基于模糊神经网络的岩性智能识别及其应用

Rock intelligent identification based on fuzzy neural network and its application

在对岩石分类的基础上进行岩性识别。提出了基于神经网络驱动模糊推理方法，目的在于将神经网络和模糊推理二者所长能够较好地结合起来，使得分析系统既具有推理功能，又具有很强的学习能力，这样更能适合变化复杂的牙轮钻进过程，更能接近人的思维方式，为难以模型化的牙轮钻进自动控制以及未来的智能钻机研究开辟了一条有效途径。在具体的分类、学习推理和计算应用中，比较使用了改进的BP网络、采用L－M算法的BP网络及径向基函数网络（RBFN）等。

In addtion, for fock property identification, fuzzy reasoning method based on neural network driven is first introduced The advantages of neural networks and fuzzy reasoning have been combined together to make the analysis system have the reasoning function and high self-study ability which is more suitable for complex drilling process. It is more close to human thinking pattern and opens a new path for drilling automation and future study for intelligent drills. In the specific classification, reasoning and computing are based on improved BP networks, BP networks with L-M algorithm and Radial Basis Function Natwork(RBFN). The results are satisfied.