PSI - Issue 16

Borys Paton et al. / Procedia Structural Integrity 16 (2019) 176–183

181

Borys Paton et al. / Structural Integrity Procedia 00 (2019) 000 – 000

6

Such a mode, on the one hand, allows easily finding on the control computer hard disc the data for the required time, and guaranteeing reliable storage of information, reducing the load on computer processor and RAM. On the other hand, the system is sufficiently reliable for acquisition and storage of significant amounts of information in one file. One of the indices of system reliability is providing the computer capacity for processing and storage of larger amounts of data than at standard tests. Therefore, several tests were performed, the longest of which envisaged continuous analysis of information and its saving in one file during 24 hours, and several more files of more than 10 h duration. Fig. 4 shows some elements of monitoring program window, after 1 h of pipeline monitoring in the working condition at system initial standard settings. Bars shown in the diagram indicate the locations of AE event concentration in each monitored cluster. Numbers on the flags show the quantity of AE events in each cluster, flag colour shows the danger level of the fracture process, running in the cluster. Danger of the situation in a certain pipeline section at this moment of time is shown by colour above the cluster diagram in the prediction window in a special table for the three antennas, monitoring the entire pipeline system. Coordinates of the most dangerous locations in the entire pipeline of 240 m total length (by antennas 1, 2) and the predicted destructive pressures in these pipeline areas, are shown. Antenna 3 monitors pipeline section from 12Kh1MF steel 20 m long of 377 mm diameter with 17 mm wall thickness. a b

Fig. 4. AE activity of pipeline in working condition 1 h after the start of monitoring at initial system settings (a), without filter application on antenna 3 (b).

The lower part of the displays shows instantaneous values of continuous emission, recorded by each monitoring AE transducer. Emission intensity is ranged by bar height and colour. Note that the data in Fig. 4 were obtained before setting optimization. In this connection, the presented information, including the readings of indicators and results of fracture prediction, was considered as the initial stage for further system calibration. Obtained results are important by the availability of AE data proper and fundamental possibility of providing danger indication, and failure prediction, so as to use this as a basis to perform the required correction of system settings in such a way that the results would correspond to physical state of the controlled material. Fig. 5 schematically shows part of the pipeline monitored by antennas 1 and 2 (run #1 and 2 of 120 m length each). Transducers from 1 to 14 are shown. Given below is the conditional distance scale from 0 to 100 m in each section. Data given in Fig. 5 were obtained after detailed study of the influence of interference and technological noise on monitoring results. Proceeding from the data of monitoring and high-temperature testing of samples, data were obtained on the most effective methods of filtering by the parameters of recorded AE events, and then the respective filters were applied to the already acquired data. Necessary correction of fracture prediction settings was also performed. Data given in Fig. 5 are quite demonstrative, as AE recoding with appropriate analysis and prediction of fracture was performed for quite a long time that served as additional check of system performance as a whole, at rather large incoming information flows, as well as correctness of selected settings.