论文部分内容阅读
为进一步了解漓江流域含沙量日变化规律及降雨对含沙量的影响,监测并分析了2013年漓江监测点日均含沙量及其中上游地区的降雨量,统计数据表明2013年丰水期漓江的日均含沙量明显高于枯水期,丰水期含沙量占全年的85.3%以上,且日均含沙量与漓江中上游地区的日均降雨量显著正相关。同时,利用2013年漓江监测点日均含沙量及其中上游地区的降雨量数据构建BP神经网络模型,对比2013年6月实测和预测含沙量结果表明,以测量当天、前一天、前两天降雨量为输入神经元进行训练而获得的预测值的平均相对误差绝对值为7.16%、合格率为90%,符合预测精度要求,说明建立的BP模型结构合理,预测效果较好。研究结果可为漓江含沙量监测预报及漓江流域生态环境治理提供技术参考。
In order to further understand the diurnal variation of sediment concentration in the Lijiang River Basin and the effect of rainfall on sediment concentration, the daily average sediment concentration of the Lijiang River monitoring point in 2013 and the rainfall in the upper reaches of the Lijiang River monitoring point were monitored and analyzed. The statistical data show that in 2013, The average daily sediment concentration of Lijiang River was significantly higher than that of dry season. The sediment concentration in flood season was over 85.3% of the whole year, and the daily average sediment concentration was positively correlated with the daily average rainfall in the middle and upper reaches of Lijiang River. At the same time, BP neural network model was constructed by using the daily average sediment concentration in the monitoring points of Lijiang River in 2013 and the rainfall data in the upper reaches of the Lijiang River. Compared with the measured and predicted sediment concentrations in June 2013, The average relative error of predicted value obtained from daily rainfall for inputting neurons was 7.16% and the pass rate was 90%, which accorded with the prediction accuracy requirements. The results showed that the BP model was reasonable and the prediction effect was good. The results can provide technical reference for the monitoring and forecast of sediment concentration in the Lijiang River and ecological environment control in Lijiang River Basin.