基于蒙特卡罗方法的中国中长期煤炭需求预测

为避免传统预测方法的参数取值主观性问题,采用参数随机产生的蒙特卡罗方法预测中国中长期煤炭需求。首先分析了经济增长、能源结构和产业结构三个主要煤炭需求影响因素,并基于1980~2015年间各影响因素及煤炭消费的历史数据和最小二乘法的多元线性回归拟合煤炭需求方程。在此基础上,构建各影响因素的概率分布,采用蒙特卡罗方法模拟1981~2015年的煤炭需求,发现仿真结果可以较好拟合现实,可作为仿真预测的有效工具。结合经济新常态和能源结构调整的现状,控制参数取值范围进行蒙特卡罗仿真预测,结果显示,2016~2025年的煤炭需求呈先上升后下降趋势,并于2020年达到需求的峰值40.25亿吨,这些结果对于煤炭产业的科学决策有重要的作用。

Chinese Medium and Long-term Coal Demand Forecast Based on Monte Carlo Method

In order to avoid the subjectivity of parameters value of traditional prediction methods, Monte Carlo method, which can randomly generate parameters value, is used to predict the medium and long-term coal demand in China. Firstly, three main influencing factors of coal demand such as economic growth, energy structure and industrial structure are analyzed. And then the coal demand equation is fitted with Ordinary Least Squares based on the historical data of coal consumption and its influencing factors from 1980 to 2015. On this basis, the probability distribution of each influencing factor is constructed, and the coal demand of 1981~2015 is simulated by Monte Carlo method. It is found that the simulation results can better fit the reality and can be used as an effective tool for simulation prediction. Combining with the “new normal” of China's economy and its energy structure adjustment, the Monte Carlo simulation prediction is carried out by controlling the range of parameters value. The results show that the coal demand from 2016 to 2025 will first increase and then decrease, and reach the peak demand of 4.025 billion ton in 2020. These results play an important role in the scientific decision-making of the coal industry.