80 research outputs found
Strategic Engineering: Transforming P&ID Documents into Digital Twins via Machine Learning and Cloud Computing
No full textIn strategic engineering, Piping and Instrumentation Diagrams (P&IDs) historically stand as static and manually curated documents. Their transition into dynamic digital twins through advanced machine learning techniques and cloud computing frameworks marks a significant evolution. Recognizing the pivotal role of P&IDs within sectors such as petrochemical, power generation, and manufacturing, transitioning these diagrams into interactive entities offers the potential to substantially amplify operational efficacy and refine decision-making paradigms. Developed during an intensive phase at a leading technological institution, the innovative methodology adroitly combines sophisticated machine learning algorithms with the robust infrastructure of Amazon Web Services to streamline the intricate process of P&IDs digitization. The approach harnesses a spectrum of techniques, including object detection, image processing, classification models, and optical character recognition, ensuring proficient discernment of symbols, extrapolation of lines, and demarcation of interconnections. Utilizing a diverse array of AWS services, a scalable and efficient digitization pipeline emerges. The culmination produces a comprehensive CSV file and an interactive digital twin endowed with rich visual attributes, both primed for integration into auxiliary systems. A subsequent cost-benefit analysis underscores the favorable equilibrium between system performance and financial expenditure. Despite the intricate challenges encountered, the demonstrated outcomes advocate for the synergistic integration of machine learning and cloud computing in P&IDs digitization, setting a precedent for future endeavors in industrial digital transformation
SOPHIA: An Event-Based IoT and Machine Learning Architecture for Predictive Maintenance in Industry 4.0
No full textPredictive Maintenance (PdM) is a prominent strategy comprising all the operational techniques and actions required to ensure machine availability and to prevent a machine-down failure. One of the main challenges of PdM is to design and develop an embedded smart system to monitor and predict the health status of the machine. In this work, we use a data-driven approach based on machine learning applied to woodworking industrial machines for a major woodworking Italian corporation. Predicted failures probabilities are calculated through tree-based classification models (Gradient Boosting, Random Forest and Extreme Gradient Boosting) and calculated as the temporal evolution of event data. This is achieved by applying temporal feature engineering techniques and training an ensemble of classification algorithms to predict Remaining Useful Lifetime (RUL) of woodworking machines. The effectiveness of the proposed method is showed by testing an independent sample of additional woodworking machines without presenting machine down. The Gradient Boosting model achieved accuracy, recall, and precision of 98.9%, 99.6%, and 99.1%. Our predictive maintenance approach deployed on a Big Data framework allows screening simultaneously multiple connected machines by learning from terabytes of log data. The target prediction provides salient information which can be adopted within the maintenance management practice
Crystal structure of the narrow-spectrum OXA-46 class D beta-lactamase: relationship between active-site lysine carbamylation and inhibition by polycarboxylates.
No full textClass D beta-lactamases represent a heterogeneous group of active-site serine beta-lactamases that show an extraordinary panel of functional features and substrate profiles, thus representing relevant models for biochemical and structural studies. OXA-46 is a narrow-spectrum enzyme belonging to the OXA-2 subgroup which was found in a Pseudomonas aeruginosa clinical isolate from northern Italy. In this work, we obtained the three-dimensional structure of OXA-46, which shows the overall fold of active serine beta-lactamases and a dimeric quaternary structure. Significant differences with currently available structures of class D beta-lactamases were found in the loops located close to the active site, which differ in length and conformation. Interestingly, the three subunits present in the asymmetric unit showed some structural heterogeneity, only one of which presented a carbamylated lysine recognized as an important functional feature of class D enzymes. The carbamylation state of residue Lys75 appeared to be associated with different shapes and dimensions of the active site. Moreover, a tartrate molecule from the crystallization buffer was found in the active site of the noncarbamylated subunits, which interacts with catalytically relevant residues. The OXA-46 crystal asymmetric units thus interestingly present the structures of the free carbamylated active site and of the ligand-bound uncarbamylated active site, offering the structural basis for investigating the potential of new scaffolds of beta-lactamase inhibitors
Autophagy, inflammation and innate immunity in inflammatory myopathies.
