ICA and PIND based Remainder Particle Detection Technique for Space-borne Equipment

Space-borne equipment is a major component of the satellite system, which plays a very important role in the process of satellite launch and reception. Therefore, the satellite system has high requirements for the reliability and safety of space-borne equipment. The remainder particle are metal and non-metal particles inside the space-borne equipment that affect the normal operation of the space-borne equipment. When the space-borne equipment is in a weightless state, the remainder particle will produce irregular movement accordingly cause huge safety hazards to the space-borne equipment. Particle impact noise detection (PIND) by way of vibrating the space-borne equipment cause remainder particle in the space-borne equipment to collide with the inner wall of the space-borne equipment. Then, the acoustic emission sensor is used to receive the sound generated by the vibration of the remainder particles. We can analyze the received sound signal to determine whether the space-borne equipment contains remainder particle. However, how to avoid the interference of environmental noise has always been a thorny issue in the process of using PIND to detect remainder particle. In the process of detecting noise, there may be environmental noise that will inevitably affect the detection of remainder particle. In order to filter out the environmental noise, we recorded mixed audio of the remainder particle. Then the ICA algorithm is used to separate the remainder particle signal. Through many experiments, we learn that the recognition rate of this judgment method reaches 100%.