Suppliers＇ selection in supply chain management （SCM） has attracted considerable research interests in recent years. Recent literatures show that neural networks achieve better performance than traditional statisti...Suppliers＇ selection in supply chain management （SCM） has attracted considerable research interests in recent years. Recent literatures show that neural networks achieve better performance than traditional statistical methods. However, neural networks have inherent drawbacks, such as local optimization solution, lack generalization, and uncontrolled convergence. A relatively new machine learning technique, support vector machine （SVM）, which overcomes the drawbacks of neural networks, is introduced to provide a model with better explanatory power to select ideal supplier partners. Meanwhile, in practice, the suppliers＇ samples are very insufficient. SVMs are adaptive to deal with small samples＇ training and testing. The prediction accuracies for BPNN and SVM methods are compared to choose the appreciating suppliers. The actual examples illustrate that SVM methods are superior to BPNN.展开更多
Robustly stable multi-step-ahead model predictive control (MPC) based on parallel support vector machines (SVMs) with linear kernel was proposed. First, an analytical solution of optimal control laws of parallel SVMs ...Robustly stable multi-step-ahead model predictive control (MPC) based on parallel support vector machines (SVMs) with linear kernel was proposed. First, an analytical solution of optimal control laws of parallel SVMs based MPC was derived, and then the necessary and sufficient stability condition for MPC closed loop was given according to SVM model, and finally a method of judging the discrepancy between SVM model and the actual plant was presented, and consequently the constraint sets, which can guarantee that the stability condition is still robust for model/plant mismatch within some given bounds, were obtained by applying small-gain theorem. Simulation experiments show the proposed stability condition and robust constraint sets can provide a convenient way of adjusting controller parameters to ensure a closed-loop with larger stable margin.展开更多
The origin and influence factors of sand liquefaction were analyzed, and the relation between liquefaction and its influence factors was founded. A model based on support vector machines (SVM) was established whose in...The origin and influence factors of sand liquefaction were analyzed, and the relation between liquefaction and its influence factors was founded. A model based on support vector machines (SVM) was established whose input parameters were selected as following influence factors of sand liquefaction: magnitude (M), the value of SPT, effective pressure of superstratum, the content of clay and the average of grain diameter. Sand was divided into two classes: liquefaction and non-liquefaction, and the class label was treated as output parameter of the model. Then the model was used to estimate sand samples, 20 support vectors and 17 borderline support vectors were gotten, then the parameters were optimized, 14 support vectors and 6 borderline support vectors were gotten, and the prediction precision reaches 100%. In order to verify the generalization of the SVM method, two other practical samples' data from two cities, Tangshan of Hebei province and Sanshui of Guangdong province, were dealt with by another more intricate model for polytomies, which also considered some influence factors of sand liquefaction as the input parameters and divided sand into four liquefaction grades: serious liquefaction, medium liquefaction, slight liquefaction and non-liquefaction as the output parameters. The simulation results show that the latter model has a very high precision, and using SVM model to estimate sand liquefaction is completely feasible.展开更多
This paper focuses on the applications of the support vector machines in solving the problem of blind recovery in digital communication systems.We introduce the technique of support vector machines briefly,the develop...This paper focuses on the applications of the support vector machines in solving the problem of blind recovery in digital communication systems.We introduce the technique of support vector machines briefly,the development of blind equalization and analyze the problems which need to be resolved of the blind problems.Then the applicability of support vector machines in blind problem is highlighted and deduced.Finally,merit and shortage of blind equalization using support vector machines which is already exist to be discussed and the direction of further research is indicated.展开更多
针对现有的工业控制系统异常检测分类方法大多无法有效处理类不平衡和重叠耦合的问题,提出了一种基于干扰样本分布优化的工控异常检测改进SVM模型(Improved SVM Model Based on Adaptive Differential Evolution with Sphere, SJADE_SV...针对现有的工业控制系统异常检测分类方法大多无法有效处理类不平衡和重叠耦合的问题,提出了一种基于干扰样本分布优化的工控异常检测改进SVM模型(Improved SVM Model Based on Adaptive Differential Evolution with Sphere, SJADE_SVM),该模型将基于超球体覆盖的自适应差分进化过采样技术与支持向量机相结合。首先,通过改进超球体覆盖算法和构建概率公式,来识别和排除干扰样本;然后,改进合成少数派过采样技术,通过对安全样本采样,缓解类不平衡和重叠耦合问题;最后,使用自适应差分进化算法优化样本的位置和属性,同时使用SVM进行分类。在6个真实工控数据集和4个UCI公开数据集上共设计3组实验,包括与逻辑回归和高斯朴素贝叶斯等异常检测分类算法的性能对比、改善样本分布方法的实验对比以及算法的运行时间对比。实验结果表明,该模型在F-score和G-mean评价指标上分别提高了38.29%和10.54%,分类效果稳居前三,且在α=0.05的非参数双侧Wilcoxon符号秩检验和Friedman检验等统计实验中表现出显著的性能优势。展开更多
文摘Suppliers＇ selection in supply chain management （SCM） has attracted considerable research interests in recent years. Recent literatures show that neural networks achieve better performance than traditional statistical methods. However, neural networks have inherent drawbacks, such as local optimization solution, lack generalization, and uncontrolled convergence. A relatively new machine learning technique, support vector machine （SVM）, which overcomes the drawbacks of neural networks, is introduced to provide a model with better explanatory power to select ideal supplier partners. Meanwhile, in practice, the suppliers＇ samples are very insufficient. SVMs are adaptive to deal with small samples＇ training and testing. The prediction accuracies for BPNN and SVM methods are compared to choose the appreciating suppliers. The actual examples illustrate that SVM methods are superior to BPNN.
