Research On Identification Method Of Bolt Looseness And Damage Of Main Material Of Transmission Tower

The development of digital and intelligent technologies in power systems has put forward higher requirements for the identification of the health status of transmission towers.The main material of the transmission tower is its main bearing component.Bolt looseness is a common damage form of the main material of transmission tower.The looseness and damage of bolts will lead to the decline of the bearing capacity of transmission tower and even the collapse of transmission tower,which seriously endangers the operation safety of power system.Therefore,it is very necessary to study the identification method of bolt looseness and damage of main material of transmission tower.This paper takes 220 k V catou tower as the research object,and the main research contents are as follows:(1)Aiming at the problem that the bolt connection slip in the transmission tower has a great influence on the axial stiffness of the rod,and the elastic modulus of the rod is quite different from the mechanical properties of the actual tower structure,the whole tower rod is divided into seven categories.Based on the static test data,a BP neural network is proposed as the optimization algorithm,and the residual between the simulation value and the test measurement value is used as the objective function,and the finite element model of the rod is revised,so that the revised transmission tower finite element model is more efficient.Reflect the actual tower structural characteristics.(2)On the basis of the finite element correction model of the transmission tower,by reducing the axial stiffness of the main material rod end element to simplify the simulation of the bolt loosening damage of the main material of the tower,the relational expression of the damage reduction factor is given,and the main material of the tower is It is divided into different sub-structures according to the geometrical positions,and the fourth-order and second-order natural frequency change ratios before and after the damage are used as the input parameters of the neural network.Loosening damage location and identification method of main material bolts of transmission towers in the network.The results show that the probabilistic neural network has a good identification effect on the damage location of the main material members of the substructure model I.The average accuracy rate of the single damage case is over 80%,and the average accuracy rate of the double damage case is over 83%.(3)On the basis of damage localization and identification,a BP neural network damage quantitative identification method optimized by genetic algorithm(GA)is proposed.The natural frequency change rate is used as the quantitative identification index of damage,which is used as the input parameter,and the damage degree is used as the output parameter.The numerical simulation results show that the average recognition accuracy rate of this method can reach more than 90%,which shows that the GA-BP neural network algorithm can effectively identify the looseness and damage of rod bolts.