Research On Post-evaluation Of Urban Road Infrastructure Construction Process Based On Neural Network

The urban road is one of the main components of the municipal infrastructure construction project in the city,which has obvious influence on the development of the city and People's Daily life.In recent years,China's urban economy has been developing rapidly,while the scale and speed of urban road construction have been accelerating.In this way,certain problems of capital management will arise in the construction of urban road infrastructure projects,resulting in poor project construction effect and poor construction quality.Especially in the process of road infrastructure construction in some large cities,it is difficult to achieve effective management of project funds due to the long construction cycle,large investment and large number of participants.Once capital waste or improper use of capital occurs in the construction process,it will bring huge economic losses to the country and enterprises,and may even affect the normal use of the project after completion.Therefore,in the process of urban road infrastructure project construction,it is necessary to timely evaluate the project construction and management effect.In this paper,based on the analytic hierarchy process(AHP)in the study and the neural network with the combination of urban road infrastructure project construction process after the evaluation system,through the research of this paper can determine the factors influencing the project funds management and the construction effect,and then these factors to make use of the role of the neural network training,so as to determine the validity of this study.Firstly,the artificial neural network,BP neural network and RBF neural network are analyzed and compared,and the RBF neural network model which is more suitable for the research object of this paper is selected.The analytic hierarchy process(AHP)is used to determine the status of each indicator in the construction project system.The decomposition of the construction project management process and characteristic factor analysis are conducted to determine the type of characteristic factor and select the characteristic vector.The weight value of each factor is obtained and the influence of these characteristic factors on the road infrastructure construction is analyzed.After processing the feature vectors,the RBF neural network model is built based on the theoretical analysis of the neural network,and the neural network model is trained and tested.In Urumqi city high-speed(eastern)as an example verifies the method,the model application research,through the utility of the model were analyzed,and put forward existing problems,such as construction period changes frequently and the shortage of construction cost control in the process of quality problems,finally pointed proposes the solution countermeasure,in order to improve project management efficiency.Through in this paper,the study found that the effect of using neural network to evaluate the effect of urban road infrastructure projects,and the neural network can be predicted according to the construction of data error,thus improve the rationality of urban road infrastructure projects,the project funds management efficiency,and finally a good technology,capital management,a reasonable balance of the relationship between quality,cost,schedule,construction of the desired goal,provide a good theoretical guidance for the Urumqi city road construction.