The Study Of Multdimensional Modal Analysis And Vibration Field Inversion Based On ?-OTDR Vibration Data

The full distributed optical fiber sensing technology has also been born with the rapid development of optical fiber communication industry.It combines two advantages of anti-electromagnetic interference characteristics of optical wave transmission and material properties of optical fiber media.There are many different types of technologies and the focus is different.Phase Optical Time Domain Reflection monitors distributed vibration signals.It has technical advantages unmatched by other sensing systems.Therefore,it is very suitable for the technical characteristics of large-scale engineering monitoring.It has good objective demand in the fields of civil engineering,petrochemical,electric power,communications,aviation and so on.It has attracted extensive attention from experts and scholars.In the current ?-OTDR technology research,researches are mainly focused on laboratory or simulation environments.The foci of the research are more on the development of acquisition instruments and the identification of event signals.They lack the combination with practical engineering applications.This situation affects the better application of?-OTDR technology.Therefore,the article is based on the combination of theory and engineering.The vibration signal collected by ?-OTDR technology is analyzed and combed.In practical engineering applications,?-OTDR technology has some problems such as Lack of Distributed Expression,Fiber State Interference and Data Expansion.The summary mainly includes the following three aspects.(1)The vibration signal collected by ?-OTDR technology contains two dimensions of time and length.However,the current modal analysis of vibration signals is only for vibration signal analysis of the Time Axis.The study does not involve vibration mode analysis of Length Axis and Combination of Length Axis and Time Axis.(2)Optical fiber exists as a sensing unit in ?-OTDR technology.However,the current research does not consider the phenomenon that the fiber itself has pre-stress,pre-temperature and thermal coupling.The phenomena can cause physical effects such as tensile compression,thermal expansion and contraction on the optical fiber.It indirectly affects the vibrational modes of distributed signals.(3)With the development of ?-OTDR technology,the distance of monitoring is longer,the spatial resolution of monitoring is smaller,the frequency of monitoring is smaller,and the monitoring time is longer.These factors will cause the ?-OTDR monitoring data to grow geometrically.Data expansion seriously affects the analysis efficiency of distributed signal vibration modes.Simply improving hardware performance is far from meeting the needs of technological development.According to the key problems in practical application of the above ?-OTDR technology,the research content of this paper is mainly divided into four parts.Firstly,Theoretical Analysis of Vibration Mode and Vibration Field Inversion in ?-OTDR.It provides theoretical support for subsequent research.Secondly,Research on Multidimensional Element Solution of Distributed Vibration Mode.It solves the multidimensional features of the analyzed distributed vibration modes.Thirdly,Optimization Research on Key Elements of Vibration Field Inversion under Thermal Decoupling.Aiming at the key elements of inversion proposed in the first part,the influence of prestress,pre-temperature and thermal coupling is considered.Thermal decoupling is performed on the existing solution method.Forthly,Research on Anti-data Expansion Technology under Compressed Sensing.The computational efficiency and compression storage space are greatly improved by compressing the swell data in ?-OTDR.1.Theoretical Analysis of Vibration Mode and Vibration Field Inversion in ?-OTDR.The distributed vibration signal in ?-OTDR technology is a two-dimensional signal.There are three indicators of Natural Frequency,Damping Ratio and Modal Mode on the time axis.The important indicator Signal-to-Noise Ratio was introduced by considering the complex noise situation in actual engineering.The ?-OTDR signal is characterized by a sampling function on the length axis.Therefore,two important indicators Scope and Vibration Point are obtained according to the position of the symmetry axis and the distance between the zero crossings in the SA signal.There must be deviation between the calculated value and the actual value in the joint calculation of the two dimensions Time Axis and Length Axis.This deviation can be measured by another important indicator of Time and Space Difference.On the basis of vibration modal analysis of ?-OTDR technology,the vibration field inversion is based on vibration Source point,Scope and Vibration Data.The source point determines the center of the vibration field.Scope determines the range of vibration.The Vibration Data is distributed in a circular shape which the Source Point is the center and Scope is the diameter.2.Research on Multidimensional Element Solution of Distributeed Vibration ModeAccording to the above,the multi-dimensional elements of the vibration mode of the?-OTDR distributed vibration signal are obtained.The paper proposes the corresponding solution process which consists of three parts.The first part is the research on extracting signal method based on 2D-TESP,2 Dimension-Time Encode Signal Processing,method.It is compatible with the first and second derivatives of ?