元胞自动机和克隆选择算法的蓝藻水华预测模型

A cyanobacterial blooms prediction model based on celluar automa and clonal selection algorithm

针对蓝藻水华频发影响水体质量和生态环境的问题,该文以太湖为例,探讨了元胞自动机和克隆选择算法在蓝藻水华预测方面的可行性。使用遥感经验算法计算水体叶绿素a浓度;将叶绿素a浓度分布数据作为预测模型的输入数据,将基于平衡方程的预报模型与元胞自动机进行耦合,对蓝藻水华时空变化过程进行了动态模拟;在预测方面实现了蓝藻水华空间分布情况的可视化,并采用克隆选择算法优化了预测模型中的重要参数。研究结果表明:利用克隆选择算法优化的元胞自动机模型能够在蓝藻水华预测中直观地显示蓝藻水华的时空变化特征;能够在短时间内重复性地预演蓝藻水华的变化过程,为决策者提供有用的参考。

In this paper, a cyanobacterial blooms prediction model based on cellular automata （CA） and clonal selection algorithm was introduced using Taihu Lake as the research object. Remote sensing experience algorithm was used to calculate chlorophyll a concentration distribution, and the chlorophyll a concentration distribution data was imported into a predictive model as the input data, based on balanced equation forecas- ting model with CA coupling of cyanobacteria Hua spatial and temporal changes in the dynamic simulation process. The visualization of forecasted spatial distribution of cyanobacteria blooms was achieved, and clonal selection algorithm was used to optimize the predictive model. The results showed that the cellular automata model optimized by clonal selection algorithm could achieve the visualization on the intuitive prediction of spatial and temporal variation of cyanobacterial blooms, and could be used repeatedly to preview the cyanobacterial blooms change process.