Joint Chance Constrained Process Optimization through Neural Network Approximation

• Published in: Computer Aided Chemical Engineering Vol. 49; pp. 1237 - 1242
• Main Authors: Yang, Shu-Bo, Moreira, Jesús, Li, Zukui
• Format: Book Chapter
• Language: English
• Published: 2022
• Subjects: Artificial Neural Network; Joint Chance Constrained Optimization; Machine Learning and Big Data; Sample Average Approximation; Stochastic Optimization; Machine Learning and Big Data; Joint Chance Constrained Optimization; Artificial Neural Network; Sample Average Approximation; Stochastic Optimization
• ISBN: 9780323851596; 0323851592
• ISSN: 1570-7946
• DOI: 10.1016/B978-0-323-85159-6.50206-2

A neural network-based approach is proposed in this work for joint chance-constrained optimization (JCCP) problems. In the proposed approach, a joint chance constraint (JCC) is first reformulated as a quantile-based inequality to reduce the complexity in approximation. Then, the quantile function (QF) in the inequality is replaced by an empirical QF through sample average approximation. The empirical QF is further approximated by a ReLU artificial neural network (ANN). Afterwards, the ReLU ANN is incorporated into the optimization model that enables the JCCP to be solved as a deterministic optimization problem. To demonstrate the proposed approach, a case study on ethylene glycol (EG) production process yield maximization is studied. The results show that the proposed approach can efficiently solve a nonlinear JCCP problem with non-conservative constraint satisfaction.