Integrated navigation system has strong nonlinear characteristic in dynamic environment, so we propose an optimal multi-sensor fusion algorithm based on sequential UKF to improve the navigation accuracy of GNSS/CNS/SINS integrated navigation system. Firstly, we establish the nonlinear state equation and the linear measurement equations of two subfilters for GNSS/CNS/SINS integrated navigation system. Then, by simplifying the measurement updating process of the standard UKF algorithm, we design a simplified UKF algorithm that has the same filtering accuracy as the standard UKF algorithm and has characteristics of low computation. Finally, we propose the sequential UKF optimal fusion algorithm for multi-sensor integrated navigation system, by combining the sequential filtering algorithm with the simplified UKF algorithm. The simulation results show that the sequential UKF algorithm not only improves the real-time performance of the system, but also has higher filtering accuracy than the conventional centralized Kalman filter algorithm and the classical centralized linear UKF algorithm.

The attitude combination of celestial navigation/Strapdown inertial(CNS/SINS) integrated navigation system can make the attitude angle converge and restrain the divergence of position and velocity effectively. To improve the accuracy of the integrated system, we design the UKF algorithm of CNS/SINS integrated navigation system. Firstly, this paper simplifies the measurement update process of UKF algorithm and reduces the computation of UKF, based on the establishment of nonlinear state equation and linear measurement equation of CNS/SINS integrated navigation system. Then, we propose one calculation method of attitude covariance matrix in system state covariance matrix based on platform angle error, which derives the expression of platform angle error vector corresponding to prediction error of attitude measurement value in UKF algorithm, and then establishes the UKF algorithm of CNS/SINS integrated navigation system. Finally, the simulation results show that the proposed algorithm can significantly improve the accuracy of navigation parameters of the integrated navigation system as compared with the linear Kalman filter algorithm and EKF algorithm, and that the proposed algorithm has high robustness to the change the initial attitude angle error of the filter.

This paper studies the random model of Huawei P30 smartphone positioning with GPS/BDS/GLONASS/Galileo and other multi-system. It focuses on the evaluation of pseudorange noise when using multi-system mobile phone observations and analyzes the influence of different random models on the pseudorange single point positioning accuracy of mobile phone. The results show that the satellite pseudorange noise of different systems of Huawei P30 smartphone varies greatly. The mean square error of pseudorange noise of GPS, BDS, GLONASS and Galileo are 5.30 m, 2.75 m, 7.92 m and 1.07 m, respectively. When using the random model of signal-to-noise ratio+inter system weighting, the positioning performance of mobile terminal is the best. Compared with the traditional altitude angle model, the pseudorange single point positioning accuracy in E, N and U directions is improved by 36.12%, 25.79% and 31.30%.

This paper analyzes the influence of antenna phase center offset models and the combination of new and legacy frequencies on the BDS-3 precise orbit determination and positioning. The results show that generally the orbit determination and positioning based on the CSNO PCO model are slightly better than the IGS protocol model.Compared to the effects caused by the differences between CSNO and IGS PCO models, the selection of combination frequencies of B1C/B2a and b1I/b3I has a more significant impact on precise point positioning. In addition, taking the IGS B1I/B3I PCO model as a reference, the accuracy of positioning coordinates in E, N and U component using the CSNO B1C/B2a PCO model are improved by 5%, 13% and 14% respectively, so it is recommended to be used in BDS high-precision data processing.

We are concerned with the problem of gradient vanishing and ease of falling into local extremum of ordinary neural network. To enable the neural network prediction model to be more accurate in prediction, we apply the global optimizing feature of multi-verse optimizer(MVO) to retrieve the reliable neuron threshold and connection weight between each layer of BP neural network. We build the prediction model of GNSS height anomaly fitting based on the MVO-BP method, then we carry out the feasibility test of the algorithm by adopting a limited amount of height anomaly data in practical engineering. The results show that MVO-BP method is more accurate and versatile than the conventional BP neural network method and the multifaceted function method, and it has a certain reference value for the acquisition of normal height in practical engineering measurements.

