This study was performed to compare storm surges/tide simulated by the regional and coastal storm surges/tide forecast system （RTSM （regional tide/storm surges model）, CTSM （coastal tide/storm surges model）） usi...This study was performed to compare storm surges/tide simulated by the regional and coastal storm surges/tide forecast system （RTSM （regional tide/storm surges model）, CTSM （coastal tide/storm surges model）） using two different inputs from weather models （RDAPS （Regional Data Assimilation and Prediction System） and KWRF （Korea Weather and Research Forecasting）） during two typhoons that occurred between 2007 and 2008. Both the RDAPS and KWRF are the operational weather forecasting system in KMA （Korea Meteorological Administration）. The horizontal resolutions of RDAPS and KWRF are 30 and 10 km, respectively. The storm surges/tide was hind casted using sea wind and pressure fields of two Typhoons which was approaching Korean Peninsula. The CTSM using input from KWRF simulate very well the storm surges/tide pattern in the complex coastal areas. The result showed that the storm surges by the coastal storm surges/tide model with high resolution input was in well agreement with the observed sea level occurred by high tide and storm surges in the coastal areas.展开更多
针对知识化制造环境下的自适应调度问题,提出基于状态-动作不确定性偏向Q学习(state-action uncertainty bias based Q-learning,简称SAUBQ学习)的知识化制造自适应调度策略.该策略针对传统Q学习收敛速度慢,训练时间长等问题,引入信息...针对知识化制造环境下的自适应调度问题,提出基于状态-动作不确定性偏向Q学习(state-action uncertainty bias based Q-learning,简称SAUBQ学习)的知识化制造自适应调度策略.该策略针对传统Q学习收敛速度慢,训练时间长等问题,引入信息熵的概念定义了状态不确定性测度,据此定义了Q学习动作偏向信息函数,通过对Q学习奖励函数采用启发式回报函数设计,将动作偏向信息利用附加回报的方式融入学习系统,并证明了算法的收敛性和最优策略不变性.在学习过程中,Q学习根据偏向信息调整搜索空间,减少了Q学习必须探索的有效状态-动作对数目,同时偏向信息根据Q学习结果不断进行调整,避免了不正确的误导.经仿真实验比较,结果表明,该策略具有对动态环境的适应性和大状态空间下收敛的快速性,提高了调度效率.展开更多
文摘This study was performed to compare storm surges/tide simulated by the regional and coastal storm surges/tide forecast system （RTSM （regional tide/storm surges model）, CTSM （coastal tide/storm surges model）） using two different inputs from weather models （RDAPS （Regional Data Assimilation and Prediction System） and KWRF （Korea Weather and Research Forecasting）） during two typhoons that occurred between 2007 and 2008. Both the RDAPS and KWRF are the operational weather forecasting system in KMA （Korea Meteorological Administration）. The horizontal resolutions of RDAPS and KWRF are 30 and 10 km, respectively. The storm surges/tide was hind casted using sea wind and pressure fields of two Typhoons which was approaching Korean Peninsula. The CTSM using input from KWRF simulate very well the storm surges/tide pattern in the complex coastal areas. The result showed that the storm surges by the coastal storm surges/tide model with high resolution input was in well agreement with the observed sea level occurred by high tide and storm surges in the coastal areas.
文摘针对知识化制造环境下的自适应调度问题,提出基于状态-动作不确定性偏向Q学习(state-action uncertainty bias based Q-learning,简称SAUBQ学习)的知识化制造自适应调度策略.该策略针对传统Q学习收敛速度慢,训练时间长等问题,引入信息熵的概念定义了状态不确定性测度,据此定义了Q学习动作偏向信息函数,通过对Q学习奖励函数采用启发式回报函数设计,将动作偏向信息利用附加回报的方式融入学习系统,并证明了算法的收敛性和最优策略不变性.在学习过程中,Q学习根据偏向信息调整搜索空间,减少了Q学习必须探索的有效状态-动作对数目,同时偏向信息根据Q学习结果不断进行调整,避免了不正确的误导.经仿真实验比较,结果表明,该策略具有对动态环境的适应性和大状态空间下收敛的快速性,提高了调度效率.
