Search

Thermal fatigue cracking is a common problem for boiler waterwall tubes in power plants, resulting the risk of unscheduled plant shutdown. Developing an effective degradation model to predict the future condition of waterwall tubes is vital to assess boiler remaining life. Even though many numerical studies are available in the literature for understanding the physical mechanism (thermal fatigue cracking), the development of predictive models for future risks that can inform decision making are limited. One of the key impediments for developing predictive models is the lack of data. In the case of boiler critical subsystems, very few failures (if any) are recorded in its history. In addition, the number of inspections is limited due to access difficulties. Thus, in many practical cases, the condition data available for degradation modelling is sparse with both infrequent and (spatially) incomplete inspections. This paper presents the development of a stochastic degradation model prediction of thermal fatigue cracking severity in a boiler waterwall when the data is limited. The time evolution of the condition indicator for each waterwall tube is modelled using Markov Models and Bayesian identification algorithms are used to tractably address the large amount of missing data. The methodology presented in this paper employs a Markov Chain Monte Carlo (MCMC) scheme to account for the parameter uncertainty in the presence of missing data. Moreover, this paper also incorporates a novel grouping strategy in which the neighbouring components are grouped to tackle the problem of (spatial) data sparsity. The degradation modelling and prediction tools developed in this paper will support renewal decision making by using available onsite asset data and knowledge. A real-world case study of a boiler waterwall operating in an Australian power industry is also presented. It is found that the waterwall degradation appears to be slowing, but an additional inspection is recommended to confirm the trend. [ABSTRACT FROM AUTHOR]

• Optimization of safety-threshold for complex industrial processes. • Consider joint probabilities of multiple process variables leading to an accident. • Enables dynamic risk assessment based on multiple real-time process variables. Modern engineering systems give paramount importance to safety in order to avoid or mitigate hazardous accidents which can lead to huge economic losses, environmental contamination, and human injuries. This paper proposes an integrated approach that uses both Hidden Markov Model and Bayesian Network to estimate an optimum safety-threshold for complex industrial processes. In order to estimate the safety threshold, the proposed approach considers different cost factors and the joint probabilities of multiple process variables leading to an accident. In addition to the system level threshold, it also estimates the safety-threshold for components. This helps in identifying the component that needs maintenance to enhance system performance and safety. Furthermore, it proposes a dynamic risk assessment methodology based on multiple real-time process variables. The optimum safety-thresholds are estimated using Genetic Algorithm which aims at minimizing the system running cost over a finite time horizon. A case study on Tennessee Eastman Chemical Process is presented to demonstrate the proposed methodology for optimizing process safety-threshold. [ABSTRACT FROM AUTHOR]

Highlights • System functional reliability analysis in spite of sparse historical failure data. • Monitored system process variables used for real-time system reliability evaluation. • The measurement process uncertainties are incorporated using virtual evidence. • The system DBN model enables fault identification at component level. Abstract This paper presents a novel methodology to estimate and predict the functional reliability of a system using system functional indicators and condition indicators of components. Instead of 'system reliability', the paper uses the terminology 'system functional reliability' because the functional indicators used in the methodology principally represent the system performance level or system functionality. The proposed model relates the degradation state of components to the system functional state. The model allows the use of system functional indicators and condition data of components in continuous time domain. The proposed methodology uses both Hidden Markov Model and Dynamic Bayesian Network for estimating and predicting system functional reliability. HMM helps in mapping the continuous data into hidden state probabilities while the system DBN helps in finding the posterior system state probability by considering the component dependencies within a system. The study is also extended to show how the external covariates can be incorporated into the proposed model. Since the external covariates accelerate the degradation of a component, the component state transition probability in the second model is adjusted to vary with respect to the covariates. A case study based on Tennessee Eastman Chemical Process is conducted to demonstrate the proposed methodology for system functional reliability estimation and prediction. Another simulation based case study is presented to describe how the external covariates are included in the presented methodology. [ABSTRACT FROM AUTHOR]

• Joint inspection and renewal planning for a system of multiple non-identical components. • Considering impact of preventive decommissioning of components that causes future performance loss of the system. • Optimal plan is competitive with condition-based maintenance approaches. • Case study of sugar production boiler to evaluate benefit of optimized strategy. In this paper, a system of multiple degrading components in series reliability is considered where the system performance is a monotonically increasing function of the number of working components. By decommissioning failed components, the system is restored to working order, but the performance loss may only be recovered when the entire system is renewed. The failure risks are managed by inspecting components and conducting preventive decommissioning based on the result. Thus, a complete maintenance and inspection policy will specify when and what components to inspect/decommission and when to renew the entire system. A joint inspection, decommissioning, and renewal plan is developed that minimizes the cost rate and this policy is applied to a boiler heat exchanger operating in a sugar mill. The results show that the proposed policy yields significant savings compared to a benchmark fixed-interval inspection policy and is competitive with a sophisticated condition-based approach. Moreover, in contrast to the condition-based benchmark, the proposed policy provides the entire inspection and renewal plan early in the system's life rather than re-planning after each inspection. Thus, the proposed approach is quite practical–it provides additional predictability in direct expenditures compared to the best condition-based policy and has statistically similar total costs. [ABSTRACT FROM AUTHOR]