Research on Sampling and Life Prediction of Storage Products Based on Uncertainty Theory

Products with the characteristics of “long-term storage, small amount of testing, once-used” are called storage products. For the characteristics and engineering requirements of storage products, this paper proposes a quantum clustering-based sampling method for storage products. First, this paper selects a series of evaluation indexes under the framework of evaluation index system from four perspectives. Secondly, experienced experts were selected to score the selected indicators. Then, the uncertainty consensus model is used to process the scores of several experts, which reduces the large amount of epistemic uncertainty due to the small sample information. Third, the quantum clustering algorithm is introduced for the division of sampling matrices, and a sampling method is established for sample extraction. Finally, the practicality and operability of the sampling method of this paper are verified by taking a storage product as an example. It provides a new method for small sample sampling of storage products with epistemic uncertainty, and improves the accuracy of storage product reliability evaluation to a certain extent.