Deterioration State Space Partition And Maintenance Decision Modeling For Multi-unit Systems Based On Markov Process

The purpose of maintenance decision-making is to select the appropriate maintenance method,arrange a reasonable maintenance plan,ensure the safety and reliability of the system and reduce the cost of system operation and maintenance.For large and complex systems such as wind turbines,the components in the system are interdependent,and often exhibit multi-state characteristics during the failure evolution process during the life cycle.The performance and efficiency of each state are not the same,so this Such systems need to carry out multi-unit systems maintenance decision-making modeling and optimization research with multiple states.In the modeling of maintenance decision-making,the deterioration process of components in the system is usually regarded as a continuous random process.However,there are many unvalued states that do not need to be distinguished in the actual monitoring status of certain systems or components.The division of the area can accurately reflect its operating performance,and continuous state modeling can easily lead to combinatorial explosion problems.Discrete state modeling can effectively reduce the redundancy of state data in continuous modeling and reduce the complexity of the calculation process.Therefore,this paper studies the problem of joint deterioration state space partition modeling and maintenance decision-making based on Markov process for a multi-unit systems composed of multiple components of the same type with independent deterioration and the same deterioration process.The main research contents are as follows:(1)The problem of multi-component system degradation state space partition modeling based on Markov process is researched.Describe the degradation process of a single component as a Markov process with discrete time states,and divide it into multiple regions with several discrete status points according to maintenance needs,and then combine the degradation status spaces of multiple components to form multiple components The multi-state degradation space division model of the system is used to determine the corresponding relationship between the space area and the maintenance demand combination.Based on this,the probability expression of each maintenance demand combination is derived;(2)The stationary probabilistic characteristics analysis and modeling of multi-component systems are researched.Based on the analysis of the multi-state deterioration space transition characteristics,the Markov process theory is used to solve the stationary probability expression of each state of a single component in the system.Correctness and effectiveness;(3)The maintenance decision of multi-component system based on the degradation state space division is researched.Based on the established multi-unit systems deterioration space division model,the average cost rate optimal decision model for multi-unit systems with multi-state deterioration characteristics is established to solve the optimal detection under the condition that the system’s long-term average cost rate is minimum Intervals,opportunities,and preventive maintenance thresholds;(4)The problem of maintenance decision of wind turbine is researched.Take the wind turbine as an example to verify the correctness and applicability of the maintenance decision model.The results show that the model can assist the actual multi-unit systems to give a more economical and reasonable maintenance strategy.