Abstract：Expanded Logarithmic Mean Decomposition Index (LMDI) method through introducing variables of urbanization and residential consumption. Decomposed regional energy consumption and residential energy consumption into six effects as energy structure, energy intensity, population scale, urbanization, residential consumption, and consumption inhibit. And then, this paper analyzedits influence on the overall regional energy consumption and residential energy consumption. Results show that: From 2003 to 2012, impact of urbanization on regional energy consumption and residential energy consumption of Chinese three areas was significantly higher than population size; The "eastern gathered" phenomenon of population causedeastern region getting the largest Population Scale Effect; Driving force of residential consumption on regional energy consumption and residential energy consumption was much higher than the other five effects; Due to the comparative advantage of residential consumptioncompared with government consumption, investmentand net export, the decrease of ratio of consumptionpromoted the growth of regional energy consumption; The progress of energy utilization technology slowed the growth of regional energy consumption.
