基于遗传算法-BP神经网络法的深埋地下水水质评价

为探究开封市地下水水质特征及成因,依据开封市31眼深度600～1 600m地下水开采井的水质检测资料,系统研究了各亚含水层的水化学特征,利用遗传算法-BP神经网络法进行了水质评价,并从埋深、富水性两个方面分析了水质特征的分布规律。结果表明,开封市600～800m亚含水层地下水水质最好,800～1 400m次之,1 400～1 600m亚含水层水质最差,随着埋深的增加,水质变差,富水性越强、水质越好。可见遗传算法-BP神经网络法能够客观地描述地下水水质综合情况,避免了人为主观影响,使评价结果更为合理。

Evaluation of Deep Buried Groundwater Based on Genetic Algorithm and BP Neural Network

In order to explore the characteristics and causes of groundwater quality in Kaifeng City, the hydrochemical characteristics of sub-aquifers were systematically studied based on the water quality test data of 31 groundwater exploitation wells with depths of 600-1 600 m in Kaifeng City. The hybrid genetic algorithm and BP neural network method was used to evaluate water quality, and the distribution law of water quality characteristics was analyzed from two aspects of burial depth and water-rich. The results show that the groundwater quality of 600-800 m sub-aquifer is the best in Kaifeng City, followed by 800-1 400 m, and 1 400-1 600 m is the worst. With the increase of burial depth, water quality becomes worse, water richness becomes stronger, and water quality becomes better. It can be seen that the hybrid genetic algorithm and BP neural network method can objectively describe the comprehensive situation of groundwater quality, avoid the subjective influence of human, and make the evaluation results more reasonable.