Design Of Digital Correlation Detection System For Rotor Icing

Anti-icing for helicopter has been an important issue which the helicopter researchers keep on studying and resolving, and it is also one of the technical difficulties from the original helicopter. However, the infrared icing detection method which is described in this thesis provides a good solution to this problem. In view of the helicopter flight environment, the infrared radiation energy of rotor icing which is received by infrared sensors has become very weak after absorbed through the atmosphere and refracted by the particles, besides the aerodynamic characteristics of the rotor and the atmospheric interference, which make the SNR of the sensor output signal less than testing requirements. Therefore, the digital multi-point average algorithm is used in this thesis to design a digital correlation detection system which can detect the useful periodic signal overflowed by noise. This technique can detect the weak signal which the general approaches are difficult to measure, so that the precision of weak signal measurement is greatly improved.The characteristics of the helicopter rotor icing signal and the sources of noise and interference are described in this thesis. After being compared with other several related detection techniques, the digital multi-point average algorithm is chose as the final algorithm according to the signal characteristics and the performance requirements, and verified correct and available through a simulation. Meanwhile, the digital-related hardware circuit of the detection system based on TMS320LF2407A is designed according to the related characteristics of icing infrared sensors and the performance targets which is required by the system. And then the algorithm software is programmed according to the basic principles of digital multi-point average algorithm and the system's hardware design. Finally this system is used to detect the actual signal form icing test platform, and the test results and actual waveform are basically the same.The results show that using digital multi-point average algorithm to detect the weak period signal which is overflowed by the noise is practicable. The digital multi-point average device is superior to analog device in dynamic characteristics, linear distortion, zero drift and so on. So digital devices have a broad prospect for its wide application in industry, such as weak-light, small displacement, tiny vibration, micro temperature difference, micro electric capacitance, weak magnetic, weak conductance, feeble current and so on.