Research On Micro-LIBS Identification Technology And Methods Of Natural Gold Associated Minerals

Gold is a precious currency metal,and it has an irreplaceable role in modern industrial production.Gold exploration has always been highly valued by countries around the world.In natural gold deposits,gold is often associated with a variety of minerals such as sulfides and silicates.It is an important prospecting method to obtain ore information by identifying and analyzing gold associated ores.Probes,electron microscopes,and other micro-area analysis instruments have advantages in determining the composition of mineral elements and geological analysis,but these instruments often need to be carried out in the laboratory,which is difficult to meet the needs of on-site analysis.Laser Induced Breakdown Spectroscopy(LIBS)technology,which has in-situ and rapid advantages,can analyze almost all elements,and it has important application potential in mineral micro-area analysis.This article focuses on the key issues that limit the application of portable LIBS instruments in the detection of natural gold associated ores,and studies core technologies and methods such as micro-area analysis autofocus,on-site rapid identification,and instrument miniaturization to improve the practicality of LIBS technology in this scenario value.In order to realize LIBS micro-area detection,this article first introduces the micro-objective lens into the optical system of the portable LIBS instrument.The micro-objective lens can converge the laser beam finer,thereby achieving a smaller detection range and reducing the interference of impurities on the detection results.However,the introduction of a micro-objective lens will shorten the interval of the laser energy density reaching the excitation threshold.It is very necessary to design a set of fast and accurate focusing methods.In this paper,a two-color laser-assisted combined image contrast focusing module is designed to achieve precise focusing.Compared with the traditional focusing method,the focusing accuracy of this method reaches 20 ?m,the focusing range is expanded to 2400?m,and the focusing speed is accelerated by 2/3,which meets the needs of micro-zone LIBS detectionThe composition of natural mineral samples is complex,and there is a high degree of similarity in the elemental composition of similar samples.How to obtain accurate recognition results from complex spectral data is the focus of research.The classification method of artificial intelligence can make up for the shortcomings of the traditional artificial feature extraction classification method,which can improve the recognition accuracy.In this paper,field-programmable gate array(FPGA)is used to accelerate the convolutional neural network classification algorithm,and the intelligent mineral classification based on LIBS data is realized on the embedded computing platform.Compared with the cortex-A9 processor,the calculation time is shortened by more than two thirds The classification accuracy of this solution is close to that of the PC platform,which greatly improves portability and reduces energy consumption,and is especially suitable for portable LIBS detection instrument applicationsIn order to improve the portability of the LIBS instrument,a miniaturized laser is used,combined with the micro-focusing module to achieve the required energy density to excite the gold associated minerals and reduce power consumption.However,the energy of small lasers fluctuates greatly,which leads to the instability of LIBS spectral signals excited by different laser pulses.In this paper,the energy monitoring module is designed to monitor the energy of a single laser in real time,and the impact of energy fluctuations on the accuracy of classification and recognition is evaluated.The data normalization method is used to reduce the singularity of the data according to the characteristics of the data with energy fluctuations.The recognition accuracy of single-shot laser spectral data of 30 kinds of gold associated minerals has increased to more than 94%.This work is of great significance for classification based on single-shot laser spectral data.Compared with the conventional multiple averaging method,this working mode can greatly improve the endurance of portable instrumentsThe relevant research in this paper is expected to be applied to the in-situ analysis of natural gold associated ore analysis to improve on-site exploration efficiency.