Data fusion for improved circularity through higher quality of prediction and increased reliability of inspection

The paper presents an innovative approach to seaport security problems. In particular the authors propose the modelling & simulation and data fusion integration to provide an efficient tool to test and improve the container inspection reliability taking into consideration - at the same time - the impact of different security levels on system performances. In this context the opportunity given by new standards and normative, in terms of sharing information, highlights the possibility to use simulation as well as data fusion for analyzing different aspects (related to security) enhancing the container selection approach based on container risk evaluation (as strongly required, for instance, by Customs-Trade Partnership Against Terrorism, C-TPAT).

The paper presents an innovative approach to seaport security problems. In particular the authors propose the modelling & simulation and data fusion integration to provide an efficient tool to test and improve the container inspection reliability taking into consideration - at the same time - the impact of different security levels on system performances. In this context the opportunity given by new standards and normative, in terms of sharing information, highlights the possibility to use simulation as well as data fusion for analyzing different aspects (related to security) enhancing the container selection approach based on container risk evaluation (as strongly required, for instance, by Customs-Trade Partnership Against Terrorism, C-TPAT).

A promising application of Process Analytical Technology to the downstream process of monoclonal antibodies(mAbs) is the monitoring of the Protein A load phase as its control promises economic benefits. Different spectroscopic techniques have been evaluated in literature with regard to the ability to quantify the mAb concentration in the column effluent. Raman and Ultraviolet (UV) spectroscopy are among the most promising techniques. In this study, both were investigated in an in-line setup and directly compared. The data of each sensor were analyzed independently with Partial-Least-Squares (PLS) models and Convolutional Neural Networks (CNNs) for regression. Furthermore, data fusion strategies were investigated by combining both sensors in hierarchical PLS models or in CNNs. Among the tested options, UV spectroscopy alone allowed for the most precise and accurate prediction of the mAb concentration. A Root Mean Square Error of Prediction (RMSEP) of 0.013 g L-1 was reached with the UV-based PLS model. The Raman-based PLS model reached an RMSEP of 0.232 g L-1 . The different data fusion techniques did not improve the prediction accuracy above the prediction accuracy of the UV-based PLS model. Data fusion by PLS models seems meritless when combining a very accurate sensor with a less accurate signal. Furthermore, the application of CNNs for UV and Raman spectra did not yield significant improvements in the prediction quality. For the presented application, linear regression techniques seem to be better suited compared with advanced nonlinear regression techniques, like, CNNs. In summary, the results support the application of UV spectroscopy and PLS modeling for future research and development activities aiming to implement spectroscopic real-time monitoring of the Protein A load phase.

Visual inspection plays an important role in aviation maintenance. Human reliability analysis (HRA) in this field is necessary and can bring benefits to better human error management. Because of the lack of safety data, the present paper aims to introduce the Bayesian network (BN) approach to perform HRA in visual inspection, which permits the utilization of multi-disciplinary sources of objective and subjective information. In this paper, significant influence factors of visual inspection are identified according to the Human Factors Analysis and Classification System–Maintenance Extension. Then a network representing the visual inspection performance model is constructed. Expert opinions, data fusion from accident reports, and related literature are utilized in the step of obtaining parameters. Two canonical models used in probabilistic network model building, the Noisy-OR gates and the Recursive Noisy-OR rule, are applied to generate conditional probabilities from parameters obtained by the absolute probability judgment technique. Through the BN inference, the inspection reliability can be assessed, and some conclusions and recommendations are drawn that could provide theoretical base and data support to make interventions for safety management of visual inspection.

Vehicle dynamic safety warning system based on data fusion is researched and designed in this paper. Dynamic information of vehicle, road and environment can be collected in real time. By analyzing and processing these data, active safety warning information is achieved for individual vehicle on different work condition. Security characteristic parameter, road conditions and traffic meteorological information can be detected automatically by the vehicle dynamic security supervision system. Active warning and supervision of comprehensive vehicle safety performance are achieved. The warning system owns the function of data acquisition, operation, wireless data transmission, information distribution and parameter preserving. Warning information is transferred by wireless network in real time. Research work in this paper is applied to road traffic control area. This vehicle warning system is of value to removing incipient fault of running vehicle, improving reliability, rational maintaining and proper inspection.

Photovoltaic (PV) power installed capacity of China has been growing rapidly with marked improvements in renewable energy accommodation during recent years. it is necessary to improve the accuracy of power forecasts, therefore the underlying electrical grid can be operated in a cost efficient way. This paper aims to improving the accuracy of short-term PV power predictions. Firstly, Measured power data, satellite-based data and numerical weather prediction data are utilized. The data sets of these sources are preprocessed and fused with machine learning techniques to get the sequence feature information. In particular, support vector machine based on data fusion (SVM-DF) is proposed to run as the main regression model. SVM-DF is an extension of the support vector machine and is capable of learning regression functions in continuous space by identifying structures in the mapping of input to output data. One advantage over Artificial Neural Network (ANN) is its ability to construct non-linear dependencies between the input and output data sets. Compared with predicting the PV power through classical ANN model, the SVM-DF approach acquire a more accurate dataset. Predictions of the higher quality can be achieved by SVM-DF with access to PV measurements and weather forecasts. To study further possible improvements in prediction quality, a study on NWP weather parameters, is performed to evaluate their suitability as input features for PV power forecasting. The results confirm the importance of data fusion which make use of spatio-temporal correlations between stations. Results showed that the SVM-DF model performed better than ANN model.

