数据分析提高油田开采效率

Analytics can help optimize oil and gas equipment performance and predict failures

Today, the oil and gas industry, like many others, is embracing a data-driven mindset – what’s driving the shift?

The downturn in oil and gas prices put pressure on producers to cut costs and maximize use of their existing, aging operating assets. But with operations more complex than ever, optimizing oil and gas equipment can be an impossible task when you are unable to answer critical questions about health and performance, such as: Are we prepared to react to ongoing market changes? Why did a piece of equipment fail? How long until it fails again?

Hoping to capitalize on the promise of the digital oilfield, oil and gas producers are seeking to implement new IoT technologies, and connect their equipment, devices and systems, and integrate their data. Most critically though, producers are looking for ways to transform all that data into operational intelligence.

Analytics involve monitoring large amounts of data from multiple sources, collecting and contextualizing that data into information, and then presenting that information to your workers in the form of useful, actionable insights.

This requires seamless connectivity across your oilfield. The most basic job of analytics is to provide context to data. This could include well behavior and performance trends from artificial lift systems or can also include real-time performance from surface processing skids, remote wellhead-monitoring systems or insights into downhole and reservoir conditions. Furthermore, analytics can help users investigate anomalies and troubleshoot issues.

Oil and gas companies that want to create a roadmap for deploying analytics should first understand the four analytics categories, as well as their complexities and benefits.

1. Descriptive and diagnostic analytics are the starting point for oil and gas producers. At the most basic level, this data informs workers how things are performing and the conditions in which they’re operating. If a failure occurs, these analytics can help technicians quickly diagnose the problem, and help make sure they send the right person to the right place, with the right tools to resolve the problem.

2. Predictive analytics are the target goal for many producers today. These analytics look for known conditions and relationships within the data that indicate a device or piece of equipment is approaching failure. Producers can use this to identify and prevent equipment failures before they happen or prepare operations and maintenance for an impending equipment failure, providing a tremendous opportunity to reduce costly downtime. Note: predictive analytics won’t always be absolute. Sometimes, they are configured around the probability of various failures. For example, a production team could set a threshold for a failure probability of 80 percent. If the software detects a failure probability at this threshold, it will trigger an alarm condition and can notify users that a failure is impending and requires service.

3. Prescriptive analytics are the most complex analytical techniques for equipment maintenance and the ultimate goal for most oil and gas producers. Prescriptive analytics build upon predictive analytics to not only anticipate equipment failure, but also recommend corrective action based on historical failure data and trends. Workers can use the recommended actions to correct issues and optimize performance of assets in real-time. When paired with techniques like augmented reality, operators or maintenance personnel can be guided through a set of digital instructions to repair a failed piece of equipment or change operational parameters to improve process performance.

At the highest analytics level, users can view their entire operations on a single dashboard. This provides real-time, dynamic KPI’s versus targets, and a prioritized list of required actions to meet business goals. In addition, current performance can be compared to past performance. Operators can drill deep into each site, viewing well-pad performance, and drilling even further into a well or pump’s performance.

Rod-pump control analytics, for example, can use downhole card information to provide insights into the downhole conditions and identify if equipment is running optimally and if intervention is required. Plunger-lift analytics are retrieved from the field and displayed as a timeline of well-performance indication and influence key operator decisions, such as if pressure should be increased. Gas-lift analytics can help operators to know if they’re over injecting gas and identify the proper ratio of oil and gas injection.

In the vast oil reserves of Western Canada, ARC Resources has multiple assets deployed for the exploration, development and production of conventional oil and natural gas. At one of their large multi-well pad sites, the company wanted to optimize control operations and gain access to valuable production information to support business decision-making and equipment troubleshooting.

The company’s optimization team deployed a Rockwell Automation ConnectedProduction solution using analytics software to mitigate wax build up and reduce chemical injection on their wells.

The analytics solution helped ARC Resources gain production visibility with contextualized information for simplified equipment troubleshooting. The system helps guide operators through the entire well optimization lifecycle and helps operators better respond to potential system and equipment issues. Since implementing analytics, ARC Resources has reduced operational costs by approximately $30k a year per well, and plans to include this as a key part of their optimization strategy.

Tap into the Potential of the Digital Oilfield

Analytics in a connected oilfield environment can help you track performance trends, quickly identify and react to events affecting production and equipment uptime, and proactively take actions to prevent downtime and lost production. When these analytics are combined with business systems, producers can quickly adjust production in response to market changes and business needs.

It has never been easier to deploy analytics. The foundation you need in place is well established and new technologies can put the power of your data directly into the hands of the workers who need it and contextualize it for them.