PSI - Issue 28

G.N. Gusev et al. / Procedia Structural Integrity 28 (2020) 2328–2334 Author name / Structural Integrity Procedia 00 (2019) 000–000

2332

5

and connected to the PC via RS-485/USB AC4 interface adapters ("OVEN”). The equipment was interrogated using OpenSCADA 0.8 packages (Debian 8 OS). In order to minimize the distortion of the sensor signal, the input signal units were located in close proximity to the strain control points and connected with shielded cables. The measuring equipment was also protected against moisture and dust. The choice of these software products is determined by two factors: a) software programs are distributed freely, and b) the OPC server software provided by the survey module vendor is not able to poll more than one channel per device with the available 4 channels at the time of creation of the data collection system.

Fig. 4. Sensor locations in zone 2 (with buckling).

The software operates according to the following scenario: 1. Interrogation of sensors every20 s using the Modbus RTU protocol. 2. Transformation of the signals received by sensors into physical quantities (strains). 3. Checking the excess of the critical and warning values of physical quantities specified in normative and design documents. If one of these values is exceeded, a sound signal is sent. 4. Display of sensor readings on the mimic diagram. 5. Storage of sensor data. The 21-day experiment was performed under the supervision of the contracting authority in compliance with the standards GOST R 52728-2007 "Method of full-scale tensometry". In addition, a safety work cycle was carried out. During the experiment, along with the fulfillment of the Construction Regulations SNiP 12-03-2001 "Labor Safety in the Construction (Part 1), General requirements ", SNiP 12-04-2002 "Labor Safety in the Construction (Part 2). Construction" and the safety handbook "Guidelines for the safety of vertical cylindrical steel tanks for oil and petroleum products ", the tank was inspected for cracks and other defects. Also, the operator exercised mimic diagram control over of the values of deformation recorded at the examined joints. Tests were carried out according to the plan described in Table 1. At the beginning of the experiment, the jump signals were registered at a height of more than 5 m. This is due to the fact that the examined zones of the tank took an equilibrium shape under hydrostatic pressure. In the mid-stage of the experiment, the strain values in these joints were normalized, and the absence of sharp increments was registered. This, among other things, has led to the conclusion that the recorded zones of local buckling are the zones where the part of the tank shell loses its local stability, which cannot be reflected in the mathematical model and is, in turn, the most important factor in assessing the initial stress-strain state of the structure. Table 1. Schedule of tests