Research On Method Of Quickly Obtaining Crane Structure Stress Spectrum Based On Dynamic Incremental Load Samples

With the continuous expansion of China’s industrial production and the increase in the degree of automation,cranes,as important equipment for material handling,are becoming more and more widely used in modern production processes,and their requirements are becoming higher and higher.It can achieve vertical lifting and horizontal transportation of heavy objects within a certain range,so it is widely used in railway freight yards,mines,nuclear power stations,ports,construction sites and other places.The use of cranes can reduce physical labor,greatly improve production efficiency,and promote mechanization and automation in industrial production processes.Cranes work under harsh environments and complex loads for a long time,and safety accidents will occur during the work,causing loss of property and personnel.Therefore,the prediction of crane fatigue life is an important part of structural safety assessment.When predicting the fatigue remaining life,the random load spectrum that the structure bears throughout the life cycle is the basic key data for calculating the fatigue life.At present,there are two methods to obtain the random load spectrum: field measurement and simulation analysis.Although the former is accurate,the cost is high,the cycle is long,and it is even difficult to achieve in some working environments;the latter is fast in obtaining load spectrum and low in cost,but its calculation results have low reliability.With the continuous upgrading of Internet technology,many factories are now setting up real-time monitoring systems for crane operation data,which can more easily and quickly obtain various parameters during crane operation.If these data can be effectively combined with computer simulation to obtain The load spectrum can have the advantages of both accuracy and fastness.It is used to calculate the remaining fatigue life and safety assessment closer to the reality.Therefore,this paper explores a method for quickly obtaining the crane load spectrum using the dynamic incremental data of the real-time monitoring system.First,the data monitored by the crane in real time is processed to form a dynamic incremental real-time data sample,which provides data support for thenext research.Then use ANSYS finite element software to model the bridge crane,modify and improve the finite element model through experimental test data,use the finite element model to simulate the bridge crane with different lifting weights and trolley positions to obtain various points of the crane To construct a 3D-Map stress database.Then,a radial basis neural network model is established,and the radial value neural network is used to predict the stress value of the obtained dynamic incremental real-time data samples to form a time-stress curve.When the radial basis neural network model is used for prediction,the input layer is the lifting load of the crane and the trolley position,and the output layer is the equivalent stress value of any desired position.Finally,two-dimensional stress spectrum data was acquired for the stress-time curve by the rain flow counting method.Then use the damage tolerance-based fracture mechanics method for life analysis and remaining life assessment.In this paper,the equivalent stress value is calculated by two methods of finite element software and radial basis neural network,and the data of the two are compared.The error is within the allowable range,which verifies the feasibility of the neural network calculation,and its operation speed is much faster than Finite element simulation,so we use neural network to calculate equivalent stress value,and obtain stress spectrum and fatigue life assessment,this method is feasible and effective.