An Optimal Lubrication Oil Replacement Method Based on Selected Oil Field Data

The regular replacement of lubricating oil plays a key role in improving machine reliability and reducing unexpected failures of an oil lubricated system. This paper proposes a condition-based maintenance problem with selected oil field data to determine the optimal time of the lubricating oil replacement. The selected oil field data contain health information about the lubricating oil, so the degradation state of the oil can be predicted and the future health condition can be evaluated. The proposed lubricating oil replacement problem is modeled with the evaluated oil health condition in a Markov decision process framework and then, a method for constructing a health index for the lubricating oil is proposed based on information theory to fuse the multiple oil field data and build a degradation progression prediction model. Finally, the proposed method for condition-based lubricating oil replacement is illustrated in a practical case study. The possible applications of an optimal policy for lubricating oil replacement are much wider. For instance, the method can be used as an input to optimize an operational plan and further reduce the maintenance costs.