基于单/多参数的DC-DC模块电源健康评估技术研究

Research on Health Evaluation Model of DC-DC Power Module Based on Mahalanobis Distance Method

随着电子产品的快速发展,系统设备对电源系统的性能以及可靠性提出了更高的要求,将基于状态的维修方法引入到电源可靠性保障领域十分必要.实现模块电源基于健康管理智能维护的首要基础是要准确获得电源的健康状态.针对两种型号的DC-DC模块电源提出了一种基于分布间距度量的单个或多个监测参数的健康评估方法.基于DC-DC电源的4个评估参数的历史监测数据,文中采用mRMR特征选择的方式提取出相关敏感退化特征,然后采用基于CV值(置信值)单参数及基于马氏距离多参数的健康评估方法,对两款直流28V转5V以及28V转3.3V的DC-DC模块电源进行健康状态评估,试验结果验证了评估算法的正确性和实用性.这种电源评估算法对于轻度退化程度以上的电源评估问题具有较高的准确率,该研究成果对DC-DC模块电源的健康评估以及解决电源产品的可靠性保障提供一种切实可行且具有工程化应用前景的方法.

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 method 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,a health assessment method of single or multiple monitoring parameters was proposed 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 power supplies of 28 V to 5 V and 28 V to 3.3 V. The experimental results prove the correctness and practicability of the algorithm proposed in this paper. The algorithm 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 guarantee of power products.