Multi-Value Missing Processing Method of Integrated Modular Avionics System Operation Monitoring Data

The Integrated Modular Avionics system has high integration and complex functions. Its working status needs to be monitored in real time during operation. During flight, the external environment is harsh and changeable, and there may be mechanical vibration, electrical noise, series mode interference, and common mode interference, temperature changes and other factors affect the sensor, resulting in the inability to obtain part of the data measured by the sensor and a multi-value missing problem. This paper is based on the multi-value missingness of the Central Processing Unit (CPU) temperature data of a certain type of aircraft's IMA system. Through working mechanism analysis, the nonlinear regression analysis method is used to perform regression analysis on the CPU temperature and CPU utilization data, so as to realize the estimation of CPU temperature data missing value, improve the quality of IMA system operating status data and provides guarantee for subsequent anomaly detection, fault diagnosis and other work.