Stochastic Model Analysis of Preventive Maintenance based on Wind Turbine Gearbox Space-Time Evolution

Abstract

Due to rapid scientific and technical developments, condition-based maintenance (CBM) has been broadly applied in modern large-scale wind enterprises. However, in practice, many large-size modern wind plants still primarily use time-based maintenance (TBM) and only use CBM for auxiliary maintenance. Hence, this study proposes a scheme for combining TBM and CBM to maintain the same physical system, which can effectively make full use of the advantages and overcome the shortcomings of the two approaches. To achieve this aim, this study firstly analyzes the general operating characteristics of TBM and CBM and then establishes a unified stochastic differential equation (SDE) model. Moreover, by combining the SDE with field monitoring data, a proportional hazard model (PHM) is then proposed to solve this model. The results show that the stochastic model can better reflect the state of the device than ordinary differential equation. Model analysis shows that TBM depends on global information and possesses a fixed maintenance time-interval, whereas CBM relies only on the local sample track and generates sporadic repair times. This shows that TBM is a planned maintenance activity, whereas CBM is unplanned. And the system possesses better long-term global behavior under TBM, although there are prominent short-term benefits under CBM, indicating that the combination of the two strategies is necessary and effective. Finally, examples of the model’s application verify the effectiveness and correctness of the proposed model, which is suitable for the current maintenance modes in many enterprises.

Keywords

• stochastic differential equation(sde)
• condition-based maintenance(cbm)
• time-based maintenance(tbm)
• proportion hazard model(phm)
• combination
• state