Incremental analysis of load handling device deflection considering lubrication degradation for predictive maintenance

Abstract

In the Industry 4.0 era, unmanned automated factories heavily rely on load handling devices (LHDs), which are indispensable but susceptible to failures. Deflection is a crucial health indicator, with excessive deflection causing issues such as tilting and reduced positioning accuracy. This underscores the necessity for reliable deflection prediction methods in LHDs, a currently under-researched area, essential for informing predictive maintenance strategies. This study introduces an incremental analysis method that addresses the impact of internal cumulative wear and lubrication degradation on deflection increase. It involves iteratively updating a multiparametric model to represent the cumulative wear process and utilizes an empirical model based on the Stribeck curve to describe lubrication degradation. To enhance accuracy, particle filtering (PF) is employed to revise friction coefficients derived from the empirical model. Experimental data from an LHD operating in a load-carrying mission validates the accurate and robust nature of this method, particularly when compared with non-friction coefficient revising counterparts. As a typical application case study for preventive maintenance, this method is utilized to simulate various scenarios of lubrication replenishment and quantify the effect of timing on LHD lifespan enhancement through lubricant replenishment. The results facilitate the optimization of maintenance schedules and offer a new approach for updating lubrication-related parameters in digital twin models of equipment-level products, in line with preventative maintenance strategies.