Multi-objective Maintenance Optimization And Its Applicationfor Complex Electromechanical Products

For complex electromechanical products such as wind turbines,CNC machine tools,and radars,due to their high acquisition cost and poor working condition,once the failure occurs,it will result in huge downtime loss and maintenance cost.Scientific maintenance strategy is not only an important means to improve the product availability,but also an important guarantee to attract customers and promote products.At present,the maintenance of complex electromechanical products is still facing many problems,mainly reflected in the following aspects: the lack of scientific estimation for the maintenance period and maintenance service cost,the difficulty of timely and effective maintenance service,and the insufficient fit between maintenance strategy and engineering practice.Therefore,it is of great theoretical value and engineering application prospect to carry out the maintenance optimization for complex electromechanical products.This thesis takes the complex electromechanical products as the research object,based on the maintenance policy of unequal inspection period,multi-stage preventive maintenance(PM)and multi-objective maintenance,and considering the optimization objectives of product reliability,availability and operation and maintenance(O&M)cost,an optimization of multi-objective maintenance decision-making for complex electromechanical products is studied.The major work in the thesis is as follows:(1)Availability-based maintenance optimization is carried out under unequal inspection period and imperfect repair.Taking electromechanical products(such as wind turbines)as the object,a maintenance optimization method of unequal inspection period is proposed.By introducing age reduction factor and considering imperfect repair,a maintenance decision model with availability as the optimization objective is established.Case study is completed with the key components of wind turbines.The results show that compared with the maintenance model based on equal inspection period,the maintenance model based on unequal inspection period can improve the system availability effectively.(2)Multi-stage PM optimization is implemented based on the reliability.A multi-stage PM optimization method is proposed for electromechanical systems,such as wind turbines.The age reduction factor is used to describe the effect of imperfect PM in each stage,and a multi-stage PM decision-making model is derived with the cost rate as the optimization objective.Case study and sensitivity analysis are completed with the gearbox,rotor and bearing of the wind turbines.The results show that compared with the traditional one-stage or two-stage PM,the multi-stage PM considering reliability can reduce the cost rate of the system effectively,meanwhile it can also optimize the PM period and the number of PMs in each stage.(3)Multi-objective imperfect PM optimization is accomplished by considering the failure classification.By introducing the hybrid failure rate to model the imperfect PM,the failures occurring in the imperfect PM period are divided into soft failure and hard failure,and preventive inspection and minimal repair are adopted for the two types of failures.The quasi-renewal coefficient is used to describe the relationship between the imperfect PM periods.With the system availability and cost rate as the optimization objectives,a multi-objective maintenance mathematical model is established.The Pareto optimal solution of the multi-objective maintenance model is solved by using NSGA-II.Case study and sensitivity analysis are completed with wind turbines.The results show that the Pareto frontier of multi-objective optimization is consistent with the results of single-objective optimization.Furthermore,compared with the results of single objective optimization,the results of multi-objective optimization can better meet customers' preferences and adapt to different maintenance resource constraints.Therefore,it can provide more feasible maintenance solutions.The main contributions of this thesis lie in systematically carrying out the multi-objective maintenance optimization of electromechanical products based on availability,reliability and cost,and also in the improvement of maintenance policies and methods for electromechanical products from the aspects of unequal inspection period policy,imperfect PM modeling,multi-stage maintenance and failure classification.The research enriches the maintenance decision-making model and method,and provides a theoretical basis for making scientific and reasonable maintenance policies.