欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
根据权重求取的不同约束准则,对变形并联组合预测模型的构建进行研究。分别构建了非最优、最优权、灰色综合关联度定权、熵权、神经网络非线性5类变形并联组合预测模型,讨论了5类约束准则下的权重求取特点,并比较了预测的效果。分析表明,基于神经网络的非线性组合预测具有良好的特性和较高的预测精度。
英文摘要：
The parallel combination deformation prediction models were studied. According to the weighted constraint criterion, five kinds of deformation parallel combination prediction models were established, including sub- optimal weight combination, optimal weigh combination, synthesis relational grade of grey-weight, entropy-weight and artificial neural network-weight. Based on the characteristics of the models, the effects of parallel combination prediction under different constraint criterion are compared. The results of analysis with real data show that the nonlinear combination prediction based on neural network is the best one.
同期刊论文项目
同项目期刊论文
Application of Lodrigues Matrix in Coordinate Transformation of Similar Material Simulation Experime
The Research on Forest Resources Change Detection Based on C5.0 Algorithm and Neighborhood Correlati
Study of Earthquake Disaster Population Risk Based on GIS a case study of Wenchuan earthquake region
Based on GIS and MATLAB Technology Forecast the Dynamic Variation of the Groundwater in Minqin Oasis
期刊信息
位置：