模糊神经网络在UV-LIGA工艺优化中的应用

将模糊神经网络理论和算法应用于负性光刻胶(SU-8)加工高分辨率和高深宽比微结构的工艺研究,在正交试验的基础上对网络进行训练,建立了光刻图形质量与前烘时间、前烘温度、曝光量、后烘时间之间的预测模型。该模型采用五层前向模糊神经网络,学习算法为梯度下降法。进行了实验,实验结果表明,前烘温度与前烘时间对光刻质量影响最大。对120~340 μm厚的光刻胶,前烘温度取95℃,前烘时间100 min时,图形的相对线宽差最小;超声搅拌能缩短显影时间,显著改善图形质量,试验结果与计算结果十分吻合。将模糊神经网络应用于UV－LIGA工艺中,能实现光刻加工微结构的工艺参数优化。

Application of fuzzy neural network to optimizing UV-LIGA process

The theory and the algorithm of the fuzzy neural network were applied in the research of the fabrication of high resolute microstructure using negative (SU-8), the artificial neural network were trained based on orthogonality experiment, and the prediction model was built between the fabrication structure quality and the main technology parameters. The model was a Feedforward Fuzzy neural network possessing five layers,and Gradient Descent was adopted as learning algorithm. The experiment results show that the temperature and the soft bake time is the most important factor of the structure quality. when thickness of the photoresist is ranged from 120~340 μm,the temperature and time of the soft bake is 95℃ and 100 min,respectivly,then the optimal microstructure with the least relative line width discrepancy can be obtained. In addition,the developing time is reduced and the image quality is greatly improved through ultrasonic agitation. The predict results are in good agreement with the experimental results,and the lithographic process could be optimized with fuzzy neural network.