We propose a combined prediction model of building deformation based on fractional Fourier transform(FrFT) and support vector machine(SVM). First, FrFT is used for multi-scale analysis of deformation time series, and the complex time series is decomposed into a series of subsequences with relatively simple structure. Then, we use SVM to establish prediction models for each subsequence, and the final prediction results are obtained by comprehensively superimposing the prediction results of each subsequence. At the same time, considering the difficult problem of parameter selection of SVM model, we propose an improved fruit fly optimization algorithm(IFOA) to globally optimize it and improve the prediction performance. Taking the measured deformation data of a concrete dam in Southwest China as an example, the results show that the proposed combined prediction model can fully mine the trend and regularity information hidden in the data and obtain high prediction accuracy.

The performance of the traditional gross error detection algorithm for GNSS deformation monitoring is limited by the data length, so we propose a first-order derivative gross error elimination method based on wavelet analysis. First, we use the new algorithm to remove gross errors from the original signal, and then use generalized continuation interpolation to supplement the residual defects. The actual measurement results show that the execution time of the new algorithm is 0.01 times that of the 3-times medium-error method (3σ method), and the deviation from the real value after interpolation is only 0.03 mm, which is much better than other traditional gross error detection methods.

This paper analyzes the capability of SWOT to monitor SLV in South China Sea using numerical simulation observations and ECCO2 model data. Based on the SWOT simulator, we comprehensively analyze the simulation errors of SWOT, in which phase error and roll error are dominant errors; generally the farther away from the nadir points, the larger the error. The root mean square error between SWOT and the truth value can be reduced from 4 cm to 2 cm after eliminating the error by spatial smoothing, which improves the reliability of the results. Further analysis indicates that the spatial pattern of SWOT error is mainly affected by the along-track observation mode and is independent of the original SLV signal amplitude. Then, we further analyze one track of SWOT with grid cell sizes of 10 km, respectively. The geostrophic current results show that the simulated results with a resolution of 10 km agree with the output of the ocean model better, while the result with 2 km interpolation is significantly affected by noise.

Using mobile gravity observation data in Jiangsu from 2019 to 2021, we analyze the variation characteristics of the gravity field at different time scales before the Changzhou M4.2 earthquake in Jiangsu. The results show that: 1) The 0.5 a scale gravity field change image better reflects the change process of this earthquake’s decline-turn up-continuous rise-reverse change, and the evolution of seismogenesis near the zero contour; 2) The 0.5 a and 1 a scale gravity fields before the Changzhou M4.2 earthquake both show that the earthquake occurred near the zero-contour line of gravity change.

In order to investigate the influence of GRACE-FO K-band ranging system and laser ranging interferometer on the time-variable gravity field model, during January 2019 to August 2021 the unconstrained 90 d/o monthly gravity field solutions APM_KBR and APM_LRI are derived by dynamic methods. In terms of degree variance from the gravity field model and the spatial distribution of equivalent water height, the accuracy of the two types of ranging data is at the same level under the condition of complete and good quality. However, satellite maneuvers or other satellite events in that month will lead to missing LRI data, resulting in a slightly lower signal-to-noise ratio of time series of the equivalent water height obtained by APM_LRI model in that month. From the perspective of post-fit residual, the RMS of APM_LRI is about 50% lower than that of APM_KBR, and the post-fit residual of LRI is significantly lower than that of KBR measurement at high frequency (≥60 CPR). In general, under the existing conditions, due to the influence of background models as well as other factors, the degree variance and the equivalent water height of gravity field from the two kinds of ranging data are basically the same, but the LRI ranging data has lower noise in the post-fit residual. In other words, LRI ranging data can be directly applied to detect and analyze tiny geophysical signals in the future.

