Fatigue Life Assessment Of Tower Crane Based On Neural Network To Obtain Stress Spectrum

With the rapid development of our country’s economy and construction industries,tower cranes,as an indispensable construction hoisting equipment in the field of construction,have greatly improved construction efficiency and saved labor costs.Since the tower crane works outdoors,the working environment is complicated and the conditions are harsh,and safety accidents are extremely prone to occur.Once an accident occurs,it will cause irreversible losses and threaten the life and safety of on-site construction personnel.Therefore,it is of great significance to study the fatigue remaining life of the tower crane in the actual working environment.Tower cranes,due to their complex working environment and randomness of the lifting load,greatly increase the difficulty of obtaining the fatigue load spectrum of the tower crane.At this stage,there are two main ways to obtain the fatigue load spectrum.One is field collection.The data obtained by this method is accurate,but there are problems such as long cycle,high cost and difficulty.The other is through computer simulation.This method has a short cycle and high efficiency,but the accuracy of the data obtained needs to be improved.Based on the above reasons,this article takes an in-service tower crane as the research object,and proposes a method to quickly obtain the tower crane’s stress spectrum and calculate the fatigue remaining life of the tower crane.The main research contents are as follows:(1)The commonly used fatigue analysis methods are introduced.Combined with the actual operating conditions of tower cranes,most of them are subjected to alternating loads,and fatigue cracks are easily formed under the action of alternating loads.Therefore,the fracture mechanics method is selected to calculate the remaining fatigue life.(2)In order to obtain the location of the dangerous point of the tower crane,a static analysis was carried out.Through the actual parameters of the tower crane,the components with less force in the structure were simplified,and the ANSYS finite element analysis model of the tower crane was established.Combining the lifting characteristic curve of the tower crane and the cause of the accident,five typical operating conditions are determined.The statics analysis was carried out on 5 working conditions,and the position of the dangerous point was determined.(3)In order to simulate the actual operating conditions of the tower crane and obtain the stress time history data at the dangerous point of the tower crane,the transient dynamics analysis was carried out.Five typical operating conditions of the tower crane in actual operation are simulated,and the equivalent stress values of dangerous points under these operating conditions are obtained by calculation.(4)Because the finite element analysis needs to be recalculated when the working conditions are changed,it takes a long time and a large workload.Therefore,a method to quickly obtain the stress time history of dangerous points under various working conditions is proposed through the neural network model.A BP neural network and a radial basis function neural network were established respectively,and the number of iterations and convergence of the two were compared.(5)Train the neural network model to realize the input of the weight of the hoisting load and the position of the luffing trolley into the neural network model,and output the equivalent stress value of the dangerous point.The prediction results of the neural network and the finite element analysis results are compared,and the feasibility of the neural network model is verified,and the calculation speed of the neural network is better than that of the finite element calculation.(6)The operating data of the tower crane is actually recorded for a period of time,and the recorded operating data is input into the neural network model as a prediction sample.The stress time history of the dangerous points is counted by rain flow counting method,and the two-dimensional stress spectrum of the dangerous points is obtained.Finally,the residual fatigue life of each dangerous point is calculated by the fracture mechanics method.In this paper,the fatigue remaining life of the tower crane is estimated,and the analysis results provide a reliable basis for the long-term safe use and later maintenance of the crane,save engineering costs,reduce the number of safety accidents to a certain extent,and protect the life safety of on-site workers,Provide a guarantee for safe construction operations.