As-designed Model Research Articles

Towards Automatic Segmentation and Recognition of Multiple Precast Concrete Elements in Outdoor Laser Scan Data

To date, to improve construction quality and efficiency and reduce environmental pollution, the use of precast concrete elements (PCEs) has become popular in civil engineering. As PCEs are manufactured in a batch manner and possess complicated shapes, traditional manual inspection methods cannot meet today’s requirements in terms of production rate of PCEs. The manual inspection of PCEs needs to be conducted one by one after the production, resulting in the excessive storage of finished PCEs in the storage yards. Therefore, many studies have proposed the use of terrestrial laser scanners (TLSs) for the quality inspection of PCEs. However, all these studies focus on the data of a single PCE or a single surface of PCE, which is acquired from a unique or predefined scanning angle. It is thus still inefficient and impractical in reality, where hundred types of PCEs with different properties may exist. Taking this cue, this study proposes to scan multiple PCEs simultaneously to improve the inspection efficiency by using TLSs. In particular, a segmentation and recognition approach is proposed to automatically extract and identify the different types of PCEs in a large amount of outdoor laser scan data. For the data segmentation, 3D data is first converted into 2D images. Image processing is then combined with radially bounded nearest neighbor graph (RBNN) algorithm to speed up the laser scan data segmentation. For the PCE recognition, based on the as-designed models of PCEs in building information modeling (BIM), the proposed method uses a coarse matching and a fine matching to recognize the type of each PCE data. To the best of our knowledge, no research work has been conducted on the automatic recognition of PCEs from a million or even ten million of the outdoor laser scan points, which contain many different types of PCEs. To verify the feasibility of the proposed method, experimental studies have been conducted on the PCE outdoor laser scan data, considering the shape, type, and amount of PCEs. In total, 22 PCEs including 12 different types are involved in this paper. Experiment results confirm the effectiveness and efficiency of the proposed approach for automatic segmentation and recognition of different PCEs.

Influence of manufacturing geometric defects on the mechanical properties of AlSi10Mg alloy fabricated by selective laser melting

The tensile behavior of bulk AlSi10Mg components, fabricated by selective laser melting (SLM), was investigated by uniaxial tensile testing and image-based finite element simulation. The initial morphological features of the structures were imaged by micro X-ray tomography. Moreover, the reconstructed model and as-designed model were compared to quantify the process-induced defects, which remained unavoidable due to complex manufacturing processes. The un-melted AlSi10Mg powders, sticking to the melting pool after condensation, intensified the deviation of side edges. Futhermore, the unevenly distributed process-induced defects resulted in anisotropic mechanical properties of AlSi10Mg alloy. Two finite element models were developed from X-ray tomography images and CAD model, which were simulated by finite element solver ABAQUS/Standard to discuss the effect of initial morphological features on the mechanical behavior of these samples. The geometric defects have slightly reduced Young's modulus and yield strength, but remarkably increased the equivalent plastic strain of the bulk structures. Furthermore, the ultimate strength and elongation, predicted by the image-based finite element model and the ductile failure criterion, was much lower than the values predicted by the as-designed model due to the influence of geometric defects.

A closed-loop machining system for assembly interfaces of large-scale component based on extended STEP-NC

Machining process is usually needed to fettle the assembly interfaces of large-scale component before the final assembly. There are several issues that make the process time and effort consuming, such as hard-to-machine material, difficulties in positioning and alignment, errors between as-built component, and as-designed model. Shop floor optimization is essential for machining this kind of large and complex workpiece but is limited by information loss from CAD/CAM to CNC (Computer Numerical Control) using traditional G&M codes. STEP-NC has been proven to perform excellently in data exchange between CAD/CAM and CNC but it does not support some special processes in machining the assembly interfaces. To solve these matters, this paper proposes a closed-loop machining system based on extended STEP-NC data models which cover aligning and laser tracker measuring process. The system consists of: (1) a STEP-NC compliant CAD/CAM system which exports STEP-NC programs, (2) a STEP-NC high-level process controller which translates STEP-NC programs and makes closed-loop optimization, and (3) CNCs and mechanical systems which perform machining and measuring tasks. In the end, a vertical tail of a large passenger plane is used as a study case to demonstrate the advantages of the proposed data models and systems. This paper contributes to machining the assembly interfaces of large-scale components, and the future development of STEP-NC in large-scale manufacturing.

