Research On Detection Method Of Coal Adulteration Rate Based On AC Impedance Measurement And Levenberg-Marquardt Algorithm

Coal is an important petrochemical energy and industrial raw material in China.The detection of coal quality is of great significance for coal mining and processing.Traditional laboratory testing methods have complex testing procedures and low testing efficiency;existing real-time online testing methods have expensive equipment and high testing costs.Aiming at the shortcomings of the current coal quality detection technology,it is proposed to use the AC impedance technology to detect the coal quality to realize the rapid detection of coal quality.The main research contents of this thesis are as follows:Taking Ningxia anthracite as a sample,the differences in AC impedance characteristics of standard high-quality coal and adulated coal are verified,and the influence of different moisture content and adulteration rate on the output value of the impedance measurement module is determined,indicating that the adulteration rate and moisture content of coal are important factors affecting its AC impedance characteristics.Based on the principle of coal AC impedance characteristics,the design of coal AC impedance detection system with STM32F103ZET6 microcontroller as the core processor is completed.The detection system mainly includes AC impedance measurement module,moisture content determination module,microcontroller control module and input and output module.In order to explore the relationship between the AC impedance characteristics of coal and the adulteration rate and moisture content,and then identify the adulated coal,two multivariate data fusion models are established in this thesis.Firstly,a coal detection model based on the Levenberg-Marquardt algorithm is established by taking the moisture content of the tested coal and the adulteration rate of coal gangue as the input,and the output value of the coal impedance measurement module as the output.The optimization of parameters ensured the decrease and the convergence speed of the algorithm.Secondly,in view of the influence of moisture content on the detection of adulated coal,a coal detection model based on surface fitting algorithm is proposed.Taking the moisture content of the tested coal and the output values of the coal impedance measurement module as input,an information fusion model based on binary nonlinear regression is constructed,and a fusion model solving algorithm based on the least squares method is given.The two models are compared and analyzed.The root mean square error RMSE of the coal detection model based on the Levenberg-Marquardt algorithm is 0.0061,and the coefficient of determination R2 is 0.9898.This model can better describe the relationship between the output value of the coal impedance measurement module and its moisture content and adulteration rate.The coal adulteration rate is calculated by the Newton iteration method,so as to realize the rapid detection of adulated coal.