Method for extracting longest common substring of time series data

The invention discloses a method for extracting a longest common substring of time series data. The method comprises the following steps: step 1, reading to-be-compared time series data; step 2, selecting a time characteristic parameter of the time sequence data as a calculation parameter, and performing differential transformation, quantitative processing and symbolization processing on the timesequence data to obtain a symbolized sequence; step 3, establishing an equivalent character table of time sequence data according to the symbolized sequence; step 4, according to the equivalent character table, searching and storing a common substring of the time sequence data by adopting a dynamic sliding window mode; and step 5, extracting the longest common substring by judging the length of the common substring. According to the longest common substring extraction method for the time sequence data, the longest common substring of the time sequence data can be quickly extracted, and the longest common substring is also effective under the condition that the data part of the time sequence is lost.