针对中国土地复垦起步晚,新技术与新理论应用较少的问题,研究利用人工神经网络来辅助土地复垦的决策。介绍了利用自组织映射神经网络的自动分类功能对进行矿区土地复垦条件分类,为因地制宜地采取复垦措施提供依据。然后,基于 BP 神经网...针对中国土地复垦起步晚,新技术与新理论应用较少的问题,研究利用人工神经网络来辅助土地复垦的决策。介绍了利用自组织映射神经网络的自动分类功能对进行矿区土地复垦条件分类,为因地制宜地采取复垦措施提供依据。然后,基于 BP 神经网络,选取评价因子,通过对已有经验的学习,对复垦土地适宜性进行评价,并与传统的方法相比较,研究结果表明,利用人工神经网络辅助矿区土地复垦决策是可行的。展开更多
Chloride ion transport in reclaimed soil solutions filled with fly ash （FA） was investigated by measuring the hydraulic parameters （i,e. water retention curves and hydraulic conductivity） of three substrates, name...Chloride ion transport in reclaimed soil solutions filled with fly ash （FA） was investigated by measuring the hydraulic parameters （i,e. water retention curves and hydraulic conductivity） of three substrates, namely GSL, GFA, and CFA. Similar simulations were carried out under certain weather conditions. The different boundary conditions of chloride transport were also discussed from FA texture, cover soil thickness, groundwater table level, and initial chloride concentration. Furthermore, the sensitivities of chloride ions to these effect factors were analyzed. The results show that the different top soil thickness and initial chloride concentration have no effect on salinity of topsoil solution in the monitoring points, but they can clearly change the chloride concentration of FA layers. The sensibilities from top soil thickness and initial chloride content are exceedingly weak to the salinity balance based on two dimensions of the time and concentration. While the different FA texture and groundwater table not only affect the salinity equilibrium process of the whole reclaimed soil profile, but also change its balance state. Generally, coarse FA particles and high groundwater table can defer the salinity balance process of the reclaimed soil solution, and they also increase the chloride concentration of FA layer solutions, and even topsoil ones.展开更多
文摘针对中国土地复垦起步晚,新技术与新理论应用较少的问题,研究利用人工神经网络来辅助土地复垦的决策。介绍了利用自组织映射神经网络的自动分类功能对进行矿区土地复垦条件分类,为因地制宜地采取复垦措施提供依据。然后,基于 BP 神经网络,选取评价因子,通过对已有经验的学习,对复垦土地适宜性进行评价,并与传统的方法相比较,研究结果表明,利用人工神经网络辅助矿区土地复垦决策是可行的。
基金Supported by the National Natural Science Foundation of China （51274013） the International Cooperation Projects of Anhui, China （10080703026） Acknowledgements We thank the work group of soil information systems and solute transport models from Osnabrueck University of Applied Sciences, Germany for providing us their experimental work.
文摘Chloride ion transport in reclaimed soil solutions filled with fly ash （FA） was investigated by measuring the hydraulic parameters （i,e. water retention curves and hydraulic conductivity） of three substrates, namely GSL, GFA, and CFA. Similar simulations were carried out under certain weather conditions. The different boundary conditions of chloride transport were also discussed from FA texture, cover soil thickness, groundwater table level, and initial chloride concentration. Furthermore, the sensitivities of chloride ions to these effect factors were analyzed. The results show that the different top soil thickness and initial chloride concentration have no effect on salinity of topsoil solution in the monitoring points, but they can clearly change the chloride concentration of FA layers. The sensibilities from top soil thickness and initial chloride content are exceedingly weak to the salinity balance based on two dimensions of the time and concentration. While the different FA texture and groundwater table not only affect the salinity equilibrium process of the whole reclaimed soil profile, but also change its balance state. Generally, coarse FA particles and high groundwater table can defer the salinity balance process of the reclaimed soil solution, and they also increase the chloride concentration of FA layer solutions, and even topsoil ones.
