【摘要】 道路交通安全是城市发展中的重要问题之一,可靠、准确的预测模型是道路交通安全的基础和难点。以万车事故率、万人事故率2个相对指标为目标层,引入免疫系统的抗体选择机制,构造基于免疫机制的免疫粒子群优化算法(ICPSO),以避免标准粒子群算法(PSO)在迭代过程中容易过早收敛而出现局部最优的缺陷。利用优化方法对最小二乘支持向量机预报模型(LSSVM)进行参数寻优,建立优化最小二乘支持向量机(ICPSO_LSSVM)预测模型,并对湖州市的万车事故率和万人事故率2个指标进行对比分析,以验证所建预测模型的有效性和可行性。更多还原

【Abstract】 Traffic safety is one of the key points in urban development,reliable and accurate prediction model is the foundation and key point for traffic safety. In this paper,the two relative indices of accident rate per 10,000 vehicles and the accident rate per 10,000 people are introduced as the target level,so to introduce antibody selection mechanisms for the immune system,and constructed an immune particle swarm optimization algorithm based on immune mechanism( ICPSO),To avoid local optimization because of premature convergence of the standard particle swarm algorithm( PSO) in Iterative process. The authors utilize optimization method to seek the best parameters for Least Squares Support Vector Model( LSSVM),so to set up optimized Least Squares Support Vector machine( ICPSO_ LSSVM) prediction model.Comparison analysis was carried out to accident rate per 10,000 vehicles and the accident rate per 10,000 people in Huzhou City,to verify effect and feasibility of the preset prediction model.更多还原