基金Project(2002CB312200) supported by the National Key Fundamental Research and Development Program of China project(60574019) supported by the National Natural Science Foundation of China
文摘Robustly stable multi-step-ahead model predictive control (MPC) based on parallel support vector machines (SVMs) with linear kernel was proposed. First, an analytical solution of optimal control laws of parallel SVMs based MPC was derived, and then the necessary and sufficient stability condition for MPC closed loop was given according to SVM model, and finally a method of judging the discrepancy between SVM model and the actual plant was presented, and consequently the constraint sets, which can guarantee that the stability condition is still robust for model/plant mismatch within some given bounds, were obtained by applying small-gain theorem. Simulation experiments show the proposed stability condition and robust constraint sets can provide a convenient way of adjusting controller parameters to ensure a closed-loop with larger stable margin.
基金Supported by the National Creative Research Groups Science Foundation of P.R. China （NCRGSFC： 60421002） and National High Technology Research and Development Program of China （863 Program） （2006AA04 Z182）
文摘The origin and influence factors of sand liquefaction were analyzed, and the relation between liquefaction and its influence factors was founded. A model based on support vector machines (SVM) was established whose input parameters were selected as following influence factors of sand liquefaction: magnitude (M), the value of SPT, effective pressure of superstratum, the content of clay and the average of grain diameter. Sand was divided into two classes: liquefaction and non-liquefaction, and the class label was treated as output parameter of the model. Then the model was used to estimate sand samples, 20 support vectors and 17 borderline support vectors were gotten, then the parameters were optimized, 14 support vectors and 6 borderline support vectors were gotten, and the prediction precision reaches 100%. In order to verify the generalization of the SVM method, two other practical samples' data from two cities, Tangshan of Hebei province and Sanshui of Guangdong province, were dealt with by another more intricate model for polytomies, which also considered some influence factors of sand liquefaction as the input parameters and divided sand into four liquefaction grades: serious liquefaction, medium liquefaction, slight liquefaction and non-liquefaction as the output parameters. The simulation results show that the latter model has a very high precision, and using SVM model to estimate sand liquefaction is completely feasible.
基金supported in part by the National Natural Science Foundation of China(No.60772060 )the project of NJUPT(No.NY207056)
文摘This paper focuses on the applications of the support vector machines in solving the problem of blind recovery in digital communication systems.We introduce the technique of support vector machines briefly,the development of blind equalization and analyze the problems which need to be resolved of the blind problems.Then the applicability of support vector machines in blind problem is highlighted and deduced.Finally,merit and shortage of blind equalization using support vector machines which is already exist to be discussed and the direction of further research is indicated.
文摘针对现有的工业控制系统异常检测分类方法大多无法有效处理类不平衡和重叠耦合的问题,提出了一种基于干扰样本分布优化的工控异常检测改进SVM模型(Improved SVM Model Based on Adaptive Differential Evolution with Sphere, SJADE_SVM),该模型将基于超球体覆盖的自适应差分进化过采样技术与支持向量机相结合。首先,通过改进超球体覆盖算法和构建概率公式,来识别和排除干扰样本;然后,改进合成少数派过采样技术,通过对安全样本采样,缓解类不平衡和重叠耦合问题;最后,使用自适应差分进化算法优化样本的位置和属性,同时使用SVM进行分类。在6个真实工控数据集和4个UCI公开数据集上共设计3组实验,包括与逻辑回归和高斯朴素贝叶斯等异常检测分类算法的性能对比、改善样本分布方法的实验对比以及算法的运行时间对比。实验结果表明,该模型在F-score和G-mean评价指标上分别提高了38.29%和10.54%,分类效果稳居前三,且在α=0.05的非参数双侧Wilcoxon符号秩检验和Friedman检验等统计实验中表现出显著的性能优势。
