DISTRIBUTED TIME SEQUENCE PATTERN RETRIEVAL METHOD FOR MASSIVE EQUIPMENT OPERATION DATA

摘要

A distributed time sequence pattern retrieval method for massive equipment operation data, which relates to the technical field of time sequence analysis. The method comprises: first, configuring a retrieval reference time sequence pattern and the number of the most similar patterns which need to be retrieved; reading run time sequence data of historical equipment to be retrieved into a distributed system so as to construct a distributed data set X; then, constructing an index for elements of X; constructing multiple auxiliary distributed data sets according to the length of a reference time sequence, connecting X and an auxiliary data set, and constructing a distributed data set Z, each element thereof being a short time sequence; calculating the Euclidean distance between the reference time sequence and the elements of the Z, so as to construct a distributed data set R; sorting R, taking the smallest k elements, and returning the index of a corresponding element; and obtaining the corresponding element in the data set Z according to the index. The present method improves the similarity retrieval efficiency of a large amount of equipment running time sequence data.