied in a large-scale data rectification problem because there are too many binary variables to be considered. In this paper, a strategy is proposed to generate a list of gross error candidates with reliability factors. The list of candidates are combined into the MILP objective function to improve the efficiency and accuracy through reducing the number of binary variables and giving accurate weights for suspected gross errors. Industrial examples are provided to show the efficiency of the algorithm.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
