摘要
根据橡胶充气芯模混凝土填料施工的特点,详细分析了影响其工期的因素,结合支持向量回归机(SVR)和粒子群算法(PSO)的优点,建立基于PSO-SVR的橡胶充气芯模混凝土填料工程工期预测模型。通过对某选煤厂混凝土填料施工的数据进行仿真,结果显示:PSO-SVR模型的预测效果优于基于交叉验证的支持向量回归机(CV-SVR)模型以及基于遗传算法的支持向量回归机(GA-SVR)模型。
According to the construction characteristics of concrete filling engineering using inflatable rubber mandrel, this paper analyzes the factors affecting project duration and it establishes a duration forecasting model based on PSO-SVR, which integrates the advantages of support vector regression and particle swarm optimization algorithm. Data of the concrete filling construction of a coal preparation plant is simulated. The results show that forecast effect on the PSO-SVR model is better than that of CV-SVR model and GA- SVR model.
出处
《价值工程》
2016年第4期70-72,共3页
Value Engineering
关键词
橡胶充气芯模
粒子群算法
支持向量回归机
工期预测
inflatable rubber mandrel
Particle swarm optimization algorithm
support vector regression
duration forecast
