Comparison and Verification of Reliability Assessment Techniques for Fuel Cell-Based Hybrid Power System for Ships

Abstract

In order to secure the safe operation of the ship, it is crucial to closely examine the suitability from the design stage of the ship, and to set up a preliminary review and countermeasures for failures and defects that may occur during the construction process. In shipyards, the failure mode and effects analysis (FMEA) evaluation method using risk priority number (RPN) is used in the shipbuilding process. In the case of the conventional RPN method, evaluation items and criteria are ambiguous, and subjective factors such as evaluator’s experience and understanding of the system operate a lot on the same contents, resulting in differences in evaluation results. Therefore, this study aims to evaluate the safety and reliability for ship application of the reliability-enhanced fuel cell-based hybrid power system by applying the re-established FMEA technique. Experts formed an FMEA team to redefine reliable assessment criteria for the RPN assessment factors severity (S), occurrence (O), and detection (D). Analyze potential failures of each function of the molten-carbonate fuel cell (MCFC) system, battery system, and diesel engine components of the fuel cell-based hybrid power system set as evaluation targets to redefine the evaluation criteria, and the evaluation criteria were derived by identifying the effects of potential failures. In order to confirm the reliability of the derived criteria, the reliability of individual evaluation items was verified by using the significance probability used in statistics and the coincidence coefficient of Kendall. The evaluation was conducted to the external evaluators using the reestablished evaluation criteria. As a result of analyzing the correspondence according to the results of the evaluation items, the severity was 0.906, the incidence 0.844, and the detection degree 0.861. Improved agreement was obtained, which is a significant result to confirm the reliability of the reestablished evaluation results.