Fe-AI compound at the interface of steel-mushy AI-20Sn bonding plate was studied quantitatively. The relationship between ratio of Fe-AI compound at interface and bonding parameters (such as preheat temperature of ste...Fe-AI compound at the interface of steel-mushy AI-20Sn bonding plate was studied quantitatively. The relationship between ratio of Fe-AI compound at interface and bonding parameters (such as preheat temperature of steel plate, solid fraction of AI-20Sn slurry and rolling speed) was established by artificial neural networks perfectly. The results show that when the bonding parameters are 505℃ for preheat temperature of steel plate, 34.3% for solid fraction of AI-20Sn slurry and 10 mm/s for rolling speed, the reasonable ratio of Fe-AI compound corresponding to the largest interfacial shear strength of bonding plate is obtained. Its value is 72%. This reasonable ratio of Fe-AI compound is a quantitative criterion of interfacial embrittlement, that is, when the ratio of Fe-AI compound at interface is larger than 72%, interfacial embrittlement will occur.展开更多
Fe-Al compound at the interface of steel-mushy Al-28Pb bonding plate was studied quantitatively. The relationship between ratio of Fe-Al compound at interface and bonding parameters such as preheating temperature of s...Fe-Al compound at the interface of steel-mushy Al-28Pb bonding plate was studied quantitatively. The relationship between ratio of Fe-Al compound at interface and bonding parameters such as preheating temperature of steel plate, solid volume fraction of Al-28Pb slurry and rolling speed, was established by artificial neural networks perfectly. The results show that when the bonding parameters are 546 ℃ for preheating temperature of steel plate, 43.5% for solid volume fraction of Al-28Pb slurry and 8.6 mm/s for rolling speed, the reasonable ratio of Fe-Al compound corresponding to the largest interfacial shear strength of bonding plate is obtained as 71.5%. This reasonable ratio of Fe-Al compound is a quantitative criterion of interfacial embrittlement, that is, when the ratio of Fe-Al compound at interface is larger than 71.5%, interfacial embrittlement will occur.展开更多
10 quantum chemical descriptors of 21 aromatic compounds have been calculated by the semi-empirical quantum chemical method AM1. The Quantitative Structure-Biodegradability Relationships (QSBR) studies were performe...10 quantum chemical descriptors of 21 aromatic compounds have been calculated by the semi-empirical quantum chemical method AM1. The Quantitative Structure-Biodegradability Relationships (QSBR) studies were performed by the multiple linear regression (MLR), principal component regression (PCR) and back propagation artificial neural network (BP-ANN), respectively. The root mean square error (RMSE) of the training and validation sets of the BP-ANN model are 0.1363 and 0.0244, the mean absolute percentage errors (MAPE) are 0.1638 and 0.0326, the squared correlation coefficients (R^2) are 0.9853 and 0.9996, respectively. The results show that the BP-ANN model achieved a better prediction result than those of MLR and PCR. In addition, some insights into the structural factors affecting the aerobic biodegradation mechanism were discussed in detail.展开更多
With the artificial neural network(ANN) method combined with the multiple linear regression(MLR),based on a series of quantum chemical descriptors and molecular connectivity indexes,quantitative structure-activity...With the artificial neural network(ANN) method combined with the multiple linear regression(MLR),based on a series of quantum chemical descriptors and molecular connectivity indexes,quantitative structure-activity relationship(QSAR) models to predict the acute toxicity(-lgEC50) of substituted aromatic compounds to Photobacterium phosphoreum were established.Four molecular descriptors that appear in the MLR model,namely,the second order valence molecular connectivity index(2XV),the energy of the highest occupied molecular orbital(EHOMO),the logarithm of n-octyl alcohol/water partition coefficient(logKow) and the Connolly molecular area(MA),were inputs of the ANN model.The root-mean-square error(RMSE) of the training and validation sets of the ANN model are 0.1359 and 0.2523,and the correlation coefficient(R) is 0.9810 and 0.8681,respectively.The leave-one-out(LOO) cross validated correlation coefficient(Q L2OO) of the MLR and ANN models is 0.6954 and 0.6708,respectively.The result showed that the two methods are complementary in the calculations.The regression method gave support to the neural network with physical explanation,and the neural network method gave a more accurate model for QSAR.In addition,some insights into the structural factors affecting the acute toxicity and toxicity mechanism of substituted aromatic compounds were discussed.展开更多
采用人工神经网络(Artificial Neural Network,ANN)中的误差反向传播(Error Back Propagation,BP)神经网络分别建立了35种香水百合头香成分的三种结构与色谱保值之间的定量关系模型,其中两个为一维定量模型(分别是结构与色谱保留时间一...采用人工神经网络(Artificial Neural Network,ANN)中的误差反向传播(Error Back Propagation,BP)神经网络分别建立了35种香水百合头香成分的三种结构与色谱保值之间的定量关系模型,其中两个为一维定量模型(分别是结构与色谱保留时间一维定量模型、结构与Kovats保留指数一维定量模型),另一个为二维定量模型(结构与保留时间和Kovats保留指数同时预测的二维定量模型).以量子化学参数作为网络输入、以保留时间和Kovats保留指数作为输出,采用内外双重验证的办法分析和检验所得模型的稳定性.结果表明,BP-ANN模型稳健性与多元线性回归(Multiple Linear Regression,MLR)模型相当,但外部预测集拟合效果BP-ANN优于MLR.另外,BP-ANN构建的一维和二维模型相当,而二维模型更加节省时间.展开更多
基金This project was supported by the National Natural Science Foundation of China(No.50274047)Beijing Jiaotong Univer sity Foundation.
