Research On Cnstruction Cost Prediction Of Building Based On Denoising AutoEncoder

Construction engineering is an important part of China’s construction real estate industry.It is related to the development of China’s national economy and reflects the development level of the national economy.The construction industry accounts for a large share of China’s national economy.In recent years,the construction industry’s share in the national economy has continued to increase.According to the National Bureau of Statistics,with the structural reform of the supply side,the investment in real estate development in 2017 was 1,097.9 billion yuan,an increase of 7.0% over the previous year.Among them,residential investment was 751.48 billion yuan,an increase of 9.4%.Construction engineering is the core part of the construction industry.Therefore,the construction cost prediction research of construction engineering is of great significance.For construction companies,cost management is an important part of the company’s core competitiveness.In the case of stable income and expenditure,the lower the cost means the higher the profit of the enterprise.The company has more funds invested in project bidding,technology research and development,and technological innovation,which makes the brand of the company have higher value.To enhance the market competitiveness of the brand.From the perspective of the construction project itself,the construction cost of construction directly determines the profitability of a project.Construction project cost prediction is an important part in construction project management.It can be used as an important basis for project construction organization design,economic benefit comparison of construction plan,bulk material procurement,etc.,and its accuracy directly affects the project’s profitability.The prediction theory and mathematical model of construction project construction cost include linear regression,time series method and grey theory.In recent years,there are many researches on engineering cost prediction based on artificial neural network methods at home and abroad,and it is also the mainstream forecasting method in science and engineering field.However,these research methods all have the problem that the theoretical combination is not matched or the influencing factor index system is not perfect.The engineering cost prediction is only studied from a static perspective,and the prediction accuracy is not high.The deep learning theory based on the combination of artificial neural network and big data has become a new field in the research of machine learning.At present,there is no scholar at home and abroad to apply deep learning to the field of construction cost prediction.In the systematic analysis of traditional construction engineering cost prediction methods,this paper selects the powerful learning prediction principle based on the stack noise reduction automatic encoder under the deep learning network,which can accurately and quickly predict the construction engineering cost.At the same time,this study constructs a more comprehensive quantitative system of construction engineering cost influencing factors for high-rise building projects,and makes up for the shortcomings of current construction engineering cost prediction.Based on the deep learning network,the construction cost prediction of the construction project,considering the influencing factors more in line with the attributes of the project itself,with high accuracy and more stable prediction results,therefore,has high research value.