Neural-fuzzy control system application for monitoring process response and control of anaerobic hybrid reactor in wastewater treatment and biogas production

Estimation of Solid Waste Biomethanization: Development of Novel Methodology and Anfis Models

“…Recently, ANFIS has been used in the modeling of complex systems such as MSW biomethanization and the anaerobic digestion of primary sludge (Table 7), for which the selection of the input variables and the study of the data play a significant role in both the reliability and the quality of the estimation (Waewsak, Nopharatana, Chaiprasert, 2010;Ojolo et al, 2008). However, ANFIS is not the only way to model an input-output data set; FIS could also be identified as using the least-squares method (Takagi and Sugeno 1985) as well as genetic algorithm-based methods (Abu Qdais BaninHani, Shatnawi, 2009;KurtulusOzcan et al, 2008) and the fuzzy c-means method (Sugeno and Yasukawa, 1993).…”

“…In the search for generic, simple, and applicable models, artificial intelligence (AI) tools have been demonstrated to be very useful and capable of reproducing human characteristics such as flexibility, uncertainty tolerance, ability to consider imprecision or data voids, and facilitation of result interpretation (Jang and Sun, 1995). Currently, there is a wide variety of studies in the environmental field that have employed the characteristics of this tool (Afshar-Kazemi et al, 2011;Erdirencelebi and Yalpir, 2011;Mullai et al, 2011;Waewsak Nopharatana, and Chaiprasert, 2010;Noori et al 2009). When attempting to estimate the methane generation rate, several reports have been made that employ fuzzy logic (Abdallah et al, 2009;Garg, Achari, and Joshi, 2006), neural networks (Ogwueleka and Ogwueleka, 2010;Jalili and Noori, 2008;Ozkaya, Demir, and Bilgili, 2007), and genetic algorithms (Kurtulus Ozcan et al, 2008); however, only a much smaller number of studies use combinations of these techniques (Abdallah et al, 2011;Abu Qdais, BaninHani, and Shatnawi, 2009;Cakmakci, 2007;Nath and Das, 2011).…”

Automatic process control in anaerobic digestion technology: A critical review

“…A wide variety of control strategies were studied to manipulate feeding rate, which included proportional integral (PI) control (Steyer et al, 1999;Alvarez-Ramirez et al, 2002), PID control (García-Diéguez et al, 2011), adaptive control (Hilgert et al, 2000;Luo et al, 2014), robust adaptive (Petre et al, 2013), fuzzy logic (Murnleitner et al, 2002;Djatkov et al, 2014), neural network (Holubar et al, 2002(Holubar et al, , 2003, and neural fuzzy (Waewsak et al, 2010). The feeding rate can be used to simultaneously regulate the retention time and organic loading rate, allowing microbial communities in the system to adapt to the disturbances.…”

“…The system was applied in various anaerobic reactor configurations (i.e., anaerobic fluidized bed reactor, anaerobic filter, and UASB) and could accurately predict VFA, TOC, methane production rate in advance time of one hour. The similar control system was then adapted in automatic control of anaerobic hybrid reactor (Waewsak et al, 2010) (Table 3). By using the artificial neural network component to predict the disturbance, and fuzzy logic component to control the influent flow rate, the reactor was regulated under organic and hydraulic overloading.…”

Control strategy for maximum anaerobic co-digestion performance

“…A number of key process parameters have been suggested as diagnosis indicator of process stability: pH and biogas output flow rate (Steyer et al, 1999); H 2 -gas and methane flow rate (Rodríguez et al, 2006); pH, alkalinity and total volatile acids (Waewsak et al, 2010); concentration of volatile fatty acids (VFAs) in the effluent and methane flow rate (García-Di eguez et al, 2011). Boe et al (2010) compared the behaviour of different process indicators (biogas flow, pH, VFA, dissolved hydrogen, methane content in biogas) under organic and hydraulic overloads and concluded that VFA was the most effective indicator.…”