中文摘要：

研究影响水体污染物扩散的因素,使用计算机快速地对水体污染物进行模拟,能够为水体污染物的控制提供决策支持。将元胞自动机与并行计算相结合,能够高效地模拟水体污染物扩散。分析了元胞自动机的构成,综合考虑了污染物浓度、风向、风速、水流方向、水流速度、水体边缘等因素对污染物扩散的影响,设计了面向水体污染扩散的元胞自动机模型。分析了MPI（Message Passing Interface）在并行计算方面的优势,设计了面向水体污染物扩散的并行元胞自动机模拟算法流程,实现了水体污染物扩散的模拟。对比分析了面向水体污染物的元胞自动机串行计算、单机并行计算和多机并行计算的执行时间,结果表明,运用并行元胞自动机可以节省计算时间,提高计算效率。

英文摘要：

The traditional water pollutant spreading simulation requires a lot of time because of the large area of water bodies. This research aims to combine the cellular automata and parallel computing to achieve water pollutant spreading simulation. The components of CA are described in detail. The effect of pollutant concentration, wind direction, wind velocity, water flow direction, water flow velocity and the boundary of the water body have been taken into account. Overall considering above factors, water pollutant spreading simulation model is constructed.Parallel computing is realized by using this model through Message Passing Interface（ MPI）. The algorithm of parallel cellular automata to simulate the water pollutants spreading is described. Water pollutants spreading simulation model based on parallel cellular automata is established. The executing time of pollutant spreading in different computing scale including serial computing on single-core processer, parallel computing on multi-core processers of one computer, and parallel computing on multi-core processers of two computers is listed. The comparison result of executing time also be illustrated. The comparison result of serial algorithm time efficiency and parallel algorithm time efficiency show that parallel algorithm can improve the time efficiency of simulation results.