Building Carbon Emission Scenario Prediction Using STIRPAT and GA-BP Neural Network Model

Sensen Zhang; Zhenggang Huo; Chencheng Zhai

Building Carbon Emission Scenario Prediction Using STIRPAT and GA-BP Neural Network Model

Sensen Zhang; Zhenggang Huo; Chencheng Zhai

As a major province of energy consumption and carbon emission, Jiangsu Province is also a major province of the construction industry, which is a key region and potential area for carbon emission reduction in China. The research and prediction of carbon emission in the construction industry is of great significance for the development of low-carbon policies in the construction industry of other cities. The purpose of this paper is to study the influencing factors of the whole life cycle carbon emissions of buildings in Jiangsu Province, and to predict the carbon emissions of buildings in Jiangsu Province based on the main influencing factors. This paper uses the energy balance sheet splitting method, STIRPAT model, gray correlation method and GA-BP neural network model to study and predict the carbon emissions of construction industry in Jiangsu Province. The research results show that the resident population, urbanization rate, steel production, average distance of road transportation, and labor productivity of construction enterprises have a catalytic effect on construction carbon emissions: GDP per capita and added value of tertiary industry have a suppressive effect: construction carbon emissions reached the historical peak in 2012: the prediction results show that the future construction carbon emissions in Jiangsu province generally show a decreasing trend. The research results of this paper provide a possibility to refine the study of construction carbon emission, and also provide a basis and guidance for subsequent research on construction carbon emission.

Afficher plus [+]