During the service life of tankers, deterioration of the material occurs. Overloading and corrosion due to environmental factors inevitably lead to a deterioration in performance. If the damage is not detected, it can spread and lead to catastrophic failure. This undetectable mal-function causes changes in measured values and process responses.
Because data from various process sensors are required for the ship's system management, measurements in the cargo tank of an LNG carrier are critical. The increasing use of computers and the tighter economic and technological limits imposed by plant regulations present challenges to the design of a modern and accurate sensor for the process variables relevant to measurement processes.
In the report on the analysis of maritime accidents, issued by the European Maritime Safety Agency (EMSA), it is stated that "human actions" account for approximately 78 percent of maritime accidents. The presented data are derived from events reported in the EMCIP and have been investigated, revealing that about 8 per-cent of accidents were directly linked to technological deficiencies. In the reports on the investigation of maritime accidents, 44 analysed accidents are associated with failures, of which 3.7 percent (21 accidents) were caused by grounding, 2.1 percent (12 accidents) by collision, and loss of control/contact in 1.9 percent (11 accidents).
To define or explain the occurrence of a failure, one must know its cause and the magnitude of the event. Outcomes that emerge during the identification of the nature or cause of a fail-ure are crucial in detecting and locating the source of the problem that arises in the operation of sensors, as well as determining why the issue occurred. Undetected phenomena can impact the results of the measurement process or all segments of automation, or only a specific segment. A comprehensive analysis presented in the thesis enhances the significance of under-standing the processes in the cargo tanks of tankers and the process variables as their key parameters. The essence of the work lies in defining the most common failures of measuring sensors in order to detect and identify the source of random errors. The occurrence and in-creasing frequency of such errors strongly indicate sensor failures within the controlled process.
|
