Modelling

The digital factory provides undoubtedly great potential for future production systems in terms of efficiency and effectivity. A key aspect on the way to realize the digital copy of a real factory is the understanding of complex indoor environments on the basis of three-dimensional (3D) data. In order to generate an accurate factory model including the major components, i.e., building parts, product assets, and process details, the 3D data that are collected during digitalization can be processed with advanced methods of deep learning. For instance, the semantic segmentation of a point cloud enables the identification of relevant objects within the environment. In this work, we propose a fully Bayesian and an approximate Bayesian neural network for point cloud segmentation. Both of the networks are used within a workflow in order to generate an environment model on the basis of raw point clouds. The Bayesian and approximate Bayesian networks allow us to analyse how different ways of estimating uncertainty in these networks improve segmentation results on raw point clouds. We achieve superior model performance for both, the Bayesian and the approximate Bayesian model compared to the frequentist one. This performance difference becomes even more striking when incorporating the networks’ uncertainty in their predictions. For evaluation, we use the scientific data set S3DIS as well as a data set, which was collected by the authors at a German automotive production plant. The methods proposed in this work lead to more accurate segmentation results and the incorporation of uncertainty information also makes this approach especially applicable to safety critical applications aside from our factory planning use case. Full article

Residual stress (RS) is the most challenging problem in metal additive manufacturing (AM) since the build-up of high tensile RS may influence the fatigue life, corrosion resistance, crack initiation, and failure of the additively manufactured components. While tensile RS is inherent in all the AM processes, fast and accurate prediction of the stress state within the part is extremely valuable and results in optimization of the process parameters to achieve a desired RS and control of the build process. This paper proposes a physics-based analytical model to rapidly and accurately predict the RS within the additively manufactured part. In this model, a transient moving point heat source (HS) is utilized to determine the temperature field. Due to the high temperature gradient within the proximity of the melt pool area, the material experiences high thermal stress. Thermal stress is calculated by combining three sources of stresses known as stresses due to the body forces, normal tension, and hydrostatic stress in a homogeneous semi-infinite medium. The thermal stress determines the RS state within the part. Consequently, by taking the thermal stress history as an input, both the in-plane and out of plane RS distributions are found from the incremental plasticity and kinematic hardening behavior of the metal by considering volume conservation in plastic deformation in coupling with the equilibrium and compatibility conditions. In this modeling, material properties are temperature-sensitive since the steep temperature gradient varies the properties significantly. Moreover, the energy needed for the solid-state phase transition is reflected by modifying the specific heat employing the latent heat of fusion. Furthermore, the multi-layer and multi-scan aspects of metal AM are considered by including the temperature history from previous layers and scans. Results from the analytical RS model presented excellent agreement with XRD measurements employed to determine the RS in the Ti-6Al-4V specimens. Full article

Tissues of the brain, especially white matter, are extremely heterogeneous—with constitutive responses varying spatially. In this paper, we implement a high-resolution Finite Element (FE) head model where heterogeneities of white matter structures are introduced through Magnetic Resonance Elastography (MRE) experiments. Displacement of white matter under shear wave excitation is captured and the material properties determined through an inversion algorithm are incorporated in the FE model via a two-term Ogden hyper-elastic material model. This approach is found to improve model predictions when compared to experimental results. In the first place, mechanical response in the cerebrum near stiff structures such as the corpus callosum and corona radiata are markedly different compared with a homogenized material model. Additionally, the heterogeneities introduce additional attenuation of the shear wave due to wave scattering within the cerebrum. Full article

In recent years, the crop growth modeling community invested immense effort into high resolution global simulations estimating inter alia the impacts of projected climate change. The demand for computing resources in this context is high and expressed in processor core-years per one global simulation, implying several crops, management systems, and a several decades time span for a single climatic scenario. The anticipated need to model a richer set of alternative management options and crop varieties would increase the processing capacity requirements even more, raising the looming issue of computational efficiency. While several publications report on the successful application of the original field-scale crop growth model EPIC (Environmental Policy Integrated Climate) for running on modern supercomputers, the related performance improvement issues and, especially, associated trade-offs have only received, so far, limited coverage. This paper provides a comprehensive view on the principles of the EPIC setup for parallel computations and, for the first time, on those specific to heterogeneous compute clusters that are comprised of desktop computers utilizing their idle time to carry out massive computations. The suggested modification of the core EPIC model allows for a dramatic performance increase (order of magnitude) on a compute cluster that is powered by the open-source high-throughput computing software framework HTCondor. Full article

In this work, a novel high-order formulation for multilayered piezoelectric plates based on the combination of variable-order interior penalty discontinuous Galerkin methods and general layer-wise plate theories is presented, implemented and tested. The key feature of the formulation is the possibility to tune the order of the basis functions in both the in-plane approximation and the through-the-thickness expansion of the primary variables, namely displacements and electric potential. The results obtained from the application to the considered test cases show accuracy and robustness, thus confirming the developed technique as a supplementary computational tool for the analysis and design of smart laminated devices. Full article

