Analysis And Research On Energy Consumption Per Unit GDP Of Regional Economy

Energy is an important strategic material, and it becomes the primary material concerning the people’s livelihood and the social economic development because of.the scarcity and nonrenewable of energy. This thesis researches on the relationship between economic growth and energy consumption in order to discover the change rule of energy consumption per unit GDP. Then, we focus on constructing reasonable and effective prediction, warning methods and system of energy consumption per unit GDP. The methods and system are helpful to transform the development mode, promote the scientific development of regional economy and build a resource conserving and environment friendly society.This thesis studies the impact factors to energy consumption per unit GDP. The prediction and warning model of energy consumption per unit GDP in regional economy is established based on analytics statistics of a large number of relevant data.The main work of this thesis includes five aspects as follows:1) According to the reality, this thesis analyzes the relevant data which has been recorded and published by Bureau of Statistics. We mainly analyzes the electricity consumption data, GDP data, industrial structure, energy consumption structure and other data related to energy consumption per unit GDP. The electricity consumption data, historical energy consumption data and historical GDP statistics data are selected as the base of construction prediction model. And the data are preprocessed for modeling the energy consumption per unit GDP.2) Because of the sparse data, the support vector machine (SVM) algorithm is chosen as the modeling algorithm. In order to improve the accuracy of the prediction model, the particle swarm optimization (PSO) algorithm is utilized to optimize the parameters of SVM. So the algorithm proposed in the thesis can obtain prediction model with good accuracy for different problems. The experimental results show that the algorithm proposed in the thesis can follow the variation of actual situation with small prediction error which meets the actual demand.3) The thesis analyzes the prediction results of energy consumption per unit GDP and shows warning message based on the given range of the energy consumption per unit GDP. The warning of energy consumption per unit GDP is shown in different forms to provide data support for decision makers.4) We research the relevant factors that may affect energy consumption per unit GDP, and analyze the relationship between the industrial structure, energy consumption structure and energy consumption per unit GDP of Regional Economy qualitatively and quantitatively. Based on the analysis, a reasonable theoretical direction for improving the energy consumption efficiency and reducing energy consumption per unit GDP is given.5) Combined with the prediction and warning algorithm, the energy consumption per unit GDP prediction and warning system of Regional Economy is given in this thesis. Besides of data management, the system has good prediction ability of energy consumption per unit GDP. The system meets the accuracy requirement of energy consumption per unit GDP, provides energy consumption warning function with good human machine interaction.