Prediction compost criteria of organic wastes with Biochar additive in in-vessel composting machine using ANFIS and ANN methods

Abstract

In-vessel composting machine with the agitating system, circulating aeration system, and heating system on vegetable and food waste with coco peat additives and biochar obtained from coco peat was investigated. The composting process was tested at 55 °C, at three fresh inlet air rates of 20%, 30%, and 50%, three initial carbon-to-nitrogen (C/N) ratios of 18, 22, 26, and the addition of coco peat biochar of 5%, 10% w.b. (wet basis). To predict compost evaluation indicators of Electrical conductivity (EC), pH, C/N & GI, artificial neural network (ANN), and neural-fuzzy inference systems were used. The evaluation of the output parameters of compost showed high efficiency of the process. The amount of EC, acidity, and GI increased for all treatments, and the C/N ratio decreased. Also, the initial C/N ratio of 22 and fresh inlet air (FIA) of 30% were considered as the optimal setting conditions of the device. Treatment containing 5% biochar in the C/N of 22 resulted in the highest germination index of 93.55%. The best values of the coefficient of determination for the output parameters of the compost production process (EC, pH, C/N & GI) in the artificial neural network were 0.9252, 0.9863, 0.9691, and 0.9909 respectively. Moreover, the best values of the coefficient of determination in the fuzzy neural inference system for the output parameters of the compost include EC, pH, C/N and GI were 0.999, 0.999, 0.994, and 0.992, respectively. Also, the lowest values of MAE and RMSE in the fuzzy neural inference system for the output parameters of the compost include EC, pH, C/N, and GI were 0.0308, 0.0001, 0.2420, and 0.003 for MAE; and 0.0021, 3.66E−05, 0.1908 and 0.0041 for RMSE, respectively.

Keywords

• Artificial neural network
• Composting machine
• Determination coefficient
• Germination index