Identification Methods Based On Fractional Differentiation And Cellular Automata For The Electrochemical Corrosion Damage Of Steelbar

Reinforced concrete (RC) structure is one of the most important structural styles atpresent and in the long future. Corrosion of the reinforcing steel is the major factorwhich deteriorates the durability of RC structrues。The electrochemical essencedetermines that the corrosion monitoring online could be achieved effectively anddirectly based on electrochemical. To realize the corrosion monitoring based onelectrochemical theory ultimately, some key scientific problems in the recognitionmethods of the electrochemical corrosion characteristics should be broken throughpreviously. Specifically, integrated and fast identification approach of the generalcorrosion features where dispersion and diffusion effects are considered in theequivalent circuit (EC) should be established, and the quantitative prediction of thegrowth and evolution of the pitting based on the electrochemical noise (EN) intrinsicinformation should be explored.From the structural health monitoring (SHM) perspective, the author deploys theresearch to resolve the key scientific issues mentioned above for the on-lineelectrochemical corrosion monitoring of RC structures. The conjunction of theoryinvestigation, numerical simulation and experimental verification had been applied here.Firstly, the algorithm for the response in time domain of the transfer function ofelectrochemical corrosion system has been established based on the complex variablefunction approximation. Then the influence of the series elements in the complex EC onthe response in the time domain with the transient potentiostatic step excitation has beeninvestigated. Secondly, Fractional Derivative (FD) theory had been applied to establishthe integrated and fast identification approach to recognize the electrochemicalcharacteristics of the complex EC containing dispersion and diffusion elements in timedomain. Then, the FD algorithm has been verified numerically based on the results ofcomplex variable function approximation. Furthermore, the features of EN with coupledcomplex conditions for pitting corrosion have been investigated in detail based on theWavelet Energy Spectroscopy (WES) algorithm. Finally,3-D Cellular Automata（CA）model drived by the real-time monitoring information of EN for the growth andevolution prediction of pitting corrosion has been explored, and its accuracy has beenverified by the3-D Ultra-depth Video Microscope experiments.Corrosion monitoring of the reinforcing steel is bound to provide the solidscientific basis for the safety assessment, the rational scheme of corrosion control and maintenance, and the life circle design based on performance of the significant RCstructures. This paper paves the theoretical way to realize the on-line corrosionmonitoring of the major RC structures.