Three ways to reduce maintenance costs with analytics
Oil and gas producers predict process integrity and reliability
Whether attempting to maximize production from existing, aging assets or navigating the complexities of finding and tapping into new reserves in more difficult environments, exploration and production operators experience daily production challenges. Maintenance issues abound as companies strive to increase production while guaranteeing safety, flow assurance and equipment reliability.
Optimized, sustainable maintenance strategies—and improved performance and availability of production equipment— depends on the detection and diagnosis of the root causes of poor performance and unplanned downtime. Read on to discover three ways that large oil and gas companies are using analytics to ensure ongoing improvements for facility reliability and integrity:
Improving product quality
To pinpoint and correct the glycol consumption, the company turned to SAS® Predictive Asset Maintenance for facility integrity and reliability. The effort prevented production deferment, generated cost savings, created new business strategies and improved health, safety and environmental performance.
Glycol is an additive that keeps gas from liquefying during the production process for gas feedstocks, but special care must be taken with its use. In the drying process, too much glycol may penetrate the gas and decrease quality. Too little adds humidity and other impurities. Either sends customers elsewhere in search of higher-quality gas feedstock.
The challenge for oil and gas producers is to identify process malfunctions and detect leaks. Glycol cannot be measured during the drying process, but in theory the amount of glycol added should approximate the amount extracted.
When less glycol than expected is present, producers must find where the loss occurred by spotting problems among the valves, vacuum seals, gauges, pipes and tanks that make up the production process. That's a difficult, time-consuming chore for consultants and engineers. And the results are often mixed – and disappointing.
SAS takes a different approach. Chemical theory – what goes in must come out in similar quantity – is used in conjunction with the limited number of data collection points spread along the production process.
SAS Predictive Asset Maintenance revealed that the feedstock quality might require adjustments to glycol levels during the drying process. The solution predicted where in the process glycol needed adjustment, depending on other factors.
The glycol adjustment predictions have several direct benefits to the gas producer. Glycol replacement costs are reduced. Leaks along the process are quickly identified using data inputs and in-process measurements.
Meanwhile, by avoiding negative effects on the environment, the company saves the cost of penalties and fines. And it maintains its reputation for higherquality, consistent feedstock production, which drives demand and, in turn, increases margins.
With SAS, the company learned it can control production processes using predictive sciences that integrate data points with chemical theory. Realizing the value of applied predictive analytics, the company uses SAS Predictive Asset Maintenance in other processes and streamlines its energy consumption rates, finding additional savings along the way.
Resolving tricky situations
Failure to separate the sulfur and transform it into blocks could violate environmental regulations and increase the likelihood of fines and penalties. It also prevents opportunities to sell gas and sulfur.
As the cause remained a mystery, managers worried about health, safety and environmental (HSE) risks. They knew that the problem could pose the risk of leaks or even explosions. Meanwhile, the temporary fix – using an electric motor to blow air into the process – was a costly and inefficient remedy.
As expected when a piece of equipment fails, everyone's attention is focused on that machine. Was it installed correctly? Was there a malfunctioning part? Was there a material flaw? Different groups in the company – process and reliability engineers, operators and maintenance crews – each had their opinions. Yet they could not coordinate their assessments to find the root cause and mitigate the problem. Their siloed structure prevented them from taking a holistic view.
Using the SAS Predictive Asset Maintenance solution and the facility integrity and reliability methodology, it was quickly determined that the process itself was flawed. Oxygen was being pulled into the recovery unit, creating water. While the variations in the acid feedstock were requiring the oxygen variations, the turbine itself was not designed to match them. This process flaw was causing the turbine to trip and shut down.
The ability to gather data from historians and reports, correlate parameters and capture occurrence signatures helped to identify potential problem areas within the process itself. By fixing the process, the company has improved production uptime rates and gained better insight into preventive maintenance or replacement needs.
The company learned that equipment failures might be the effect of process anomalies occurring separately from the impact observation. SAS proved that the high volume of mostly unused historic data could help diagnose problems holistically, implement predictive surveillance and improve process automation.
Now realizing the value of applied predictive analytics to such production processes, the company uses SAS Predictive Asset Maintenance in other processes at larger refineries and gas plant locations.
Preventing production loss
Any degradation in the sensors or the bearings themselves can cause the compressor to trip, thereby stalling production and initiating repair work. Or, sensor degradation can wrongly inform plant operators, causing errant process decisions, false alarms and other delays. Various catastrophic situations can arise. Replacement is costly and time-consuming.
On the other hand, any planned increase in production requires additional compressors and sensors that produce volumes of new data. Manual administration is not an option, yet failure to appropriately integrate the new data might trigger alarms or misinform operational processes.
In view of these challenges, the company applied the SAS methodology for facility integrity and reliability, along with the SAS Predictive Asset Maintenance solution to develop data-driven models that can predict sensor and magnetic bearing degradation.
The models point maintenance teams to equipment that may fail given the signals they produce, before any significant problem or loss of efficiency occurs.
Such heightened understanding of overall facility integrity and reliability helps oil and gas producers prevent production process deviations. As a result, they may reduce operating costs, develop new business strategies for maintenance and improve health, safety and environmental performance.
With SAS, the company can assemble data from many platforms and systems to create a picture of what has happened and to predict what's going to happen. Such foresight increases production uptime by informing smart, timely maintenance decisions.
The company uses SAS to monitor potential weak points and proactively maintain smooth production processes by applying predictive sciences that intersect analytics, physics and mechanical theory.
That's possible because all systems and equipment have a metering, monitoring or surveillance system that produces staggering volumes of data. SAS provides a comprehensive view across all systems and equipment, taking into account interdependencies to accurately monitor overall performance.
Seeing the value of applied predictive analytics to such production processes, the company plans to implement SAS Predictive Asset Maintenance in other processes at other locations.
SAS enables predictive, preventive maintenance of your assets with minimal disruption to production. As a result, you can maximize the use of maintenance resources to meet operational goals for profitability, safety and environmental compliance.