基于神经网络的提拉法钛单晶生长过程建模

用神经网络建立提拉法钛单晶生长过程的经验模型，并通过试验验证模型的有效性。改进钛单晶生长试验设备，采集建立经验模型所需的无噪实验数据。建立前馈神经网络预测器，建模提拉法钛晶体生长过程非线性动态特性，用自适应BP算法训练神经网络，以加快网络的学习和收敛。

Modeling of Czochralski Single Crystal Growth Process Using Neural Network

First known attempt to empirically modeling and experimentally verifying the growth of ilmenite single crystals using Czockralski process is presented. Czochralski is an industrial crystal pulling process extensively used for silicon and germanium single crystal growth. The experimental apparatus for ilmenite growth process has been significantly improved, and applied to acquisition of noise-free experimental data for empirical modeling. A feedforward multilayer perceptron is used to develop a single-step predictor, modeling the thermal response of the Czochralski single crystal growth process of ilmenite. The training of the neural network is performed using adaptive back-propagation, an accelerated learning algorithm.