| The moisture measurement technology of infrared moisture meter is a synthetic technology relating to many subjects. During the past 30 years, much prominent progress has been achieved. However, many influence factors especially the temperature have influence on the output signal of infrared moisture meter which belongs to the weak signal, and the existing infrared moisture meter have some defects such as just carrying simple filter and temperature compensation on it, so the measurement precision, the capability of adapting the environment and anti-noise are still not satisfying, as well as its performance and maintenance. In view of the above, the wavelet soft threshold value filter and the information fusion technology are introduced; some explorations and attempts are made on processing the information intelligently, which give a new data processing basis of the infrared moisture meter. The main content and achievements can be summed up as follows:1. The basic components and operation principle of infrared moisture meter, designing and assembling the light path and hardware circuit of the system are introduced in detail, the program is compiled and debugged.2. Standard samples that are needed for neural network training are made.3. The theory about the wavelet denoising method is introduced, as well as the methods of wavelet soft threshold selection.4. The basic concepts and several ways about information fusion are illustrated, especially the information fusion technology based on neural network.5. Considering the signal's SNR is relatively low and it belongs to weak signal, some feasible simulations of the wavelet soft threshold filter with MATLAB wavelet analysis tool packet are done, and the C program is compiled and executed.6. The temperature's influence on the moisture sensor is very obvious, so bi-sensors information fusion techniques based on BP neural network isproposed, and be simulated with MATLAB, the effect of it is ideal.
|
PDF Full Text Request