Based on GRACE/GRACE-FO gravity satellite data from 2004-2021, we invert the spatial and temporal changes of terrestrial water storage in the Yellow river basin and construct both a drought index model and a flood factor model to analyze and study the extreme climate phenomena in the Yellow river basin. The results show that the terrestrial water storage in the Yellow river basin decreased at a rate of 0.56 cm/a during 2004 to 2021 considering cyclical seasonal characteristics, showing a surplus in summer and autumn and a deficit in spring and winter. The drought index model monitored 22 extreme droughts and 37 severe droughts in the Yellow river basin, and the scope of drought events covered the entire Yellow river basin. The flood factor model detected 118 flood events in the Yellow river basin, with floods mostly occurring in summer and autumn when rainfall was more abundant, during which the capacity of terrestrial water storage of the Yellow river basin was weaker. The meteorological results detected by the drought index model and flood factor model constructed with GRACE/GRACE-FO gravity satellite data are consistent with actual physical observations, reflecting the climactic conditions occurring in the Yellow river basin and providing a favorable tool for the study of extreme climate.

We study the observational data of the geoelectric field in northeast China and combined with the Songyuan M_S5.7 earthquake on May 28, 2018, in Jilin. At the same time, we comparatively analyze the background, imminent and post-earthquake response time series changes of the natural electric field E_SP and the telluric field E_T around the epicenter. In the study area, we calculate the dominant direction of charge movement by applying the charge movement seepage model of rock fissure water. The results show that the natural electric field E_SP in Suihua, Lindian, Zhaodong, and Wangkui show obvious small-amplitude sudden jumps or large-scale jumps before the earthquake, and 45° extrusion or tension anomalies feature occurred in the dominant direction of charge movement within 3 months before the earthquake.

In order to explore their seismogenic structures and their genesis, based on the observation data of the Beijing-Tianjin-Hebei Seismic Network, we use the CAP method to invert the focal mechanism solutions and centroid depths of the Luanzhou M_S4.3 earthquake, the Changli M_S4.2 earthquake and their M_S3.3 aftershocks. We use the local seismic depth phase sPL to obtain more accurate moments. According to the depth of the center, combined with the double-difference positioning method, we obtain the source distribution results of the two sequences, and analyze the seismogenic structure and genetic correlation. The results show that: 1) The strike, dip and slip angles of nodal plane Ⅰ of the Luanzhou M_S4.3 earthquake are 211°, 85°, and 168°, respectively, and nodal plane Ⅱ are 302°, 78°, and 5°, respectively. It is a strike-slip type, the focal depth is 8 km, the focal distribution of the earthquake sequence is NNE, and the short-axis section seismic distribution shows a near-vertical dip and a high dip angle. It is considered that the seismogenic fault plane is a nodal plane Ⅰ. The nodal plane Ⅰ of the aftershocks are 189°, 68°, 161° and 190°, 61°, 170° respectively; the nodal plane Ⅱ are 286°, 72°, 23° and 285°, 81°, 29°, respectively. The focal dislocation type is both strike-slip type, the focal depth is 10.5 km, the focal distribution of the earthquake sequence is NNE direction, the short-axis section shows a near-vertical dip, high dip angle, and the seismogenic fault plane is considered to be the nodal plane Ⅰ. 2) Based on the source parameters of the Luanzhou M_S4.3 earthquake, the Changli M_S4.2 earthquake and the Luanzhou M_S7.1 earthquake, combined with regional geological structural data, we believe that the seismogenic structures of these three earthquakes are not preexisting in the shallow crust. It may be related to the low-velocity layer in the crust in the source area, and the interlaced inclusions of high-velocity and low-velocity bodies in the crust are the common basis for these earthquakes.

