Analytics use case examples demonstrate uses, benefits
Search for improved data use to address industry challenges
As the oil & gas industry strives to digitally transform amidst the worst downturn on record, analytics applications are enabling operators to improve operations and cut costs through the monitoring of offshore assets. By detecting and anticipating performance issues for critical assets such as pumps and compressors, maintenance activities can be prioritized and scheduled. This reduces the likelihood of catastrophic failure, increases revenue by improving uptime, and saves costs by enabling proactive maintenance.
Subject matter experts (SMEs) of all disciplines are now able to perform advanced self-service analytics by seamlessly accessing data from disparate data sources. This empowers them to implement advanced condition monitoring of offshore assets for near real-time situational awareness of asset health by examining historical failure and performance data.
This article demonstrates how analytics are leveraged to address some common asset management pain points.
Issues with traditional tools
In the oil & gas industry, it is essential to maintain and prolong the lifetime of critical assets including pumps, valves, heat exchangers, compressors, and others. Better insights into these assets provide a clearer understanding of when they are not operating efficiently and help determine the root cause of degraded performance. These insights also enable maintenance to be performed proactively prior to catastrophic failure, saving time and money.
Common current approaches for analyzing performance and predicting maintenance require SMEs to manually combine data from multiple sources in a spreadsheet and then spend hours, if not days, formatting and filtering the content and removing non-relevant data (for example, when equipment was out of service). Spreadsheets are not designed for these data wrangling and analysis tasks, resulting in a host of potential problems (Figure 1).
The following use cases demonstrate how SMEs leverage analytics to optimize asset maintenance and improve performance for pumps, compressors, valves, and heat exchangers.
Solving pump problems
A large U.S. oil & gas company was experiencing unplanned pump downtime of about 10 hours per event for a centrifugal fixed pump. The company wanted to better monitor all its pumps’ health, avoiding unplanned downtime while not incurring unscheduled maintenance inefficiencies.
Using analytics, condition-based monitoring analysis allows the company to continuously monitor and better understand asset health across hundreds of pumps. The application used data from a historian which was cleansed to remove downtime periods. The two major pump performance indicators considered were percent time in an undesired mode of operation and percent violation from the threshold (Figure 2). Then an Organizer Topic was used to create visualizations displaying current health status of all pumps as a Daily Health Score.
This approach is now used by the company’s operators, reliability engineers, process engineers, and data science teams. They access the data, which flows into a predictive model in Seeq, enabling them to predict pump failures before they happen.
Addressing compressor conundrums
Site maintenance teams often find it difficult to gauge compressor performance and plan maintenance based only on operating hours. Even when operating hours appear to be well within range, unplanned downtime is sometimes imminent, with the physical condition of the compressor beyond corrective maintenance. This leads to lost revenue, along with environmental and safety issues.
By means of the solution, compressor polytropic efficiency and head were determined from first-principles calculations using the Formula tool (Figure 3). These functions were then distributed in a scatter plot, along with the actual efficiency and head of the compressor, to visually identify current compressor health status.
The models developed were used to predict when maintenance would be required based on status and long-term projections. The result is maintenance performed just when needed, but not before, with site maintenance and outage planning teams now able to optimize compressor maintenance plans.
Optimizing valve operation
Critical valves, such as those used in pipeline and subsea applications, are challenging for characterizing valve performance, and for anticipating valve health issues in near real-time. Oilfield operations need to know when maintenance is needed, but calendar-based scheduling is not optimal because it results in unnecessary effort and costs if work is performed too early. If work is performed too late, unplanned downtime results, along with lost revenue and increased environmental and safety risk.
Using advanced analytics, SMEs implemented condition-based monitoring analysis to monitor valve health across their entire fleet. They utilized historical data to accurately create a predictive maintenance forecast by forecasting valve failures and then used the model with new data as it was written to the plant historian.
Overall valve health indicators (Figure 4) were created by monitoring valve performance with various key metrics, such as stroke time and static friction. This led to early identification of bad actors and optimization of maintenance activities. This reduced the likelihood of valve failure, increasing production and the manpower required for each event.
The analytics solution can also reliably provide context and insight into alerts received from other software, such as an asset management system or a historian.
Improving heat exchanger performance
A major U.S. oil & gas firm was operating its heat exchangers on a set maintenance schedule, taking them offline every few months for cleaning. This resulted in inefficiency and incurred unnecessary cost. They needed to address this issue by implementing risk-based heat exchanger maintenance planning to optimize processing rates, while cutting operating and maintenance costs.
Using advanced analytics, the refinery’s SMEs created a predictive model to anticipate when maintenance would be required, with analysis scaled out to monitor heat exchangers at other sites (Figure 5).
The model also empowered SMEs to investigate how changes in operating conditions could be made to extend maintenance intervals. Additionally, the solution made it easy to compare the current cycle with previous cycles to determine periods of accelerated fouling and diagnose root causes.
Using the analytics application to monitor heat exchanger performance in place of time-consuming spreadsheets eliminated weeks of work for SMEs. The refinery expects to save millions of dollars per year due to improved turnaround planning and other opportunities.
The engineering solution drove long-term improvements for the refinery, including reduced production loss, at a savings of roughly $10,000 per year. The solution also decreased the impact of unplanned rate reductions from heat-transfer constraints, enabling the refinery to avoid losing millions of dollars in opportunities from crude intermediate processing margins.
Current conditions and anticipated future market disruptions are forcing the oil & gas industry to optimize operations across the board. Advanced analytics can be used to improve operations and maintenance, saving valuable SME time, cutting costs, and increasing production.
Original content can be found at Oil and Gas Engineering.