Automatic as-built modeling for concurrent progress tracking of plant construction based on laser scanning

The application of automatic as-built modeling based on laser scanning can potentially facilitate progress tracking and control in industrial plant construction. Although notable work has been conducted in the as-built modeling field, the level of automation and ability for programs to recognize semantic information is low. Semantic information, such as an installation schedule for industrial components, is vital for identifying actual construction progress. Unfortunately, as the current practices lack the ability to use robust process mapping to turn such information into corresponding as-built models, the current successful rate of recognition remains low. To fill these gaps, this article describes a new as-built modeling process for industrial components by incorporating segmentation and three-dimensional object recognition techniques from computer vision fields. Following the generation of the as-built model, the tracking process is able to identify schedule delays through deviation analysis between the as-built and four-dimensional as-designed models. The modeling process can be integrated in a concurrent construction environment, which provides precise feedback for planners and site managers to simultaneously maintain the quality of construction plans. A case study is conducted, which demonstrates that the developed process enables as-built modeling with semantic information and automatic construction progress tracking. With a certain number of as-built components of a dehydration module being captured, a successful recognition rate of over 90% is achieved. Furthermore, the processing time of the case study lies within an acceptable time period, which supports efficient progress tracking. The results show the feasibility of the developed process, which promises to save time and labor costs during construction.

Adaptive 3D Imaging and Tolerance Analysis of Prefabricated Components for Accelerated Construction

Tolerance analysis of prefabricated components poses challenges to effective quality control of accelerated construction projects in urban areas. In busy urban environments, accelerated construction methods quickly assemble prefabricated components to achieve workflows that are more efficient and reduce impacts of construction on urban traffic and business. Accelerated constructions also bring challenges of “fit-up:” misalignments between components can occur due to less detailed tolerance assessments of components. Conventional tolerance checking approaches, such as manual mock-up, cannot provide detailed geometric assessments in a timely manner. This paper proposes the integration of an adaptive 3D imaging and spatial pattern analysis methods to achieve detailed and frequent “fit-up” analysis of prefabricated components. The adaptive 3D imaging methods progressively adjust imaging parameters of a laser scanner according to the geometric complexities of prefabricated components captured in data collected so far. The spatial pattern analysis methods automatically analyze deviations of prefabricated components from as-designed models to derive tolerance networks that capture relationships between tolerances of components and identify risks of misalignments.

Variances of Cylinder Parameters Fitted to Range Data.

Industrial pipelines are frequently scanned with 3D imaging systems (e.g., LADAR) and cylinders are fitted to the collected data. Then, the fitted as-built model is compared with the as-designed model. Meaningful comparison between the two models requires estimates of uncertainties of fitted model parameters. In this paper, the formulas for variances of cylinder parameters fitted with Nonlinear Least Squares to a point cloud acquired from one scanning position are derived. Two different error functions used in minimization are discussed: the orthogonal and the directional function. Derived formulas explain how some uncertainty components are propagated from measured ranges to fitted cylinder parameters.

On-site visualization of building component erection enabled by integration of four-dimensional modeling and automated surveying

This research develops a new methodology for seamless integration of automated construction surveying with four-dimensional (4D) modeling in order to improve current practices of building component positioning and erection in terms of efficiency and quality. The building production models are represented in 4D and generated in consideration of construction engineering constraints, such as lifting capacity of tower cranes, construction method and activity sequence. The surveying data include identification, surveying time and coordinates of a limited quantity of tracking points that are marked on a building component. The data are processed using a special algorithm to derive transformation matrices, which encode movements and rotations of a solid object in the 3D space. As a result, the 3D model of the building component is updated to mirror its actual motion in the site during installation operations. Furthermore, by comparing the as-designed model and the actual model of the building product, any deviations between them are determined in terms of position offsets and rotation angles, which facilitate follow-up adjustment operations. A software system named as 4D-PosCon (acronyms of four-dimensional positioning controller) was prototyped based on the proposed methodology. Laboratory experiments were designed and carried out, validating the proposed methodology and demonstrating the prototype system of 4D-PosCon. In conclusion, the resulting 4D visualization is effective to facilitate positioning control in erecting a building component by providing intuitive perception and accurate comprehension of the relative orientation and position of the building component in reference to its final as-designed state.