The results of estimation show that BP neural network can estimate farmland drainage accurately with the largest relative error only of-2.45%.
估算结果表明利用BP神经网络可以准确地估算农田排水量，最大相对误差仅为-2.45%。
2
Some key technical problems have been amended, for example, leakage of hydraulic pressure bursa, control of drainage, analysis of apparatus system error, design of accessorial equipments, and so on.
The influential factors of calculation error for groundwater table are discussed by using the iteration method to find solution of numerical model in the mine drainage forecast of deep drawdown.