Optimal sensor placement and maintenance strategies for mass-flow networks

A mathematical programming model has been developed in this study to determine the best measurement locations in a given process network and also the optimal numbers of redundant and spare sensors used in a corrective maintenance program. The model solution yields the maximum system availability under a set of user supplied limitations on life-cycle cost and/or estimator's precision. Genetic algorithms were used to identify the optimum in an evolutionary process. The usefulness of the proposed approach is demonstrated with extensive case studies.