Construction Cost Estimation Research Based On Gray BP Neural Network

With the rapid development of the construction market in our country,the social investment in fixed assets increases gradually,which is accompanied by the serious loss of control of market investment.Therefore,rapid and accurate cost estimation engineering is the focus of attention of many scholars and practitioners.This paper mainly study on the cost estimation system of construction engineering,and seeks an effective and feasible method for the cost estimation,to improve the running efficiency of the whole system.Firstly,this article gives an overview of the theory of project cost,provides a basis for the basic theoretical system of construction project cost estimation,and describes in detail the role of project cost and the main factors affecting the project cost.Secondly,the basic principles of BP neural network and grey theory are described,which provides a technical foundation for the modeling of construction cost estimation system.Finally,an optimization model based on grey BP neural network method is established.In order to solve construction project cost estimation,a more accurate knowledge framework system is established.The main function of BP neural network is to converge slowly and easily fall into local minimum.The disadvantages of the value and the prediction are not accurate.It is proposed that the data of the input layer of the BP neural network is processed once in grey,and then the output layer is processed by using grey one-time accumulation,thereby improving the BP neural network method.In order to verify the validity and feasibility of the model,in the early stage of the project,according to the 40 samples of the collected residential projects,30 were selected as training samples,and the remaining 10 were test samples,through the quantification of engineering eigenvectors and the gray-scale accumulation and normalization of training samples and test samples,MATLAB software was used to establish a dynamic and rapid estimation of the construction cost model.The test results showed that the grey BP neural network model's prediction results meet the requirements.And then,through specific engineering examples,the GM(1,1)model,the BP neural network method and the gray BP neural network method in the grey system theory are used for prediction respectively,and the prediction results are analyzed and compared to obtain the BP with the optimized grey theory.Neural network model is feasible in the construction industry.This paper proposes to establish a construction cost estimation model based on gray BP neural network.This model provides technical support for cost estimation of construction projects that can effectively reduce engineering costs and increases work efficiency,and estimates the development of construction engineering cost in the direction of automation and intelligence has a certain practical significance.