In this paper we address the problem of Software-MCU assignment in car electronic system design. This is a difficult optimization problem due to complex characteristics of the car electronic system; further, designer preferences make the problem more difficult. In order to solve this combinatorial optimization problem we propose a method which is based on the Lagrangian method. Experimental results show that the search can effectively find the feasible solutions. We also introduce short and long term memories to the Lagrangian method. There are several ways of integrating the short and long term memories. We examine these ways of integration in the experiments, and decide the most effective one. When we use this effective integration method, we can improve the efficiency of the Lagrangian method further.

抄録全体を表示

The 1D multilayered physical snowpack model Snow Metamorphism and Albedo Process (SMAP), which was originally designed for climate studies, is now updated by incorporating a detailed water movement scheme, realistic snow settlement process and limitation for the Richardson number to ensure minimum turbulent exchanges even under highly stable atmospheric conditions.　The evaluation of the updated version of SMAP was first performed using the data obtained at Sapporo, Japan, during the 2007-2009 winters and the effectiveness of these updates was demonstrated in terms of snow depth and snow surface temperature.　However, we pointed out that the choice of maximum Richardson number should be further examined.　To test the reliability of SMAP under different climate conditions, we applied it to Naga oka, Japan, during the 2011-2012 winter.　At Nagaoka, we performed snow-soil-coupled simulations, because ground heat flux was not available during the study period.　For this purpose, we developed a soil submodel for SMAP. Consequently, we confirmed that the updated version performed better than the old version in terms of mass balance simulations at Nagaoka too.　Although mass balance-related parameters of the snowpack simulated by the updated version agreed well with observations during the accumulation period, the model substantially overestimated snow depth, as well as column-integrated snow water equivalent during the ablation period.　By discussing the reasons for these discrepancies, we highlighted that further investigation on snow-melt processes for thick seasonal snowpack is necessary.

抄録全体を表示