Options
A Novel Cycle Partitioning Approach to Reliability Based Optimal Sensor Placement for Linear Flow Processes
Journal
Computer Aided Chemical Engineering
ISSN
15707946
Date Issued
2022-01-01
Author(s)
Prakash, Om
Bhushan, Mani
Abstract
In literature, system reliability for sensor placement design for linear mass flow processes has been defined in terms of cycles (Prakash et al., 2020a). The computation of system reliability involves computation of probability of a union of a set of events, which is NP-hard, thereby limiting its applicability to large processes. To reduce this computational complexity, in the current work, we propose to partition the set of cycles, instead of partitioning the original process. Thus, our approach involves approximating the system reliability function of the original process, rather than approximating the original process. Towards this end, we use a bipartite spectral graph partitioning algorithm to obtain the partitions of cycles. The resulting sensor placement design approach is applied to a case study to show its efficacy.
Volume
49
Subjects