Analysing GNSS network data for deformation monitoring in multi-scale, multi-temporal civil engineering applications

变形监测是制定土木工程基础设施复原力和可持续性战略以及防治地质灾害措施的关键要素。全球导航卫星系统站点网络提供了一种监测工程结构运动的手段，能够在几秒到几年的时间尺度上探测到局部(例如单点)和较大规模(例如全国)的结构运动。全球导航卫星系统网络数据分析的大多数战略基于：(1)全球导航卫星系统站速恒定；(2)对全球导航卫星系统数据周期信号进行建模；(3)采用最小二乘法调整测量误差。这种战略是确定性方法，采用特定的条件/约束(例如恒定速度)来模拟全球导航卫星系统站的行为。此外，对每个全球导航卫星系统站点进行单独分析，而不检查附近和更远的其他全球导航卫星系统站点的行为，在这些站点上可能发生类似的行为。这一研究项目将使用有待开发的技术，在分析全球导航卫星系统网络数据时采用确定性较低的方法(例如人工智能)，并采用将全球导航卫星系统数据视为网络一部分的空间分析方法。假设这些办法能够提取更多关于全球导航卫星系统数据行为的信息，这些数据要么与全球导航卫星系统特定的站有关，要么与全球导航卫星系统网络的一个区域有关。该项目的目的是制定一种方法，对全球导航卫星系统网络数据在时间和空间两个领域进行分析，并确定各站的全球导航卫星系统数据作为单独的全球导航卫星系统站和作为网络的一部分的行为。这一方法将应用于各种工程项目的变形监测情景；(I)多尺度项目，从小型土木工程结构(例如桥梁)到包括地质灾害(例如地面运动)在内的较大规模；以及(Ii)多时间变形，从短期变形(例如桥梁振动)到长期变形(例如地面沉降)。

Deformation monitoring is a key element to develop strategies of resilience and sustainability in civil engineering infrastructure and measures against geohazards. A network of GNSS stations provides a means to monitor the motions of engineering structures and enables both localised (e.g. single points) and larger scale (e.g. national) structural movements to be detected, over time scales of seconds to years. Most of the strategies in GNSS network data analysis are based on (i) constant GNSS station velocities, (ii) modelling of GNSS data periodic signals and (iii) least squares methods to adjust measurement errors. Such strategies are deterministic approaches, where specific conditions/constraints are applied (e.g. constant velocity) to model the behaviour of the GNSS stations. Furthermore, each GNSS station is analysed individually, without examining the behaviour of other GNSS stations, both nearby and further afield, at which similar behaviour may be occurring. This research project will use to-be-developed techniques, following less deterministic approaches (e.g. artificial intelligence) in the analysis of the GNSS network data and adopting spatial analysis methods which treat the GNSS data as part of a network. It is hypothesised that these approaches could enable the extraction of more information regarding the behaviour of the GNSS data, either related locally to a GNSS-specific station or related to a region of the GNSS network. The aim of this project is to develop a methodology where the GNSS network data will be analysed in both the time and space domains and identify the behaviour of the GNSS data of the stations, as individual GNSS stations and as part of the network. This methodology will be applied to deformation monitoring scenarios from a broad range of engineering projects; (i) multi-scale projects, from small scale civil engineering structures (e.g. bridges) to larger scale including the field of geohazards (e.g. ground motion) and (ii) multi-temporal deformations, from short-term (e.g. bridge vibration) to long-term deformation (e.g. land subsidence).