灰色马尔可夫组合预测模型的改进与应用

发布时间：2018-10-09 08:52

【摘要】：本文以灰色马尔可夫预测模型为研究对象,结合灰色系统理论、马尔可夫链理论,将新陈代谢和加权的思想加入到灰色马尔可夫预测模型中,改进了灰色马尔可夫预测模型,并进行了数据分析。灰色预测模型以少量数据信息为研究对象,原理简单、运算方便,该模型对少量信息建模有较好的预测精度。马尔可夫预测模型适用于数据波动较大的随机过程,该模型预测的对象要求具有马氏性。论文首先将两个模型组合起来,形成灰色马尔可夫预测模型,应用灰色预测模型反映数据序列的发展总趋势,对数据进行趋势化处理的基础上应用马尔可夫模型进行预测,希望通过发挥他们各自的优势,达到提高预测精度的目的,但是通过实例验证,预测结果并不理想,于是论文提出了一种基于遗传算法的加权方法,对灰色马尔可夫预测模型进行了改进,在新陈代谢的基础上建立了加权的灰色马尔可夫预测模型,并进行了实例分析。实例研究表明,用改进前和改进后的模型对降水量进行预测,其结果均与对原始数据划分的状态有关,而在相同的状态划分下,改进后的模型有更好的预测效果。
[Abstract]:In this paper, the grey Markov prediction model is taken as the research object, combining the grey system theory and Markov chain theory, the idea of metabolism and weight is added to the grey Markov prediction model, and the grey Markov prediction model is improved. The data are analyzed. The grey prediction model takes a small amount of data information as the research object, the principle is simple, the operation is convenient, the model has good prediction precision for the small amount of information modeling. Markov prediction model is suitable for random processes with large data fluctuation, and the objects predicted by the model require Markov property. Firstly, the two models are combined to form the grey Markov prediction model. The grey prediction model is used to reflect the general trend of the development of the data series, and the Markov model is used to predict the data on the basis of the trend processing. It is hoped that the prediction accuracy can be improved by giving full play to their respective advantages, but the prediction results are not satisfactory through examples. Therefore, a weighted method based on genetic algorithm is proposed in this paper. The grey Markov prediction model is improved and the weighted grey Markov prediction model is established on the basis of metabolism. The case study shows that the prediction of precipitation by the improved model and the improved model is related to the state of the original data partition, and the improved model has better prediction effect under the same state partition.
【学位授予单位】：西安建筑科技大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：O211.62

【参考文献】