Generating Real-valued Failure Data for Prognostics Under the Conditions of Limited Data Availability

Abstract

Data-driven prognostics solutions underperform under the conditions of limited failure data availability since the number of failure data samples is insufficient for training prognostics models effectively. In order to address this problem, we present a novel methodology for generating real-valued failure data which allows training datasets to be augmented so that the number of failure data samples is increased. In contrast to existing data generation techniques which duplicate or randomly generate data, the proposed methodology is capable of generating new and realistic failure data samples. To this end, we utilised the conditional generative adversarial network and auxiliary information pertaining to the failure modes. The proposed methodology is evaluated in a real-world case study involving the prediction of air purge valve failures in heavy trucks. Two prognostics models are developed using gradient boosting machine and random forest classifiers. It is shown that when these models are trained on the augmented training dataset, they outperform the best prognostics solution previously proposed in the literature for the case study by a large margin. More specifically, costs due to breakdowns and false alarms are reduced by 44%.