| Laser Induced Breakdown Spectroscopy(LIBS)is a new type of spectral analysis technology.This technology is used to analyze the slag composition in molten steel smelting process to explore the feasibility and reliability of LIBS technology for real-time online analysis of slag.Taking the optimization of the experimental conditions of LIBS technology and the quantitative analysis of slag as the main research content,the following work has been done on the application of LIBS technology to detect slag,The basic principles and development status of laser-induced breakdown spectroscopy technology were introduced,and the basic principles of quantitative analysis;the laser collection position,focus position,and acquisition delay of the experimental system were optimized to improve the signal-to-noise ratio of the spectral signal.This article briefly introduced several commonly used LIBS calibration analysis methods,analyzing their respective scope of application and the feasibility of the application of LIBS technology in the quantitative analysis of molten liquid slag composition.Through comparison,it is found that the method of establishing a neural network training model and a large number of data samples can be used to train the network to build a reasonable and reliable network model,which can be applied to the quantitative analysis and measurement of slag of unknown analysis objects.The GA-BPANN was used to train the relevant data of Ca to quantitatively analyze the high temperature molten slag.The content of Ca in the slag was calculated,and the calculation results and XRF determination were performed.The results(true values),the training results of BP-ANN,and the results of the free calibration method are compared.The experimental results show that the GA-BP-ANN model can better determine the content of Ca in the slag,and the measurement results are more accurate and lower error,it is a LIBS quantitative analysis method with great development potential.It is also proved that using LIBS technology for real-time online detection of slag composition has great development potential.Figure 31,Table 8,Reference 64... |