Abstract: With the rapid development of electronic product technology, the equipment requires higher performance and reliability of power supply, so it is necessary to introduce state-based maintenance into the field of power supply reliability. The key to intelligent maintenance of modular power supply based on health management is to get the health status of power accurately. For two types of DC-DC module power supply, this paper presents a health assessment method of single or multiple monitoring parameters based on the distance between parameter distributions.Based on the historical monitoring data of 4 parameters of the DC-DC power supply, this paper uses the mRMR feature selection method to extract the related sensitive degradation characteristics, and uses the CV value (confidence value) single parameter and the martensitic distance multi parameter health assessment method to evaluate two DC-DC powers with 28VDC to 5VDC and 28VDC to 3.3VDC. The experimental results prove the correctness and practicability of the algorithm proposed in this paper. The algorithm proposed in this paper has higher accuracy for degraded power assessment.The results of this paper provide a practical and engineering method for the health assessment of DC-DC power supply and the reliability assurance of power products.