Modeling and Analysis of Linkage Fluctuation Network for Complex Industrial Process

Industry process includes a variety of monitoring variables, there are strong relevance between variables. Time series from different monitoring points can reflect the relationship of system's linkage fluctuation. Using coarse graining method, monitoring series has been changed into a sequence of characters. The sequence which is sliding by a certain length has been converted into a continuous modals of linkage fluctuation. The complex network of linkage fluctuation is composed of all the modals and the link edges between them. Transformation law for modals has been analyzed by degree distribution and structure entropy of complex networks theory. This method has been applied to analyze typical fault of monitoring data from large-scale coal chemical compressor units. It is indicated that the modals both in the normal operation and fault condition follows a power-law distribution, transmission is mainly finished by few modals, but the structure entropy of network have obvious difference. The results not only can be used for complex industrial process fault warning but also provides an idea for assessment of process recovery.