联合BP神经网络和粒子群的结构有限元模型修正
|
摘要
为解决有限元模型修正中寻优速度慢和算法易于陷入局部最优的问题,提出了采用BP神经网络作为代理模型,采用粒子群算法作为优化算法的模型修正方法。为验证算法的有效性和高效性,采用所提方法对一个钢桁架进行了模型修正,修正结果表明,所提算法能够准确找到结构损伤位置,并且对杆件的损伤程度的识别精度较高,并且与传统的直接调用有限元模型计算结构响应的方法对比,所提方法在寻优速度上具有明显提升。
In order to solve the problem that the optimization speed was slow and the algorithm is easy to fall into the local optimum in the finite element model modification,BP neural network was used as the proxy model,and the particle swarm optimization algorithm was used as the model correction method of the optimization algorithm.In order to verify the validity and efficiency of the algorithm,a steel truss model was modified by the proposed method.The correction results show that the proposed algorithm can accurately find the structural damage location and the recognition accuracy of the damage degree of the member is high.And compared with the traditional direct call finite element model to calculate the structural response,the proposed method has a significant improvement in the speed of optimization.
In order to solve the problem that the optimization speed was slow and the algorithm is easy to fall into the local optimum in the finite element model modification,BP neural network was used as the proxy model,and the particle swarm optimization algorithm was used as the model correction method of the optimization algorithm.In order to verify the validity and efficiency of the algorithm,a steel truss model was modified by the proposed method.The correction results show that the proposed algorithm can accurately find the structural damage location and the recognition accuracy of the damage degree of the member is high.And compared with the traditional direct call finite element model to calculate the structural response,the proposed method has a significant improvement in the speed of optimization.
引文
[1] 秦世强,蒲黔辉,施洲.高速铁路桥梁健康监测数据采集系统设计研究[J].铁道标准设计,2010(10):67-71.
[2] 王戒躁,钟继卫,王波.大跨桥梁健康监测系统设计构成及其进展[J].桥梁建设,2009(S2):7-12.
[3] 任伟新,彭雪林.青洲斜拉桥的基准动力有限元模型[J].计算力学学报,2007(5):609-614.
[4] 张刚刚.基于RBF神经网络的桥梁损伤识别方法研究[D].西安:长安大学,2004.
[5] 鲍诺,王春洁.基于响应面优化的结构有限元模型修正[J].北京航空航天大学学报,2014(7):927-933.
[6] 李阳.基于代理模型的桥梁随机有限元模型修正算法研究[D].哈尔滨:哈尔滨工业大学,2014.
[7] 鲍诺,王春洁.基于响应面优化的结构有限元模型修正[J].北京航空航天大学学报,2014,40(7):927-933.
[8] 胡俊亮,颜全胜,郑恒斌,等.基于Kriging模型的钢管混凝土连续梁拱桥有限元模型修正[J].振动与冲击,2014(14):33-39.
[9] 窦毅芳,刘飞,张为华.响应面建模方法的比较分析[J].工程设计学报,2007(05):359-363.
[10] 雷开友,邱玉辉,贺一.一种优化高维复杂函数的PSO算法[J].计算机科学,2006,33(8):202-205.
[11] 林玉锋,陈璟,陈少飞,等.一种新颖的混合响应面优化方法[J].计算机应用研究,2012,29(6):2180-2183.
[12] 何宣虎.基于人工蜂群算法的最优潮流相关技术研究[D].北京:北京交通大学,2016.
[2] 王戒躁,钟继卫,王波.大跨桥梁健康监测系统设计构成及其进展[J].桥梁建设,2009(S2):7-12.
[3] 任伟新,彭雪林.青洲斜拉桥的基准动力有限元模型[J].计算力学学报,2007(5):609-614.
[4] 张刚刚.基于RBF神经网络的桥梁损伤识别方法研究[D].西安:长安大学,2004.
[5] 鲍诺,王春洁.基于响应面优化的结构有限元模型修正[J].北京航空航天大学学报,2014(7):927-933.
[6] 李阳.基于代理模型的桥梁随机有限元模型修正算法研究[D].哈尔滨:哈尔滨工业大学,2014.
[7] 鲍诺,王春洁.基于响应面优化的结构有限元模型修正[J].北京航空航天大学学报,2014,40(7):927-933.
[8] 胡俊亮,颜全胜,郑恒斌,等.基于Kriging模型的钢管混凝土连续梁拱桥有限元模型修正[J].振动与冲击,2014(14):33-39.
[9] 窦毅芳,刘飞,张为华.响应面建模方法的比较分析[J].工程设计学报,2007(05):359-363.
[10] 雷开友,邱玉辉,贺一.一种优化高维复杂函数的PSO算法[J].计算机科学,2006,33(8):202-205.
[11] 林玉锋,陈璟,陈少飞,等.一种新颖的混合响应面优化方法[J].计算机应用研究,2012,29(6):2180-2183.
[12] 何宣虎.基于人工蜂群算法的最优潮流相关技术研究[D].北京:北京交通大学,2016.