High-performance Distributed Fiber Optic Monitoring And Condition Assessment Methods For Infrastructure

Infrastructures are of vital importance to support a nation's development.Moreover,infrastructures are featured with long service life,harsh working environment,complicated failure modes and significant loss once crush.To ensure safe operation of infrastructures,during the past 30 years,structural health monitoring techniques and methods were proposed and studied by worldwide researchers including monitoring structural responses,detecting damage and degradation,assessing structural safety conditions.Conventional structural monitoring technologies are based on discrete sensors which have an inevitable disadvantage of limited sensing locations,especially when monitoring large scale transportation infrastructures.Long distance sensing technologies with highly densed sensing points and affordable cost are much more preferred in these circumstances.At the same time,structural health monitoring methods base on distributed sensors need to be developed.In this thesis,high performance distributed fiber optic sensors are employed to develop advanced structural health monitoring technologies for infrastructures.Taking advantage of their super long sensing range and highly densed sensing points,full length structural behaviours can be obtained.This thesis is aimed at developing new methods for infrastructure monitoring and condition assessment.The specific sections are listed as follows:(1)Measurement characteristics of the Brillouin time domain analysis system(BOTDA)are studied and signal processing methods of feature extraction and evaluation for distributed structural behaviours are proposed.Firstly,the effects of sensing fiber's non-uniform Brillouin frequency shift distribution within the spatial resolution distance of a BOTDA system on the shape of the measured Brillouin gain spectrum are studied by simulating the stimulated Brillouin scattering process.Secondly,the measured structural response distributions are proposed to be categorized into two types,i.e.the regional structural behaviour and the local structural behaviour.For each type,feature extraction and characterization are carried out respectively to provide insight into the structure's behaviours from different angles.Moreover,a pattern matching method for evaluating the similarity between two structural response distributions is proposed.Finally,a structural abnormality recognition method is proposed by considering the response values and the distribution patterns of both the regional and the local structural behaviours.(2)Strucutral condition assessment methods for large scale bridges base on full length strain distributions measured by distributed fiber optic sensors are studied.A distributed strain monitoring system for a large suspension briges using Brillouin distributed fiber optic sensors is developed and the spatial distribution characteristics of strain along the full length of the main grider of the suspension bridge are analysed.A bridge structural condition assessment method is proposed by considering both the strain values and the distribution patterns of the main grider during a load test.A method for calculating displacement distribution of the grider based on strain distribution is proposed.The above methods were applied and validated in a large suspension bridge in China.(3)Settlement monitoring methods for subgrades using distributed fiber optic sensors are proposed.Firstly,the coupling mechanisms of subgrade and embedded strain sensing fibers are studied,the effects of subgrade settlement on strain distributions of sensing fibers are analysised.A subgrade settlement monitoring and assessment method which can cover long distance is proposed.The local strain distributions of the embedded strain fibers are utilized to detect non-significant settlement and the regional strain distributions are utilized to detect significant settlement.A settlement displacement measuring method using fixed point strain fibers is proposed.Secondly,a settlement monitoring system of a permafrost subgrade was designed and built.The feasibility of permafrost subgrade monitoring using distributed optical fiber sensors was verified using field monitoring data.Finally,short-term and long-term strain distributions of the subgrade were collected,the spatial distribution characters of the longitudinal strain distributions of the subgrade are are analysed.Settelement of the subgrade is assessed by the method proposed and the validity of the method is verified.(4)Railway track bed monitoring method using self-sensing sleeprs is studied.Self-sensing prestressed concrete sleepers with embedded fibre optic sensor arrays were developed and manfactured by a standard industrial production line.Prestress loss of the sleepers during manufacturing process and before installation were monitored.Damage detection abilities of the self-sensing sleepers in terms of crack detection and identifying loss of ballast support at the rail seat locations are studied.(5)Distributed dynamic strain monitoring method using distributed fiber optic sensors is studied.A fiber optic high spatial resolution distributed dynamic strain sensing system,namely SA-DPP-BOTDA,is developed and successfully applied to a pile shaking table test and a suspension bridge field monitoring test.Sensing characteristics and accuracy of the SA-DPP-BOTDA system is studied.