Research On Real-time Dynamic Monitoring Algorithm Of Landslide Based On Multi-system GNSS

China is one of the countries with serious geological landslide disasters in the world.Scientific and effective disaster early warning technology is an important measure to protect people's lives and property.By monitoring the displacement of the landslide body,the steady state of the landslide can be reflected,which is a common method for landslide monitoring.GNSS-RTK technology can directly acquire the three-dimensional coordinates of surface landslide monitoring points.It has the advantages of high precision,automation and all-weather.It has become the main technical means of landslide monitoring.The solution pressure caused by multi-system combination positioning,environmental disturbance around the landslide body,and observation noise have a great influence on the positioning accuracy.Therefore,the actual engineering application puts forward higher requirements for the landslide monitoring algorithm.Based on the theory of relative positioning of BDS and GPS,this paper focuses on the research of key algorithms such as multi-system star selection,noise attenuation and multi-path modeling.The main research results and innovations of this paper are as follows:(1)In this paper,an improved six-satellite selection method considering the height angle threshold is proposed in view of the deficiency of the traditional star selection method in practical application.In the BDS/GPS combined real-time dynamic positioning solution,the root mean square error and operation time are taken as the measurement accuracy and efficiency standards.The height angle recommended threshold of the improved six-satellite selection method is determined by numerical experiments,and the improved six-satellite is compared.The method and other star selection methods locate the results at baselines of different lengths.The results show that compared with the traditional non-selected star strategy,the improved six-satellite selection method has the same plane positioning accuracy,the elevation direction positioning accuracy is increased by 40%,and the solution time is increased by 70%.The improved six-satellite selection method can effectively weaken the influence of low-altitude satellites on BDS/GPS combined positioning.Under the condition of ensuring no loss of positioning accuracy,the efficiency of understanding and calculation can be significantly improved,and the accuracy and timeliness of real-time positioning can be considered.In addition,the improved six-satellite selection method proposed in this paper is also applicable to other modes of GNSS real-time combined positioning.(2)Aiming at the complex noise that exists in landslide monitoring and it is difficult to eliminate by modeling or differential method,a denoising method based on EEMD method is proposed.Through the measured landslide observation data,the advantages of EEMD method in reducing modal aliasing phenomenon compared with EMD are verified,and the denoising efficiency and reliability are improved.The results show that compared with the EMD method,the EEMD method can effectively reduce the occurrence of modal aliasing,improve the automation of data processing,and is more suitable for complex monitoring environments.The EEMD method is used in both the double-difference observation range and coordinates.The denoising of the domain is slightly improved,and the standard deviation is improved by 12.3%,46.9%,and 10.1% in the E,N,and U directions,respectively,compared with the unde-noised results.The noise increased by 8.8%,9.5%,and 8.7%,respectively.Therefore,the EEMD method is used to synthesize the denoising methods of different data fields,which can effectively reduce the influence of random noise and instantaneous strong noise,and accurately describe the displacement variation of the landslide body,and provide reliable coordinate change information for the subsequent establishment of multipath error model.(3)After a short baseline difference,most of the error is eliminated or weakened,but multipath errors still exist,which is a major factor limiting short baseline accuracy.Generally,the landslide monitoring environment has stability,so the multipath error has a certain periodicity.In this paper,based on the EEMD method to denoise the original data,the stellar-day filtering method is used to extract the multi-path error establishment error model of the previous time period,and compensated to the corresponding later time period result according to the periodic characteristics.in.The experimental results show that the RMS of E,N and U in the three directions after deducting the multipath error are 0.22 cm,0.18 cm and 0.68 cm,respectively,and the lifting ranges are 89%,92% and 87%,respectively.The accuracy is obviously improved.(4)In order to study the influence of earthquake on landslide,this paper uses GNSS-RTK technology to compare and analyze the time series of displacement change of monitoring points on loess landslide and earth landslide before and after Jiuzhaigou earthquake.According to the change of relative position of monitoring points before and after earthquake,the effect of earthquake on landslide is analyzed..The results show that in the natural state without earthquake triggering,the daily coordinate change of the loess landslide in Nanyang of Xiangyang is 1~2mm,and the daily coordinate change of the landslide of Ziyang landslide is about 1.5cm.The impact of the landslide on the annual accumulation of 221 days is up to 221 days.5mm,the amplitude is 2.5 times of the normal change,and then sinks for 5 days at 2mm/d;the abnormal rise of the Ziyang landslide reaches 2.5cm on this day,the amplitude is 1.5 times of the normal change,and then continues at 2cm/d.Shock for 5 days.After the earthquake,the oscillating intensity of the landslide increased obviously,and the slope body fluctuated abnormally,which was consistent with the theory of the progressive failure effect of the slope caused by the earthquake.Based on GNSS-RTK technology,it can effectively monitor the acceleration and even the collapse of landslide caused by vibration in real time,and provide decision-making basis for landslide warning and landslide emergency response after earthquake.