Robust and highly reliable oil leakage detection with optimal performance in faulty and normal conditions

Abstract

This paper proposes the design of robust and highly reliable oil leakage detection with optimal performance in faulty and normal conditions. Flow rates and sound are two parameters considered for the design of this project. The purpose of this design is to ensure that faults do not result in malfunctioning and system failure and to achieve the best performance even with minimum number of input and output devices working properly. The fault detection and isolation (FDI) problem is an inherently complex one and for this reason, the immediate aims are to preserve the stability of the process and if it is possible, to control and continue the process in a slightly degraded manner. The role of the FDI algorithms is to ensure that the control equipment must make correct decision. This system has two flow stations that are connected by a pipeline and flow rate difference is used to detect leakage. First flow station has 2 sound alarms and pumps as output devices and 2 flow meters and start button as input devices. Also the second flow station has the same arrangement as the first flow station except that it does not have a pump. Both flow stations use mobile phones as modems to send and receive short message services. The essence of the two flow meters and two sound alarms at each of the flow stations is to ensure optimal performance and robustness even if one or more of the devices failed. Once, the system starts, the first flow station pumps crude oil to the second flow station while the former also sends the flow rate at its end M1 to the later. Then, the second flow station measures its flow rate as M2 and compares it with M1 so as to check whether the difference is still within the allowable threshold. If the difference between M1 and M2 is within the allowable threshold, the system continues working but once the difference is above the recommended threshold, the second flow station sends short message services (sms) to the first flow station and equally alerts its workers using sound alarm. Then the first flow station switches off the pump and equally alerts its worker that leakage has occurred using the sound alarms. The developed system showed that at sensors (input devices) and actuators (output devices) reliabilities of 50% and above, the system efficiency was up to 95%. This system would go a long way to ensure robust oil leakage detection in our oil industries, if implemented.