随着军事信息化、智能化进程不断深化发展,面对突发情况时通常需要在极短时间内完成事件处置方案拟制,因此对事件处置智能化的需求极为急迫。结合Platt scaling概率校准模型,针对非均衡数据分布和多类属输出概率校准拓展问题,对核最小...随着军事信息化、智能化进程不断深化发展,面对突发情况时通常需要在极短时间内完成事件处置方案拟制,因此对事件处置智能化的需求极为急迫。结合Platt scaling概率校准模型,针对非均衡数据分布和多类属输出概率校准拓展问题,对核最小二乘支持向量机模型进行改进,提出一种校准概率最小二乘支持向量机(Calibration Probabilistic Least Squares Support Vector Machine, CPLS-SVM)分类算法;设计一种基于CPLS-SVM算法的事件处置案例匹配方法,提高复杂案例样本环境下的匹配精度;通过构建基于校准概率最小二乘支持向量机,并应用到事件处置实例模拟数据进行仿真实验,验证算法的有效性和性能优势。展开更多
The occurrence of social security events is uncertain, and the distribution characteristics are highly complex due to a variety of external factors, posing challenges to their rapid and effective handling. The scienti...The occurrence of social security events is uncertain, and the distribution characteristics are highly complex due to a variety of external factors, posing challenges to their rapid and effective handling. The scientific and reasonable requirement evaluation of the emergency force to deal with social security events is very urgent. Based on data analysis, this paper uses the neural network, operational research, modelling and simulation to predict and analyze social security events, studies the usage rule of emergency force and deployment algorithm, and conducts simulation experiments to evaluate and compare the different force deployment schemes for selection.展开更多
A new generation of numerical prediction system GRAPES (a short form of Global/Regional Assimilation and PrEdiction System) was set up in China Meteorological Administration (CMA). This paper focuses on the scientific...A new generation of numerical prediction system GRAPES (a short form of Global/Regional Assimilation and PrEdiction System) was set up in China Meteorological Administration (CMA). This paper focuses on the scientific design and preliminary results of the numerical prediction model in GRAPES, including basic idea and strategy of the general scientific design, multi-scale dynamic core, physical package configuration, architecture and parallelization of the codes. A series of numerical experiments using the real data with horizontal resolutions from 10 to 280 km and idealized experiments with very high resolution up to 100 m are conducted, giving encouraging results supporting the multi-scale application of GRAPES. The results of operational implementation of GRAPES model in some NWP centers are also presented with stress at evaluations of the capability to predict the main features of precipitation in China. Finally the issues to be dealt with for further development are discussed.展开更多
During the past few years, most of the new developed numerical weather prediction models adopt the strategy of multi-scale technique. Therefore, China Meteorological Administration has devoted to developing a new gene...During the past few years, most of the new developed numerical weather prediction models adopt the strategy of multi-scale technique. Therefore, China Meteorological Administration has devoted to developing a new generation of global and regional multi-scale model since 2003. In order to validate the performance of the GRAPES (Global and Regional Assimilation and PrEdiction System) model both for its scientific design and program coding, a suite of idealized tests has been proposed and conducted, which includes the density flow test, three-dimensional mountain wave and the cross-polar flow test. The density flow experiment indicates that the dynamic core has the ability to simulate the fine scale nonlinear flow structures and its transient features. While the three-dimensional mountain wave test shows that the model can reproduce the horizontal and vertical propagation of internal gravity waves quite well. Cross-polar flow test demonstrates the rationality of both for the semi-Lagrangian departure point calculation and the discretization of the model near the poles. The real case forecasts reveal that the model has the ability to predict the large-scale weather regimes in summer such as the subtropical high, and to capture the major synoptic patterns in the mid and high latitudes.展开更多
文摘随着军事信息化、智能化进程不断深化发展,面对突发情况时通常需要在极短时间内完成事件处置方案拟制,因此对事件处置智能化的需求极为急迫。结合Platt scaling概率校准模型,针对非均衡数据分布和多类属输出概率校准拓展问题,对核最小二乘支持向量机模型进行改进,提出一种校准概率最小二乘支持向量机(Calibration Probabilistic Least Squares Support Vector Machine, CPLS-SVM)分类算法;设计一种基于CPLS-SVM算法的事件处置案例匹配方法,提高复杂案例样本环境下的匹配精度;通过构建基于校准概率最小二乘支持向量机,并应用到事件处置实例模拟数据进行仿真实验,验证算法的有效性和性能优势。
文摘The occurrence of social security events is uncertain, and the distribution characteristics are highly complex due to a variety of external factors, posing challenges to their rapid and effective handling. The scientific and reasonable requirement evaluation of the emergency force to deal with social security events is very urgent. Based on data analysis, this paper uses the neural network, operational research, modelling and simulation to predict and analyze social security events, studies the usage rule of emergency force and deployment algorithm, and conducts simulation experiments to evaluate and compare the different force deployment schemes for selection.
基金Key Technologies Research and Development Program (Grant No. 2001BA607B02)National Key Technology Research and Development Program (Grant No. 2006BAC02B03)National Natural Science Foundation of China (Grant No. 40575050)
文摘A new generation of numerical prediction system GRAPES (a short form of Global/Regional Assimilation and PrEdiction System) was set up in China Meteorological Administration (CMA). This paper focuses on the scientific design and preliminary results of the numerical prediction model in GRAPES, including basic idea and strategy of the general scientific design, multi-scale dynamic core, physical package configuration, architecture and parallelization of the codes. A series of numerical experiments using the real data with horizontal resolutions from 10 to 280 km and idealized experiments with very high resolution up to 100 m are conducted, giving encouraging results supporting the multi-scale application of GRAPES. The results of operational implementation of GRAPES model in some NWP centers are also presented with stress at evaluations of the capability to predict the main features of precipitation in China. Finally the issues to be dealt with for further development are discussed.
基金National Key Technology Research and Development (Grant Nos. 2006BAC02B01 and 2006BAC03B03)National High-Tech Research & Development Program of China (Grant No. 2006AA01A123)National Natural Science Foundation of China (Grant No. 40505023)
文摘During the past few years, most of the new developed numerical weather prediction models adopt the strategy of multi-scale technique. Therefore, China Meteorological Administration has devoted to developing a new generation of global and regional multi-scale model since 2003. In order to validate the performance of the GRAPES (Global and Regional Assimilation and PrEdiction System) model both for its scientific design and program coding, a suite of idealized tests has been proposed and conducted, which includes the density flow test, three-dimensional mountain wave and the cross-polar flow test. The density flow experiment indicates that the dynamic core has the ability to simulate the fine scale nonlinear flow structures and its transient features. While the three-dimensional mountain wave test shows that the model can reproduce the horizontal and vertical propagation of internal gravity waves quite well. Cross-polar flow test demonstrates the rationality of both for the semi-Lagrangian departure point calculation and the discretization of the model near the poles. The real case forecasts reveal that the model has the ability to predict the large-scale weather regimes in summer such as the subtropical high, and to capture the major synoptic patterns in the mid and high latitudes.
