Model independent parametric decision making

TY - JOUR

T1 - Model independent parametric decision making

AU - Banerjee, Ipsita

N1 - Funding Information: The authors gratefully acknowledge financial support from National Science Foundation under the NSF CAREER program CTS-9983406 and from the donors of the Petroleum Research Fund administered by the ACS.

PY - 2004/11

Y1 - 2004/11

N2 - Accurate knowledge of the effect of parameter uncertainty on process design and operation is essential for optimal and feasible operation of a process plant. Existing approaches dealing with uncertainty in the design and process operations level assume the existence of a well defined model to represent process behavior and in almost all cases convexity of the involved equations. However, most of the realistic case studies cannot be described by well characterised models. Thus, a new approach is presented in this paper based on the idea of High Dimensional Model Reduction technique which utilize a reduced number of model runs to build an uncertainty propagation model that expresses process feasibility. Building on this idea a systematic iterative procedure is developed for design under uncertainty with a unique characteristic of providing parametric expression of the optimal objective with respect to uncertain parameters. The proposed approach treats the system as a black box since it does not rely on the nature of the mathematical model of the process, as is illustrated through a number of examples.

AB - Accurate knowledge of the effect of parameter uncertainty on process design and operation is essential for optimal and feasible operation of a process plant. Existing approaches dealing with uncertainty in the design and process operations level assume the existence of a well defined model to represent process behavior and in almost all cases convexity of the involved equations. However, most of the realistic case studies cannot be described by well characterised models. Thus, a new approach is presented in this paper based on the idea of High Dimensional Model Reduction technique which utilize a reduced number of model runs to build an uncertainty propagation model that expresses process feasibility. Building on this idea a systematic iterative procedure is developed for design under uncertainty with a unique characteristic of providing parametric expression of the optimal objective with respect to uncertain parameters. The proposed approach treats the system as a black box since it does not rely on the nature of the mathematical model of the process, as is illustrated through a number of examples.

KW - HDMR

KW - parametric analysis

KW - uncertainty

UR - https://www.scopus.com/pages/publications/6344278102

UR - https://www.scopus.com/pages/publications/6344278102#tab=citedBy

U2 - 10.1023/B:ANOR.0000045280.55945.e8

DO - 10.1023/B:ANOR.0000045280.55945.e8

M3 - Article

SN - 0254-5330

VL - 132

SP - 135

EP - 155

JO - Annals of Operations Research

JF - Annals of Operations Research

IS - 1-4

ER -