At present,railway transportation is still the main form of transportation in China.China Railway is the world’s number one in terms of passenger turnover,cargo delivery volume and transportation density.With the long-term and high-load operation of railways,the rails will inevitably be weak.Cracks and other defects,these have buried a huge hidden danger for the safe operation of the railway,so there is an urgent need for a reliable and accurate rail detection equipment to ensure the safe operation of the railway.In this paper,a complete rail crack detection system is designed.The system can realize the detection of rail crack defect signals by using new acoustic emission technology and neural network model.The main work of this paper is as follows:First of all,this paper first built a rail crack detection experimental platform,which simulates the wheel-rail contact state under real conditions,so that the signal data in the real environment can be obtained in the laboratory environment,and the platform control software is designed.Control the relative running speed between the wheel and the rail;secondly,design and implement the crack acoustic emission data acquisition software for the PCI-2 acoustic emission instrument,through which the communication can be communicated with the PCI-2 acoustic emission device,and the waveform stream signal is realized.The data is collected,and the waveform stream data is parsed and converted by the analysis of the data protocol.Finally,in order to detect the collected signal and observe whether it contains defects,this paper designs and implements the rail crack detection software based on QT platform.The software can directly read and preprocess the parsed signal data,and then send it to the trained neural network model for detection,and the network’s predicted output is processed by clustering and sliding window to obtain the final result. |