Machine learning for quantile regression of biogas production rates in anaerobic digesters.

Anaerobic digestion is a well-established tool at wastewater treatment plants for processing raw sludge; it can also be used to generate renewable energy by harvesting biogas in anaerobic digesters. Operational parameters, such as temperature, are usually set by plant operators according to expert knowledge. To completely utilize the potential of operational management, in this study, we calibrated a novel Temporal Fusion Transformer based on six years of life-scale time series data together with categorical features such as public holidays. The model design allows for the interpretability of the output in contrast to traditional data-driven techniques, using multi-head attention. In addition to forecasting the median biogas production rates for the following seven days, our model also yields quantiles, making it less prone to strong fluctuations. We used three well-known statistical techniques as benchmarks. The mean absolute percentage error of our forecasting approach is below 8 %.

On the effect of the inlet configuration for anaerobic digester mixing.

Sludge recirculation mixing in anaerobic digesters is essential for the stable operation of the digestion process. While often neglected, the configuration of the sludge inlet has a substantial influence on the efficiency of the mixing process. The fluid is either injected directly into the enclosed fluid domain or splashes onto the free surface of the slurry flow. In this paper, the aim was to investigate the effect of the inlet configuration by means of computational fluid dynamics-using ANSYS Fluent. Single-phase and multi-phase models are applied for a submerged and splashing inlet, respectively. To reduce the high computational demand, we also develop surrogate single-phase models for the splashing inlet. The digester mixing is analyzed by comparing velocity contours, velocity profiles, mixing time and dead volume. The non-Newtonian characteristics of the sludge is considered, and a [Formula: see text] model is employed for obtaining turbulence closure. Our method is validated by means of a previous study on the same geometry. Applying a submerged inlet configuration, the resulting dead volume in the tank is estimated around 80 times lower than for the case of a splashing inlet. Additionally, by emulating the multi-phase model for splashing inlet configurations with a single-phase one, the simulation clock time reduced to 15%.