Drones Help Detect Corrosion under Insulation

The thermal IR and multispectral imaging sensors are attached below the body of a multi-rotor

Corrosion under insulation (CUI) is one of the most common forms of corrosion found in the oil and gas industry. Many components such as piping systems, pressure vessels, tanks, and other equipment are insulated for personnel protection and/or to keep fluids at appropriate temperatures for process efficiency. The underlying metal substrate, however, is vulnerable to accelerated localized corrosion when moisture gets trapped in the insulating material. Although there may be visible signs of rust on the protective cladding that can indicate CUI, the corrosion isn’t observable until the insulation is removed and the substrate is exposed. Often the insulated components, such as process piping, are not easily accessible for visual inspection. To address the challenges of detecting CUI, ALS Oil & Gas—Pipeline & Asset Integrity Monitoring (Houston, Texas) and Unmanned Ad-Hoc Industries (UAI) (Spring, Texas) have partnered to provide remote imaging technology with unmanned aerial vehicle (UAV) surveillance to inspect insulated piping and components for indications of CUI. The methodology is part of an overall Intelligent Data Collection program employed by the two companies to efficiently gather corrosion-related inspection data with multiple state-of-the-art, highly advanced sensor technologies using various manned and unmanned platforms, and then process and analyze the data using interpretive software.
Ryan Pullen with UAI readies a drone for takeoff. Photo courtesy of UAI.

According to NACE International member Dean Lioliou, director of sales with ALS Oil & Gas—Pipeline & Asset Integrity Monitoring, corrosion costs the oil and gas midstream market ~$600 million annually. Of that, ~60% of the corrosion costs can be attributed to CUI. “When we look at the cost of CUI to the pipeline industry, it is staggering,” says Lioliou. “If we were able to simply tackle CUI and minimize it by a small amount, say 20%, the costs that could be saved by organizations, including savings due to other issues such as environmental costs, loss of product, and loss of access, are incredible.”

CUI often goes undetected because it is hidden from sight, and it is not always obvious where CUI may be occurring. Although protective cladding over insulation may appear intact, it could have leaks because of mechanical damage, degradation of joint sealants, and loose or missing inspection port sealing caps, which enable water or moisture to enter the insulation. Because migrating water tends to flow to low points in the insulation system, it can be difficult to predict where water will contact the metal surface. “Unless you open up the insulation and visually inspect the structure, you won’t know that you have CUI,” says Lioliou.

Because removing all insulation material and examining the substrate underneath is cost prohibitive, a common way to inspect for CUI has been to remove small portions of the insulation at select locations that may be at risk for CUI, and use nondestructive testing techniques on the surface of the structure to determine if there is metal loss. This method, however, creates a potential entry point for moisture ingress where the cladding is opened. Additionally, since CUI is generally localized, active corrosion may not be found if a particular piece of insulation removed is not covering the specific area where corrosion is occurring.

By using a combination of remote thermal infrared (IR) and multispectral imaging sensors to capture images of insulated components, anomalies that may be indicative of CUI can be detected on piping and other equipment, says Paul Ramirez, president and COO of UAI. He explains that multispectral images can capture image data at specific frequencies across the electromagnetic spectrum and can detect very small differences in the way light is reflected from the surface of an object. Thermal IR provides images that illustrate temperature differences—from a component’s lowest temperature to its highest—and indicate if the thermal signature on a portion of a pipeline or component is different from the rest of the component.

Mounting these sensors on a UAV allows inspection of areas that are inaccessible from the ground and typically require scaffolding or some type of manlift for access. Typically, a small ~20-lb (9-kg) multi-rotor UAV vehicle that can stay airborne for ~30 min at a time is used to inspect piping and equipment in the confined spaces of an oil and gas processing facility. The two sensors are colocated on the UAV and calibrated with each other so that corresponding visual and thermal images are captured of the same area of the structure. During an inspection, Ramirez comments, a ground-based UAV pilot flies the vehicle around the structures to be assessed. For a 12-in (205-mm) diameter pipe, the UAV is typically 25 to 50 ft (7.6 to 15 m) away from the structure. To scan the underside, the pilot changes the angle of the sensors while flying the drone overhead so images can be captured of almost the entire pipe circumference, depending on its availability (i.e., if it isn’t blocked by another structure). “In most cases, the best approach is to capture data from a more oblique angle to assess the pipe’s condition on both the top and bottom,” he says.

Images in the form of video from the sensors on the UAV are sent to the ground station computer. The multispectral images show how the surface of the component appears visually, including any potential trouble spots such as degradation, damage, or rust stains on the cladding. The thermal IR images show visual temperature variances. As an example, the IR images may be predominantly red, indicating warmer temperatures because of the hot fluid inside the pipe. If there is an area where water or moisture has infiltrated the insulation, which can lead to CUI, the temperature there will be cooler and depicted as a different color and that area will have a different thermal signature than the rest of the pipeline. “By looking at the pipe from a different perspective— the IR wave lengths—you can detect possible indications of corrosion before it’s apparent in ordinary light,” says Ramirez. “The ability to do that is a potential game changer.”

To note when and where potential problem areas are seen, the operator presses a button when an anomaly is sighted and the associated software program captures the image’s geographical coordinates, time stamps it, and saves it in a corresponding data folder. A baseline can be established with the first inspection, and subsequent inspections several months later can identify and compare any changes in the visual images as well as the thermal signatures, and determine if the anomaly appears stable or a trend suggests that potential problem areas are intensifying. This provides the asset owner with specific locations to inspect for CUI based on actual sensor data. “Instead of playing a guessing game and undoing cladding over the entire facility, this gives them a prioritized view of where to look first,” Ramirez says.

For a recent project in a midstream oil and gas facility, a drone equipped with thermal IR and multispectral imaging sensors was able to inspect ~5 miles (8 km) of insulated pipeline for indications of CUI in about four and a half days. For this facility, says Lioliou, opening up the insulation and inspecting the pipes for CUI once or twice a year would be a very daunting task—one that more than likely would not get done. By using the sensors with the drone technology, the company was able to pinpoint areas where CUI was most likely to be present and open the insulation for inspection only in those locations. At the very least, Lioliou says, the owner may simply need to replace insulation that is wet. Alternatively, he notes, the asset owner may find corrosion and have the opportunity to proactively repair the damage before a failure occurs.

Contact Dean Lioliou, ALS Oil & Gas— e-mail: Dean.Lioliou@alsglobal.com; or Paul Ramirez, UAI—e-mail: paul.ramirez@ globaluai.com.

Related Articles