Complex industrial abnormal working condition causal reasoning and early warning method based on time sequence knowledge graph

Abstract

The application relates to the technical field of industrial process control and abnormality monitoring, in particular to a complex industrial abnormal working condition causal reasoning early warning method based on a time sequence knowledge graph, which comprises the following steps: collecting multi-source heterogeneous time sequence data of a complex industrial process under normal working conditions and various abnormal working conditions; constructing a static field knowledge graph ontology containing multiple levels of nodes; quantifying the dynamic causal strength between entities at all levels and integrating the same as a relationship edge attribute into the static graph to form a time sequence knowledge graph which can be dynamically updated; constructing a time sequence knowledge graph reasoning model to model and predict the propagation path and evolution trend of abnormal working conditions; and generating interpretable early warning information containing the root cause, path, trend and disposal suggestions; the application can decompose the input-output mapping of a traditional black box model into intermediate sub-processes with physical meanings by constructing a multi-level causal evolution chain, and quantifies the influence of the nonlinear coupling effect between states on the final quality.

Images