| China’s "Chang’e 4"(CE-4)and "Chang’e 5"(CE-5)are equipped with lunar penetration radar(Lunar Penetrating Radar,LPR)and lunar soil detection respectively.Radar(Lunar Regolith Penterating Radar,LRPR)to achieve in-position detection of the lunar surface on the moon.The lunar exploration radar system uses electromagnetic waves for detection,and can non-destructively reveal the geological structure and structure of the near-lunar surface.Its high-resolution and multi-level information acquisition capabilities make it a cuttingedge technology for exploring the moon,and has great advantages in the field of lunar exploration.Significance and significant advantages.The study of lunar radar data helps us to deeply explore the evolution rules,medium composition and geological evolution process of the moon,thereby providing key clues for further understanding of the formation and changes of the moon.Whether it is assisting in landing site assessment,searching for potential resources,exploring the evolution of the solar system,or promoting the in-depth scientific research,lunar exploration radar has played an indispensable role.As the world’s first spacecraft to achieve soft landing and patrol exploration on the far side of the Moon,the dual channel single offset lunar exploration radar system carried by Chang’e-4 provides humanity with an opportunity to gain a deeper understanding of the near lunar surface structure of the Feng Kamen impact crater on the far side of the Moon.Its main task is to land on the surface of the far side of the Moon,explore geological and resource information on the far side of the Moon at a deeper level,and improve the lunar archives.The Chang’e-5 probe,due to undertaking China’s first lunar sampling and return mission,is equipped with a multi offset high-resolution lunar weathering layer penetration imaging radar installed on the lander platform.Its main task is to provide information support for drilling and sampling processes,while measuring the thickness of the lunar weathering layer and underground structures on land.The key scientific issue of this article is how to use the lunar exploration radar data of Chang’e-4 and Chang’e-5 as the research objective,obtain accurate geological structures from lunar soil to shallow lunar surface through data processing,denoising,imaging inversion,and infer their evolution based on the evolution history of the moon.On this basis,combined with the existing scientific research status,based on the different radar systems and scientific objectives carried by Chang’e-4 and Chang’e-5,the following four specific scientific questions have been refined:(1)How to use the electrical parameters of lunar soil and rocks obtained by predecessors when facing multiple lunar radar systems and their corresponding detection targets,Consider multiple geological factors to establish an electrical medium model that can reflect different types of lunar exploration radar systems;(2)How to further denoise and image the data from the Chang’e-4 lunar exploration radar in the existing data processing process,and use the lunar exploration radar data to analyze and explain the geological structure and evolution process of the Feng Kamen impact crater;(3)How to develop an inversion method that is suitable for lunar soil detection radar,in order to obtain the electrical properties and structural structure of lunar soil,in the face of limited data volume and low signal-to-noise ratio of Chang’e-5 lunar soil detection data;(4)how to quickly and accurately identify rocks in lunar soil using data from the Chang’e-5 lunar soil detection radar in the face of the drilling and sampling task in the lunar exploration plan,in order to help the drilling and sampling task proceed smoothly.Based on the above scientific issues,this paper has completed the following research work on the Chang’e-4 lunar exploration radar data and Chang’e-5 lunar soil detection radar data:(1)Based on different radar systems,a multi-scale comprehensive electrical model of the moon from the lunar soil to the shallow lunar crust was established.In the process of establishing the model,random functions were used to construct the terrain,abrasive technology was used to create irregular stones,fractal methods were used to construct random media,and the influence of the lunar soil dielectric constant changing with depth was considered.The established model can better simulate the real electrical changes in the near-lunar surface and shallow lunar crust.The forward modeling results calculated by it contain more details and can provide a basis for the processing and interpretation of lunar exploration radar data.