PILOTING update: Large-scale Testing and Experiments in a Refinery
At-a-glance
Current European refineries and civil infrastructures, like tunnels and bridges, are ageing, and therefore gradually become deteriorated. It is important to increase the efficiency and quality of inspection and maintenance activities in order to keep the necessary safety levels in these ageing infrastructures. To overcome this important challenge, PILOTING, an EU-funded project, offers an integrated platform adapting, integrating, and demonstrating novel robotic solutions.
The first demonstration of novel robotics solutions took place at a refinery and the experiment focused on testing the end-to-end solution, from inspection planning to gathering and harmonizing the inspection data and visualizing the mission results using intuitive 3D digital tooling.
Test Environment
The selected refinery is a large industrial complex which offers multiple opportunities to test inspection procedures, making it an ideal candidate for the PILOTING project. A wide range of inspection categories such as large structure inspection, inspection of pipework at height, internal vessel inspection, ground monitoring and manipulation of equipment can be tested at site. For the piloting demonstration the following use cases were included:
• Refinery ground monitoring
• Refinery vessel inspection
• Refinery pipes inspection
• Refinery large structures inspection
Inspection Plan
Before arriving at site, authorized users created and defined test plans for the different test cases using the Intelligence and Visualization Portal (IVP). A digital representation of the asset was put in place by using the modelling tool of the IVP, to describe the inspection locations, tasks to be performed and specific instructions on performing them (e.g.: visual, UT, cleaning). Details of the robotic system and sensor payloads used are also included in the inspection plan. This enables a better traceability, reliability, and repeatability of the robotic missions.
Gathering of Inspections Data
The main goal of asset owners for introducing new procedures for inspection and maintenance tasks is risk reduction, by removing people from dangerous situations. New procedures should also improve the quality of the captured data and reduce the cost and resources needed to perform routine tasks. Different robotic solutions were used inspecting the different types of assets. These solutions have the advantage of removing people from confined spaces and working at heights as well as from traveling long distances while performing I&M tasks.
The AeroX system was used to perform contact inspection of a storage tank wall. The drone performed seven tests in total (ranging approximately from 10 to 15 minutes in duration), all of which related to two unique experiment profiles: a longitudinal ultrasonic scan along the vertical axis, covering almost all the way up to the tank’s top and a path that traversed an area defined by a singular rectangle. Altogether, three and four trials were done for the first and second experiment profiles respectively.
The Hybrid platform, equipped with a UT satellite crawler robot, was selected to perform UT thickness measurements on the pipes. The drone was initially placed in a location with sufficient landing clearance and from this location the satellite deployed from its bay. All 10 measurements could be captured within one deployment and within the 3 meters tether range of the drone. Upon completion, the satellite returned to its bay and the drone retrieved from the area.
For pressure vessel inspection the BIKE robot, a magnetic crawler, was selected to enter a confined space. The BIKE robot drove on the entire vessel shell, including vertically up, overhead as well as doing the transition from cylinder to head section successfully. A total of seven localization accuracy checks in various locations were performed. Accuracy was clearly within 10 cm, the biggest measured deviation was 7.5 cm.
Refinery ground monitoring was performed by two robots, RISING and UGV; the former having the ability to climb stairs and the latter featuring autonomous valve detection and manipulation. Both robots navigate autonomously taking pictures, thermal and IR measurements in predefined locations. In addition, 5 connected static sensors were installed to monitor flow anomalies and eventual leaks in piping systems.
Data transmission, transformation and storage
The mission data originating from the different robotic solutions are harmonized and posted to the centralized. DMS platform, triggering later the Artificial Intelligence (AI) pipelines for data analytics. The results of the mission can be visualized in the IVP and analyzed by a qualified inspector to generate a standardized and comprehensive inspection report, including findings and recommendations to the asset owners.
Lessons Learned
The experiment was the first test large-scale industrial experiment of the PILOTING project. Overall, the experiment can be seen as a success, testing was conducted safely, on time and according to initial requirements. The experiments were performed with a fully integrated end-to-end PILOTING I&M platform,
including mission planning, execution, and analysis.
The project partners walked away for the experiment with confidence as it was proven that it is possible to run multiple robotic solutions through an agnostic control station and get reliable and relevant data through an online platform. The partners also received relevant end user feedback for improvements to be taken back to apply before the next demonstration.
“After a week of experiments, we have obtained good results and a lot of experience that
we will apply to the next version.”Dr. Antidio Viguria Jiménez, CATEC
Quasset is a part of PILOTING, a fourteen-member consortium that focuses on the use of robotic technologies to increase efficiency and quality of inspection and maintenance of aging infrastructures. PILOTING is a multi-year, European Horizon 2020 project working towards advancing the adaptation, integration, and demonstration of robotic solutions in an integrated platform which will be tested and evaluated in three large-scale pilots: refineries, bridges/viaducts, and tunnels with the involvement of all actors in the value chain.