DOI: 10.11799/ce202602022
Key technologies for multi-feature data-driven lubricant maintenance decision-making in shearer reducers
LOU Jiajia
PANG Xinyu
XIANG Peidong
MA Fanjie
LIU Minshuai
ZHANG Zhandong
Abstract：To address the reliance of shearer gearbox oil maintenance on manual experience and the difficulty of quantitatively evaluating multi-feature data,a multi-feature data-driven oil maintenance decision system is developed,which extracts 14 characteristic parameters from lubricant temperature,physicochemical indicators,and ferrography data.And three key technologies are proposed:a wear trend recognition method based on Recurrent Neural Network(RNN),which captures temporal dependencies using sliding windows;a wear stage classification method based on Bidirectional Gated Recurrent Unit(Bi-GRU),which learns bidirectional temporal relationships of features;and a fault analysis and recommendation approach combining three-level quantitative scaling and hierarchical logical reasoning,establishing correlations between degradation parameters and maintenance suggestions.Testing shows that the RNN-based wear trend recognition accuracy reaches 90.7%,and the Bi-GRU-based wear stage classification accuracy is 91.49%.Field validation results align with expert assessments.Additionally,bench tests demonstrate the system's generalization capability.The developed system effectively integrates multi-source heterogeneous data,enabling intelligent identification of wear status,fault analysis,and maintenance recommendations for shearer reducers,significantly reducing reliance on manual experience and enhancing the objectivity and reliability of maintenance decisions.
Keywords：multi-featureshearer reducerRNNBi-GRUthree-level quantitative scalehierarchical logical reasoning
Publication Date：2026-02-20
Online Publishing Date：2026-03-23(First online date of this platform, not the publication date of the document)
Pages：8( 176-183 )