基于ARIMA模型的矿区重金属污染时间序列预测

Time Series Prediction of Heavy Metal Pollution in Mining Areas Based on ARIMA Model

矿区重金属污染具有时间序列的特征,因此可以采用时间序列ARIMA模型对重金属污染进行预测。对南方某铜硫矿,在1995年1月—2008年6月重金属月监测数据的基础上,运用ARIMA模型建立了矿区尾矿库废水总排放口Zn浓度的预测模型,结果表明,ARIMA(1,1,2)模型能较好地拟合2008年1月—2008年6月重金属污染变化规律,经实际计算结果验证所建模型,误差在5%左右,经检验其精度满足要求。预测结果显示,该矿区未来重金属Zn仍然处在污染状态。

Heavy metal pollution in mining areas possesses the character of time series,so time series ARIMA model can be used to forecast heavy metal pollution.Based on monitoring data on heavy metals for a certain copper-sulfide ore in the South during January 1995 and June 2008,ARIMA model is adopted to build the Zn total prediction in wastewater outfall of the mine tailings.The results indicate that ARIMA （1,1,2） model approximates the alteration rule of heavy metals from January 2008 to June 2008,and the prediction error is around 5% between the calculated and the predicted data of mining areas.The forecasting results show that heavy metal Zn contents in the mining area will be still in the state of pollution in future.