Abstract:To optimally plan maintenance of wind turbine blades, knowledge of the degradation processes and the remaining useful life is essential. In this paper, a method is proposed for calibration of a Markov deterioration model based on past inspection data for a range of blades, and updating of the model for a specific wind turbine blade, whenever information is available from inspections and/or condition monitoring. Dynamic Bayesian networks are used to obtain probabilities of inspection outcomes for a maximum likelihood estimation of the transition probabilities in the Markov model, and are used again when updating the model for a specific blade using observations. The method is illustrated using indicative data from a database containing data from inspections of wind turbine blades.
In order to make wind energy more competitive, the big expenses for operation and maintenance must be reduced. Consistent decisions that minimize the expected costs can be made based on risk-based methods. Such methods have been implemented for maintenance planning for oil and gas structures, but for offshore wind turbines, the conditions are different and the methods need to be adjusted accordingly. This paper gives an overview of various approaches to solve the decision problem: methods with decision rules based on observed variables, a method with decision rules based on the probability of failure, a method based on limited memory influence diagrams and a method based on the partially observable Markov decision process. The methods with decision rules based on observed variables are easy to use, but can only take the most recent observation into account, when a decision is made. The other methods can take more information into account, and especially, the method based on the Markov decision process is very flexible and accurate. A case study shows that the Markov decision process and decision rules based on the probability of failure are equally good and give lower costs compared to decision rules based on observed variables.
This paper presents a computational framework for risk-based planning of inspections and repairs for deteriorating components. Two distinct types of decision rules are used to model decisions: simple decision rules that depend on constants or observed variables (e.g. inspection outcome), and advanced decision rules that depend on variables found using Bayesian updating (e.g. probability of failure). Two decision models are developed, both relying on dynamic Bayesian networks (dBNs) for deterioration modelling. For simple decision rules, dBNs are used directly for exact assessment of total expected life-cycle costs. For advanced decision rules, simulations are performed to estimate the expected costs, and dBNs are used within the simulations for decision-making. Information from inspections and condition monitoring are included if available. An example in the paper demonstrates the framework and the implemented strategies and decision rules, including various types of condition-based maintenance. The strategies using advanced decision rules lead to reduced costs compared to the simple decision rules when condition monitoring is applied, and the value of condition monitoring is estimated by comparing the lowest costs obtained with and without condition monitoring.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.