Exploration of energy saving potential in China power industry based on Adaboost back propagation neural network

Abstract The power industry of China is an energy-intensive one. In recent years, the demand for energy-saving oriented researches has been increased to provide recommendations to the final energy abatement policy making for the industry. Hence, a study which analyzes the typical energy consumption influencing factors and the potential energy-saving capability for the current power industry of China is needed. In this paper, the energy consumption significant influence factors have been picked out and the potential energy-saving capability has been estimated through the analysis conducted under several different scenarios. The results of this work can contribute to the assessment of energy abatement spaces for the power industry of China from now to the year of 2025. At the beginning of the study, 8 factors including social, economic and technical aspects which affect the energy consumption significantly have been selected by Back Propagation Mean Impact Value (BP-MIV) network approach. Then the 8 filtered factors are employed as input in the Back Propagation Adaptive Boosting (Adaboost-BP) neural network model. The Adaboost-BP was trained and tested on existing data set. Being compared with the results on the same data set from grey prediction and regular BP network, the results of Adaboost-BP showed better accuracy over the other two methods. In the final step of this work, the potential energy-saving capability of China's power industry through the year 2025 was predicted using the trained Adaboost-BP network under 3 different scenarios. The results show that the energy consumption estimates can reach at most 319.657 million tons of standard coal under conventional condition at the year of 2025, with 307.8087 and 294.0182 million tons under moderate and advanced conditions respectively. This means that, 11.8483 million tons and 25.6388 million tons of standard coal could be potentially saved at the year of 2025 under moderate scenario and advanced scenario respectively. This work also presented a guideline for energy saving policies making for the power industry of China.

[1]  Lisha Yang,et al.  The potential estimation and factor analysis of China's energy conservation on thermal power industry , 2013 .

[2]  Ling Jie,et al.  Adaboost detector based on multiple thresholds for weak classifier , 2009 .

[3]  Yi-Ming Wei,et al.  Forecasting China’s regional energy demand by 2030: A Bayesian approach , 2017 .

[4]  Boqiang Lin,et al.  China's energy demand and its characteristics in the industrialization and urbanization process , 2012 .

[5]  Júlia Seixas,et al.  Energy savings potential in urban rehabilitation: A spatial-based methodology applied to historic centres , 2017 .

[6]  Peng Jiang,et al.  Forecasting energy demand using neural-network-based grey residual modification models , 2017, J. Oper. Res. Soc..

[7]  S. Iniyan,et al.  A review of energy models , 2006 .

[8]  Boqiang Lin,et al.  Scenario Prediction of Energy Consumption and CO 2 Emissions in China’s Machinery Industry , 2017 .

[9]  Ehsanolah Assareh,et al.  Forecasting Energy Demand in Iran Using Genetic Algorithm (GA) and Particle Swarm Optimization (PSO) Methods , 2012 .

[10]  Lin Yun Zhang,et al.  Forecasting Chinese Energy Supply and Demand Situation with BP Neural Network , 2013 .

[11]  LI Xiao-li.  Research on Risk Recognition of Real Estate Projects Based on MIV-BP Neural Network Test , 2013 .

[12]  Shanlin Yang,et al.  Energy conservation and emission reduction of China’s electric power industry , 2015 .

[13]  Wei Sun,et al.  Using a back propagation neural network based on improved particle swarm optimization to study the influential factors of carbon dioxide emissions in Hebei Province, China , 2016 .

[14]  L. Galindo,et al.  Short- and long-run demand for energy in Mexico: a cointegration approach , 2005 .

[15]  Sun Feng-qi.  A New Soft Sensor Modeling Method Based on Improved AdaBoost Algorithm for Molten Steel Composition in LF , 2009 .

[16]  Song Li,et al.  Prediction of Market Demand Based on AdaBoost_BP Neural Network , 2013, 2013 International Conference on Computer Sciences and Applications.

[17]  Boqiang Lin,et al.  Estimates of electricity saving potential in Chinese nonferrous metals industry , 2013 .

[18]  Z. Cao,et al.  Energy Demand Forecasting Based on Economy-related Factors in China , 2014 .

[19]  Boqiang Lin,et al.  Estimates of energy demand and energy saving potential in China's agricultural sector , 2017 .

[20]  Henry J. Kelley,et al.  Gradient Theory of Optimal Flight Paths , 1960 .

[21]  G. Dombi,et al.  Prediction of rib fracture injury outcome by an artificial neural network. , 1995, The Journal of trauma.

[22]  Chao Feng,et al.  Analysis of energy efficiency and energy savings potential in China’s provincial industrial sectors , 2017 .

[23]  Feng Xiating,et al.  Classification and prediction of rockburst using AdaBoost combination learning method , 2008 .

[24]  Boqiang Lin,et al.  Understanding the rapid growth of China's energy consumption: A comprehensive decomposition framework , 2015 .

[25]  Geoffrey E. Hinton,et al.  The appeal of parallel distributed processing , 1986 .

[26]  Does China’s Energy Development Plan Affect Energy Conservation? Empirical Evidence from Coal-Fired Power Generation , 2015 .

[27]  Boqiang Lin,et al.  Principles, effects and problems of differential power pricing policy for energy intensive industrie , 2011 .

[28]  F. Li-don.  GA-SA model forecasting China's energy demand , 2015 .

[29]  Chunping Xie,et al.  Estimation on oil demand and oil saving potential of China's road transport sector , 2013 .

[30]  Arash Ghanbari,et al.  A Cooperative Ant Colony Optimization-Genetic Algorithm approach for construction of energy demand forecasting knowledge-based expert systems , 2013, Knowl. Based Syst..

[31]  Xianchun Tan,et al.  China’s regional CO2 emissions reduction potential: A study of Chongqing city , 2016 .

[32]  Nima Hamta,et al.  Predictive Modeling Techniques to Forecast Energy Demand in the United States: A Focus on Economic and Demographic Factors , 2016 .