In order to investigate the spatial distribution, internal structure, and boundary contact relations of target granite rock at Inner  Mongolia Alxa pre-selected high-level radioactive waste geological  repository site, seismic detecting work is carried out. Near-surface velocity tomography inversion profiles and seismic reflection profiles are obtainedby acquiring and processing the seismic data.This seismic exploration work aimsat technical problems such as ruggedness surface work area near-surface velocity model-building, tomographic statics correction, and strong seismic scattered wave pressing. To this end, we adoptakind of MSFM(multi-stencils fast marching) based first-arrival travel-time tomography inversion technology, and an improved vector resolution noise removal approach.Meanwhile, based on routine high resolution reflection seismic exploration data processing flow, a set of customized fine data processing flow is set up.

 Using digital waveform data recorded by regional seismic networks in Yunnan between 2011 and 2015, attenuation characteristics and site response respectively are calculated. The frequency dependent Q(f) is obtained by the iterative grid-search technique, as described by Atkinson and Mereu, based on trilinear geometrical spreading model. Our results demonstrate that for Yunnan the associated model for regional quality factor for frequencies can be expressed as  Q(f)=193.8f0.528. Additionally, we ascertain the source spectra determined by the model of Brune and the site responses of seismic stations derived by Moya’s method using genetic algorithms. The site response results of 46 stations demonstrate that the seismic signal amplification of Station bedrock is not the same in different frequency bands. The site response in Yunnan region shows obvious amplification with low frequency and attenuation with high frequency. Site amplifications are apparent in excellent frequency band (1~4 Hz), mostly between 1 and 10 times, and in the high frequency (10 Hz or more) site attenuation is mostly between 0.05 and 1 times.

With data from the Ruichang-Yangxin magnitude 4.6 earthquake sequence recorded by the digital seismic network of Jiangxi and its adjacent areas, we relocate the sequence, study the relatively large earthquake’s focal mechanism, discuss the spatial distribution feature and seismogenic structure of the sequence and discuss the relationship between two moderate-strong earthquakes in this region. Our results show that the strike of the Ruichang-Yangxin magnitude 4.6 earthquake sequence is  approximately NE 60°, that the main seismogenic structure is the Longtang-Pijiashan buried fault, which is a subfault of the Tanlu fault. Furthermore, the Jiujiang magnitude 5.7 earthquake sequence and the Ruichang-Yangxin magnitude 4.6 earthquake sequence have a close relationship in spatio-temporal distribution.

On March 21th, 2008, a MS7.3 earthquake occurred in Yutian of Xinjiang. In this paper, InSAR technology is used to obtain the coseismic and postseismic deformation. The inversion of coseismic slip distribution shows that the largest slip achieves 5.4 m and that the slip is dominant in the south of the fault, near 0-5 km depth beneath the ground surface. The earthquake is a normal faulting event with a slight sinistral strike-slip component. Postseismic deformation indicates that there are differences in movement in the north section of seismogenic fault, and the largest cumulative deformation difference is about 15 cm in 782 d after the earthquake. According to the further analysis, the afterslip of postseismic fault may be the main mechanism for deformation. The degree of postseismic afterslip is relatively small in two years and the slip region decreases greatly and for the most part is located in shallow areas. The energy of north fault is relatively exhaustive released, while there is little energy remaining in south part of the fault.

The Lajishan area is an earthquake tectonic window to reflect on tectonic activity and earthquake activity. The Lajishan fault has caused about 20 moderate earthquakes on the both sides of Lajishan mountain in documented history. Lajishan fault is still an active fault zone, and has structural conditions of medium-strong earthquakes in the region. The Lajishan area is at 3 500 m above sea level and is not convenient for study. It remains doubtful whether the Lajishan regional crustal deformation is still in the uplift state. This paper uses 1983-1995 leveling data, combined with 1999-2007 and 2009-2014 GPS data, to analyze the three dimensional crustal deformation of this area. Research shows that the west section of Lajishan fault has a vertical rate of 1±0.5 mm/a ,which proves that the area is still in the uplift state. GPS data also show horizontal compression in this area. The Lajishan area remains a crustal horizontal shortening deformation. The crustal shortening rate of the region has become smaller, and shows an obvious left-lateral strike-slip component after 2008. The results are corroborated by satellite imagery interpretation and field geological survey results. Under the control of extrusion environment, tectonic deformation in the area is still ongoing. This area is also influenced by the Riyueshan right-lateral strike-slip fault. We should be concerned and put attention to seismic risk at the western end of Lajishan fault.

