In the operation of large complex systems, a substantial amount of process data is collected and stored. The extraction, utilization, and identification of features from complex data are crucial for system modeling, simulation system construction, and the design of control algorithm. The experimental data of the flight environment simulation system (FESS) was investigated. Firstly, an information entropy calculation method based on the probability cloud space was proposed, which could effectively identify the dynamic characteristics of the rate of change in the signal amplitude of the experimental data. Secondly, the information entropy was used to supportthe research on the adaptive tracking differentiator algorithm of integral step parameter and the signal correlation mining algorithm combined with the Apriori algorithm. The results show that the developed adaptive tracking differentiator algorithm can achieve superior signal extraction quality for both steady and dynamic states within the constraints of computational cycles. The proposed signal correlation mining algorithm can effectively reflect the internal patterns among different experimental parameters.
Zhang et al. (Sun,) studied this question.