Statistical operating strategies for charging batch reactors

Operating Strategies for Charging Batch Reactors under Multiple Constraints

“…Owing to imperfect control and operation of the equipments, the actual amount of reactant charged into the reactor is usually not exactly the same as the target value, i.e., where ∆ i is the error due to the charging system and, for convenience, it is assumed to be a normally distributed random variable. Two types of error models have been proposed in an earlier study (Tsai and Chang, 1996). For the sake of brevity, detailed descriptions of these model are omitted in the present paper.…”

“…In this study, it is further assumed that the charging errors ∆ i and these measurement errors Ξ i (l) are statistically independent. The detailed descriptions of measurement error models can be found in Tsai and Chang (1996).…”

“…By making use of the techniques suggested by Tsai and Chang (1996), the variances of all measurement errors can be estimated first. Then, maximum likelihood estimates of the rest of the parameters associated with the charging and measurement errors can be determined accordingly.…”

“…Furthermore, any inconsistency identified in the measurement data obtained from redundant sensors is usually resolved on the basis of operation experience. In a previous work, statistics-based strategies have been proposed for charging the batch reactors (Tsai and Chang, 1996). Specifically, various simple error models were first formulated according to the characteristics of different types of charging and measurement methods.…”

Efficient sampling technique for optimization under uncertainty

“…These profiles even if reoptimized using on-line measurements often contain nonnegligible uncertainties due to the system state being inferred from indirect, so-called model-based measurements, which can be subject to both stochastic measurement noise and structural measurement-model mismatch (Terwiesch, 1995). The need to take these uncertainties into consideration in the design and planning stage is well recognized (Reklaitis et al, 1989;Cott and Machhietto, 1989;Watzdorf et al, 1993;Mignon et al, 1995;Tsai and Chang, 1996;Ierapetritou and Pistikopoulos, 1996). This new sampling technique provides an efficient and generalized approach for handling uncertainties in batch process design, optimization, scheduling and planning.…”