XRF element quantitative analysis method based on AOG-BP neural network

Through the BP neural network optimized by alternating generations (AOG-BP), the problem of mutual interference of element peaks caused by the background effect in XRF element quantitative analysis was solved, the accuracy and efficiency of element quantitative prediction were improved, and the operation process was simplified.

CN115541641BActive Publication Date: 2025-09-23YANGTZE DELTA REGION INST OF UNIV OF ELECTRONICS SCI & TECH OF CHINE (HUZHOU)
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202210992195.7
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-08-18
Publication Date
2025-09-23
Estimated Expiration
2042-08-18

AI Technical Summary

Technical Problem

In the existing technology, the mutual interference of element peaks caused by the background effect makes the inversion prediction of element content in XRF element quantitative analysis more difficult. The optimization algorithm introduces random numbers and conditional branch structures, making the convergence process cumbersome and inefficient.

Method used

The BP neural network with alternating generations optimization (AOG-BP) is adopted. By alternating between the whale optimization algorithm and the particle swarm optimization algorithm, and combining the fitness function of the determination coefficient R2 and the absolute value of the error, the weights and thresholds of the BP neural network are optimized to avoid fast convergence and falling into the local optimum.

Benefits of technology

It improves the accuracy and efficiency of element quantitative prediction, simplifies the operation process, and realizes efficient element component value inversion and quantitative analysis.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN115541641B_ABST
    Figure CN115541641B_ABST
Patent Text Reader

Abstract

The present invention discloses an XRF element quantitative analysis method based on AOG-BP neural network, which is to optimize the weights and thresholds of BP neural network by using AOG in the field of element quantitative analysis of X-ray fluorescence spectrum element detection technology, and use the optimized weights and thresholds to construct BP neural network to invert the content of the element to be tested, and finally use the determination coefficient R 2 The prediction effect of AOG-BP was evaluated. The method is simple and easy to understand, requires minimal pretreatment, and has a fast prediction speed and high accuracy. It can quickly and effectively predict the elements contained in the analyte.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Coal seam gas content prediction method based on PSO-BP model and seismic attribute parameters

    CN111045080A

  • XRF element quantitative analysis method based on SSA-BP neural network

    CN114839210A