| There are many methods can detect the flaws of the steel channels,in which the magnetic flux leakage testing method has more prominent advantages. It has very sensitive detection results for steel flaws. The highest advantage of this detection method is it has high detection rate which the characteristics most needed by the industrial automation lines. By this method quality control of the detection is safe and reliable to meet production requirements. Therefore, magnetic flux leakage detection method becomes the most common way among so many detection methods.The research comes from commissioned projects, the task is to develop a monitoring system for steel magnetic flux leakage pipeline. This system includes a data acquisition board and a PC monitoring software. First,existing magnetic detection monitoring system's shortcomings and problems are analyzed in detail. Then, the problems of the existing system are illustrated and a set of more effective data transmission scheme is proposed for the problems. By this scheme ultra-long-distance signal transmission can be achieved and the data acquisition board and PC monitoring software are developed. Data acquisition board mainly achieves the uploading of the detection signal from pipeline acquisition and downloading of monitoring system's controlling signal. The PC monitoring software is developed by Visual Studio C#2010 and MySQL database tool and can achieve data processing, displaying, recording, querying, pipeline controling and so on. The interface of PC monitoring software is achieved by multi-UART card which can open up several buffer areas and achieve the receiving and processing of the data parallelly. Therefore, it can increase the utilization of CPU and speed up the data processing which can meet the high real-time requirement of the monitoring system. Old magnetic flux leakage monitoring system can't meet the requirement of technical indicators during processing the test data and a lot of invalid test data appears especially the transverse test data. The designed method of data collection, integration, processing can effectively solve these problems and achieve the desired results.Because of mechanical perturbations in the magnetic flux leakage pipeline detection system and voltage disturbances of the excitation coil to the total flux, the detection results will be influenced seriously and it is very unfavorable to judging the flaws of steel. The wavelet thresholding method is usually used in the field of signal immunity, however in the method the pseudo-Gibbs phenomenon can't be avoided which limits its application. To solve the problem of pseudo-Gibbs, in this paper a new de-noising method shifting mean wavelet threshold denoising method is proposed which can reduce the error between data and estimates and achieve better approximation. By this method, the filtered data curve becomes more smooth and pleasing. The fast algorithms of this method are In-depth studied, which is more suitable to practical signal processing in the engineering. In addition, a magnetic flux leakage signal processing wavelet compression method is designed to solve the problem that the amount of the magnetic flux leakage detection signal is not suitable to transmit and record.After data compression which reduces the data storage space, the characteristics are maintained and the impact to the judging of steel flaws is not large.Finally, the paper details the problems of magnetic flux leakage monitoring system encountered during commissioning and gives the final test results. |
PDF Full Text Request