RELIABILITY-BASED MULTI-OBJECTIVE OPTIMIZATION APPLIED TO CHEMICAL ENGINEERING DESIGN

Fran S Lobato,Aldemir A Cavalini Jr,Márcio A Da Silva,Valder Steffen Jr

doi:10.1590/0104-6632.20190361s20170392

Fran S Lobato, Aldemir A Cavalini Jr + Show 2 more

Open Access

https://doi.org/10.1590/0104-6632.20190361s20170392

Copy DOI

Abstract

ABSTRACT Chemical engineering optimization represents a significant challenge due to the complexity of the mathematical models that are frequently required in this area. These models are normally associated with nonlinear equations that represent mass, energy, and momentum balances, which are submitted to physical, constitutive, environmental, and design limitations. The design of chemical systems is generally carried out by considering the model, the vector of design variables, and system parameters as deterministic values, i.e., small variations in these quantities do not affect the objective function. In this contribution, a new methodology based on a double loop iteration process to evaluate the influence of uncertainties on chemical engineering design is proposed. The inner optimization loop is used to find the solution associated with the highest probability value by using the so-called Inverse Reliability Analysis and the outer loop is the regular optimization loop used to determine the vector of design variables. For this aim, the Multi-Objective Optimization Water Cycle Algorithm is improved, adopting a mechanism of neighborhood exploration. For illustration purposes, the proposed methodology is applied to mathematical functions and to chemical engineering design. The obtained results demonstrate that the proposed strategy represents an interesting alternative to reliability design in chemical engineering.

Highlights

In chemical engineering design, the model, the vector of design variables, and the vector of parameters are considered as deterministic quantities, i.e., the possible influence of uncertainties on the resulting design is disregarded
The strategy named as e-MOWCA (Multi-objective Optimization Water Cycle Algorithm) is based on the extension of the Water Cycle Algorithm proposed by Sadollah et al (2015c,d) to the multi-objective context
This extension consists in the incorporation of the neighborhood exploration operator, in association with Inverse Reliability Analysis (IRA) to deal with uncertainties

Summary

Introduction

In chemical engineering design, the model, the vector of design variables, and the vector of parameters are considered as deterministic quantities, i.e., the possible influence of uncertainties on the resulting design is disregarded. The computational cost increases since the optimization procedure is applied many times to evaluate the influence of the reliability parameter on the value of the objective function. In the outer optimization loop, e-MOWCA is carried out to define the vector of design variables, which is necessary to solve the reliability-based multi-objective optimization problem.

Results

Conclusion