FU Lihua, WANG Gang. Nonlinear compensation of eddy current sensor based on wavelet neural network[J]. Journal of Mine Automation, 2015, 41(9): 74-77. DOI: 10.13272/j.issn.1671-251x.2015.09.019
Citation: FU Lihua, WANG Gang. Nonlinear compensation of eddy current sensor based on wavelet neural network[J]. Journal of Mine Automation, 2015, 41(9): 74-77. DOI: 10.13272/j.issn.1671-251x.2015.09.019

Nonlinear compensation of eddy current sensor based on wavelet neural network

  • To eliminate the nonlinear error and improve the measurement accuracy of eddy current sensor, a nonlinear compensation method of eddy current sensor based on wavelet neural network and genetic algorithm was proposed. The method uses nonlinear mapping ability of wavelet neural network to make input and output of the sensor linearization, and uses genetic algorithm to search the optimal initial value of the wavelet neural network to strengthen network nonlinear approximation ability and convergence capability, which can significantly improve nonlinear compensation effect of eddy current sensor. The experimental results show that accuraty of the sensor is improved greatly after the compensation, the maximum absolute error of output voltage of the sensor is 15.55 mV, the maximum relative error is 1.36%, and nonlinear error is 0.34%.
  • loading

Catalog

    /

    DownLoad:  Full-Size Img  PowerPoint
    Return
    Return