基于GM(1,1)模型的空气质量变化趋势预测及分析

在掌握区域空气环境质量现状资料的基础上,运用具有适用性广,预测准确率高等优点的灰色系统预测模型对大连市未来五年空气中SO2的浓度变化趋势进行预测,对其模型的精度和可行性进行了分折和检验,并与目前常用的指数平滑法的预测结果进行了对比。研究结果表明:建立基于GM(1,1)模型的大气质量变化趋势预测模型是可行的,其模型的平均相对误差比指数平滑法减少了44％,因此预测精度高,能够满足预测要求。

Prediction and analysis of trend of air quality based on GM(1,1) model

Based on the actual data of regional atmospheric quality, GM (1,1) model is applied to forecast the trend of annual average concentrations of SO2 in Dalian in the future five years for its broad applicability and high simulated precision. And its precision and feasibility are validated by comparison with index level and smooth method used presently. The result shows that the forecasting model is feasible to predict the trend of air quality based on GM(1,1) model, whose relative error is reduced by 44% compared with index level and smooth method's. So the prediction by the GM(1,1) model achieves a high precision, which is qualified for the forecast.