Bueno Pedro, Yañez Remedios, Caparrós Sebastian, Díaz Manuel Jesús
Departamento de Ingeniería Química, Facultad de Ciencias Experimentales, Universidad de Huelva, Huelva, Spain.
J Air Waste Manag Assoc. 2009 Jul;59(7):790-800. doi: 10.1080/10473289.2009.10465778.
A neural fuzzy system was used to investigate the influence of environmental variables (time, aeration, moisture, and particle size) on composting parameters (pH, organic matter [OM], nitrogen [N], ammonium nitrogen [NH4(+)-N] and nitrate nitrogen [NO3(-)-N]). This was to determine the best composting conditions to ensure the maximum quality on the composts obtained with the minimum ammonium losses. A central composite experimental design was used to obtain the neural fuzzy model for each dependent variable. These models, consisting of the four independent process variables, were found to accurately describe the composting process (the differences between the experimental values and those estimated by using the equations never exceeded 5-10% of the former). Results of the modeling showed that creating a product with acceptable chemical properties (pH, NH4(+)-N and NO3(-)-N) entails operating at medium moisture content (55%) and medium to high particle size (3-5 cm). Moderate to low aeration (0.2 L air/min m kg) would be the best compromise to compost this residue because of the scant statistical influence of this independent variable.
采用神经模糊系统研究环境变量（时间、曝气、湿度和粒径）对堆肥参数（pH值、有机质[OM]、氮[N]、铵态氮[NH4(+)-N]和硝态氮[NO3(-)-N]）的影响。目的是确定最佳堆肥条件，以确保在铵损失最小的情况下获得质量最高的堆肥。采用中心复合实验设计来获得每个因变量的神经模糊模型。这些由四个独立过程变量组成的模型被发现能够准确描述堆肥过程（实验值与使用方程估算的值之间的差异从未超过前者的5-10%）。建模结果表明，要制备具有可接受化学性质（pH值、NH4(+)-N和NO3(-)-N）的产品，需要在中等湿度（55%）和中等至高粒径（3-5厘米）下进行操作。由于该自变量的统计影响较小，适度至低曝气（0.2升空气/分钟·米·千克）将是堆肥该残渣的最佳折衷方案。
