Pareto Optimal Solutions Visualization Techniques for Multiobjective Design and Upgrade of Instrumentation Networks

Instrumentation networks are essential for the proper operation of chemical plants. This paper addresses the problem of designing and upgrading instrumentation networks. A multiobjective approach is selected to solve the optimization problem, instead of the traditional minimum cost formulation. This paper proposes two methodologies for the visualization of Pareto optimal solutions (POSs) for the design and upgrade of sensor networks: one is based on projections of the POS onto specific two-dimensional surfaces, and the second is the representation of the problem in parallel coordinates systems. The proposed network design algorithms enable the decision maker to see several candidate solutions for the network configuration, and at the same time, they allow the number of candidates to be narrowed to satisfy the decision maker’s specifications.

[1]  Tony Perris Process plant performance: measurement and data processing for optimization and retrofits : by Frantisek Madron (Ellis Horwood, 1992, ISBN 0 13 723875 4, 300 pp, £46.00) , 1994 .

[2]  Paolo Fiorini,et al.  Configuration space representation in parallel coordinates , 1989, Proceedings, 1989 International Conference on Robotics and Automation.

[3]  Ivo F. Sbalzariniy,et al.  Multiobjective optimization using evolutionary algorithms , 2000 .

[4]  E. Wegman Hyperdimensional Data Analysis Using Parallel Coordinates , 1990 .

[5]  Miguel J. Bagajewicz,et al.  New MILP formulation for instrumentation network design and upgrade , 2002 .

[6]  Daniel C. H. Yang,et al.  Mobility analysis of planar four-bar mechanisms through the parallel coordinate system , 1986 .

[7]  D. Chmielewski,et al.  On the theory of optimal sensor placement , 2002 .

[8]  Miguel J. Bagajewicz Process Plant Instrumentation: Design and Upgrade , 2000 .

[9]  Miguel J. Bagajewicz,et al.  Reallocation and upgrade of instrumentation in process plants , 2000 .

[10]  José Alberto Hernández,et al.  An Evolutionary Approach for the Design of Nonredundant Sensor Networks , 2001 .

[11]  Kaisa Miettinen,et al.  Nonlinear multiobjective optimization , 1998, International series in operations research and management science.

[12]  K. Papalexandri,et al.  A Parametric Mixed-Integer Optimization Algorithm for Multiobjective Engineering Problems Involving Discrete Decisions , 1998 .

[13]  B. Koehret,et al.  Optimal selection of sensor location on a complex plant, using a graph oriented approach , 1994 .

[14]  Donald J. Chmielewski Convex Methods in Sensor Placement , 2001 .

[15]  Miguel J. Bagajewicz,et al.  On the impact of corrective maintenance in the design of sensor networks , 2000 .

[16]  Alfred Inselberg,et al.  Convexity algorithms in parallel coordinates , 1987, JACM.

[17]  Miguel J. Bagajewicz,et al.  Design and retrofit of sensor networks in process plants , 1997 .