Abstract: In view of problems that traditional lubricating oil analysis method exists long test and analysis period and high cost, and cannot reflect lubricating oil state and running state of coal mining equipment systematically, a multi-parameter fusion diagnosis method was proposed. The method constructs multi-parameter pattern vector of lubricating oil by collecting viscosity, density, dielectric constant and temperature of the lubricating oil used in coal mining equipment, and judges equipment state through analysis on pattern vector of state to be detected and standard pattern vector. It discovers internal correlation and variation law of pollution state parameters of the lubricating oil by comparing with ferrography analysis result of the lubricating oil. The experimental result shows that the method can realize accurate and rapid diagnosis of pollution state of lubricating oil and running state of coal mining equipment.