(343a) The Industrial Implementation of Validated Dynamic Simulation and Optimisation Tools Towards Superior Beer Fermentation

Beer fermentation is an extremely complex biochemical phenomenon (Vanderhaegen et al., 2006), whose mathematical representation and parameter estimation are hitherto pursued only via reduced order models (Dochain, 2003), which consider only the most essential subset of important biochemical reactions via selective aggregation. Subsequent process optimisation studies have been successfully performed on the basis of these foregoing models, primarily addressing batch time minimisation or ethanol yield maximisation (de Andrés-Toro, 1998; Carrillo-Ureta, 2001), or both (Rodman and Gerogiorgis, 2016). A necessary assumption of process optimisation studies is that all parameter datasets published in the original fermentation modelling papers are accurate and consistent; even more so, the complexity (and deterioration) of yeast performance perplexes further industrial endeavours.

Our recent work has involved the comprehensive computation and visualisation of attainable envelopes for beer ethanol maximisation and batch duration minimisation, under explicit flavour-modifying compound concentration constraints, pursued via simulation-based optimisation (Rodman and Gerogiorgis, 2016), as well as by rigorous dynamic optimisation, via an NLP formulation employing orthogonal collocation on finite elements (Rodman and Gerogiorgis, 2017). More recently, we have also explored the problem of obtaining highest ethanol in minimum time, employing a multi-objective formulation via the Strawberry stochastic evolutionary algorithm, achieving the detailed visualisation of Pareto fronts (Rodman et al., 2018). Therein, a clear operational asymmetry is observed: while acceptable fermentation rapidly diminishes below a minimum batch duration threshold, the subsequent improvement in final ethanol concentration changes very little with increasing batch time, rendering long fermentations meaningless (and implying a "sweet spot", which is highly dependent upon several underlying biochemical characteristics).

The present paper illustrates the industrial implementation of these novel computational tools for optimising fermentation operations in a major multinational brewing corporation, under the auspices of a Royal Academy of Engineering (RAEng) Industrial Fellowship (2017-2018). By combining the published results of our foregoing studies with additional sensory (product), but even more so operational (plant management) constraints, we have arrived at superior temperature manipulation profiles for industrial fermentors, which achieve higher final ethanol concentration, reduced batch duration, and constraint satisfaction for key beer flavour modifiers (esters and diacetyls). In contrast to our previous studies, these optimal fermentation T(t) protocols also consider a minimum number of turning points, reflecting the prominent control room culture (and the operational preference) to avoid frequent adjustments - a fact well justified by the limited instrumentation and online data acquisition, which in turn is due to the limited accessibility, and the multiphase, inhomogeneous nature of the fermenting mixture.

REFERENCES

Vanderhaegen, B., et al., 2006. The chemistry of beer aging - a critical review. Food Chem. 95(3): 357-381.

de Andrés-Toro, B., et al., 1998. A kinetic model for beer production under industrial operational conditions. Math. Comput. Simulat. 48(1): 65-74.

Carrillo-Ureta, G., P. Roberts, and V. Becerra, 2001. Genetic algorithms for optimal control of beer fermentation. Proc. IEEE Int. Symp. Intell. Control, 391-396.

Dochain, D., 2003. State and parameter estimation in chemical and biochemical processes: a tutorial, J. Proc. Control 13(8): 801-818.

Gee, D. A., Ramirez, W. F. 1988. Optimal temperature control for batch beer fermentation. Biotechnol. Bioengineering, 31: 224-234.

Rodman, A. D., Gerogiorgis, D.I., 2016. Multi-objective process optimisation of beer fermentation via dynamic simulation, Food Bioprod. Proc. 100(A): 255-274.

Rodman, A.D. and Gerogiorgis, D.I, 2017. Dynamic optimization of beer fermentation: sensitivity analysis of attainable performance vs. product flavor constraints. Comput. Chem. Eng. 106(C): 582-595.

Rodman, A.D., Fraga, E.S. and Gerogiorgis, D.I., 2018. On the application of a nature-inspired stochastic evolutionary algorithm to constrained multi-objective beer fermentation optimisation. Comput. Chem. Eng. 108: 448-459.