Process simulation and BPNNM prediction for chemical looping co-gasification of rice husk and textile wastes as cement alternative fuels

Chemical looping gasification (CLG) can inherently split the traditional gasification into two processes to produce high-quality syngas, avoiding the N₂ dilution for syngas. CLG of solid wastes has gained attention for its satisfactory performance with waste valorization. The chemical looping co-gasification (CLCG) performances of rice husk and textile wastes are investigated, which are typical solid wastes used in industry as alternative fuels. A thermodynamic process model of CLCG is established, and the effects of different operating parameters are quantitatively analyzed. Furthermore, a multi-input and multi-output back propagation neural network model (BPNNM) is trained using process model results for the performance prediction. Key findings reveal that increasing equivalence ratios of oxygen carrier and steam (α_OC/F and α_steam/F) significantly affect gasification efficiency. Specifically, increasing α_OC/F to 0.5 decreases gasification efficiency to 60.82%. Conversely, increasing α_steam/F from 0.1 to 0.5 leads to a slight decrease in gasification efficiency from 85.95 to 84.80%, while simultaneously increasing hydrogen concentration in syngas from 39.91 to 46.28%. Elevating the gasification temperature from 650 to 850 °C can raise the η from 81.52% up to 86.00%. The blending ratio of the rice husk and textile waste also dramatically affects gasification efficiency, with efficiency decreasing from 92.27 to 74.41% as the blending ratio R_r increases from 0 to 1. The tests of random conditions demonstrate that the trained BPNNM can be a very accurate tool for the prediction of syngas compositions and gasification indicators in CLCG.