Prognostication And Health Management Of Integrated Reconnaissance, Interference, Detection And Communications

Prognostic and health management of the integrated reconnaissance,interference,detection and communications is the core technology of the evaluation mode in the system of “reconnaissance and interference and detection and communication and evaluation integration”,and it is the key to whether the integrated system can meet the real-time operational requirements.This paper takes the system design of “reconnaissance and interference and detection and communication and evaluation integration” as the background of the project,and studies the integrated system fault prediction and health management strategy,real-time data acquisition and feature extraction of integrated systems and real-time fault prediction of integrated systems.The main contents and results of this paper are as follows:1)Research on a strategic plan for integrated reconnaissance,interference,detection and communications fault prediction and health management.Firstly,the preliminary design of the structure of the integrated system was carried out,and the PHM system architecture of the multi-platform integrated system was proposed according to the preliminary structural design of the integrated system.Then,a test-based integrated system fault prediction and health management scheme is proposed.The fault prediction uses a combination of model-based fault prediction and data-driven fault prediction.Finally,an integrated system comprehensive diagnosis scheme combining in-machine diagnosis and off-board diagnosis is proposed.2)The strategy and practical scheme for fault diagnosis and location of integrated reconnaissance,interference,detection and communications front panel components are researched.If the system is idle,open the array integration test,and process the integration test data according to the designed fault diagnosis and fault location data processing flow.If the system is working,the fault real-time diagnosis and location method based on support vector machine classification is used to further establish a hierarchical diagnosis model with fault tree structure.3)Research on real-time data acquisition and feature extraction of integrated reconnaissance,interference,detection and communications.In-depth exploration and research on the measurement methods and feature extraction of several types of data parameters common in integrated systems.The feature extraction of parameters such as cyclic spectrum is simulated and analyzed,which shows that the extracted features meet the requirements of integrated system health state management.4)A method for real-time fault prediction of integrated reconnaissance,interference,detection and communications is researched.Since most of the faults of the integrated system are from analog circuits,this paper selects the analog transmit channel,analog receive channel and detection mode of the integrated system as the real-time fault prediction research object of the integrated system,and according to the above characteristic parameters of the integrated system The study established the system state model of the research object here.Then according to a proposed model-based integrated system real-time fault prediction method,combined with particle filter EOL prediction algorithm,the fault prediction and life prediction of the integrated system simulation launch channel and detection mode are carried out.Finally,the effectiveness of the proposed fault prediction scheme is verified by simulation experiments.