The suction friction angle index φ~b of unsaturated soil is predicted and studied based on the method of BP nerural network with TRAINLM training function.

The shear strength index of Xi\'an loess is predicted based on BP neural network with TRAINGDX training function in the gradient descent method.

The water quality of 17 wells in the area polluted were classified in different degrees with different train epochs by using correlation functions in MATLAB and area polluted was determined.

The paper mainly discusses BP neural network structure and choice of sample trainning parameter, which is based on neural netwrok generator.

The paper mainly discusses BP newral network structure and selection of samples trainning parameters based on neural network generator developed by the authors and is devoted to study of how to overcome the disadvantages of conventional BP neural network.

In order to getting the effective training data of chemical engineering modeling,two algorithms that fuzzy C-means and fast global fuzzy C-means clustering were used.

Also,the technologies of channel probing/ranging, parameter training and handshake procedure are introduced.