Research On Maintenance Decision Model Of Rural Cement Pavement In Henan Province Based On Particle Swarm Optimization Algorithm

In recent years,the Henan Provincial Government has provided active support and encouragement to the construction of rural roads,continuously increasing investment and accelerating the pace of rural road construction.As of the end of 2021,the length of rural roads in Henan Province reached 233000 kilometers,accounting for 86% of the total length of roads in Henan Province.Rural roads occupy an important position in the highway network of Henan Province,and the maintenance pressure brought by the increasing mileage of rural roads has become an important problem faced by the highway management and maintenance department.In response to the insufficient understanding of the importance of preventive maintenance of road surfaces,the lack of a systematic evaluation system for road maintenance decision-making,and the lack of green maintenance concepts in Henan Province during the road maintenance process,this paper constructs a road maintenance fund allocation model based on the optimization particle swarm theory to solve the problems in maintenance decision-making and obtain better decision-making results.The main work and conclusions include:(1)A survey was conducted on the overall situation of cement pavement on rural roads in Henan Province,and the proportion of various levels of roads in rural roads in Henan Province was analyzed.From an administrative level perspective,the main component of rural roads in Henan Province is village roads,accounting for 62.66%;In terms of technical level,fourth-class highway is the main road,accounting for 81.02%.It indicates that rural roads in Henan Province are at a low level in terms of grade,so maintenance pressure is high,and scientific maintenance of rural roads is imperative.(2)Compared with the equal length segmentation method and the variable length segmentation method,the clustering method has superior similarity partitioning performance,and is usually applied in deep learning research in the image field,urbanization level partitioning,vector machine intrusion detection technology,and other fields.This article applies this method to road segmentation research.Taking the service performance of cement pavement of rural roads in Henan Province as the research object,combined with the quantitative characteristics of pavement diseases,PCI(pavement damage condition index),RQI(pavement driving quality index),crack area,broken plate area and other indicators were selected,and the European distance and average distance methods were used to standardize the data,and then the hierarchical clustering method was used to divide the road sections for similarity.The results obtained three clustering clusters and divided the road sections into three categories,achieving good segmentation results.(3)In view of the defects of elementary particle swarm optimization algorithm,such as easy to fall into local optimization,slow convergence speed in the later period,and the characteristics of inertia weight w that have a great impact on the algorithm,this paper uses the method of adaptive adjustment of w to optimize the algorithm,so that the weight w can automatically adjust with the change of particle fitness value,solving the above defects of elementary particle swarm optimization algorithm.(4)An optimized maintenance fund allocation model was constructed using the optimized particle swarm optimization algorithm,and based on empirical analysis of maintenance decisions for rural cement roads in Henan Province.Using the maintenance benefit value and the repaired PCI value as the objective functions of the algorithm,with the minimum number of maintenance sections and maintenance funds as constraints.On the basis of analyzing the objective function and constraint conditions,a maintenance decision model is constructed to solve the Pareto optimal solution and obtain an optimized fund allocation plan.At the same time,a comprehensive analysis was conducted on all schemes to obtain pavement fund allocation schemes under different needs,confirming the practicality and feasibility of the model,and providing a theoretical basis for the maintenance department to choose appropriate maintenance schemes based on actual maintenance situations.