Research On Damage Identification Technique For Anchor Joint Of Messenger Wire In Electrified Railway Based On Synergetics

With the development of electrified railway in recent years, as the main load-bearing parts of pantograph catenary system, messenger wire can be damaged by natural disasters and other reasons, which takes a heavy toll on people’s lives and properties. And also different types of damages online imag ing monitoring need to separate different modal stress waves. So studying the damage identification method for messenger wire is becoming an urgent problem in structural health monitoring area. Facing the problem of damage identification of messenger wire in the complex environment, a new identification technique for anchor joint of messenger wire in electrified railway based on synergetics is put forward in this paper. It provides a new way of thinking and method for messenger wire damage identification, and will lay the foundation for the structure identification in the other more complex environments. It has important academic and engineering applications value. Research work is supported by the national natural science funding project(51305141), Guangdong province natural science funding project(2014A030313248) and Guangzhou technology plan project. This paper is organized as follow:(1) According to the messenger wire structure, Synergetics was introduced innovatively into structural healthy monitoring and a dynamic cooperative identification method was proposed, the evolution process of damage recognition was studied.(2) A dynamic decision-making mechanism is put forward, the damage identification based on active acoustic emission was researched by Sim Sonic simulation. A 3-D messenger wire model has been established to extract the features from received signals, and a SNN-based damage identification model was established.(3) Aiming at raising the correct rate of recognition, a feedback-based optimization mechanism with feature evaluation method and eigenvector reconstruction approach is introduced. The simulation results show that the new technology can identify different damage types in messenger wire, and different reconstruction approaches can improve recognition rate in varying degrees.(4) Algorithm was designed for dynamic decision and feedback-based optimization mechanism, and according to the algorithm a software was designed, the experimental platform was set up.(5) Experiments were conducted in disturbance and icing environment. The experiment results showed that in disturbance environment, the proposed new method can identify different depths of cracks well, the recognition rate is 100%. And in disturbance and icing environments, recognition rates of damage types reached 91.67% and 85% separately. The identification rate increased when f eedback-based optimization was introduced.