Computational comparison between transfer function and state variable modeling for slug stabilization in a biphasic separator

• Main Author: Furtado, Maximiliano Matheus
• Publication Date: 2020
• Format: Bachelor thesis
• Language: por
• Source: Repositório Institucional da Udesc
• dARK ID: ark:/33523/0013000006zch
• Download full: https://repositorio.udesc.br/handle/UDESC/17463
• topic: Engenharia do petróleo; Engenharia de reservatório de óleo; Escoamento

During the oil production, mainly in offshore systems, it is common to have a slug flow regime in the pipeline, this situation can present risks to the production process, reducing or interrupting production and possibly causing damage to equipment. Aiming to solve this problem, this work presents a comparison between transfer function and state variable modeling applied to the case study of a horizontal biphasic separator in a simplified model. The model is controlled by two valves, one to control the liquid outlet and the other to control the gas outlet. The control valves always operate with an opening of 30% to 80% of their maximum opening, increasing the equipment life cycle. A PID controller governs the valve performance, this control is optimized with a particle swarm optimizer through an objective function. This function uses the root of the sum of the square of errors to calculate the efficiency of the solution. In this case study, it is observed that the code with state variables presents better performance in relation to the transfer function modeling and that the optimization with PSO is effective in obtaining an optimal solution for a two-phase separator