Laser scanning provided existing-conditions information to support a major revamp/modification of the Pulai-A offshore production platform. Business benefits of laser scanning reported for the project were:Schedule and risk management. Positional accuracy in capturing existing conditions was critical to minimize design and fabrication errors. Because the revamp required shutdown of the facility, avoiding errors and rework that would disrupt the schedule was essential. Laser scanning was selected to minimize rework by reducing the risk of interferences.
Completeness. The completeness of laser scan data greatly minimized the risk of having to revisit the site to capture information that was either missed the first time, or not understood to have been needed.
Safety. The need for survey staff to erect scaffolding and climb over structures was reduced and in some cases eliminated with laser scanning.
Visualization. The scanning service provider reports that using point-cloud data for site visualization builds clients’ confidence that laser scan data is accurate and trustworthy as a basis for design work.
Asset management. Laser scan data was used to create a 3D CAD deliverable for the owner’s use in managing the asset.The contractor chose laser scanning over photogrammetry for existing-conditions capture
The 27-year-old Pulai-A offshore production platform is owned by Petronas Carigali, a subsidiary of Petroliam Nasional (Petronas), the national oil company of Malaysia. The revamp, which involved four modules, was carried out in September and October 2003 by Malaysia Mining Corporation Berhad (MMC).
The project required accurate existing-conditions information in order to minimize rework offshore that could disrupt the project schedule and lengthen the shutdown period. However, available existing-conditions documentation consisted of general arrangement drawings that were no longer considered reliable. The drawings had been produced by manual methods, and most were no longer readable.
To remedy the lack of usable and reliable as-built documentation, MMC engaged Plant Design Engineers Sdn bhd (PDE) to survey existing conditions on Pulai-A in June 2003. Although MMC had employed close-range photogrammetry in some past projects, PDE reports, with this project MMC elected to use laser scanning.Risk management motivated PDE to adopt laser scanning
PDE is a specialist service provider for the oil-and-gas, power and petrochemicals industries. It offers as-built services (laser scanning), 3D modeling services, living-quarters design, and minor engineering and project management services. Established in 1999, the company reports it has a particular interest in brownfield work where the documentation of existing conditions may not exist at all, or may be limited to old paper drawings which it says are almost invariably out of date or incomplete.
According to Ron Bouman, technical director at PDE, the company has executed four projects using laser scanning to date. In addition to Pulai-A, PDE has carried out one scanning project for the Senoko repowering project in Singapore (through Alstom Power) and two scanning projects for Talisman, and is currently undertaking a scanning project in Sumatra for Caltex.
PDE reports that over several years, it became apparent to the company that errors associated with incomplete and inaccurate as-built surveys increased its exposure to risk. This was what motivated it to begin using laser scanning to capture existing conditions.
The company owns a Cyrax 2500 scanner from Leica/Cyra, and executes most of its design work using AVEVA’s PDMS system. (Bouman notes that he formerly worked for AVEVA.) Bouman reports that in selecting the Cyrax (now HDS2500) scanner, a key factor was the Leica name – a “safe choice,” he says. Leica has a local presence, and PDE had a relationship with the local distributor. PDE’s principals also knew the Leica representative. PDE reports that it also evaluated MENSI‘s offering.Four people surveyed four modules with a Cyrax 2500 in 90 hours of scanning
On the Pulai-A platform, PDE used laser scanning to survey four modules: the M1 process module, M2 wellhead module, M3 workshop/laboratory/storage, and M6 compressor module.
The survey crew consisted of four personnel: two surveyors, one chainman and one piping engineer. The main survey equipment included a Cyrax 2500 laser scanner and a Leica ATR (Automatic Target
Recognition) reflectorless total station. Scanning operations and establishing the primary survey control started concurrently, splitting the crew into two teams.
To establish the main survey control network, four rounds of angles to each target station were measured, using a Leica TCRA705 total station and its onboard sets-of-angles program together with the ATR functionality. The final least squares adjusted network had error ellipses no larger than 3mm, which was more than acceptable for the eventual positioning of the scan targets using the reflectorless capability of the total station. The survey control required two days to put in place, with ongoing establishment of secondary control stations to coordinate the scan targets.
Scanning went on for the full seven days the survey crew was scheduled to be on the platform. At the beginning, scanning was carried out for 10 hours a day. Once the survey control was put in place, freeing up additional manpower, scanning was extended to 16 hours a day. In all, some 90 hours of scanning were carried out, resulting in over 250 scan positions, at an average time of about 20 minutes per scan setup.
During scanning, overlapping registration was carried out to ensure complete coverage, and to reduce the time spent at the end of each day registering scans. The majority of scans were captured at a scan density of 10mm x 10mm. One of the factors typical of scanning on offshore platforms is that ranges are substantially less than 10m, and can be as close as 1m.Scanning had to be carried out in 7-day window dictated by the weekly helicopter trip to the platform
PDE reports that careful planning was required due the location of the site, 200 kilometers offshore. All equipment had to be flown out on a helicopter that visited the platform just once a week. The seven-day time window for the survey was primarily dictated by this transportation schedule.
For each scan scene, at least one and in some cases two or three overlapping scans were made, to ensure that the point cloud was registered in a relative sense. As is typical on offshore platforms, the
scanners had to be moved many times in order to capture all the detail behind the congestion on the platform.
Once the final survey control was accepted and coordinates of the scan targets were established, only then was the relative registration turned into an absolute one.
Ongoing registration of the data was continuous. Together with a review of the data, this made sure that no areas were missed and that critical tie-in locations could be modeled to the correct accuracy, according to PDE.
Scanned geometry was modeled in Cyclone, exported to PDMS via MicroStation DGN, then converted to intelligent PDMS models
After returning to shore, PDE began processing the point-cloud data. To shorten the time needed for this, two operators using Leica/Cyra’s Cyclone software worked in parallel on the scan database. They began by modeling the structural steel to ensure that a grid similar to design drawings was indeed correct. No major deviations from project grid dimensions were found, so modeling of additional pipework, flanges and vessels continued.
The resulting Cyclone models were transferred to PDMS via MicroStation DGN. To speed up the process of converting this geometry into intelligent PDMS models, PDE has used AVEVA’s PML (Programming Macro Language) to develop routines that automate this activity. Line identification numbers were also added in PDMS.
The ability to view point clouds directly in PDMS would be very helpful, PDE reports
PDMS is the corporate design standard for Petronas Carigali for most projects, according to PDE. For small projects, 2D CAD may be used; the choice is typically left to the EPC, it reports.
The deliverable for this project was an accurate 3D PDMS model with lines/instruments tagged. Point-cloud data was used to create a detailed Cyclone model, which was then transferred from Cyclone to PDMS via MicroStation DGN. The resulting PDMS model is based on primitives. Line identification numbers were then added in PDMS, resulting in what PDE calls a “semi-intelligent” PDMS model. This was then converted to an intelligent PDMS model using PML routines developed by PDE.
Bouman of PDE reports that better integration of point-cloud data with PDMS would be very helpful. Particularly useful, he says, would be a capability to view point clouds directly inside PDMS.
This material is excerpted from Spar Point’s just-released publication
Capturing Existing Conditions with Terrestrial Laser Scanning: A Report on Opportunities, Challenges and Best Practices for Owners, Operators, Engineering/Construction Contractors and Surveyors of Built Assets and Civil Infrastructure