| Magnetic resonance sounding?MRS?can directly determine the position of aquifers,water content,media porosity,and electrical conductivity at a subsurface depth of 0–150 m.This method has been successfully applied in groundwater exploration in China,thereby providing effective technical support to alleviate the drought by specifically searching for underground water sources in arid regions.However,the existing detection technology is incapable of meeting the actual work requirements with the gradual expansion of MRS application to the environment where electromagnetic noises are serious,and problems,such as poor anti-interference capability,easy saturation of amplifiers,and incapability to detect effective signals,are observed.Therefore,a detection method and system that can acquire effective MRS signals in strong noise environments should be developed.MRS signals are susceptible to noise interferences because they are expressed in nanovolts.The detection of effective signals from noises has constantly been the focus in this field.Currently,a relatively mature set of MRS signal detection steps has been developed that is divided into four processes,namely,low-noise sensor for signal capture,data acquisition system for signal recording,digital filter algorithm for noise suppression,and curve fitting for signal extraction.For the effective signal detection under strong noise conditions,this study analyzes the influence of various types of noises on each detection process and investigates the key technologies for collecting reliable data and extracting effective signals.A real-time anti-saturation detection method for exploring groundwater by MRS in strong noise environments is proposed,and a new MRS system is developed that is verified by field experiment.The main contents of this paper are organized as follows:?1?The magnetic resonance responses of different aquifer structures and the wideband characteristics of MRS signals are investigated,and the influence of different noises on the detection effect of MRS signals is analyzed.On the basis of nuclear magnetic resonance theory,a mathematical model for magnetic resonance responses in the coil is established,and the basic principle and working process of MRS are elaborated.The magnetic resonance response of different depths,thicknesses,and water contents of aquifers is analyzed by a numerical modeling method,which provides a theoretical basis for MRS.The bandwidth of MRS signals is theoretically obtained based on their characteristics with amplitude modulation,which defines the constraint conditions in designing a detection system.Then,the effect of various types of noise on the MRS detection system is investigated to provide the basis for determining the overall solution by classifying the noises in strong interference environments.?2?A real-time anti-saturation sensing technology based on wideband LC matching network is proposed and investigated to improve the system sensitivity and suppress the preamplifier saturation caused by low-frequency harmonic noises.A broadband matching network with a gain of 4.6 dB is constructed using an LC resonance circuit and a passive bandpass filter by combining the characteristics of the pickup coil.An equivalent noise model is established to optimize the MRS sensing system based on the matching network.A new sensor system with real-time suppression of low-frequency harmonic interferences is developed with a bandwidth of 1030 Hz?covering the entire country?and a noise floor density of 1.78 nV/?Hz?1/2@2kHz.The sensitivity of the new system is redefined and evaluated using Allen variance.The test results show that the minimum detectable signal is 3.12 nV.?3?A real-time anti-saturation data acquisition technology based on instantaneous floating-point amplification is proposed and investigated to improve the dynamic range of the MRS acquisition system and inhibit the secondary amplifier saturation caused by peak and harmonic noises.Different gain adjustment techniques are analyzed to meet the requirements of noise mutation characteristics and relative conversion accuracy of A/D.Among the gain adjustment techniques,the instantaneous floating-point amplification technique has the advantages of high-speed regulation and automatic optimum gain selection.Therefore,this study uses this principle to design a data acquisition system with a dynamic range of192.65 dB and a regulation rate of 50 kS/s.On the basis of the instantaneous value of each subsample,the new system adaptively adjusts the gain in real time during the subsample time to ensure that the tail code is between 1/2 full scale and full scale of the A/D converter.To suppress the saturation of the amplifier,the new system also effectively improves the dynamic range of the acquisition system and the relative conversion accuracy of A/D and ensures the reliability of data collection in strong noise environments.?4?A lock-in amplifier based on modeling recovery is proposed and investigated to solve the low accuracy of extracted signals and the incapability to detect effective signals caused by signal losses due to multi-step noise canceling,incomplete filtering,and noise residual in MRS data processing.Although the lock-in amplification technique has a strong filtering effect to detect a single-frequency signal,it cannot be directly used to process the raw data due to the broadband characteristic of MRS signals.Therefore,a mathematical model for MRS signal detection by the lock-in amplifier is established,and a method for restoring the original MRS signals is proposed by analyzing the output results and performing mathematical transformation.Indoor and field test results show that this method can extract high-precision MRS signals from raw data with strong background noises and provide technical guarantee for subsequent inversion interpretations.?5?A new MRS instrument system that integrates signal sensing,data acquisition,and data processing is developed based on the proposed key technologies.The comparison test between the new system and JLMRS-II?a representative MRS instrument developed by Jilin University?shows that the sensitivity of the new system is improved from 5 nV to 3.12 nV,the dynamic range is increased from 144.49 dB to 192.65 dB,and the noise margin in the passband is increased from 20000 nV to 2×106 nV.In addition,the comparison test exhibits the good performance of real-time anti-saturation and the effectiveness of extracted signals for the new system.The developed real-time anti-saturation detection method and system is applied in detecting groundwater in Changchun's suburb.The results show that the new MRS system can effectively work under strong noise interference environments,and the inversion results are identical with the drilling data,which verify the validity and practicability of the new method and system.In summary,this study expands the application of MRS and provides the foundation for groundwater detection by MRS in areas with severe noise interferences in China. |
PDF Full Text Request