Copyright © 2018 Authors. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. International Journal of Engineering & Technology, 7 (2.21) (2018) 316-318 International Journal of Engineering & Technology Website: www.sciencepubco.com/index.php/IJET Research paper Development of ANN models for optimization of methane yield from floating dome digester S. Sathish 1* , A. Parthiban 2 , R. Balakrishna 3 , R. Anandan 4 1 Department of Mechanical Engineering, Hindustan Institute of Technology & Science, Chennai, India. 2 Department of Mechanical Engineering, Vels Institute of Science, Technology & Advanced Studies(VISTAS), Chennai, India. 3 Department of Computer Science & Engineering, Vels Institute of Science, Technology & Advanced Studies(VISTAS), Chennai, India. 4 Department of Computer Science & Engineering, Vels Institute of Science, Technology & Advanced Studies(VISTAS), Chennai, India. *Corresponding author E-mail:Sathishamg88@gmail.com Abstract The development of methane generation is mainly based on a desirable combination of operating parameters. The essential objective of this analysis systematically analyzes the prediction of methane yield with different operating parameters. Current work is to analyse the reaction of Temperature (T), Agitation time (AT), pH, value and Substrate Loading Rate (SLR) which are all considered to be the different factors. Artificial Neural Network (ANN) is the modern method aid to solve complex issues that could not be addressed by conventional methods. In this work examine the study employ the ANN as a tool for prediction of methane from floating dome anaerobic digester with press mud. The result showed that ANN model is found the value of methane yields much closed to theoretical methane yield. It is obtained the percentage of predicted value of methane is 58% and theoretical value of methane is 62 % with the temperature of 45ºC and agitation time of 20 min, pH value of 7.2 and substrate loading rate of 120 kg. Keywords: Substrate loading rate, temperature, agitation time, methane. 1. Introduction Biological operations has been incorporated in Anaerobic digestion process, in the case loss of oxygen for crack up organic element along with that conversion CH4 and CO2 causes stabilization of these materials which is almost solid slag [1]. Anaerobic digestion gives appealing prospects as well as providing results to global interest as substitute energy creation, managing people, and animal community along with the safety of industrial wastes, governing environmental pollution. For reprocessing of nutrients back to the soil, Biogas technology has been efficiently utilized. Ecological shortcomings occurs due to the flaming of non-commercial fuel resources, like manure as well as agricultural slag, where these sources are utilised as fuel and not as fertilizer, because the resources like mini-nutrients, nitrogen, potassium, nutrients are actually vanished from the ecosystem [2]. It is an important technology; the improvement of computational model for anaerobic digestion applications can assist in operation and control the process of anaerobic digestion as well as enhance the methane production. Ann is [3] fit the process particularities, these models and tools have to be calibrated and modified with the biogas plants own data. They are used to predict, design and optimize the process. Artificial neural network (ANN) Fig. 1: ANN architecture Fig. 1 illustrate the architecture of Artificial Neural Networks where, Zin = a1.f1 + a2.f2 + a3.f3… an.fn i.e., Net input Zin = ∑   .    Biological neural networks (BNN) are the base from where the Artificial Neural Network (ANN) is a dynamic measuring model formed. ANNs is the abbreviation stands for “artificial neural systems,” or “parallel distributed processing systems,” or “connectionist systems.” ANN brings in a big gathering of entities that are linked in certain format to grant transmission among entities. These entities, which are commonly known as nodes or neurons, are smooth operators which process alongside. The key basic of ANN models are supported on an information processing paradigm and the learning processes of a biological neural system, in this model offer the different methods of analysing data and make out patterns within the data in comparison with traditional mechanistic approaches. The approach of ANN model is