Current world climate events have put the spotlight on the impending climate crisis in the minds of governments, consumers and investors, driving decarbonizing of major emitters such as the process industries. New digital technologies make it possible to achieve quick gains through accelerated design and day-to-day reduction in emissions from operations. Digital design techniques based on high-fidelity predictive models enable accelerated development of optimal low-carbon processes via rapid exploration of the decision space and reduced need for experimentation and pilots. Digital applications for operations make it possible to reduce daily plant emissions by coupling digital process twins containing deep process knowledge with real-time plant data, for monitoring, soft-sensing, real-time optimization and what-if decision support. This seminar describes recent advances, illustrated with examples from refineries, polymer production and development of hydrogen processes among others.
Please note that this agenda is subject to change prior to the event.
Sang Phil Han, SPSE
|09:05-09:30||Model-based digital technologies: Deploying deep process knowledge to accelerate decarbonization
Costas Pantelides, SPSE
At a time when decarbonization is driving great change in the process industries, the digital revolution is making it possible to significantly accelerate new process development and improve efficiency of operations. Prof. Pantelides describes how state-of-the-art digital design techniques use high-fidelity predictive models (digital process twins) of process physics and chemistry systematically to accelerate innovation and arrive at economically optimal process designs, based on quantified, managed technology risk. A key advance for operations is the ability to deploy such digital process twins to bring deep process knowledge into process operations and control. A new generation of Digital Applications generates value by combining real-time or historic plant data with the knowledge in the models; the wealth of new information generated is used to create additional daily value from the plant via monitoring, soft-sensing, real-time optimization and “what-if” operations decision support.
|09:30-09:55||Reducing daily plant CO2 emissions using site-wide utilities optimization and operator advisory tools
Steve Hall, SPSE
The combination of equation-based modelling technologies and a next-generation digital applications framework provide an environment for deploying systems to advise operators on minimizing CO2 emissions in real-time. This presentation shows how a model of a whole site utility system is created using SPSE’s gPROMS Utilities tool, then deployed to operations using the gPROMS Digital Applications Platform. The resulting digital twin of the utility system links to plant data systems, updating itself through machine-learning capabilities, validating actual performance and continually calculating optimal operating conditions. The objective function can be to minimize CO2 emissions or costs. CO2-reducing operating changes are highlighted in dashboards in real time, giving operators greater insight and confidence to run the process safely at the optimum point. Emissions reduction actions can be taken quickly with confidence, leading to improved environmental performance.
|09:55-10:20||Enhancing polymer production with non-linear model-predictive control to minimize waste and reduce environmental impact
Chris Leingang, SPSE
While linear model-predictive control (MPC) has been applied very effectively to large-scale continuous plants for 30 years, it cannot easily be applied to high-value batch or semi-continuous processes with multiple products and frequent grade changes, such as polymer production. One of the most exciting recent developments is viable non-linear MPC. NLMPC is now becoming a reality because of the availability of accurate, physics-based process models, the increased speed and robustness of solution techniques, and the availability of real-time data delivered by recent digitalization initiatives. The presentation describes the application of NLMPC to polymer processes for continuous quality monitoring, yield maximization and acceleration of grade change to minimize production of low-value off-spec material. The promising early results show that NLMPC has the potential to add around a week’s on-spec production to a typical plant annually, while reducing wasteful rework and emissions.
|10:20-10:45||Model-based engineering of carbon capture, utilization and storage systems
Bart de Groot, SPSE
The decarbonization of advanced economies requires fundamental changes in the chemical process industry. With many of the transformative technologies required still in their infancy, there is a need for solutions that support a reduction in GHG emissions in the immediate future. Carbon capture, utilization and storage (CCUS) can help reduce emissions from today’s fossil-based production methods. By their very nature, CCUS facilities need to be tightly integrated into other process systems. This raises questions on how best to maximise efficiency of the overall system, what impact the CCUS process has on process dynamics and control, and how best to reduce capital and operating expenditure. Questions like these require sufficiently accurate representation of the process, in order to accurately predict plant performance, and operate close to true limits.
In this presentation, we present how digital design techniques are used to map system interactions, predict process responses in highly transient scenarios, optimize equipment and system designs, and ultimately provide reassurance to all stakeholders in the carbon chain to confidently navigate the road to decarbonization.
|11:00-11:25||Enabling the hydrogen economy: Digital design solutions to accelerate hydrogen applications
Bart de Groot, Jorge Aguerrevere, SPSE
Hydrogen is widely expected to play a key role in the decarbonization of advanced economies, and technologies for more sustainable production and utilization of hydrogen are seeing an increased interest. For many of these technologies, challenges remain to scale up, reduce costs, integrate into wider process systems and increase confidence and acceptance. In this presentation, we present how digital design techniques using digital twins based on high-fidelity, predictive process models can help speed up technology development, map system interactions, determine optimal buffer sizing, especially in highly transient scenarios, optimize equipment and system designs, and ultimately provide reassurance to all stakeholders in the hydrogen economy to confidently navigate the road to decarbonization.
|11:25-11:50||Digital Applications for high-performance hydrogen production via Steam Methane Reforming
Steve Hall, SPSE
This presentation describes how SPSE has deployed a digital twin for an SMR reactor using its gPROMS Process tool and then deployed it using its gPROMS Digital Applications Platform. The high-fidelity model of the reactor fully describes its behavior and allows it to be accurately modelled, with the reactor operating conditions being optimized to maximize production. The resulting digital twin of the reactor links to live plant data systems, autotuning and updating itself through machine-learning capabilities, validating actual performance, providing soft-sensed information on internal conditions and continually calculating optimal operating conditions to maximize hydrogen production. H2 production rate and purity increases are highlighted in dashboards in real time, giving operators greater insight and confidence to run the process safely at the optimum point. Operator actions can be taken quickly with confidence, leading to improved hydrogen production.
|11:50-12:15||Bioprocessing digital twins: From process development to online decision support systems
Edward Close, SPSE
Science-based data-calibrated digital twins based on mechanistic models use process understanding combined with targeted experimentation to describe and predict the quantitative behaviour of processes for process design and optimisation, risk assessment, design space exploration and scale up/tech transfer. The formulated products industries are increasingly using bioreactor and chromatography digital twins to address challenges described above, bringing value to their businesses across R&D, engineering, and manufacturing functions.
In this presentation, we will show examples where model-based solutions for bioreactors and chromatography have provided value to industry, including enabling R&D efficiency in process design and optimization, and de-risking the move from batch to continuous processes. We will finish with SPSE’s vision on how model-based solutions can bring value in manufacturing via the deployment of digital twin solutions for online decision support.
|12:15-12:30||Discussion & Conclusion
Sang Phil Han, SPSE
To attend this seminar, registration for 2021 KIChE Fall Meeting is required. (Please send email to [email protected] for more information)