Abstract
Summary
This article proposes a methodology based on artificial intelligence to enhance methane production from waste paper. The proposed methodology combines fuzzy logic‐based modelling and modern optimization. Firstly, a robust Adaptive Network‐based Fuzzy Inference System model of methane production process through fuzzy logic modelling is created using experimental datasets. Second, a particle swarm optimizer was used to obtain the optimal process conditions. During the optimization procedure, the beating time and feedstock/inoculum ratio are employed as decision variables in order to maximize methane production. The obtained resulted from the proposed methodology are compared with those obtained by response surface methodology. The results of the comparison confirmed the superiority of the proposed methodology. The fuzzy model shows a better fitting to the experimental data compared to ANOVA. The fuzzy model showed a higher coefficient of determination and a lower value of root mean squared errors compared to ANOVA. Moreover, the proposed strategy, that is, modelling and optimization, is an effective method for increasing the biomethane yield at extended range conditions.
•A robust Adaptive Network‐based Fuzzy Inference System (ANFIS) model for methane production from waste paper was built using fuzzy logic modelling based on the experimental datasets.
•A particle swarm optimizer (PSO) was used to maximize methane production through optimizing beating time and feedstock/inoculum ratio.
•The fuzzy model shows a better fitting to the experimental data compared to ANOVA. The proposed strategy, i.e., modelling and optimization, is an effective for increasing the biomethane yield at extended range conditions.