Research On Spatial Y-shaped Steel Box Arch Rib Cable Hoisting Force Optimization Algorithm And Linear Construction Prediction Of Arch Rib

The spatial Y-shaped steel box arch rib is a new type of spatial three-dimensional arch bridge structure,the main and secondary arch ribs are synergistic force characteristics make the arch rib has higher strength and stability.In this paper,after analyzing the currently used methods for calculating the cable force,it is found that these methods cannot be applied to the calculation of the cable force of the arch rib of this special structure.At the same time,the construction of spatial Y-shaped steel box arch rib is carried out by cable lifting method,which requires high construction accuracy and requires arch rib alignment construction prediction.In this paper,we take the arch rib of a spatial Y-shaped steel box arch bridge as the engineering background,and start the research on the optimization algorithm of cable suspension method and the construction prediction of arch rib alignment.The main work and research results of the paper are as follows.(1)In this paper,we analyze the commonly used methods for calculating the cable suspension force,introduce the calculation principles and summarize the scope of application and shortcomings of each method.After comparing different calculation methods,this paper proposes the first multi-objective linear programming based on the influence matrix method as the optimization algorithm for the cable force of spatial Y-shaped steel box arch ribs.The optimization algorithm takes the minimum displacement of the arch rib as the objective function,the stress of the control section of the arch rib as the constraint condition,and solves the multiobjective linear programming model with the help of fuzzy mathematical theory.Therefore,the optimization algorithm proposed in this paper can simultaneously optimize the arch rib alignment and stress during cable suspension,and its optimization results meet the design and specification requirements.(2)This paper summarizes the feasibility and shortcomings of BP neural networks in construction prediction applications by investigating the basic principles and advantages and disadvantages of BP neural networks.The basic principle of NSGA2 genetic algorithm is introduced in detail,and it is argued that the algorithm can be used as an optimization method for BP neural network.In this paper,in order to optimize the weights and thresholds of BP neural network,we propose to use the training mean square error MSE and root mean square of weights of BP neural network as the objective function at the same time,and use NSGA2 genetic algorithm to optimize and calculate the optimal weights and thresholds.The NSGA2-BP neural network code is written with the help of MATLAB,and the code covers building the BP neural network,population iteration,fast non-dominated sorting,crowding degree calculation,and optimized BP neural network prediction.(3)In this paper,NSGA2-BP neural network is applied to the construction prediction of spatial Y-shaped steel box arch rib alignment for the first time.Firstly,the parameter sensitivity analysis of the arch rib is carried out based on the control variable method to determine the main design parameters affecting the construction alignment of the arch rib.Then the main design parameters were used as input values of NSGA2-BP neural network to predict the displacement of the buckling point control section,and the final overall prediction error of arch rib displacement did not exceed 1mm,and the maximum prediction deviation was 0.75 mm,and the prediction results met the requirements of construction prediction of spatial Y-shaped steel box arch rib.By comparing the prediction results of BP neural network and NSGA2-BP neural network for buckling point displacement,it is proved that the optimized BP neural network has high prediction accuracy and significant optimization effect.