The strengths and torque of well-boiling hydrocarbons are of utmost significance. Boiling a well is one of the most critical steps in the discovery and production of oil and gas. The well’s boiling process is expensive because the drilling depth can be as much as 7000 meters. Any delay (breakdown time) in boiling costs a lot of money for hydrocarbon firms. Various boiler parameters are continuously tracked and regulated to avoid drilling delays. This paper focuses on the vibrations occurring at the bottom hole assembly (BHA) stick-slip. Two modelling methods, the lumped parameter model and the combination of the distributed–lumped (D–L) parameter model, were used and compared to the actual measurement performance. The D–L model was found to be more precise, particularly for long strings. Using the simulations, the most comprehensive modelling methodology is introduced. Full article

Earthmoving is one of the main processes involved in heavy construction and mining projects. It requires a significant share of the project budget and can dictate its overall success. Earthmoving related activities have a stochastic nature that affects the project cost and duration. In common practice, the equipment required for a project is selected using average operating cycles, neglecting the stochastic nature of operations and equipment. Ultimately this can lead to rough estimates and poor results in meeting the projects’ objectives. This research aims to provide a decision-support tool for earthmoving operations and achieve the best arrangement of equipment based on project objectives and equipment specifications by utilizing historical data. Operation simulation is identified as an efficient technique to model and analyze the stochastic aspects of the cost and duration of earthmoving operations in construction projects. Therefore, two simulation models—namely the Decision-Support Model and the Estimation Model, have been developed in the Symphony.net modeling environment to address the industry needs. The Decision-Support Model provides the best arrangement of equipment to maximize global resource utilization. In contrast, the Estimation Model captures more of the project details and compares various equipment arrangements. In this paper, these models are created, and the modeling logic is validated through a case study employing a real-world earthmoving project that demonstrates the model’s capabilities. The decision support model showed promising results in determining the optimized fleet selection when considering equipment and shift variations, and the cost model helped better quantifying the impact of the decision on the cost and profit of the project. This modeling approach can be reproduced by others in any case of interest to gain optimal fleet management. Full article

Current and future smart cities are moving towards the zero-net energy use concept. To this end, the built environment should also be designed for efficient energy use and play a significant role in the production of such energy. At present, this is achieved by focusing on energy demand in buildings and to the renewable trade-off related to smart power grids. However, urban water distribution systems constantly carry an excess of hydraulic energy that can potentially be recovered to produce electricity. This paper presents a comprehensive review of current strategies for energy production by reviewing the state-of-the-art of smart water systems. New technologies (such as cyber-physical systems, digital twins, blockchain) and new methodologies (network dynamics, geometric deep learning) associated with digital water are also discussed. The paper then focuses on modelling the installation of both micro-turbines and pumps as turbines, instead of/together with pressure reduction valves, to further demonstrate the energy-recovery methods which will enable water network partitioning into district metered areas. The associated benefits on leakage control, as a source of energy, and for contributing to overall network resilience are also highlighted. The paper concludes by presenting future research directions. Notably, digital water is proposed as the main research and operational direction for current and future Water Distribution Systems (WDS) and as a holistic, data-centred framework for the operation and management of water networks. Full article

Weather events are arbitrary, and this makes it difficult to incorporate weather parameters into transportation models. Recent research on traffic weather interaction analysis conducted at the University of Regina, Canada reported traffic variations with cold temperatures and snowfall. The research team at the University of Regina proposed a linear association between snowfall and temperature to analyze the traffic variation on provincial highways during winter months. The variations were studies with the inclusion of the expected daily volume factor as an independent variable in the model structure. However, the study did not analyze the nature of the association between daily truck traffic volume and snowfall. Based on these drawbacks of the past studies, in this research, the objective is to focus on the effects of snow and temperature on traffic volume changes with a methodological help of Maximal Information Coefficient (MIC), which stems from the maximal information-based nonparametric exploration (MINE) statistics. The results obtained from the analysis indicate that the relationship between snow and truck traffic is non-linear. However, the study could not establish any functional relationship between snowfall and daily truck volume. It is desired to further conduct an hourly analysis to explore a new relationship between snowfall and truck volume. Full article

