Eventos Anais de eventos
ENCIT 2016
16th Brazilian Congress of Thermal Sciences and Engineering
DETERMINING COMMINGLING VOLUME AND ITS UNCERTAINTY IN BATCH TRANSFERS FROM NOISY DATA
Submission Author:
Lauro Carvalho , RJ
Co-Authors:
Felipe Bastos de Freitas Rachid, Maria Laura Martins-Costa, ROGERIO GAMA
Presenter: Lauro Carvalho
doi://10.26678/ABCM.ENCIT2016.CIT2016-0714
Abstract
The transfer of dissimilar products through a same pipeline is a common practice in the petroleum industry and is usually known as batch transfer. When this task is carried out without using scrapers or pigs to separate the products, a commingling zone develops at the products’ interface, and increases in extent as it travels along the line. By the time this zone reaches the receiving point, the commingling volume, as well as its uncertainty, must be properly identified and determined, so that it can be segregated and accommodated in a separate tank, to be shipped back to the refinery for later reprocessing. The feasibility of such operation depends heavily on the proper identification of the beginning and the end of commingling zone in real time during the transfer. It is usually done by continuously monitoring and acquiring a physical property of the products, capable to discern them, such as density and sonic velocity among others. However, in real world applications such data are not free of noise, what poses difficulties and sometimes compromises this task. To get rid of such an inconvenience, it is proposed in this paper a statistical strategy capable to not only identify the ends of the commingling zone and determine its volume for any pair of admissible contamination previously established, but also to automatically compute its uncertainty, even in noisy data. To test and validate the performance and robustness of the proposed methodology, numerical examples are presented for data with different levels of noise, which are artificially introduced in the concentration profile predicted by a classical well-known theoretical model for continuous batch transfers.
Keywords
Batch transfer, Commingling volume, Commingling volume uncertainty, Noisy data, Batch transfer, Commingling volume, Commingling volume uncertainty, Noisy data
