Multi-fidelity modeling and optimization of biogas plants

Abstract

An essential task for operation and planning of biogas plants is the optimization of substrate feed mixtures. Optimizing the monetary gain requires the determination of the exact amounts of maize, manure, grass silage, and other substrates. Accurate simulation models are mandatory for this optimization, because the underlying chemical processes are very slow. The simulation models themselves may be time-consuming to evaluate, hence we show how to use surrogatemodel-based approaches to optimize biogas plants efficiently. In detail, a Kriging surrogate is employed. To improve model quality of this surrogate, we integrate cheaply available data into the optimization process. Doing so, multifidelity modeling methods like Co-Kriging are employed. Furthermore, a two-layered modeling approach is employed to avoid deterioration of model quality due to discontinuities in the search space. At the same time, the cheaply available data is shown to be very useful for initialization of the employed optimization algorithms.Overall, we show how biogas plants can be efficiently modeled using data-driven methods, avoiding discontinuities as well as including cheaply available data. The application of the derived surrogate models to an optimization process is shown to be very difficult, yet successful for a lower problem dimension.

Multifidelity Surrogate Models for Efficient Uncertainty Propagation Analysis in Salars Systems

“…As mentioned previously, the use of LF models could be favorable in repetitive simulation tasks, particularly in cases where single runtimes of the HF model are computationally expensive. We consider here a set of common approaches presented in the literature of multifidelity surrogate modeling (Zaefferer et al, 2016;Fernández-Godino et al, 2019b). Additive and multiplicative corrections are implemented by constructing an approximation model of the difference (discrepancy) or the ratio between the HF and the LF model outputs.…”

“…It is an active research topic to identify regions where there is less correlation between the HF and LF model via informative sampling strategies and improve the prediction skills of the multifidelity surrogate (Zhou et al, 2015(Zhou et al, , 2016. In the case where the LF model is considered computationally costly, albeit faster than the HF model, then an additional surrogate model can be constructed to approximate the LF model response (Zaefferer et al, 2016). Finally, for the numerical experiments considered in this work we also focus on a special formulation of Kriging (KRG) to develop another multifidelity surrogate model.…”

Prediction of Biogas Production from Co-digestion of Winery Solid Waste and Zebra Manure using modified Gompertz Model (GM) and Logistic Equation (LE)

“…Current depletion of fossil fuels and its adverse environmental effect has recently led to an increasing trend of alternative energy security research including biofuels, with a focus on improving its economic viability [5]. However, with so many substrate mixtures to produce biogas such as municipal waste, sewage, manure, green waste or food waste, simulation models are necessary in order to optimize biogas plants efficiency [6]. In sewage treatment, biogas production has been observed to have the advantages of reducing the amount of sludge for disposal, kill most of the pathogens initially present and reducing the bad odour linked to the remaining putrescible substance in the sludge [7].…”

Application of Biogas Installations as an Alternative Method to Receive Energy

“…The station receives raw materials for biogas production (organic waste -manure, birds' dung, beet press, grain, any waste of fish and cattle production, household waste, grass, waste of dairy production, biodiesel; specially grown corn and seaweeds can also be used as raw materials). The end products are electric power, heat, carbonic and liquefied gas, water and fertilizers (Zaefferer et al, 2016). Biogas is used for electric generators without any cleaning.…”