密度峰值聚类在塔机损伤诊断中的应用研究

Study on the Application of Density Peak Clustering in Damage Diagnosis of Tower Crane

建立塔机有限元模型,获取塔机完好状态和各损伤工况的各采集点的动态位移。提出了两种模型建立方法,基于悬臂梁的双输入单输出模型和基于时域数据的动态双输入单输出模型,对基于时域数据的双输入单输出模型首先利用最小二乘法计算参数初值,进一步利用粒子群优化方法进行参数优化,提高了模型精度。以完好工况的塔机数据为基础,建立基于悬臂梁的双输入单输出模型和基于时域数据的双输入单输出模型,计算参数,建立损伤识别模型,用待检状态的位移数值拟合模型,用两种模型计算出的残差方差做损伤因子,利用密度峰值聚类方法对损伤因子进行分析,实现了对塔机的损伤判定和损伤位置的确定。这种基于密度峰值聚类的诊断方法可对塔机微小损伤进行智能诊断和位置确定,该方法只需要塔机完好状态的数据和待检状态的数据即可自动诊断,解决了塔机损伤识别中损伤数据难以获取,因而无法实现智能训练和诊断的问题。

The finite element model of the tower crane is established,and the dynamic displacement of each collection point under intact condition and damage condition is obtained.Two kinds of model building methods are proposed.One is based on cantilever beam and the other is based on time-domain data.For the dual input and single output model based on time-domain data,the least squares method is used to calculate the initial parameters of the model,and the particle swarm optimization method is used to optimize the parameters to improve the accuracy of the model.Based on the intact tower crane data,the double-input-single-output model based on cantilever beam and the double input and single output model based on time domain data are established.The parameters are calculated and the damage identification model is established.The model is fitted with the displacement value of the inspected state.The residual variance calculated by the two models is used as the damage factor.The damage factor is analyzed by density peak clustering method,and the damage judgment and damage location of tower crane are realized.This diagnosis method based on density peak clustering can intelligently diagnose and determine the location of tower crane's minor damage.This method only needs the data of intact state and the data of inspection-waiting state for automatic diagnosis.It solves the problem that intelligent training and diagnosis cannot be realized because damage data is difficult to obtain in damage identification of tower crane.