中国产业结构调整碳排放效应的多目标遗传算法

Multi-objective Genetic Algorithm for Optimizing Carbon Emission from Industrial Structure Adjustment in China

基于建立的投入产出多目标优化模型,借助多目标遗传算法NSGA-Ⅱ详细测算分析了中国产业结构调整的碳排放效应,研究思路与算法均不同于因素分解或回归分析模式。得出结论:在保证经济增长与就业水平稳定的前提条件下,合理压缩工业、适度扩张批发零售住宿餐饮业等结构性调整仍能比较显著地减少碳排放量,但目前既定的技术水平、所处的工业化阶段会一定程度上限制了这种低碳效应的充分发挥;为了充分实现产业结构的"低碳化调整",需要优化行业内部结构、产品结构与能耗结构,改进生产技术,创新开发与应用节能技术、新能源等方面相互配合、步调一致。

Based on a multi-objective optimization model of input-output,the multi-objective genetic algorithm of NSGA-II is used to calculate and analyze the carbon emission effect due to industrial structure adjustment,which is different from the factor decomposition or regression analysis methods.The conclusions are as follows： Under the premise of stable economic growth and employment level,the structure adjustment such as reasonably narrowing the industrial scale,and moderately expanding the industrial scale of wholesale and retail services,and hospitality services,may significantly reduce carbon emissions.However,the current technical and the industrialization levels may hinder the low-carbon emission effect to some degree.It is clear that other efforts such as optimization of internal industry structures,product mixes and energy consumption structures,improvement of production technologies,innovative development and applications of energy-saving technologies and new energy,should be coordinated to adequately adjust the industrial structure for carbon emission reduction.