Research On The Inversion Algorithm Of Low-field Nmr Signal

Low-field NMR technology is a new technology developed in recent years for observing and analyzing the physical parameters of samples.This indicates that NMR has moved from high-end molecular chemical structure research and medical inspection services to a wider range of industry and agriculture,such as food,agriculture,mining,chemical and other fields,and play an increasingly important role in the application and scientific research in related fields.Low-field NMR technology focuses on hydrogen nuclei in oils,water,and polymers.After obtaining the resonance signal generated by the hydrogen nuclei of the sample,the sample characteristics or distribution are analyzed and measured after signal processing,thereby indirectly analyzing and analyzing the carrier or environment?such as pore size?.The technology can deeply observe and analyze the internal information of the material without destroying the sample,and has the characteristics of fast,accurate,multi-parameter,no radiation,environmentally friendly.NMR can measure the sample directly,and the obtained relaxation time information can reflect the physical properties of the material.However,the original signal obtained directly from the acquisition is often very complex,so it needs to be converted into an easily understandable spectrum by means of inversion techniques.At present,the commonly used spectrum mainly has one-dimensional?1D?spectrum and two-dimensional?2D?spectrum.The inversion technique mainly discussed in this paper mainly involves its corresponding 1D inversion and 2D inversion technique.The traditional 1D and 2D inversion techniques have different degrees of problems.1D inversion techniques mainly have the phenomenon that the peaks of different components of the inversion spectrum overlap and the problem of difficult to reverse the short relaxation signal.2D inversion technology mainly has the problems of large amount of data,high degree of morbidity and difficulty in distinguishing adjacent peaks.In response to these problems,this paper mainly carried out the following research:?1?1D inversion algorithm:A new method of weighted iterative T₂ spectrum inversion is proposed.By applying weights to the inversion matrix to ensure that the short relaxation component is not eliminated,and a 1D inversion software is designed.?2?2D inversion algorithm:An improved 2D inversion algorithm based on LSQR and L-curve is proposed.The algorithm uses L-curve instead of manual intervention to locate the optimal solution,which improves the inversion precision.An improved 2D inversion algorithm based on BRD is proposed.The improved inversion algorithm is optimized for the selection of regularization factors and slow calculation speed of BRD algorithm.A new method based on PDCO for 2D inversion is proposed.The algorithm can simultaneously use L1 regularization and L2 regularization.The neighboring narrow peaks are well distinguished.Based on the 1D inversion software,the 2D inversion function is added.The simulation experiment and sampling experiment results show that the 1D and2D inversion methods proposed in this paper have good robustness and accuracy.These methods not only meet the needs of low-field NMR inversion technology in practical applications,but also promote the development of low-field NMR industry.