Application of a neural network model with multimodal fusion for fluorescence spectroscopy

In energy-dispersive X-ray fluorescence spectroscopy,the estimation of the pulse amplitude determines the accuracy of the spectrum measurement.The error generated by the amplitude estimation of the pulse output distorted by the measurement system leads to false peaks in the measured spectrum.To eliminate these false peaks and achieve an accurate estimation of the distorted pulse amplitude,a composite neural network model is proposed,which embeds long and short-term memory(LSTM)into the UNet structure.The UNet network realizes the fusion of pulse sequence features and the LSTM model realizes pulse amplitude estimation.The model is trained using simulated pulse datasets with different amplitudes and distortion times.For the pulse height estimation,the average relative error of the trained model on the test set was approximately 0.64%,which is 27.37% lower than that of the traditional trapezoidal shaping algorithm.Offline processing of a standard iron source further validated the pulse height estimation performance of the UNet-LSTM model.After estimating the amplitude of the distorted pulses using the model,the false peak area was reduced by approximately 91% over the full spectrum and was corrected to the characteristic peak region of interest(ROI).The corrected peak area accounted for approximately 1.32%of the characteristic peak ROI area.The results indicate that the model can accurately estimate the height of distorted pulses and has substantial corrective effects on false peaks.

Reactive transport modeling constraints on the complex genesis of a lacustrine dolomite reservoir:A case from the Eocene Qaidam Basin,China

Reactive transport modeling(RTM)is an emerging method used to address geological issues in diagenesis research.However,the extrapolation of RTM results to practical reservoir prediction is not sufficiently understood.This paper presents a case study of the Eocene Qaidam Basin that combines RTM results with petrological and mineralogical evidence.The results show that the Eocene Xiaganchaigou Formation is characterized by mixed siliciclastic-carbonate-evaporite sedimentation in a semiclosed saline lacustrine environment.Periodic evaporation and salinization processes during the syngeneticpenecontemporaneous stage gave rise to the replacive genesis of dolomites and the cyclic enrichment of dolomite in the middle-upper parts of the meter-scale depositional sequences.The successive change in mineral paragenesis from terrigenous clastics to carbonates to evaporites was reconstructed using RTM simulations.Parametric uncertainty analyses further suggest that the evaporation intensity(brine salinity)and particle size of sediments(reactive surface area)were important rate-determining factors in the dolomitization,as shown by the relatively higher reaction rates under conditions of higher brine salinity and fine-grained sediments.Combining the simulation results with measured mineralogical and reservoir physical property data indicates that the preservation of original intergranular pores and the generation of porosity via replacive dolomitization were the major formation mechanisms of the distinctive lacustrine dolomite reservoirs(widespread submicron intercrystalline micropores)in the Eocene Qaidam Basin.The results confirm that RTM can be effectively used in geological studies,can provide a better general understanding of the dolomitizing fluid-rock interactions,and can shed light on the spatiotemporal evolution of mineralogy and porosity during dolomitization and the formation of lacustrine dolomite reservoirs.