Real‐Time Optimization in the Chemical Processing Industry

Abstract

The advancements of computation capabilities and efficient flowsheet simulators have led simulation based real-time optimization (RTO) of continuous processes to become a prominent topic in the last years. This contribution discusses prerequisites to help guarantee the economic viability of an RTO project as well as the deliberations that should be performed prior to its initiation. Additionally, considerations regarding collaborations with various technical fields are highlighted and approaches to alleviate some challenges associated with the implementation of RTO are discussed. Motivation and IntroductionAn ever increasing global competition and tightening of product requirements has continuously fortified the necessity of improving production processes [1]. Over the last decades multiple advancements in the field of mathematics, thermodynamics, process systems engineering, and computational sciences have promoted the development of online model-based real-time optimization (RTO). Among these advancements are the availability of efficient flowsheet simulators, the development of improved property calculations methods, the advent of equation oriented modeling, and improved optimization algorithms for nonlinear systems. Due to these advancements, RTO has received a significant amount of attention in both the industrial and academic literature, thus, becoming an ever more prominent topic. An abundance of successful RTO implementations can be found in numerous industrial fields ranging from [2 -4] in the petrochemical industry, [5 -7] in the chemical industry, to [8,9] in the energy and water supply sector. Adding to this, concepts and overviews of RTOtopics were written repeatedly over the last decades, such as contributions by Naysmith and Douglas [1] or by Darby et al. from 2011 [10].Even though the topic is well known and many successful RTO implementations can be found, (~250 to 300 using commercially available techniques [10]) opposing opinions can still be found when it comes to an investment decision concerning RTO. In most cases, the benefits are seen and a consensus exists to carry out an investment. In other cases, however, it was concluded that RTO is not viable. This history as well as experiences of the last years affords us the opportunity to deliberate the success of RTO technology in the chemical industry and to carry out a discussion of when RTO truly becomes viable for a company. Therefore, the aim of this article is to shed light on the following arising questions from an industrial perspective: -The time when a company should continue to invest in RTO applications. -Deliberations performed prior to the implementation of RTO. -Considered collaborations to successfully realize an RTO project. -Considered factors and challenges to increase the likelihood of success of an RTO project. As background information, a brief overview of RTO's role in the control and decision making hierarchy of a plant as well as the main steps in the RTO functionality cycle is given.www.cit-journal.com

Current and future requirements to industrial analytical infrastructure—part 2: smart sensors

“…Data science solutions for model switching will choose the best model being part of the digital twin for the current, control, optimization, or estimation task. This can be supplemented with sematic sensors, in which expert knowledge is embedded-speeding up an implementation project of model-based real-time optimization (RTO) in the process industry [23].…”

Process Modeling, Simulation and Optimization: From Single Solutions to a Multitude of Solutions to Support Decision Making

“…Therefore, in these cases detailed analysis or offline optimization is in general restricted to process parts with predictive models like distillation sequences. As long as in a larger process model only about one or two unit operations are based on descriptive models and the rest is based on predictive models, these process models can be used to support process operation in online applications like condition and performance monitoring and real‐time optimization .…”

“…In general, the use of online optimization seems to be more successful than offline applications, although the effort is only justified in special cases . A robust implementation and application is challenging and expensive, therefore the improvement of such methods should support efficient setup, since the benefit in contribution margin due to the constraints of the plants is often only in the lower 1‐digit percent range.…”

Process analytical technology (PAT) as a versatile tool for real-time monitoring and kinetic evaluation of photocatalytic reactions

“…39 The easy implementation and robustness allows PAT to support the development from the laboratory to an industrial scale. 40,41 This makes PAT a promising candidate for the in situ monitoring of visible light induced photocatalytic reactions.…”