Comparative analyses is made to the precipitable water vapor (PWV) with hourly intervals derived from 9 GPS stations of the Chinese coastal GPS network, using the observation data from radiosonde, ground-based water vapor radiometer (WVR). The results show the annual correlation coefficient of precipitable water vapor between ground-based GPS and radiosonde is above 0.95 and the mean bias gradually increases from north to south. The annual average deviation of the GPS stations (Xisha is excluded) is less than 2 mm and RMSE is less than 3 mm with evitable seasonal change. Compared with ground-based water vapor radiometer, the correlation coefficient reaches an even better value of 0.989. During the synchronous observation period, their mean deviation is 1.84 mm and the RMSE is 2.06 mm with maximum values occurring in July, August and September.

In this paper, the accuracy of Bevis empirical formula, GTm-Ⅱ and GTm-Ⅲ models is validated using the sounding data in year 2014 provided by integrated global radiosonde archive. Furthermore, the variation characteristics of the accuracy for these three models with latitude and season are also analyzed. Results show that the overall precision of the GTm-Ⅲ model is superior to the other two models on a global scale. The MAE and RMS values for the GTm-Ⅲ model are 3.26 and 4.10 K, respectively in global scale. It also can be found that these three models show a higher accuracy in low-middle latitude regions than at high latitude regions and that the precision of Bevis empirical formula and the GTm-Ⅱ model display a hemispherically asymmetric pattern. The seasonal variation characteristics of the accuracy for the GTm-Ⅱ, GTm-Ⅲ models and the Bevis formula present consistent variable trends and the RMS differences between four seasons are at range of 2~3 K.

Based on GNSS data from the crustal movement observation network of China (CMONOC), we extract ionospheric delay from levelled carrier-phase observations, deduct the satellite hardware delay, then attach stochastic models of VTEC and DCB; these apply the semi-variable function as the spatial constraint of VTEC and the experienced VTEC and receiver hardware delay as the periods constraint, thus setting up the regional ionospheric grid model in real time. The results show that the inner precision of the model is about 0.3 m and the residuals are within 1 m. Taking GIM, which is the IGS ionospheric delay product, as reference, the characteristics of sequence diagram of VTEC are in accordance with GIM for a week. Furthermore, receiver hardware delay, with variations within 1.5 ns each day, are relatively stable. Compared with the regional station receiver hardware delay isolated from the original observation by using satellite hardware delay and spherical harmonic coefficient of IGS, the mean differences for a week are all within 2 ns.

An improved strategy for the prediction principle based on single difference values of satellite clock bias (SCB) is proposed. This paper analyzes prediction characteristics of the frequently-used linear polynomial model, quadratic polynomial model and grey model when they use the proposed prediction principle. The simulation results show that the proposed improvement for the prediction principle is effective and the prediction precision of the frequently-used models can get better prediction results based on the proposed prediction principle in the short-term SCB prediction.

In order to improve satellite clock bias prediction, a new prediction method is proposed considering periodic errors and stochastic characteristics. First, the given satellite clock bias is fitted by four quadratic polynomial models with one to four dominating periodic errors; the best model is then chosen to obtain the fitting residuals. Then, the prediction of the fitting residuals is modeled based on grey model, considering the stochastic characteristics of the fitting residuals. Finally, the clock bias based on the best of the four models and prediction of the fitting residuals are combined to obtain the ultimate prediction result. The precise data of satellite clock bias within 15 min from IGS are used to conduct experiments on different models. The results show that the model with two dominating periodic errors is better than the model with other dominating periodic errors and that the proposed model performs better than commonly used models in short-term prediction.

For the issue of rank deficiency in single-frequency single-epoch ambiguity resolution, a regularization matrix constraining coordinate correction parameters is constructed by singular value decomposition on coefficient matrix of double-differenced carrier phase observation equations, in order to improve the ill-condition of normal equation. Single-epoch ambiguity fixed solutions are then acquired using the LAMBDA method. Compared with fitting method of weights, the results indicate that for a 5.8m baseline, the ambiguity resolution success rate of BDS is increased by 17.61%, while the GPS, BDS and GPS/BDS of 2.34 km are increased by 4.67%, 3.56% and 3.63%, respectively. In addition, positioning precision in the East and North reaches the millimeter and millimeter to centimeter levels in Up , when the cut-off elevation angle of satellites is taken as 10°.

GPS is the main means for dynamic deformation monitoring of large-span bridges. The signal of GPS contains all kinds of ‘noises’ and the true signal can be extracted after removing the noise. The EMD and wavelet de-noising method is proposed and optimized, and the EMD de-noising and the wavelet threshold de-noising methods are compared. The results show that the optimized EMD and wavelet de-noising method is effective in dynamic deformation monitoring of large-span bridges, and the de-noising effect is best.

