Performance evaluation of ANFIS and RSM modeling in predicting biogas and methane yields from Arachis hypogea shells pretreated with size reduction

In this study, Response Surface Methodology (RSM) was used to examine the effects of temperature, hydraulic retention time, and particle size of Arachis hypogea shell on biogas and methane yields in a batch test. Further to this, an Adaptive Neuro-fuzzy Inference System (ANFIS) clustered with fuzzy c-means (FCM) was developed to predict organic dry matter biogas yield (oDMBY), fresh mass biogas yield (FMBY), organic dry matter methane yield (oDMMY), and fresh mass methane yield (FMMY). Relevant statistical metrics like root mean square error (RMSE), mean absolute percentage error (MAPE), mean absolute deviation (MAD), and correlation coefficient (R-2) were used to evaluate the performance of the developed ANFIS model. The performance of both RSM and ANFIS were compared based on the performance metrics. The R-2 values of RSM for oDMBY, FMBY, oDMMY and FMMY are 0.6268, 0.5875, 0.6109 and 0.5547 respectively; and 0.9601, 0.9486, 0.9626 and 0.9172 respectively for ANFIS model. The results revealed the better performance of the ANFIS than the RSM, with lesser prediction error and higher accuracy. It is concluded that RSM and ANFIS are practical models for predicting particle size limits in a multiple-input parameter without attempting any experiment within a short period with a tiny error rate. (c) 2022 Elsevier Ltd. All rights reserved.

Automated summary

The effects of temperature, hydraulic retention time, and particle size of Arachis hypogea shell on biogas and methane yields were examined using Response Surface Methodology (RSM). An Adaptive Neuro-fuzzy Inference System (ANFIS) was developed to predict the yields, and the performance of both RSM and ANFIS models was compared. The results showed that the ANFIS model outperformed the RSM model in terms of accuracy and prediction error.