摘要
在对Ti-Al-O-N体系热力学分析的基础上合成了AlON-TiN复相材料,XRD,TEM分析与观测结果证实了热力学分析的可靠性。运用人工神经元网络(artificial neural networks,ANN)对合成AlON-TiN的工艺参数进行了训练,训练结果良好,神经网络的预测值和实测值一致;遗传算法(genetic algorithm,GA)寻优得到最佳工艺参数:烧结温度为1830℃,保温2.45 h,TiN的含量为15%(质量分数)。对遗传算法寻优结果进行了革因素分析,随着TiN含量的增加,强度增加,而随烧结温度和保温时间的增加,强度先增加后减小。
The AlON-TiN composite is synthesized by the hot pressure sintering technique according to the thermodynamic analysis of the Ti-Al-O-N system. XRD and TEM results verify the credibility of the thermodynamic analysis. The Web-based artificial neural networks-genetic algorithm (ANN-GA) system is used to optimize the synthesis parameters and the results agree with the original targets. The predicted points are in accord with the original points. The best synthesis parameters got by GA are as follows: the sintering temperature is 1830°C, the soaking time is 2.45 h and the content of TiN is 15%. The analysis of the single factor based on GA shows that the strength of the composite increases with the increase of the TiN content, and with the increase of the soaking time and the sintering temperature the strength of the composite increases first and then decreases.
出处
《硅酸盐学报》
EI
CAS
CSCD
北大核心
2003年第8期727-731,737,共6页
Journal of The Chinese Ceramic Society
基金
国家自然科学资金(50074004
50272010)
关键词
氮氧化铝-氮化钛
热力学分析
复相材料
人工神经元网络
遗传算法
Genetic algorithms
Hot pressing
Neural networks
Optimization
Sintering
Synthesis (chemical)
Thermodynamics
Titanium nitride