Online monitoring and control of the biogas process

The demand for online monitoring and control of biogas process is increasing, since better monitoring and control system can improve process stability and enhance process performance for better economy of the biogas plants. A number of parameters in both the liquid and the gas phase have been suggested as process indicators. These include gas production, pH, alkalinity, volatile fatty acids (VFA) and hydrogen. Of these, VFA is the most widely recognised as a direct, relevant measure of stability. The individual, rather than collective VFA concentrations are recognised as providing significantly more information for diagnosis. However, classic on-line measurement is based on filtration, which suffers from fouling, especially in particulate or slurry wastes. In this project, a new online VFA monitoring system has been developed using gas-phase VFA
extraction to avoid sample filtration. The liquid sample is pumped into a sampling chamber, acidified, added with salt and heated to extract VFA into the gas phase before analysis by GC-FID. This allows easy application to manure. Sample and analysis time of the system varies from 25-40 min. depending on the washing duration. The sampling frequency is fast enough for the dynamic of a manure digester, which is in the range of several hours. This system has been validated over more than 6 months and had shown good agreement with offline VFA measurement. Response from this sensor was compared with other process parameters such as biogas production, pH and dissolved hydrogen during overload situations in a laboratory-scale digester, to investigate the suitability of each measure as a process indicator. VFA was most reliable for indicating process imbalance, and propionate was most persistent. However, when coupling the online VFA monitoring with a simple control for automatic controlling propionate level in a digester, it was found that propionate decreased so slow that the biogas production fluctuated. Therefore, it is more proper to optimise biogas production, while using propionate (or VFA) as a warning indicator for process imbalance. Moreover, in this project, the investigations of serial CSTR configuration for improving biogas production were also carried out both in lab-scale experiments and by using the ADM1 computer model. It was shown that the serial CSTR configuration with long retention time in the first reactor and short retention time in the second reactor could improve biogas production from manure and could improve effluent quality in terms of VFA concentration, compared to a conventional single CSTR reactor. The temperature of the second reactor in the serial CSTR configuration also affected the amount of extra biogas yield. The serial CSTR configuration present in this study can be applied to the existing process in the Danish centralized biogas plants and requires only small process modification.