基于相似度证据理论的模糊时间序列模型的研究

The Study of Fuzzy Time Series Model Based on Similarity of Evidence Theory

为了提高模糊时间序列模型的预测效果，利用证据理论在处理不确定信息和信息融合方面的优越性，利用贴近度作为证据之间的相似度，对模糊规则进行合成，形成基于相似度的证据理论的多因素模糊时间序列模型．方法在支持证据“与”运算的合成和对冲突证据的比例分配上，充分考虑了证据的权重．最后，通过实例的比较研究验证模型的有效性．

Fuzzy time series models are widely used in various fields. In order to improve the accuracy of forecasting, this paper deals with fuzzy logical relationships based on evidence theory owing to its advantage of tackling with the uncertainty and information fusion after we improve the evidence theory by using fuzzy sets close degree. As for the conjunctive operation on consistent evidences and distribution proportion of conflict to the propositions combined, the weight of evidence is taken into account and effectively employed. Finally, using a example verifies the feasibility and effectiveness of the proposed theoretical model.