【摘要】 针对带缺陷管道疲劳寿命的预测分析,提出了利用二阶粒子群优化的支持向量机的管道疲劳寿命预测方法。利用粒子群智能算法优化支持向量机的回归参数,可以避免按经验选取输入参数的盲目性,能有效地提高预测速度和精度。实例仿真结果表明用该方法对管道疲劳寿命预测具有很高的精确度,远优于BP神经网络的预测结果。更多还原

【Abstract】 With respect to the predictive analysis of pipe fatigue life with defects,this paper presents a prediction method of pipe fatigue life through use of the regress parameters of the support vector machine of the particle swarm intelligent algorithm so as to avoid the blind selection of input parameters according to the experience,a method which can effectively raise prediction speed and accuracy. Simulation results show that this pipe fatigue life prediction method has a high accuracy,much better than that of BP neural network.更多还原