Multi-model based process condition monitoring of offshore oil and gas production process

Abstract

Offshore oil and gas production platforms are uniquely hazardous in which the operating personnel have to work in a perilous environment surrounded by extremely flammable hydrocarbons. A failure in an equipment could quickly propagate to others resulting in leaks, fires and explosions, causing loss of life, capital invested and production downtime. A method for preventing such accidents is to deploy intelligent monitoring tools which continuously supervise the process and the health of equipments to provide context-specific decision support to operators during safety-critical situations. Such an intelligent system, which is condition driven is developed and described in this paper. Since relevant process data is unavailable in the literature, a dynamic model of an offshore oil and gas production platform was developed using gPROMS and data to reflect operating conditions under normal, fault conditions and maintenance activities were simulated. The different maintenance activities and normal conditions are explicitly considered as separate states of the process. The simulated data are then used to train principal component analysis monitoring models for each of these states. Online fault detection and identification are performed by identifying the operating state in real-time and triggering the respective model. In this paper, the dynamic model and the condition monitoring system are described.