No full textAutophagy has a large range of physiological functions and its dysregulation contributes to several human disorders, including autoinflammatory/autoimmune diseases such as inflammatory myopathies (IIMs). In order to better understand the pathogenetic mechanisms of these muscular disorders, we sought to define the role of autophagic processes and their relation with the innate immune system in the three main subtypes of IIM, specifically sporadic inclusion body myositis (sIBM), polymyositis (PM), dermatomyositis (DM) and juvenile dermatomyositis (JDM). We found that although the mRNA transcript levels of the autophagy-related genes BECN1, ATG5 and FBXO32 were similar in IIM and controls, autophagy activation in all IIM subgroups was suggested by immunoblotting results and confirmed by immunofluorescence. TLR4 and TLR3, two potent inducers of autophagy, were highly increased in IIM, with TLR4 transcripts significantly more expressed in PM and DM than in JDM, sIBM and controls, and TLR3 transcripts highly up-regulated in all IIM subgroups compared to controls. Co-localization between autophagic marker, LC3, and TLR4 and TLR3 was observed not only in sIBM but also in PM, DM and JDM muscle tissues. Furthermore, a highly association with the autophagic processes was observed in all IIM subgroups also for some TLR4 ligands, endogenous and bacterial HSP60, other than the high-mobility group box 1 (HMGB1). These findings indicate that autophagic processes are active not only in sIBM but also in PM, DM and JDM, probably in response to an exogenous or endogenous 'danger signal'. However, autophagic activation and regulation, and also interaction with the innate immune system, differ in each type of IIM. Better understanding of these differences may lead to new therapies for the different IIM types
Soccer Fields as Rainfall Detectors using Machine Learning: The case of Ghana
No full textAgriculture is an important source of income for many countries in the Global South, where it may account for as much as 25% of GDP. Precipitation is crucial for agriculture in countries like Ghana, where ~95% of farming is rainfed. Accurate rainfall observations are limited in Ghana. The sparse rain gauge network and the lack of weather radars make remote sensing methods a potentially attractive alternative source of rainfall data. Radar satellites, such as Sentinel-1, emit radiation that passes through the atmosphere and is scattered back to the satellite by the Earth’s surface. The backscatter measured by the satellite is correlated with the wetness of the soil but the existence of vegetation hinders straightforward quantification of soil moisture. By choosing sites with a simple and, more or less, constant phenology, it may be possible to eliminate the effect of vegetation on backscatter. Soccer field may qualify as sites with such a simple and constant phenology. The main objective of this study is to use the Sentinel-1 data over soccer fields and assess them as rainfall detectors. A machine learning approach will be used to reach this objective. This research assessed the stability and the generalization capabilities of a classification model (rain/no rain). The model was trained with and applied to different locations and periods (2019 & 2020). Ground observations from 53 Ghanaian (TAHMO) and 1 Greek stations were used. Soccer fields in Ghana and Greece were selected and their suitability as rainfall detectors was checked based on the correlation between modeled soil moisture and backscatter strength. The rain/no rain classification of the soccer fields was made with a stacked classifier that was trained and validated with both spaceborne and ground data. The classifier was tested on six different datasets from Greece and Ghana 2019 and 2020. The stability of the model was assessed by a Leave-p out cross-validation approach. The generalization in space was tested by using different environments. The generalization in time was tested by using different time periods. The results showed that the classification was stable. The minimum and maximum performances for the different testing datasets were 0.43 to 0.85. The median performance of the algorithm in Ghana for 2020 is 67%. The stacked classifier was found to have the best performance compared to other classifiers. Finally, the performance of the stacked classifier was competitive in comparison with the performance of the well-known IMERG algorithm. The study showed that there is a potential for using radar backscatter from suitable fields to detect rainfall. The classifier is stable and can be generalized in time and space under certain conditions.TWIGAWater Managemen
An event based machine learning framework for predictive maintenance in industry 4.0
No full textPredictive Maintenance concerns the smart monitoring of machine to avoid possible future failures, since because it is better to intervene before the damage occurs, saving time and money. In this paper, a Predictive Maintenance methodology based on Machine learning approach is presented and it is applied to a real cutting machine, a woodworking machinery in a real industrial group, producing accurate estimations. This kind of strategy is important to deal with maintenance problems given the ever increasing need to reduce downtime and associated costs. The Predictive Maintenance methodology implemented allows dynamical decision rules that have to be considered for maintenance prediction using a combined approach on Azure Machine Learning Studio. The Three models (RF, GBM and XGBM) allowed the accurately predict machine down ever gripped bearing thanks to the pre-processing phase
Semantic 3D city models as support for urban flood resilience: Experiences from Rotterdam
No full textThis paper presents a process to develop a CityGML-based 3D city model that, together with results from a flood simulation, can be used to investigate direct and indirect effects of floods on a city, its inhabitants and its critical infrastructure, and to quantify such effects by means of a Flood Resilience Score. In addition, the model can be used as a spatial planning support tool for urban planners to prioritise the redevelopment of certain areas and to test new spatial design decisions. First, a semantic 3D city model is prepared and enriched with additional building and infrastructure information. Then a Flood Resilience Score (FReSco) is defined and computed by quantifying the direct and indirect impacts of flooding on buildings, households, and critical infrastructure points using information from both the 3D city model and the flood simulation results. Lastly, a prototype of a spatial planning support tool is proposed to evaluate the flood resilience of a new environmental plan. As a case study, the neighbourhood of “Nieuw Kralingen” in Rotterdam was chosen. Overall, the outcomes of this work are meant to help cities better understand the impacts of flooding and adjust their urban planning activities accordingly. At the same time, the developed methodology also tests the strengths and limits of CityGML-based 3D city models in combination with openly available data and software.Urban Data Scienc
Up-regulation of neural and cell cycle-related microRNAs in brain of amyotrophic lateral sclerosis mice at late disease stage
No full textAbstract 1343: Abcc3 up-regulation confers protection from chemotherapy to NK cells in a murine model of malignant glioma
No full text- …
