Machine learning approach for determining and optimizing influential factors of biogas production from lignocellulosic biomass.

Machine learning (ML) was used to predict specific methane yields (SMY) with a dataset of 14 features from lignocellulosic biomass (LB) characteristics and operating conditions of completely mixed reactors under continuous feeding mode. The random forest (RF) model was best suited for predicting SMY with a coefficient of determination (R2) of 0.85 and root mean square error (RMSE) of 0.06. Biomass compositions greatly influenced SMYs from LB, and cellulose prevailed over lignin and biomass ratio as the most important feature. Impact of LB to manure ratio was assessed to optimize biogas production with the RF model. Under typical organic loading rates (OLR), optimum LB to manure ratio of 1:1 was identified. Experimental results confirmed influential factors revealed by the RF model and provided the highest SMY of 79.2% of the predicted value. Successful applications of ML for anaerobic digestion modelling and optimization specifically for LB were revealed in this work.

Removal of Hydrogen Sulfide from Swine-Waste Biogas on a Pilot Scale Using Immobilized Paracoccus versutus CM1.

Hydrogen sulfide (H2S) is a toxic and corrosive component that commonly occurs in biogas. In this study, H2S removal from swine-waste biogas using sulfur-oxidizing Paracoccus versutus CM1 immobilized in porous glass (PG) and polyurethane foam (PUF) biofilters was investigated. Bacterial compositions in the biofilters were also determined using polymerase chain reaction-denaturing gradient gel electrophoresis (PCR-DGGE). The biofilters were first tested on a laboratory scale under three space velocities (SV): 20, 30, and 40 h−1. Within 24 h, at an SV of 20 h−1, PG and PUF biofilters immobilized with P. versutus CM1 removed 99.5% and 99.7% of H2S, respectively, corresponding to the elimination capacities (EC) of 83.5 and 86.2 gm−3 h−1. On a pilot scale, with the horizontal PG-P. versutus CM1 biofilter operated at an SV of 30 h−1, a removal efficiency of 99.7% and a maximum EC of 113.7 gm−3 h−1 were achieved. No reduction in methane content in the outlet biogas was observed under these conditions. The PCR-DGGE analysis revealed that Paracoccus, Acidithiobacillus, and Thiomonas were the predominant bacterial genera in the biofilters, which might play important roles in H2S removal. This PG−P. versutus CM1 biofiltration system is highly efficient for H2S removal from swine-waste biogas.

Biogas production from high solids digestion of Pennisetum purpureum x Pennisetum typhoideum: Suitable conditions and microbial communities.

Effects of organic loading rates (OLRs), temperatures and effluent recirculation rates on biogas production from Giant Juncao Grass (GJG) using pilot-scale semi-continuously fed CSTRs were investigated. Thermophilic reactors could be stably operated at OLR up to 5.0 kg VS m-3 d-1, while damaged process stability was detected in mesophilic reactors at OLR of 4.0 kg VS m-3 d-1. Higher effluent recirculation rate (3:1) helped lessen negative effects of system being over-loaded, especially for mesophilic reactors. Microbial community analysis revealed that temperatures had the highest effect on bacterial community structure. Firmicutes were the dominant bacterial phyla found under high temperatures, while majority of archaea in all reactors belonged to the phylum Bathyarchaeota. Changes of microbial communities could partly explain system performance under different operating conditions. This study was the first to show GJG as a superior biogas feedstock to other energy crops thanks to its higher methane yields per planting area.

Holistic determination of suitable conditions for biogas production from Pennisetum purpureum x Pennisetum americanum liquor in anaerobic baffled reactor.

Holistic approach had been undertaken to determine the suitable conditions for biogas production from Pennisetum purpureum x Pennisetum americanum (Napier Pak Chong1) liquor using the anaerobic baffled reactor (ABR). Effects of factors, i.e. organic loading rates (OLRs), feeding schemes, trace element additions and effluent recirculation rates, on ABR performance were systematically investigated. Increase of OLRs and effluent recirculation rates adversely affected methane yields when capability of ABR in containing microorganisms was deteriorated. High stability of ABR performance was detected under the semi-continuous feeding scheme with trace element additions. The suitable condition for ABR was found at the OLR of 4.0 kg COD/m3.d under the semi-continuous feeding scheme with trace element additions at the effluent recirculation rate of 0.5 (QR/Q). At this condition, high methane yield (0.49 ±â€¯0.05 Nm3/kg VSadded) could be achieved using the economical ABR at relatively high OLR of 4.0 kg COD/m3.d.

Fate of estrogens during the biological treatment of synthetic wastewater in a nitrite-accumulating sequencing batch reactor.

The fate of estrone (El), 17beta-estradiol (E2), and 17alpha-ethynylestradiol (EE2) was investigated in two nitrite-accumulating sequencing batch reactors operating under strictly aerobic (SBR1) conditions at different sludge ages (SRT, 1.7 to 17.1 d) and anoxic/anaerobic/aerobic (SBR2) conditions with different phases and durations of redox conditions, using a modified GC-MS analytical method for estrogen detection to ng/L concentrations. In SBR1, > or =98% of E2 was removed (specific E2 removal rate ranged from 0.375 (at SRT 17.1 d) to 2.625 (at SRT 1.7 d) microg E2 x g MLVSS(-1) x d(-1)) regardless of SRT or DO (<1.0 to >5.0 mg/L). Removal of E1 and, to a greater extent, EE2 was adversely affected when this reactor was operated at SRT shorterthan 5.7 d. However, whereas E1 was removed efficiently as long as SRT was long enough for AOB to bring about nitritation, EE2 removal efficiency was significantly lower when SBR1 was operated at SRT longer than 7.5 d. This reduced removal of EE2 may have been caused by the inhibition and toxicity of nitrite, both to the ammonium monooxygenase (AMO) and to the microbial population generally. In SBR2, less removal of E2 was found at the lower MLVSS concentrations, and E1 was not removed by sludge with poor settling qualities. The removal of EE2 observed in SBR2 was considered to be mainly a result of sorption. However, the binding of estrogens to the sludge in this reactor was apparently not as strong as the binding observed in the sludge of the strictly aerobic SBR1, since desorption was observed during the aeration phase in SBR2.