method for searching and retrieving logical patterns from multisensor data time series, system for searching and retrieving logical patterns from multisensor data series, and computer program product

摘要

Methods and systems are provided for searching logical patterns in large multisensor data from the industrial internet. The method retrieves examples of patterns in data time series in which patterns are logically specified using a sequence of symbols. The logical symbols used are a subset of the qualitative abstractions specifically, the concepts of stable, increasing, decreasing. Patterns can include symbol sequences for multisensors, approximate duration, as well as slope values for each symbol. To facilitate efficient consultation, each time series sensor is preprocessed in a sequence of logical symbols. each position in the resulting compressed sequence is recorded in a triple index structure corresponding to the multiple logical patterns to which it may belong. Logical multisensor standards are efficiently retrieved and classified using such a structure. This method of indexing and searching provides an efficient mechanism for exploratory analysis of large multisensor data.