-OTDR distributed vibration signals.It also guarantees algorithmic computational efficiency.The second part is the research on signal separation method based on GAEMFNMF,Genetic Algorithm-Empirical Mode Decomposition-Nonnegative Matrix Factorization,method.It guarantees the complete signal model of the ?-OTDR signal in the Time and Length dimensions.The third part is the signal optimization method based on the Kalman-PID,Kalman Proportional Integral Differentiation,method.It will significantly improve the separation of the signal.3.Thermal Decoupling of Key Elements in Vibration Field InversionIn order to be compatible with different pre-stress and pre-temperature on the fiber,the two key indicators of the Vibration Source Inversion and the Scope proposed in the second part are better optimized in the ?-OTDR technology.Therefore,the paper proposes a thermal decoupling method.It uses the BOTDR,Brillouin Optical Time Domain Reflection,method to acquire the pre-stress and pre-temperature of the optical fiber.It uses the GA-RBF,Genetic Algorithm-Radial Basis Function,method to perform black box training on pre-temperature,pre-stress and thermal coupling data.The 3D-SURF,3 Dimension-Speeded Up Robust Features,method to match data for time,length,and thermal coupling.The fused method finally achieves the effect of thermal decoupling.4.Research on Compressive Sensing of Expansion Data in ?-OTDRIn order to solve the data expansion problem caused by the increase of"sampling rate/resolution"or"time/length"of ?-OTDR technology on the basis of retaining the key elements,the paper proposes a method of picture compression sensing according to the above research.It performs compressed storage of reserved features for ?-OTDR data.Because the ?-OTDR signal and the image signal are both two-dimensional signals,Imagem mode storage of the collected distributed vibration signals greatly reduces the memory sizee of the stored signals.At the same time,in order to determine the corresponding downsampling ratio and verify whether the saved image retains the signal characteristics,the paper uses hybrid features combined with particle filter method for feature recognition.The hybrid features contains texture features,shape features and self-extracting features.This method not only retains the key elements of vibration field inversion,but also prevents data xpansion.According to the above research contents,the main innovations of this paper are summarized as follows.1.This paper presents a multi-dimensional modal element analysis and solution method based on-OTDR distributed vibration data.It solves the current state of modal analysis limited to single-point vibration in existing research.The modal multidimensional elements of distributed vibration data in technology include Natural Frequency,Damping Ratio,Modal Model,Signal to Noise Ratio,Scope,Source Point,and Space Difference.In the Solution method,the first step is the signal extraction based on 2D-TESP method.The second step is using the GAEMDNMF method for signal separation based on the acquired signals.The last step is using the Kalman-PID method for signal optimzation of the separated signals.It finally achieves the purpose of distributed expression of ?-OTDR signal vibration modes.2.The paper proposes a thermal decoupling method of key elements in vibration field inversion.It solves the problem that thermal coupling affects the key elements of vibration field inversion.The method introduces BOTDR technology to analyse the stress and temperature of fiber based on ?-OTDR technology.Then it uses the GA-RBF network for thermal coupled feedback.Lastly it uses the 3D-SURF method to match the three-dimensional data composed of ?-OTDR signal and thermal coupling.It finally achieves the thermal decoupling effect of the key elements in vibration field inversion.3.It proposes a method of anti-data expansion under compressed sensing.It solves the problem of data anti-expansion of ?-OTDR technology in practical applications without relying on hardware improvement.The method realizes two aspects of Image Storage of?-OTDR Data and Particle Group Image Feature Recognition of Mixed Features.It not only compresses the storage space,but also retains the key features of the ?-OTDR signal.The organizational structure of this paper is divided into six chapters according to the above key issues,research contents and innovations.The first chapter is the Introduction.It introduces the source,background,purpose,significance,research status,innovation,main content,organizational structure,etc.of the paper.It is a general overview of the article.The second chapter is Theoretical Analysis of Vibration Mode and Vibration Field Inversion in O-OTDR.It combs a theoretical research system for the study of subsequent chapters.The third chapter is Research on Multidimensional Element Solution of Distributed Vibration Mode.It solves the key problem of distributed vibration mode expression based on the theoretical analysis in second chapter.The forth chapter is Thermal Decoupling of Key Elements of Vibration Field Inversion.It solves the influence of thermal coupling on?-OTDR vibration signal in vibration field inversion.The fifth chapter is Research on Compressive Sensing of Expansion Data in ?-OTDR.It conducts compressive sensing of expansion data and greatly improves the calculation effiency and storge space on the basis of retaining the key elements of vibration field inversion.The sixth chapter is Summary and Outlook.It is a summary and outlook of the article as a whole.