Research On Microwave Detection Of Carbon Content In Fly Ash Based On Free Space Method

With the steady and continuous progress of carbon peak and carbon neutrality work,it is imperative for coal-fired power units to carry out energy-saving and emission reduction as well as flexibility transformation in order to gradually achieve the transition towards clean,efficient,flexible,and low-carbon coal-fired power generation.To meet the requirements of power unit transformation,it is necessary to establish a reliable combustion system.The carbon content in fly ash is one of the important indicators reflecting the combustion efficiency of the boiler.Achieving accurate monitoring of carbon content in fly ash can guide boiler operators to correctly select excess air coefficient and wind speed and volume ratio,thus ensuring that the boiler operates in the optimal combustion condition,reducing the cost of electricity generation and improving the economic operation of the unit.Therefore,how to achieve accurate online monitoring of carbon content in fly ash has become an urgent issue that needs to be addressed in the current field of coal-fired power generation.Based on the above background,by comparing the measurement accuracy and applicable scenarios of carbon content detection methods in fly ash,this paper conducted research on microwave detection technology for carbon content in fly ash.The main work achievements are as follows:Regarding the problem of the rationality of fly ash composite dielectric models in microwave detection technology for carbon content,which has not been demonstrated by actual measurement data in current research,this article first explored the applicability of several classic equivalent media models to fly ash mixtures composed of zero ash and carbon.Subsequently,the electromagnetic characteristic parameters of fly ash were measured and used to deduce and verify the function relationship between its complex dielectric constant and the corresponding carbon content,as determined by the Brown model.This function relationship can be directly applied to obtain the dielectric property parameters of low-carbon-content fly ash,thereby greatly simplifying the construction process of the fly ash carbon content microwave detection simulation system.Regarding the problem of a lack of theoretical values for comparison with actual measurement results in current research on fly ash carbon content microwave detection technology,which makes it difficult to define measurement errors,this paper first selected the model of a feeding waveguide and determined the structural parameters of a horn antenna.Subsequently,relying on electromagnetic homogenization theory,equivalent homogeneous plate models of fly ash under different carbon content conditions were constructed,and the microwave detection system composed of this model and the horn antenna was electromagnetically simulated using HFSS.The simulation results provide a reliable reference basis for subsequent experiments.Regarding the problem of poor detection performance in current research on fly ash carbon content detection using the free space method,due to the poor distinction of electromagnetic characteristic parameter data for fly ash under low carbon content conditions,this paper constructed a fly ash carbon content regression prediction model based on Support Vector Regression(SVR)using electromagnetic parameter data collected from experiments.The predictive performance of the model was evaluated using multiple methods.The evaluation results show that the SVR-based regression prediction model has the ability to remove noise in small-sample electromagnetic parameter data sets,which can effectively improve the accuracy of traditional free space method for detecting fly ash carbon content.