基于PSO优化SVM的船舶流量预测算法

Prediction algorithm for the ship traffic based on PSO-SVM

通过对上海洋山深水港口的船舶流量的调研以及对船舶交通流量影响因素的分析,建立支持向量机预测模型。同时为了解决支持向量机预测模型的参数选择问题,引入了粒子群优化算法进行参数优化,建立较优的PSO-SVM预测模型。通过MATLAB仿真实验计算,将PSO-SVM模型与单纯的SVM预测模型和灰色神经网络预测模型结果进行对比分析,证明了该模型的可行性和优越性。

By researching ship traffic through the port of Shanghai Yangshan Deepwater as well as the factors under the conditions of ship traffic analysis, a support machine model is built. Meanwhile, in order to solve the problem of parameter selection of support vector machine model, the introduction of the particle swarm optimization mechanism, get on better PSO-SVM forecasting model. By MATLAB simulation calculation, the PSO-SVM model and simple SVM prediction model and gray neural network forecasting model for comparative analysis to prove the feasibility and superiority of the model.