The use of robotics for NDT/NDE is quickly gaining momentum as service providers develop sophisticated technologies for assessing critical infrastructure in the power, paper and pulp, oil and gas, and petrochemical industries. Robot crawlers and drones are able to access many areas humans cannot and provide numerous benefits over traditional inspection methods, including speed, safety, and accuracy.
Despite all these advantages, transitioning to new technology may seem cumbersome, complex, and even a little intimidating.
However, with the right team, equipment, and a well-executed plan, it can be a relatively seamless process.
So, how does a facility incorporate robotic inspections into their operations—maximizing the value they offer—without any disruptions? It takes the right combination of due diligence and planning.
The first thing to consider is whether the asset is fit for a robotic inspection.
Is the Asset Suitable for Robotic Inspection?
When determining if a critical asset is suitable for a robotic inspection, there are several considerations, including accessibility, material, and temperature. These factors will also help determine which method, robot crawlers or drones, is preferential, and if the inspection can occur while the asset is online or offline.
Accessibility: Perhaps the most important consideration is the accessibility of the asset and inspection area. In confined space entry (CSE) scenarios, robot crawlers and drones offer ways to collect quantitative and qualitative inspection data while ensuring the safety of inspection personnel. However, if entering the asset, both methods should require taking it offline prior to the inspection. Drones, while beneficial at providing aerial footage, are not well equipped to perform an inspection on the complete asset. This is because drones often have only one transducer. Conversely, robotic crawlers can contain upwards of 128 transducers and can cover the entire asset’s surface area. However, they are limited by the presence of support structures or external stiffeners that prevent uninterrupted crawling.
Material: Material type is not a huge consideration for drones because they often do not interface with the asset’s surface. Alternatively, with robot crawlers, material type is paramount. Some robots adhere to critical assets magnetically. Therefore, the surface must be ferrous. Others use vacuum suction or suctioned feet for adhesion. While this allows the robot to crawl materials beyond those that are ferromagnetic, the adhesion isn’t as strong.
Temperature: Temperature thresholds vary for each method depending on a number of factors. In general, robots can tolerate temperatures up to 500 degrees Fahrenheit (260℃). This is to protect the device from damage, but also for the safety of inspection personnel. When applicable, precautionary setups, such as water cooling, deal with unexpected heating situations. If the asset is hotter than the threshold, it will have to be taken offline to complete the inspection. On the other hand, a drone equipped with a heat tolerant transducer and couplant can collect data in environments over 900 degrees Fahrenheit (482℃).
The factors above should be used to determine if a robotic inspection is possible and if so, which method is preferred, and if it should be performed while the asset is in-service or offline. While there are numerous benefits, largely financial, to performing an in-service inspection, that is not always possible.
When an inspection is performed offline, it is advisable to have the inspection team on-site as soon as possible.
When to Inspect During a Planned Outage
During a planned outage, typically, the first contractor on site is a cleaning company that might need to perform work to prepare an asset for inspection, make it safe for entry, or do other work such as drain a water tank. Next, a scaffolding or rope-access company may come in where the drone methodology is preferred because it is often coupled with a manual inspection. At this point in the outage workflow, it is highly recommended to bring an inspection team on site as soon as possible.
Why? Well, say an inspection produces readings indicating an immediate repair must be made—waiting is not an option. Even though robotic inspections are typically 10x faster than traditional inspection methods, if they are performed during the back half of an outage, it may need to be extended to complete the repair work. If the inspection is front loaded during the outage, however, this gives operators time to make repairs and helps safeguard against additional costly and unplanned downtime.
Including the robotic inspection on the front-end of an outage leaves optimal time for analyzing the resulting asset condition data--informing both urgent repair needs and future planning.
Leveraging Data to Develop a Plan
Data is the hallmark of a robotic inspection, providing up to 1,000 times more information than traditional methods. When deciding between drones and robot crawlers, data type and quality should be considered. Drones provide aerial footage and pictures, and can even provide B-scans of assets. But, as previously mentioned, this method doesn’t result in the level of quantitative data that robot crawlers can supply. Additionally, some robots are equipped with cameras to provide the best of both worlds.
But, why is all of this data valuable? Because it allows operators the ability to view the overall condition of the asset and pinpoint precise locations for immediate repairs. Asset condition data from robot crawlers performing ultrasonic testing is presented in a color-coded 2D or 3D C-scan which is incredibly valuable for understanding urgent maintenance needs, as well as, providing insights into the early stages of damage for future monitoring.
Additionally, layering different approaches to robotic inspections, such as Rapid Ultrasonic Gridding (RUG) with Rapid Automated Ultrasonic Testing (R-AUT), combines corrosion thickness mapping data with a more granular solution for identifying different damage mechanisms. Similarly, drone inspections are often combined with manual inspections to further investigate areas with visual damage.
While the amount of data may seem excessive compared to the status quo, strategic and intentional use of the information to develop routine and preventative maintenance plans will inevitably pay off in the long run by limiting the financial cost of unplanned outages and shutdowns.
Planning for Future Inspections
For some assets, inspections should be done once a year, but the frequency is largely dependent on damage mechanisms and the likelihood of asset failure. For a circulating fluidized bed boiler operating in harsh conditions, every 12 to 18 months is typical. Some assets, such as a natural gas boiler or water tank, only need to be inspected every three to five years because they operate in much less aggressive environments. This is also dependent on state and local code/regulations—particularly assets that may cause harm if they fail, such as ammonia tanks. These will require a more rigid inspection schedule.
The key to robotic inspections is appropriate consistency—the robots are logging everything that they do (a photo and a reading), and this information is stored for viewing, year after year. Operators can look at historical data for predictive maintenance indications and perform other analytics that typically haven’t been possible with traditional data collection and visualization methods.
Wrapping it Up
In summary, there are some key questions to ask and action items to execute that will help smoothly integrate a robotic inspection into a facility’s workflow. These include fitness for inspection, the optimal time to undergo an inspection based on the asset’s unique characteristics and service schedule, and the devisal of an action plan based on the acquired data.
While robotic inspections involve advanced technologies that require skilled technicians to implement, there is no need for hesitancy based on their sophisticated nature. The right company will offer a turnkey solution, from inspection planning and implementation to data delivery and support.
