污泥处理湿地系统渗滤液中总氮和氨氮的预测模型

Prediction Models of Total Nitrogen and Ammonia Nitrogen in Leachate from Sludge Treatment Wetland System

将人工湿地与传统污泥干化床结合,形成了污泥处理湿地技术,其通过渗透和蒸发蒸腾作用对污泥进行脱水。污泥中的水分在经过污泥层和填料层下渗的过程中,水质成分发生复杂的变化,进而影响到渗滤液出水水质。根据污泥处理湿地系统中渗滤液出水总氮浓度和氨氮浓度检测数据的平稳、系列特点,利用时间序列分析ARIMA模型对系统渗滤液出水总氮和氨氮浓度进行了模拟和预测。结果表明,ARIMA(1,0,0)模型可以较好地模拟和预测污泥处理湿地渗滤液出水总氮浓度变化,预测相对误差在-11.7%~8.8%之间,实测值和预测值之比处于0.9~1.1范围内;ARIMA(6,0,0)模型能较好地模拟和预测污泥处理湿地渗滤液出水氨氮浓度的变化,预测相对误差基本在-14.5%~14.7%之间,实测值和预测值之比处于0.8~1.2范围内。

Sludge treatment wetland system was constructed by combining conventional sludge drying bed and constructed wetland. Sludge was dewatered by percolation,evaporation and transpiration in the system. Leachate compositions were changed during the process of sludge dewatering by gravity and infiltration through sludge layer and media layer in the wetland,which had an effect on the effluent quality. According to the characteristics of stationarity and serialization of the data for ammonia nitrogen and total nitrogen concentrations in leachate effluent from sludge treatment wetland system,autoregressive integrated moving average( ARIMA) model was used to simulate and forecast total nitrogen concentrations in leachate effluent. The results showed that ARIMA( 1,0,0) model could be efficiently used to simulate and predict the variation of the total nitrogen concentrations in leachate effluent,the relative errors of prediction values were in the range of- 11. 7% to 8. 8%,and the ratios of measured values to predicted values were in the range of 0. 9 to 1. 1. ARIMA( 6,0,0) model could also be efficiently used to simulate and predict the variation of the ammonia nitrogen concentrations in leachate effluent,the relative errors of prediction values were in the range of- 14. 5% to 14. 7%,and the ratios of measured values to predicted values were in the range of 0. 8 to 1. 2.