文摘Fe-AI compound at the interface of steel-mushy AI-20Sn bonding plate was studied quantitatively. The relationship between ratio of Fe-AI compound at interface and bonding parameters (such as preheat temperature of steel plate, solid fraction of AI-20Sn slurry and rolling speed) was established by artificial neural networks perfectly. The results show that when the bonding parameters are 505℃ for preheat temperature of steel plate, 34.3% for solid fraction of AI-20Sn slurry and 10 mm/s for rolling speed, the reasonable ratio of Fe-AI compound corresponding to the largest interfacial shear strength of bonding plate is obtained. Its value is 72%. This reasonable ratio of Fe-AI compound is a quantitative criterion of interfacial embrittlement, that is, when the ratio of Fe-AI compound at interface is larger than 72%, interfacial embrittlement will occur.
文摘Fe-Al compound at the interface of steel-mushy Al-28Pb bonding plate was studied quantitatively. The relationship between ratio of Fe-Al compound at interface and bonding parameters such as preheating temperature of steel plate, solid volume fraction of Al-28Pb slurry and rolling speed, was established by artificial neural networks perfectly. The results show that when the bonding parameters are 546 ℃ for preheating temperature of steel plate, 43.5% for solid volume fraction of Al-28Pb slurry and 8.6 mm/s for rolling speed, the reasonable ratio of Fe-Al compound corresponding to the largest interfacial shear strength of bonding plate is obtained as 71.5%. This reasonable ratio of Fe-Al compound is a quantitative criterion of interfacial embrittlement, that is, when the ratio of Fe-Al compound at interface is larger than 71.5%, interfacial embrittlement will occur.
基金supported by the Natural Science Foundation of Fujian Province (D0710019)the Natural Science Foundation of Overseas Chinese Affairs Office of the State Council (09QZR07)
文摘10 quantum chemical descriptors of 21 aromatic compounds have been calculated by the semi-empirical quantum chemical method AM1. The Quantitative Structure-Biodegradability Relationships (QSBR) studies were performed by the multiple linear regression (MLR), principal component regression (PCR) and back propagation artificial neural network (BP-ANN), respectively. The root mean square error (RMSE) of the training and validation sets of the BP-ANN model are 0.1363 and 0.0244, the mean absolute percentage errors (MAPE) are 0.1638 and 0.0326, the squared correlation coefficients (R^2) are 0.9853 and 0.9996, respectively. The results show that the BP-ANN model achieved a better prediction result than those of MLR and PCR. In addition, some insights into the structural factors affecting the aerobic biodegradation mechanism were discussed in detail.
基金supported by the Natural Science Foundation of Fujian Province (D0710019)the Natural Science Foundation of Overseas Chinese Affairs Office of the State Council (06QZR09)
文摘With the artificial neural network(ANN) method combined with the multiple linear regression(MLR),based on a series of quantum chemical descriptors and molecular connectivity indexes,quantitative structure-activity relationship(QSAR) models to predict the acute toxicity(-lgEC50) of substituted aromatic compounds to Photobacterium phosphoreum were established.Four molecular descriptors that appear in the MLR model,namely,the second order valence molecular connectivity index(2XV),the energy of the highest occupied molecular orbital(EHOMO),the logarithm of n-octyl alcohol/water partition coefficient(logKow) and the Connolly molecular area(MA),were inputs of the ANN model.The root-mean-square error(RMSE) of the training and validation sets of the ANN model are 0.1359 and 0.2523,and the correlation coefficient(R) is 0.9810 and 0.8681,respectively.The leave-one-out(LOO) cross validated correlation coefficient(Q L2OO) of the MLR and ANN models is 0.6954 and 0.6708,respectively.The result showed that the two methods are complementary in the calculations.The regression method gave support to the neural network with physical explanation,and the neural network method gave a more accurate model for QSAR.In addition,some insights into the structural factors affecting the acute toxicity and toxicity mechanism of substituted aromatic compounds were discussed.
文摘采用误差反传前向人工神经网络(artificial neural network,ANN)建立了24种取代芳烃的结构与其对发光菌的急性毒性之间的定量关系模型(ANN模型)。以24种取代芳烃的量子化学参数作为输入,急性毒性作为输出,采用内外双重验证的办法分析和检验所得模型的稳定性和外部预测能力。所构建网络模型的相关系数为0.9834、交叉检验相关系数为0.9780、标准偏差为0.11、残差绝对值≤0.33,应用于外部预测集,外部预测集相关系数为0.9955;而多元线性回归(multiple linear regression,MLR)法模型的相关系数为0.9786、标准偏差为0.12、残差绝对值≤0.36,外部预测集相关系数为0.9904。结果表明,ANN模型获得了比MLR模型更好的拟合效果。
文摘采用人工神经网络(Artificial Neural Network,ANN)中的误差反向传播(Error Back Propagation,BP)神经网络分别建立了35种香水百合头香成分的三种结构与色谱保值之间的定量关系模型,其中两个为一维定量模型(分别是结构与色谱保留时间一维定量模型、结构与Kovats保留指数一维定量模型),另一个为二维定量模型(结构与保留时间和Kovats保留指数同时预测的二维定量模型).以量子化学参数作为网络输入、以保留时间和Kovats保留指数作为输出,采用内外双重验证的办法分析和检验所得模型的稳定性.结果表明,BP-ANN模型稳健性与多元线性回归(Multiple Linear Regression,MLR)模型相当,但外部预测集拟合效果BP-ANN优于MLR.另外,BP-ANN构建的一维和二维模型相当,而二维模型更加节省时间.