Fuzzy Multi-dimensional Time Series Model Based On Evidential Theory Of Adjustable Parameters

Fuzzy time series prediction methods provide a framework to deal with fuzzy problems,it is widely used in real life.However,under the complexity of natural and social sciences,scholars have turned to fuzzy multi-dimensional time series.In fuzzy multi-dimensional time series,the influences between time series is very complicated,and directly influence the forecast result.Based on this,this paper uses the similarity of time series and the degree of closeness to study the relationship between fuzzy multi-dimensional time series and improve the prediction accuracy.D-S evidence theory has a good application of uncertain information on the synthesis of evidence.Scholars consider the high degree of evidence conflict when synthesizing evidence.However,in real life evidence is not only conflict,there will be mutual support between the evidence.This paper attempts to find effective ways to measure the interrelationships between multiple evidences by studying how the strength of mutual support between evidence is determined.In this paper,a new fuzzy multi-dimensional time series prediction method based on evidential theory of adjustable parameters is proposed.Because the influence degree of each time series is different,it will affect the prediction result of time series.Therefore,when using the theory of evidence to synthesize,the relationship between time series directly determines the strength of support in evidence synthesis.This paper determines the weight coefficient by calculating the strength of each evidence.This new method makes the parameters change with the change of evidence,which enhances the reliability and rationality of the evidence,objectively reflects the real evidence relationship,and also provides a new method for fuzzy multidimensional time series prediction.