Li-Zn mixed ferrites with composition formula ZnxLi0.5-x/2Fe2.5-x/2O4 (0.2≤x≤0.8) were prepared by the usual ceramic method in 1000~1150℃. The effects of Zn substitution and sintering temperature on the formation, ...Li-Zn mixed ferrites with composition formula ZnxLi0.5-x/2Fe2.5-x/2O4 (0.2≤x≤0.8) were prepared by the usual ceramic method in 1000~1150℃. The effects of Zn substitution and sintering temperature on the formation, densification, microstructure and a.c. electrical conductivity have been studied. Under the effect of changing the firing temperature and Zn content, high sintered Li-Zn ferrite bodies are achieved. More fine structure bodies having high electrical resistance are obtained at high Zn content展开更多
Brake friction materials with different zinc powder contents(0,2,4,6,8 wt.%)were fabricated via powder metallurgy method.The results indicate that with the increasing zinc powder content,the density and thermal conduc...Brake friction materials with different zinc powder contents(0,2,4,6,8 wt.%)were fabricated via powder metallurgy method.The results indicate that with the increasing zinc powder content,the density and thermal conductivity of the materials gradually increase,while the hardness decreases monotonously.With increasing zinc powder content,the curve of the nominal friction coefficient shows fluctuating trend but the lowest friction coefficient also shows an increase.However,the wear rate and braking noise of the friction material monotonously decrease with increasing zinc content.This effect may be attributed to the transformation of the tribological mechanism from adhesive wear and abrasive wear to adhesive wear.The brake friction material with 4 wt.%zinc powder exhibits both the best tribological and noise performance.展开更多
目的建立皮尔逊相关系数(Pearson correlation coefficient,PCC)和长短期记忆(long short term memory,LSTM)神经网络的反应液葡萄糖含量预测模型用以实时预测葡萄糖酸锌生产过程中反应液葡萄糖含量。方法通过葡萄糖酸锌制备实验,结合PC...目的建立皮尔逊相关系数(Pearson correlation coefficient,PCC)和长短期记忆(long short term memory,LSTM)神经网络的反应液葡萄糖含量预测模型用以实时预测葡萄糖酸锌生产过程中反应液葡萄糖含量。方法通过葡萄糖酸锌制备实验,结合PCC理论确定对反应液葡萄糖含量有较大影响的因素,对这些因素进行数据采集并将其作为神经网络的输入变量,采集反应液葡萄糖含量数据并进行处理,将其作为神经网络的输出变量,进而建立反向传播神经网络(backpropagation neural network,BP)和LSTM神经网络的反应液葡萄糖含量预测模型。结果通过100次模型迭代训练,对照BP反应液葡萄糖含量预测模型可以看出LSTM反应液葡萄糖含量预测模型在测试集的误差约为0.45%,误差较小,准确度较高。结论基于LSTM反应液葡萄糖含量预测模型显著提高了预测精度,相比现有检测方法更加智能高效,能够有效辅助生产进行。展开更多
文摘Li-Zn mixed ferrites with composition formula ZnxLi0.5-x/2Fe2.5-x/2O4 (0.2≤x≤0.8) were prepared by the usual ceramic method in 1000~1150℃. The effects of Zn substitution and sintering temperature on the formation, densification, microstructure and a.c. electrical conductivity have been studied. Under the effect of changing the firing temperature and Zn content, high sintered Li-Zn ferrite bodies are achieved. More fine structure bodies having high electrical resistance are obtained at high Zn content
基金Project(2016YFB1100103)supported by the National Key Research and Development Program of ChinaProject(KC1703004)supported by the Science and Technology Planning Project of Changsha City,ChinaProject(2018ZZTS127)supported by the Fundamental Research Funds for the Central Universities of Central South University,China。
文摘Brake friction materials with different zinc powder contents(0,2,4,6,8 wt.%)were fabricated via powder metallurgy method.The results indicate that with the increasing zinc powder content,the density and thermal conductivity of the materials gradually increase,while the hardness decreases monotonously.With increasing zinc powder content,the curve of the nominal friction coefficient shows fluctuating trend but the lowest friction coefficient also shows an increase.However,the wear rate and braking noise of the friction material monotonously decrease with increasing zinc content.This effect may be attributed to the transformation of the tribological mechanism from adhesive wear and abrasive wear to adhesive wear.The brake friction material with 4 wt.%zinc powder exhibits both the best tribological and noise performance.
文摘目的建立皮尔逊相关系数(Pearson correlation coefficient,PCC)和长短期记忆(long short term memory,LSTM)神经网络的反应液葡萄糖含量预测模型用以实时预测葡萄糖酸锌生产过程中反应液葡萄糖含量。方法通过葡萄糖酸锌制备实验,结合PCC理论确定对反应液葡萄糖含量有较大影响的因素,对这些因素进行数据采集并将其作为神经网络的输入变量,采集反应液葡萄糖含量数据并进行处理,将其作为神经网络的输出变量,进而建立反向传播神经网络(backpropagation neural network,BP)和LSTM神经网络的反应液葡萄糖含量预测模型。结果通过100次模型迭代训练,对照BP反应液葡萄糖含量预测模型可以看出LSTM反应液葡萄糖含量预测模型在测试集的误差约为0.45%,误差较小,准确度较高。结论基于LSTM反应液葡萄糖含量预测模型显著提高了预测精度,相比现有检测方法更加智能高效,能够有效辅助生产进行。
