GMDH算法的改进及其在煤炭价格系统中的应用

发布时间：2018-11-26 19:04

【摘要】：煤炭能源是关系到我国国计民生的重要生产资料,对国家经济建设发挥着重要的作用。煤炭价格是煤炭市场的一个重要因素,在市场经济条件下,煤炭价格直接反映着煤炭供需的变化,并调节着供需双方的生产经营活动,是企业进行生产经营决策、国家制定经济政策的重要依据。在这样的背景下对煤炭价格进行系统性研究具有十分重要的理论意义和实际应用价值。本文以煤炭价格系统为研究对象,选取GMDH算法对其进行建模及预测。首先对煤炭价格系统进行分析,归纳总结影响煤炭价格的因素,介绍我国煤炭价格的形成机制及国外煤炭定价方法;随后对GMDH算法原理及实现步骤进行介绍,分析算法的优缺点,针对经典GMDH算法的三点不足:需要人为划分样本、单独使用正则化准则作为外准则不能保证模型的外推性能、使用最小二乘辨识中间模型参数存在局限,提出了三点改进策略:样本分组划分方法、组合外准则方法以及采用遗传算法辨识中间模型参数,在此基础上,形成了GCG-GMDH算法;分别采用GMDH算法和GCG-GMDH算法实现煤炭价格系统模型的构建,基于构建模型,进行外推三周期的煤炭价格预测,对比预测结果发现:基于GCG-GMDH算法构件的模型的预测精度得到明显提高;为了进一步验证算法改进的有效性,应用GCG-GMDH算法对煤炭价格系统中的供给量、需求量进行建模及预测,实验结果表明,基于GCG-GMDH算法构建的模型具有良好的外推能力。
[Abstract]:Coal energy is an important means of production related to the national economy and people's livelihood, and plays an important role in the national economic construction. Coal price is an important factor in coal market. Under the condition of market economy, coal price directly reflects the change of coal supply and demand, and adjusts the production and management activities of both supply and demand. The important basis on which the state formulates economic policy. Under this background, systematic research on coal price has very important theoretical significance and practical application value. This paper takes the coal price system as the research object, selects the GMDH algorithm to model and forecast it. Firstly, the paper analyzes the coal price system, summarizes the factors influencing the coal price, and introduces the formation mechanism of the coal price in China and the foreign coal pricing methods. Then, the principle and implementation steps of GMDH algorithm are introduced, and the advantages and disadvantages of the algorithm are analyzed. Aiming at the three disadvantages of classical GMDH algorithm, we need to divide samples artificially, and using the regularization criterion as the outer criterion alone can not guarantee the extrapolation performance of the model. There are limitations in identifying the parameters of the intermediate model by using the least squares, and three improved strategies are put forward: the method of grouping the samples, the method of combining the outer criterion and the method of identifying the parameters of the intermediate model by using genetic algorithm. On this basis, the GCG-GMDH algorithm is formed. The coal price system model is constructed by GMDH algorithm and GCG-GMDH algorithm respectively. Based on the model, the coal price forecast is extrapolated for three cycles. Compared with the prediction results, it is found that the prediction accuracy of the model based on the component of GCG-GMDH algorithm is obviously improved. In order to further verify the effectiveness of the improved algorithm, the GCG-GMDH algorithm is used to model and forecast the supply and demand of coal price system. The experimental results show that the model based on GCG-GMDH algorithm has good extrapolation ability.
【学位授予单位】：南京航空航天大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：F764.1;F426.21;TP311.13

【相似文献】