To keep up to date, manufacturing enterprises need to use the latest results from the ICT sector, especially when collaborating with external partners in a supply chain and exchanging products and data. This has led to dealing with an increasing amount of heterogeneous information exchanged between partners including machines (physical means), humans and IT in the Supply Chain of ICT Systems (SC-ICTS). In this context, interoperability management is becoming more and more critical, but paradoxically, it is not yet fully efficiently anticipated, controlled and accompanied to recover from incompatibilities issues or failures. This paper intends to present how enterprise modeling, enterprise interoperability and model driven approaches can lead, together with system engineering architecture, to contribute to developing and improving the interoperability in the SC-ICTs. Model Driven System Engineering Architecture (MDSEA) is based on Enterprise Modeling using GRAI Model and its extensions. It gives enterprise internal developments guidelines, but originally, MDSEA is not the considering interoperability that is required between partners when setting a collaboration in the frame of SC-ICTS. As a result, the MDSEA, extended with interoperability concerns, led to the design of the MDISE (Model Driven Interoperability System Engineering) framework, which capitalizes on the research on enterprise interoperability. To finish, some proposals are made to extend the Model System Tool Box (MSTB) and the use of MDISE for Cyber Physical System (CPS) that are relevant components of SC-ICTS. Full article

Nonhomogeneous material characteristics of masonry lead to complex fracture mechanisms, which require substantial analysis regarding the influence of masonry constituents. In this context, this study presents a discontinuum modeling strategy, based on the discrete element method, developed to investigate the tensile fracture mechanism of masonry wallettes parallel to the bed joints considering the inherent variation in the material properties. The applied numerical approach utilizes polyhedral blocks to represent masonry and integrate the equations of motion explicitly to compute nodal velocities for each block in the system. The mechanical interaction between the adjacent blocks is computed at the active contact points, where the contact stresses are calculated and updated based on the implemented contact constitutive models. In this research, different fracture mechanisms of masonry wallettes under tension are explored developing at the unit–mortar interface and/or within the units. The contact properties are determined based on certain statistical variations. Emphasis is given to the influence of the material properties on the fracture mechanism and capacity of the masonry assemblages. The results of the analysis reveal and quantify the importance of the contact properties for unit and unit–mortar interfaces (e.g., tensile strength, cohesion, and friction coefficient) in terms of capacity and corresponding fracture mechanism for masonry wallettes. Full article

The purpose of this study is to find the right model to plan and predict future evolution paths of an evolving software architecture based on past evolution data. Thus, in this paper, a model to represent the software architecture evolution process is defined. In order to collect evolution data, a simple formalism allowing to easily express software architecture evolution data is introduced. The sequential pattern extraction technique is applied to the collected evolution styles of an evolving software architecture in order to predict and plan the future evolution paths. A learning and prediction model is defined to generate the software architecture possible future evolution paths. A method for evaluating the generated paths is presented. In addition, we explain and validate our approach through a study on two examples of evolution of component-oriented software architecture. Full article

A solar dryer for wood was constructed and modelled based on the climatic condition of Epinal, France, during the summer, spring and winter seasons. The solar dryer was able to raise the temperature of the drying air by 38 °C in the spring and summer season with a global effective efficiency of 39%. Modelling of the drying of the log of wood was based on the global mass transfer coefficient and the geometric form of the log which was mostly cylindrical was considered. Validation was undertaken with the log covered with the bark. The coefficient of variations of numerical points with the experimental values given by the model was less than 5% with a mean average error and a mean relative error of 2.33% and 4.53%, respectively. Full article

In this paper we introduce a new method for the registration between preoperative and intraoperative computerized tomography (CT) images used in endovascular interventions for aortic aneurysm repair. The method relies on a 3D finite-element model (FEM) of the aortic centerline reconstructed from preoperative CT scans. Intraoperative 2D fluoroscopic images are used to deform the 3D FEM and align it onto the current aortic geometry. The method was evaluated on clinical datasets for which a reference CT scan was available to evaluate the registration errors made by our method and to compare them with other registration methods based on rigid transformations. Errors were estimated based on the predicted locations of landmarks positioned at different branch ostia. It appeared that our method always reduced the registration errors of at least 20% compared to gold standard 3D rigid registration and permitted to reach a global precision of 3.8 mm and a renal precision of 2.6 mm, which is a significant improvement compatible with surgical specifications. Finally, the major asset of our method is that it only requires one fluoroscopic intraoperative 2D image to perform the 3D non-rigid registration. This would reduce patient irradiation and cut the costs compared to traditional methods. Full article

Distributed monitoring sensor networks are used in an ever increasing number of applications, particularly with the advent of IoT technologies. This has led to a growing demand for unconventional analytical tools to cope with a large amount of different signals. In this scenario, the modeling of time series in similar groups represents an interesting area especially for feature subset selection (FSS) purposes. Methods based on clustering algorithms are very promising for FSS, but in their original form they are unsuitable to manage the complexity of temporal dynamics in time series. In this paper we propose a clustering approach, based on complex network analysis, for the unsupervised FSS of time series in sensor networks. We used natural visibility graphs to map signal segments in the network domain, then extracted features in the form of node degree sequences of the graphs, and finally computed time series clustering through community detection algorithms. The approach was tested on multivariate signals monitored in a 1 MW cogeneration plant and the results show that it outperforms standard time series clustering in terms of both redundancy reduction and information gain. In addition, the proposed method demonstrated its merit in terms of retention of information content with respect to the original dataset in the analyzed condition monitoring system. Full article