(2)Preprocess the Chang’e-4 lunar exploration radar data and Chang’e-5 lunar soil detection radar data,and then propose a denoising method based on the gray wolf optimization algorithm and the variational mode decomposition method for the Chang’e-4lunar exploration radar data.method.The gray wolf optimization algorithm is used to determine the best parameter combination based on the measured data.After performing variational mode decomposition on the lunar radar data,the spectral feature information is used to reconstruct the lunar radar image with a high signal-to-noise ratio.This method not only has efficient denoising capabilities,but also has the ability to optimize manual interaction and qualitative judgment,which can provide a good foundation for subsequent radar image processing and imaging.(3)According to the plane wave separation method,the diffraction data and reflection data in the Chang’e-4 radar data are separated,and then velocity continuous extension and focusing analysis are used to obtain the landing site velocity profile and migration imaging results.According to the proposed migration imaging process,not only can clearer radar images be obtained to locate underground anomalies,but it can also provide a certain basis for the interpretation of the velocity structure and electrical structure of the underground medium.Through processing the high-frequency channel data of the Chang’e-4 lunar exploration radar,it was found that the lunar soil is divided into two categories according to the relative dielectric constant value.When processing the low-frequency channel data of the Chang’e-4 lunar exploration radar,the relative dielectric constant changes of the lunar soil layer obtained are consistent with the results obtained from the high-frequency channel lunar exploration radar data.Among them,the high dielectric constant area is inferred to be a basalt layer.The obtained velocity model can provide a basis for stratigraphic structure interpretation.(4)Use Chang’e-4 lunar exploration radar data to reconstruct the geological events and evolution process of the landing area.Based on high-frequency channel lunar exploration radar data,the stratigraphic structure within 40 meters of the near-lunar surface was revealed.Affected by weathering,the spatter and local materials under the lunar soil interacted to form a complex near-lunar surface geological structure.At the same time,it was proposed that this area There may be events where meteorites hit the surface.Based on low-frequency channel lunar exploration radar data,the established stratigraphic structure of the Chang’e-4landing site was refined,and the geological activity events experienced by the Von Kármán impact crater were explained.The stratigraphic structure of the Von Kármán crater is affected by various geological processes,including basaltic magmatism,ejecta coverage,and the interaction of volcanic eruption debris and lava.The resulting stratigraphic structure and its interpretation are consistent with known major events.consistent.These results provide important scientific basis for understanding the structure of the far side of the moon.(5)Use multi-frequency time domain full waveform inversion with total variation regularization constraints on the Chang’e-5 lunar soil detection radar data to obtain the electrical parameters of the lunar regolith.During the inversion process,the Wiener filtering method is used to decompose the lunar soil detection signal into different frequency components,so as to recover the long-wavelength information of the model parameters from the low-frequency data,and then extract more details and features from the high-frequency data.Then,the total variation regularization constraint was added to the model update to effectively suppress the influence of the antenna radiation pattern,random noise and metal signal scattering on the inversion calculation.Based on the results of the inversion of the measured data,we first analyzed the abnormal stress conditions encountered during the drilling process,and then proposed a stratigraphic model below the Chang’e-5 sampling point,and based on the geological evolution of the Chang’e-5 landing site,different layers were analyzed.bit to explain.(6)An intelligent rock positioning algorithm based on deep network was proposed for the Chang’e-5 lunar soil detection radar data,and the feature extraction and nonlinear capabilities of the convolutional neural network were used to extract rock information from the lunar soil data.First,the training set for the Chang’e-5 lunar soil detection radar is synthesized,and then the degradation network and the prediction network are used to synthesize an overall closed-loop network model for rock block prediction.Through network training and prediction,the goal of quickly identifying the location and characteristics of rock blocks based on lunar soil detection radar data was achieved.This model was applied to the processing of Chang’e-5 measured data,and the shape and location of rocks contained in shallow lunar soil were discovered,as well as the distribution information of rocks in the lunar soil that changes with path and depth.This method can quickly locate rock blocks on lunar soil detection radar data and provide methods and technologies for Chang’e-6 lunar soil collection. |
PDF Full Text Request