最小曲率神经网络在油气预测中的应用

在应用地球物理领域中，人工神经网络在模式识别和油气预测方面得到较好地应用，前向网络的重要特性是能够总结，归纳已知样本隐含的函数关系。然而礤推广性能有待进一步研究，为此，本文强调了这个问题的重要性，并提出了改善网络推广性能的技术；就是在网络的学习过程中，不仅使总误差下降，还尽可能使建立的“隐函数”平滑，并用曲率表征隐函数的平滑程度，计算实例表明，本文的算法可以明显地改善网络的推广性能。最后给出了用该

Application of minimum-curvature neural network in hydrocarbon prediction

Neural network has been nicely applied in pattern recognition and hydrocarbon prediction in applied geophysics. The importance of forward neural network lies in showing the impliclt functional relation arnong known cxamplcs, hut its gcneralized application needs to be studied further. Hence we point out the importance of the problem, and put forward the technique for improving the gencralization of the neu-ral network. The technique can rcduce overall error and smooth "implicit function"as much as possible in the course of learning, the smoothing degree of the implicit function being indicated by curvature. Computation exarnple shows that our algo-rithm can obviously improve the generalization of the neural network. The real ex-ample is given of how the techniquc was applied in hydrocarbon prcdiction in Laohc Oil Field.