The permeability of complex porous materials is of interest to many engineering disciplines. This quantity can be obtained via direct flow simulation, which provides the most accurate results, but is very computationally expensive. In particular, the simulation convergence time scales poorly as the simulation domains become less porous or more heterogeneous. Semi-analytical models that rely on averaged structural properties (i.e., porosity and tortuosity) have been proposed, but these features only partly summarize the domain, resulting in limited applicability. On the other hand, data-driven machine learning approaches have shown great promise for building more general models by virtue of accounting for the spatial arrangement of the domains’ solid boundaries. However, prior approaches building on the convolutional neural network (ConvNet) literature concerning 2D image recognition problems do not scale well to the large 3D domains required to obtain a representative elementary volume (REV). As such, most prior work focused on homogeneous samples, where a small REV entails that the global nature of fluid flow could be mostly neglected, and accordingly, the memory bottleneck of addressing 3D domains with ConvNets was side-stepped. Therefore, important geometries such as fractures and vuggy domains could not be modeled properly. In this work, we address this limitation with a general multiscale deep learning model that is able to learn from porous media simulation data. By using a coupled set of neural networks that view the domain on different scales, we enable the evaluation of large ($$>512^3$$ > 512 3 ) images in approximately one second on a single graphics processing unit. This model architecture opens up the possibility of modeling domain sizes that would not be feasible using traditional direct simulation tools on a desktop computer. We validate our method with a laminar fluid flow case using vuggy samples and fractures. As a result of viewing the entire domain at once, our model is able to perform accurate prediction on domains exhibiting a large degree of heterogeneity. We expect the methodology to be applicable to many other transport problems where complex geometries play a central role.

Purpose Because of the sharply growing interest worldwide of “hard” physical-mechanical robot systems for the execution of on-site construction tasks [i.e. single-task construction robots (STCRs)], the purpose of this study is to equip development projects with a systematic design-management system model that allows to integrate the different needs and aims of stakeholders. Design/methodology/approach This paper proposes a STCR-technology management system (STCR-TMS) for the complete development cycle of STCR designs. The STCR-TMS is based on established principles from systems engineering and management and STCR-specific activities developed and tested by the authors as standalone elements in previous research work. Findings The application of the STCR-TMS revealed the practicability of the method and the underlying concepts to provide practical guidance for the development process. Additional findings indicate that the method is sufficiently generic and flexible for application to different types of robots and indifferent world regions. This research has also shown that key activities need to be addressed to increase the practicability of the STCR-TMS. Originality/value A unique characteristic of this method is the evolution with each utilization cycle. In addition, individual elements are interchangeable and can be adapted based on external circumstances. These properties allow the TMS to be applied to other fields in construction robotics. With the progression of the verification and validation of the method, know-how and certain elements can be fed into standardization activities (e.g. establishing a management system standard).

-The conventional Building Information Models (BIM) does not contain knowledge about the relation between different pieces of information. To be able to understand the information the recipient must have sufficient knowledge within the given domain. When information cannot reach to the relevant party in real time or in advance, that will cause miscommunication and delays. At this point, the Process Information Modelling (PIM), which is a process-oriented modelling approach, provides a collaborative way of planning, designing, producing, assembling, and managing the repeatable deployment process. The main objective of PIM is to formulate a process oriented database platform that allows smooth data transfer, exchange, as well as promoting seamless and constant data sharing. Digital documentation, simulation, graphical and non-graphical data are produced progressively to support the decision-making process throughout the PIM application development stage. Thus, PIM embraces real-time data processing that helps to identify issues in advance, predict malfunction, reduce production time, and influence production cost.At the moment, PIM is developed through three main phases which include design, production, and on-site assembly phase and represents a conceptual tool that provides a platform where stakeholders can access information in real-time and to make vital decisions in advance. By using a real scenario in the on-going EU Horizon 2020 research project ZERO-PLUS, the first application of the PIM concept could be demonstrated on the Freescoo HVAC system, which will not only help the project team to structure detailed guidelines within the current project but will also lay a foundation for developing a fully functional PIM application in the future.

No abstract

No abstract