论文部分内容阅读
以中国历年能源消费量为基础,分别建立灰色自忆性预测模型和数据机理自忆性预测模型;利用经验模态分解方法分析中国能源消费增长率在经济、人口、城市化下的变化情况,并建立基于经验模态分解的BP神经网络预测模型;通过遗传算法构建能源需求总量组合预测模型,求得中国清洁能源需求总量。研究结果表明:组合预测能够充分利用多个模型的丰富信息,提高预测的准确性;2020年中国清洁能源需求量将达到5.02×108~8.26×108tce,应优先开发清洁能源。
Based on the energy consumption in China’s past years, the gray self-reliance prediction model and the self-remembering prediction model of data mechanism are respectively established. Empirical mode decomposition method is used to analyze the changes of the growth rate of energy consumption in China under the economic, population and urbanization. And establish a BP neural network prediction model based on empirical mode decomposition. Construct a combined forecasting model of energy demand by genetic algorithm, and get the total demand of clean energy in China. The results show that combined forecasting can make full use of the rich information of multiple models and improve the accuracy of forecasting. In 2020, the demand for clean energy in China will reach 5.02 × 108 ~ 8.26 × 108tce, and clean energy should be prioritized.