Reduction of impacts of oil and gas operations through intelligent maintenance solution

Abstract

Impacts of oil and gas production operations are always very obvious when there is imbalanced operation, uncontrolled stoppage or catastrophic failure of the system during normal operations. These impacts may range from high flaring and venting of associated petroleum gas, oil release or spillage, equipment damage, fire outbreak to even fatality. Possible causes of imbalanced operations or system failure are categorised into process upset, system degradation, ineffective operation and maintenance procedures and human errors. Effective maintenance strategy integrates major components of the system; people (human factors), operation and maintenance procedures (process) and production plant (technology) to develop an intelligent maintenance solution that is capable of monitoring and detecting fault in the system at incipient stage before operational integrity is compromised. This paper deploys data-based analytics technique to develop condition-based predictive maintenance system to monitor, predict and classify performance of gas processing system. Exhaust gas temperature (EGT) of Gas Turbine Engine (GTE) is one of the operating and control parameters associated with efficiency of the GTE operation. The EGT is measured using several thermocouples, temperature sensors spaced equidistant around the circumference of the exhaust duct of the GTE. Neural network technique of multisensory data fusion is integrated with intelligent maintenance system to monitor performance of GTE, detect fault and classify performance of GTE to optimal, average and abnormal performance.

Preventive Maintenance: Some Operations and Technology Related Practices at a Malaysian Oil and Gas Firm

“…Maintenance of this equipment takes place when it fails completely. This kind of maintenance system is also known as primitive maintenance philosophy [19,28].…”

“…The unplanned or breakdown maintenance, which is regarded as a reactive maintenance program where the system is put into operation until it is shut down due to the system failure before the maintenance actions take place [29]. The industrial systems were operated up to the system failure, where the affected tools or equipment have been repaired or replaced, consistent with the popular maintenance philosophy "Fix it when it fails or breaks" [28].…”

“…The reactive maintenance lauds the principle of no action is taken to prevent system failures or to detect the onset of failures [27,30]. The maintenance method performed was on demand-based, generally due to major equipment breakdown or failure [28,29].…”

Innovative VS. Traditional: a framework to assess the sustainable trade-off of maintenance strategies

“…The state of the art on the adoption of technology in the O&M phase shows that the transition from a traditional maintenance strategy to an innovative one may optimize the sustainable performance of buildings, as well as the sustainable performance of the maintenance activity itself. Eyoh and Kalawsky [20] ointed out that maintenance management has always been related to the change in technological innovation: from the control of simple machines and equipment to the monitoring of complex and sophisticated systems.…”

A Process to Implement an Artificial Neural Network and Association Rules Techniques to Improve Asset Performance and Energy Efficiency

“…The potentiality of ANNs in fault diagnosis is proven, among the others, by [42] that analyzed the vibration data to predict bearings failures, while [43] addressed the engine cylinders temperature to foresee potential malfunctioning. A similar approach is also applied for controlling the exhaust gas temperature of gas turbine engine and detecting wearing problems: integrating data recorded through different sensors of the turbine, allows the monitoring of the turbine's performance and fault detection [44]. Another application belonging to the energy field regards the analysis of temperature and irradiance values to predict the performance and failures of photovoltaic plants [45]; interestingly, as implemented in a power grid company, ANNs may also be integrated in ontological models to optimize the faults diagnostics and repairing processes [46].…”