摘要
为克服传统的海洋赤潮监测算法滞后性问题,提出了一种改进粒子群SVM的海洋赤潮监测算法。首先,给出基于改进粒子群SVM的海洋赤潮监测算法的基本原理,采用改进粒子群算法来对SVM的参数进行优化。优化后的SVM模型作为初始模型,采用有标签的样本数据对SVM进行训练,得到训练好的SVM模型。测试当前数据时,将其输入到训练好的SVM模型,通过投票方法统计得到该数据对应的预测结果。仿真实验结果表明,预测结果较为精确,与其他方法相比,具有监测精度高和时间开销小等优点。
In order to solve the delaying problem of the traditional red tide monitoring algorithms, an im- proved particle swarm algorithm combing SVM for ocean red tide monitoring is proposed. First, the ele- mentary principles of the algorithm is given. Then, the improved particle swarm algorithm is used to opti- mize the parameters of SVM, the optimized SVM model is used as the initial model, then the labeled the sample data is used to train SVM, so a group of SVM model is obtained. When the current data is needed to be tested, it can be input into the trained SVM model, and the prediction result can be computed by vot- ing. Finally, in order to verify the method proposed, the simulation is operated. The result shows that the method can predict accurately, and compared with the other methods. It has the higher monitoring accuracy and short time cost.
出处
《青岛大学学报（自然科学版）》
CAS
2016年第4期30-33,共4页
Journal of Qingdao University(Natural Science Edition)
基金
三亚市院地科技合作项目(批准号:2014YD11)资助
关键词
海洋赤潮
粒子群算法
支持向量机
监测
ocean red tide
particle algorithm
support vector machine
monitor
