Abstract:Abnormal conditions are hazardous in complex process systems, and the aim of condition recognition is to detect abnormal conditions and thus avoid severe accidents. The relationship of linkage fluctuation between monitoring variables can characterize the operation state of the system. In this study, we present a straightforward and fast computational method, the multivariable linkage coarse graining (MLCG) algorithm, which converts the linkage fluctuation relationship of multivariate time series into a directed and weighted complex network. The directed and weighted complex network thus constructed inherits several properties of the series in its structure. Thereby, periodic series convert into regular networks, and random series convert into random networks. Moreover, chaotic time series convert into scale-free networks. It demonstrates that the MLCG algorithm permits us to distinguish, identify, and describe in detail various time series. Finally, we apply the MLCG algorithm to practical observations series, the monitoring time series from a compressor unit, and identify its dynamic characteristics. Empirical results demonstrate that the MLCG algorithm is suitable for analyzing the multivariable linkage fluctuation relationship in complex electromechanical system. This method can be used to detect specific or abnormal operation condition, which is relevant to condition identification and information quality control of complex electromechanical system in the process industry.

Study of Abnormal Events Relevance in Process Industry Based on Sequences Pattern Mining

Compound Fuzzy Clustering Anomaly Detection Based on Production Process Coupling

Novel approach for industrial process anomaly detection based on process mining

Accelerated Multiple Alarm Flood Sequence Alignment for Abnormality Pattern Mining

Application of Sequential Pattern Mining Algorithm in Alarm Information Processing of Power System

Development and application of data mining system for chemical process

SDG real-time fault diagnosis based on time-sequence process fragment analysis

Contamination Event Detection Method Based On The Longest Common Subsequence Analysis Using Multiple Water Quality Parameters

A method for the early prediction of abnormal conditions in chemical processes combined with physical knowledge and the data-driven model

Anomaly Detection of Pollution Control Equipment Based on AMI Data Analytics

Using multivariate pattern segmentation to assess process performance and mine good operation conditions for dynamic chemical industry

Identification of Industrial Process Time Series Events Based on Agglomerative Hierarchical Clustering*

Application Research of Multivariate Linkage Fluctuation Analysis on Condition Evaluation in Process Industry

A Review on Data-Driven Process Monitoring Methods: Characterization and Mining of Industrial Data

Analysis of Chemical Production Accidents in China: Data Mining, Network Modeling, and Predictive Trends

Process Monitoring Using a Sequence to Sequence Model

Abnormal Situation Management in Chemical Processes: Recent Research Progress and Future Prospects

Research of Detection Algorithm for Time Series Abnormal Subsequence

Non-regular alarm correlation analysis

Fused Data-Driven Approach for Early Warning Method of Abnormal Conditions in Chemical Process

Data-based identification of the related alarms in the process industries