求取压裂气井无阻流量的BP神经网络法

APPLCATION OF BP ARTIFICIAL NERVE NETWORK TO CALCULATE ABSOLUTE OPEN FLOW OF GAS WELLS AFTER FRACTURING

与现在流行的求取压裂气井无阻流量的方法相比,BP人工神经网络方法具有极大的优越性和适用性,它勿需考虑压裂后气井的复杂渗流规律,也不需数值模拟。BP人工神经网络方法在训练时只需知道目标井的压后评估数据和区块内部分试井数据,为提高经济效益、快速、高效开发低渗气藏提供了新的技术手段。

In comparison with the traditional methods, the BP artificial nerve network method possesses considerable superiority and applicability and it isn't in need of various steady percolation model based on the percolation theory of gas reservoirs after fracturing or numerical value simulation. The BP artificial nerve network method is only in want of the primary data of the target gas well when forecasting absolute open flow and much practical testing data of a few gas wells as the samples in the same reservoir, which provides a new feasible way for the exploitation of low-permeability gas wells economically, speedly and effectly.