基于集对分析的相似模型在地下水位预测中的应用

Similar Forecast Models of Underground Water Level Based on Set Pair Analysis

地下水位预测能为生态环境保护和地下水资源规划、管理提供科学依据。为了对地下水位动态变化进行准确预测,针对地下水位样本之间的相似性及影响因子与地下水位之间具有的确定、不确定性特征,提出将集对分析与相似预测相结合,从同、异、反三方面定量刻画地下水位的当前样本与历史样本之间的相似性,建立了基于集对分析的地下水位相似预测模型,并应用该模型计算地下水位预测值。实例检验结果表明:基于集对分析的地下水位相似预测模型的平均相对误差小于3%。与其他地下水位预测方法相比,该方法用于地下水位预测具有物理概念清晰,计算简便,预测精度较高的特点。

The forecast of underground water level can provide scientific basis for the ecological environmental protection, planning and management of underground water resources. According to the similarity between the samples of underground water level and the characteristics of regularity as well as the both certain and uncer- tain between the impact factors and underground water level, this study set up the similar forecast models of underground water level based on set pair analysis, which describe quantitatively the similarity among the his- torical samples of forecast object from the views of identical, discrepancy and contrary sides, combining set pair analysis with the similar forecast. Further, the predictive values of underground water level can be calculated by the model. The empirical results suggested that average relative error of the similar forecast models of under- ground water level based on set pair analysis was less than 3%. Compared with other underground water level prediction methods, this model has the characteristics of clear physical concepts, simple computation and higher prediction accuracy.