Research on estimation methods for component failure rates in district heating systems under small sample scenarios

This paper investigates and statistically analyses the failure events of the district heating systems in three typical cities in China, and estimates the failure rates of components. In view of the limitations of the direct estimation model, the Bayes theory is introduced, and integrating historical information, the general Bayes model and hierarchical Bayes model are proposed. The accuracies of these three models are compared. The accuracies of the hierarchical Bayes model for the three types of components, namely pipes, valves, and compensators, are 0.31, 0.32, and 0.29 respectively, which are the highest among the three models. By changing the sample size, it is concluded that the accuracies of the three models all show an upward trend with the increase of the sample size, and the mutual gap is gradually narrowed, verifying the applicability of the Bayes model in the case of small samples.