To explore the response law of ground motion from surface to interior in different elevations of thick mountain, based on Lengzhuguan mountain, we establish a combination model of terrain and weathering medium using discrete element software. We input wenchuan seismic wave signals from the bottom boundary to reveal the internal and external dynamic response law of the slope on both sides of the mountain. The results show that, as elevation increases, the acceleration amplification coefficients in the body and surface of the side slope of Daduhe first increase, decrease, and then increase rapidly, reaching the maximum value near the top of the slope. The slope surface of Wasigou side is affected by topographic relief and the acceleration amplification coefficient of convex slope is enlarged and the concave slope is decreased. At different elevations, the acceleration amplification coefficient decreases gradually with the increase of slope depth from the surface to the interior, and the amplification curve tends to be gentle when it is 150-200 m away from the slope surface. As the elevation increases, the acceleration amplification coefficient decreases slowly from the outside to the inside, and the gentle convergence depth of the amplification curve increases. The wave velocity of rock mass medium decreases with the increase of rock mass differentiation and the resonance effect increases the acceleration response. Meanwhile, the combination effect of slope topography and medium makes the peak acceleration amplification coefficient of slope surface near 2.0.

Based on remote sensing interpretation, field investigation, optically simulated luminescence(OSL) dating and UAV topographic mapping, we carry out a preliminary study on the development characteristics and deformation of river terraces in Tongcheng-Taihu section at the eastern foot of Dabie mountain. The results show: 1) The river terraces in the eastern foot of Dabie mountain are poorly developed; the first-class terraces are generally developed in small and medium-sized rivers, and the second-class terraces are developed in a few large rivers. 2) The formation age of T1 terrace of the river in the eastern foot of Dabie mountain is from the late Pleistocene to the early Holocene(about 25 ka BP~12 ka BP), and the formation age of T2 terrace is the middle of late Pleistocene(about 55 ka BP). 3) There has been no obvious activity at the eastern foot of Dabie mountain of Tanlu fault zone since the formation of T2 terrace of the river. The latest active age of the fault in this section is the middle Pleistocene.

Based on 1 009 Sentinel-1A images, we monitor land subsidence in the middle route by SBAS-InSAR. The results show that the land subsidence along the middle line is mainly distributed in the southeast of Hebei province, and the maximum deformation rate is -139 mm/a. There is a certain distance from the channel, which has little influence on water transport. The land deformation rate of Beijing is -133 mm/a, while that of southwest Tianjin is -81 mm/a, which is relatively small. However, as the Tianjin branch line passes through two subsidence areas, relevant departments should pay attention to it. The time-series deformation characteristics of land subsidence in Beijing area are analyzed; combining it with the analysis of relevant data, it can be concluded that the South-North Water Diversion project has effectively replenishment of groundwater reserves in Beijing, and significantly restrained the development of land subsidence in Beijing.

In this study, we use 31 Sentinel-1 images from January 2019 to December 2020 to interpret the current development of Longyao fissures and their impacts on the surrounding areas. Based on SBAS-InSAR and Stacking InSAR techniques, we obtain the time series and rate maps of deformation of Longyao fissures and their surrounding areas, and further study the current slip status of Longyao fissure using the homogeneous elastic space model. The results show that: 1) the stepwise difference of deformation rate between the north and south sides of Longyao ground fissure in 2019 to 2020 is 4 cm/a, which is lower than that of 5 cm/a in 2007 to 2011. 2) There is obvious periodic surface deformation in Guchengdian, Weijiazhuang and Guanzhuang town, but the general trend is still subsidence. 3) The modeling results show that the depth and slip rate of Longyao ground fissure are 0km and 27 mm/a at the present stage. Compared with the results from 2007 to 2011, both depth and slip rate are decreased. It indicates that Longyao ground fissure is still active, but activity has decreased.

By calculating RMS, PSD, and PDF value at 1 to 20 Hz band, the current study explores the suppressive effect on background noise, based on the comparison of underground, tunnel and surface seismological observed data from seismic stations Chifeng and Erenhot. The findings show that the underground observation has a significant suppressive effect of background noise in comparison with the tunnel and the surface over 2 Hz frequent bands, especially higher than 9 Hz frequent bands. The suppression of underground method would weaken when the frequency is less than 16 s. However, it disappears at 16 s to 2 Hz band. All three methods share a high level of coherence of background noise in the meantime. In conclusion, we can reduce background noise effectively through underground observation, so that the data quality will be optimized.