Crystallisation optimisation and control using surrogate modelling and adaptive model predictive control

Crystallisation is inherently a challenging process to control as a result of a number of interacting parameters including nucleation, agglomeration, breakage, hydrodynamics, and so on. Due to molecular nature of the underlying mechanisms, robust modelling, control and optimisation is not straightforward. This can result in inconsistent batches, higher downstream processing costs, and difficulties controlling quality long-term. Population Balance Models (PBMs) and Model Predictive Control (MPC) exist for these systems but have limitations. PBMs can be computationally complex and as such real-time control is difficult. Due to the lack of detailed understanding of the full scope of interactions in crystallisation, PBMs will be of reduced complexity than the real system. This project proposes the use of surrogate models such as physics informed neural networks and reinforcement learning, utilising their nonlinear and adaptive predictive capabilities, to capture the full scope of behaviour contained within crystallisation system data. By combining PBMs and experimental data in the training of these surrogate models, it is proposed that a robust control and optimisation method can be developed that can be applied to a range of crystallisation systems. The models can enable real-time control and optimisation of a desired crystal size distribution, whilst balancing other constraints and objectives through multi-objective optimisation. The final goal is a control system that realises an established digital twin, with limited experimental data required.

The CDT in Molecules to Product has the potential to make a real impact as a consequence of the transformative nature of the underpinning 'design and supply' paradigm. Through the exploitation of the generated scientific knowledge, a new approach to the product development lifecycle will be developed. This know-how will impact significantly on productivity, consistency and performance within the speciality chemicals, home and personal care (HPC), fast moving consumer goods (FMCG), food and beverage, and pharma/biopharma sectors.
UK manufacturing is facing a major challenge from competitor countries such as China that are not constrained by fixed manufacturing assets, consequently they can make products more efficiently and at significantly lower operational costs. For example, the biggest competition for some well recognised 'high-end' brands is from 'own-brand' products (simple formulations that are significantly cheaper). For UK companies to compete in the global market, there is a real need to differentiate themselves from the low-cost competition, hence the need for uncopiable or IP protected, enhanced product performance, higher productivity and greater consistency. The CDT is well placed to contribute to addressing this shift in focus though its research activities, with the PGR students serving as ambassadors for this change. The CDT will thus contribute to the sustainability of UK manufacturing and economic prosperity.
The route to ensuring industry will benefit from the 'paradigm' is through the PGR students who will be highly employable as a result of their unique skills-set. This is a result of the CDT research and training programme addressing a major gap identified by industry during the co-creation of the CDT. Resulting absorptive capacity is thus significant. In addition to their core skills, the PGR students will learn new ones enabling them to work across disciplinary boundaries with a detailed understanding of the chemicals-continuum. Importantly, they will also be trained in innovation and enterprise enabling them to challenge the current status quo of 'development and manufacture' and become future leaders.
The outputs of the research projects will be collated into a structured database. This will significantly increase the impact and reach of the research, as well as ensuring the CDT outputs have a long-term transformative effect. Through this route, the industrial partners will benefit from the knowledge generated from across the totality of the product development lifecycle. The database will additionally provide the foundations from which 'benchmark processes' are tackled demonstrating the benefits of the new approach to transitioning from molecules to product.
The impact of the CDT training will be significantly wider than the CDT itself. By offering modules as Continuing Professional Development courses to industry, current employees in chemical-related sectors will have the opportunity to up-skill in new and emerging areas. The modules will also be made available to other CDTs, will serve as part of company graduate programmes and contribute to further learning opportunities for those seeking professional accreditation as Chartered Chemical Engineers.
The CDT, through public engagement activities, will serve as a platform to raise awareness of the scientific and technical challenges that underpin many of the items they rely on in daily life. For example, fast moving consumer goods including laundry products, toiletries, greener herbicides, over-the-counter drugs and processed foods. Activities will include public debates and local and national STEM events. All events will have two-way engagement to encourage the general public to think what the research could mean for them. Additionally these activities will provide the opportunity to dispel the myths around STEM in terms of career opportunities and to promote it as an activity to be embraced by all thereby contributing to the ED&I agenda.