 Based on deep analysis of the optimization process of the basic fruit fly optimization algorithm, this paper supports an improved fruit fly optimization algorithm (IFOA) for search processing of a single direction. The IFOA method can process the nonlinear function that has nonzero and nonnegative extreme points. Based on this advantage, IFOA method is applied to parameter estimation of a nonlinear model. Analysis results of a practical example show that estimation accuracy of the IFOA method is superior to the linear approximation method and the nonlinear iterative method. Compared with intelligent search methods represented by a genetic algorithm, estimation accuracy is nearly equal. In addition, the IFOA method has several obvious advantages, including fewer parameter settings, ease of finding the best one, and easy programming.

In order to improve predictive accuracy ofthe multi-point grey model, background values are selected by automatic optimization weights such that the residual sum of squares of the actual and fitted values is minimal. Caseresults show that, compared with traditional multi-point grey model, prediction accuracy is greatly improved with the background values optimized grey model.

Based on penalty function and weight of adjustment, an inequality constraints EIV (ICEIV) model is presented. The model utilizes penalty function to construct the constraint weight for the constraint equations and transforms the inequality constraint into the equality constraint by the zero or infinite weight. So, it can transform the inequality constraint adjustment criteria into the classical adjustment criteria. Therefore, a new iteration algorithm of total least squares with inequality constraints is deduced by the nonlinear least squares adjustment theory; the method uses a structured matrix to consider the repetitive elements and constant terms.

In the Tidal Generating Potential (TGP) development process, the geodetic coefficient is normalized in order to maintain the relative stability of its value among different degrees and orders. At present, there are three main normalization methods: Doodson normalization, Cartwright & Tayle normalization, and Hartmann & Wenzel normalization. The specific formulas of these normalization methods in TGP development are derived from the basic theoretical formula of TGP, along with analysis of the associated Legendre’s functions and their full normalization forms. On this basis, the transformation relationships and coefficients among the three methods are obtained. At the same time, the specific values of the 2~6 degree and order normalization factors in Doodson normalization are given, while three errors in the treatise of Doodson and Roosbeek and in the IERS 2003,2010 Conventions are pointed out and corrected.

Aimed at solving the problem of quasi-geoid controlled by linear or strip of GPS-leveling points, Coons surface interpolation model is put forward. The first step is to synthesize GPS-leveling points fitting curve, and then the curve is used to construct the surface, as far as possible to reduce the traditional fitting method by point directly on the accuracy of consumption. Respectively, using the least squares method, least square collocation method, moving interpolation and Coons surface method to fit quasi-geoid controlled by linear or strip of GPS-leveling points, the result showed that the interpolation model based on Coons surface algorithm accuracy is the highest.

In this paper, we analyze the principle of tidal and atmospheric pressure effects of the water level, focusing on solving the problem of multicollinearity between the input variables. The Partial Least Squares Regression (PLS) model is used to eliminate the tidal and atmospheric pressure effects, and the results show that the response coefficients can correctly reflect the physical processes of tidal force and atmospheric pressure fluctuations. Both the magnitude of the tidal and atmospheric pressure coefficients, and the hysteresis and delay constants of atmospheric pressure, are closely related to the parameters of the aquifer (i.e., storage, porosity, permeability, bulk modulus, etc.),the crustal stress, and strain state around the borehole. Through continuous 60 days (Step size is 9 days) calculation of Chuan 03# well, the seismic and co-seismic responses, the adjustment processes after earthquakes, and even precursor anomalies are clearly identified in the tidal and atmospheric pressure response coefficients.

This paper compares the observation data of different borehole strain meters of Taian fiducial seismic station. We analyze the self-testing reliability of component borehole strain. The correlation of component borehole strain meter and volume strain meter in the normal state and during drawing water conditions is also studied. The correlation coefficient between S1+S3 and S2+S4, two orthometric components of RZB-3, is 0.999 9. The correlation coefficient between S1+S3 and S2+S4 of RZB-3 and volume strain is above 0.998. However, there is poor correlation when the instruments are disturbed by pumping.

A novel detection approach using curvature index of modal flexibility curvature and its curvature damage diagnosis index is proposed in this paper. The proposed method is based on the unique character of temperature propagation and distribution in structure; the temperature disturbance can be eliminated by differentiating a second time for flexibility curvature. A physical interpretation is demonstrated for the proposed method based on cointegration theory. The validity and effectiveness of the proposed method are illustrated through a simple supported beam simulation, and a statistical averaging technique is utilized to improve the noise resistance of these damage indexes. Thus, the proposed method has the potential to be implemented in practical applications.