Study on the Application of Density Peak Clustering in Damage Diagnosis of Tower Crane
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-out-put 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 ana-lyzed by density peak clustering method,and the damage judgment and damage location of tower crane are realized.This diag-nosis method based on density peak clustering can intelligently diagnose and determine the location of tower crane's minor dam-age.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 identifi-cation of tower crane.
万方数据
维普
