排水管道泥沙淤积深度估算方法研究

Estimating Method of Sediment Deposition Depth in Drainage Pipeline

运用C#编程语言实现了SWMM模型参数的自动提取,通过创建BP-人工神经网络实现了节点水深值与模型参数值之间的非线性拟合,基于模型参数率定的思路提出了一种排水管道泥沙淤积深度的估算方法,并且以G市某雨水排水系统为例,采用4场降雨数据对模型进行了校核与验证。结果表明,通过两场降雨数据的验证,对于管径为1.2~1.8 m的管道,淤积深度预测值与实测值之间的绝对误差均在4 cm以内;模拟结果和实测数据的水深峰现时间偏差均低于实测数据历时的5%,峰值的数值偏差均在3%以内;场次3和场次4两场降雨4个监测点的水深预测值与实测值的平均相对误差分别为3.35%、2.98%,2.75%、2.51%,7.39%、6.77%,5.53%、8.15%,说明该方法能够对排水管道淤积情况进行有效预测。

The automatic extraction of SWMM model parameters was realized by using C#programming language,and the nonlinear fitting between nodal water depth value and model parameter was realized by building a BP-artificial neural network(ANN).Based on the idea of parameter calibration of storm water management model(SWMM),a method for estimating the sediment deposition depth of drainage pipelines was proposed.The model was checked and validated with data of four rainfall events a rainwater drainage system of G city.By model validation with data of two rainfall events,the absolute error between the predicted and measured values of sedimentation depth in pipelines with diameters between 1.2 m and 1.8 m was within 4 cm.The deviation of water depth peak appearance time between the simulated results and the measured data was below 5%of the measured duration,and the deviation of the water depth peak value was within 3%.The average relative errors between the simulated and measured water depth of rainfall 3 and 4 at the four monitoring points were 3.35%and 2.98%,2.75%and 2.51%,7.39%and 6.77%and 5.53%and 8.15%,respectively,indicating that this method could effectively predict the sedimentation of drainage pipelines.