By taking the Yong River for example in this paper, based on the multiple measured data during 1957 to 2009, the change process of runoff, tide feature, tidal wave, tidal influx and sediment transport are analyzed. Th...By taking the Yong River for example in this paper, based on the multiple measured data during 1957 to 2009, the change process of runoff, tide feature, tidal wave, tidal influx and sediment transport are analyzed. Then a mathematical model is used to reveal the influence mechanism on hydrodynamic characteristics and sediment transport of the wading engineering groups such as a tide gate, a breakwater, reservoirs, bridges and wharves, which were built in different periods. The results showed the hydrodynamic characteristics and sediment transport of the Yong River changed obviously due to the wading engineering groups. The tide gate induced deformation of the tidal wave, obvious reduction of the tidal influx and weakness of the tidal dynamic, decrease of the sediment yield of flood and ebb tide and channel deposition. The breakwater blocked estuarine entrances, resulting in the change of the tidal current and the reduction of the tidal influx in the estuarine area. The large-scale reservoirs gradually made the decrease of the Yong River runoff. The bridge and wharf groups took up cross-section areas, the cumulative affection of which caused the increase of tidal level in the tidal river.展开更多
水质指标具有多元相关性、时序性和非线性的特点,为有效预测河流水质变化,针对水质数据存在缺失和异常的问题,提出基于灰色关联分析-门控循环单元(Grey Relational Analysis-Gated Recurrent Unit, GRA-GRU)的水质预测模型。以淮河流域...水质指标具有多元相关性、时序性和非线性的特点,为有效预测河流水质变化,针对水质数据存在缺失和异常的问题,提出基于灰色关联分析-门控循环单元(Grey Relational Analysis-Gated Recurrent Unit, GRA-GRU)的水质预测模型。以淮河流域水质数据为样本,使用线性插值修补缺失数据和剔除的异常数据。使用灰色关联分析计算不同水质指标间的相关性,选择高相关性的水质指标以确定输入变量,并使用门控循环单元(Gated Recurrent Unit, GRU)预测不同的水质指标。将GRA-GRU的预测结果与反向传播神经网络(Back Propagation Neural Network, BPNN)、循环神经网络(Recurrent Neural Network, RNN)、长短期记忆神经网络(Long Short Term Memory, LSTM)、GRU及灰色关联分析-长短期记忆神经网络(Grey Relational Analysis-Long Short Term Memory, GRA-LSTM)进行对比分析,结果显示GRA-GRU在不同水质指标预测上具有较好的适应性,可以有效降低预测误差。其中,与其他模型相比,GRA-GRU预测的化学需氧量在均方根误差上分别降低了3.617%、0.681%、0.478%、1.505%和0.471%。展开更多
基金financially supported by the National Science Foundation for Distinguished Young Scholars of China(Grant No.51125034)the National Natural Science Foundation of China(Grant Nos.51279046 and 50909037)the Fundamental Research Funds for the Central Universities(Grant No.2010B01114)
文摘By taking the Yong River for example in this paper, based on the multiple measured data during 1957 to 2009, the change process of runoff, tide feature, tidal wave, tidal influx and sediment transport are analyzed. Then a mathematical model is used to reveal the influence mechanism on hydrodynamic characteristics and sediment transport of the wading engineering groups such as a tide gate, a breakwater, reservoirs, bridges and wharves, which were built in different periods. The results showed the hydrodynamic characteristics and sediment transport of the Yong River changed obviously due to the wading engineering groups. The tide gate induced deformation of the tidal wave, obvious reduction of the tidal influx and weakness of the tidal dynamic, decrease of the sediment yield of flood and ebb tide and channel deposition. The breakwater blocked estuarine entrances, resulting in the change of the tidal current and the reduction of the tidal influx in the estuarine area. The large-scale reservoirs gradually made the decrease of the Yong River runoff. The bridge and wharf groups took up cross-section areas, the cumulative affection of which caused the increase of tidal level in the tidal river.
文摘水质指标具有多元相关性、时序性和非线性的特点,为有效预测河流水质变化,针对水质数据存在缺失和异常的问题,提出基于灰色关联分析-门控循环单元(Grey Relational Analysis-Gated Recurrent Unit, GRA-GRU)的水质预测模型。以淮河流域水质数据为样本,使用线性插值修补缺失数据和剔除的异常数据。使用灰色关联分析计算不同水质指标间的相关性,选择高相关性的水质指标以确定输入变量,并使用门控循环单元(Gated Recurrent Unit, GRU)预测不同的水质指标。将GRA-GRU的预测结果与反向传播神经网络(Back Propagation Neural Network, BPNN)、循环神经网络(Recurrent Neural Network, RNN)、长短期记忆神经网络(Long Short Term Memory, LSTM)、GRU及灰色关联分析-长短期记忆神经网络(Grey Relational Analysis-Long Short Term Memory, GRA-LSTM)进行对比分析,结果显示GRA-GRU在不同水质指标预测上具有较好的适应性,可以有效降低预测误差。其中,与其他模型相比,GRA-GRU预测的化学需氧量在均方根误差上分别降低了3.617%、0.681%、0.478%、1.505%和0.471%。