In mammalian cell culture producing therapeutic proteins, one of the important challenges is the use of several complex raw materials whose compositional variability is relatively high and their influences on cell culture is poorly understood. Under these circumstances, application of spectroscopic techniques combined with chemometrics can provide fast, simple, and non-destructive ways to evaluate raw material quality, leading to more consistent cell culture performance. In this study, a comprehensive data fusion strategy of combining multiple spectroscopic techniques is investigated for the prediction of raw material quality in mammalian cell culture. To achieve this purpose, four different spectroscopic techniques of near-infrared, Raman, 2D fluorescence, and X-ray fluorescence spectra were employed for comprehensive characterization of soy hydrolysates which are commonly used as supplements in culture media. First, the different spectra were compared separately in terms of their prediction capability. Then, ensemble partial least squares (EPLS) was further employed by combining all of these spectral datasets in order to produce a more accurate estimation of raw material properties, and compared with other data fusion techniques. The results showed that data fusion models based on EPLS always exhibit best prediction accuracy among all the models including individual spectroscopic methods, demonstrating the synergetic effects of data fusion in characterizing the raw material quality.

In automated NDE a region of an inspected component is typically interrogated several times, be it within a single data channel, across multiple channels or over the course of repeated inspections. The systematic combination of these diverse readings is recognised to provide a means to improve the reliability of the inspection, for example by enabling noise suppression. Specifically, such data fusion makes it possible to declare regions of the component defect-free to a very high probability whilst readily identifying indications. The paper consists of two parts: the first addresses the computational challenges associated with indication detection in large datasets, while the latter outlines an approach to combining different sections of different amplitude fields for detecting likely indications and evaluating associated probabilities, involving spatial statistics.

Predictive model-based quality inspection using Machine Learning and Edge Cloud Computing

Flow of water through stream networks directly impacts flooding and transport of sediments and pollutants in watershed systems. Hence, knowledge of streamflow is critical for water management and mitigation of flooding and drought events. Unfortunately, spatially dense networks of in situ streamflow measurements are generally unavailable and would be prohibitively expensive to deploy and maintain. Thus, a data fusion framework is needed that utilizes available data to predict streamflow. Observed data in spatial (e.g., topography and land cover), temporal (e.g., streamflow and groundwater levels), and spatiotemporal domains (e.g., rainfall) impact streamflow. Some of these quantities can be obtained from remote sensing imagery; however, combining such disparate data types using traditional data fusion methods is problematic. Physically based hydrologic models have been used to predict streamflow but often with significant uncertainty because numerous assumptions are made for many unmeasured input and parameter values. Traditional Bayesian inference approaches suffer from superlinear increases in computational complexity as the number of data sets to be fused grows. In this paper, a scalable spatiotemporal approach based on Bayesian networks (BNs) is presented for estimating streamflow. An information-theoretic methodology based on conditional entropy is employed to quantify the impact of adding nodes in the BN in terms of information gained. The framework offers the flexibility of embedding knowledge from hydrologic models calibrated for the study area by introducing them as additional nodes in the network, thereby improving prediction accuracy. Posterior probabilities of estimates and the associated entropy provide valuable information on the quality of predictions and also offer directions for future watershed instrumentation.

In recent years with the digital transformation and industry 4.0, the Digital Twin (DT) has become a topic of relevance as a response to the need to solve problems in industries, providing visual aid in decision-making in production processes, through the interconnection between the data provided in physical space and virtual spaces, the fusion of data through the use of different tools can transform the real world and provide different services such as program verification, process optimization, quality prediction, which can be applied to each of the areas that make up organizations, DT is a technology to be used very shortly, improving the quality of processes and products. A literature review based on the aspects of sensing, data management, modelling, and services, is used for identifying the main characteristics of DT that are addressed in quality management.

This article is aimed at demonstrating the feasibility of combining water quality observations with modeling using data fusion techniques for efficient nutrients monitoring in the Shenandoah River (SR). It explores the hypothesis; “Sensitivity and uncertainty from water quality modeling and field observation can be improved through data fusion for a better prediction of water quality.” It models water quality using water quality simulation programs and combines the results with field observation, using a Kalman filter (KF). The results show that the analysis can be improved by using more observations in watersheds where minor variations to the analysis result in large differences in the subsequent forecast. Analyses also show that while data fusion was an invaluable tool to reduce uncertainty, an improvement in the temporal scales would also enhance results and reduce uncertainty. To examine how changes in the field observation affects the final KF analysis, the fusion and lab analysis cross-validation showed some improvement in the results with a very high coefficient of determination.

This article is aimed at demonstrating the feasibility of combining water quality observations with modeling using data fusion techniques for efficient nutrients monitoring in the Shenandoah River (SR). It explores the hypothesis; “Sensitivity and uncertainty from water quality modeling and field observation can be improved through data fusion for a better prediction of water quality.” It models water quality using water quality simulation programs and combines the results with field observation, using a Kalman filter (KF). The results show that the analysis can be improved by using more observations in watersheds where minor variations to the analysis result in large differences in the subsequent forecast. Analyses also show that while data fusion was an invaluable tool to reduce uncertainty, an improvement in the temporal scales would also enhance results and reduce uncertainty. To examine how changes in the field observation affects the final KF analysis, the fusion and lab analysis cross-validation showed some improvement in the results with a very high coefficient of determination.

In order to study the quality analysis method of key links in smart energy meters, the author proposes a data fusion based quality analysis and prediction method for smart energy meters. This method is based on the relevant data of key links in the electric energy meter, and selects the data of the electric energy meter in research and development design, material procurement, production and manufacturing, acceptance testing, installation and operation, dismantling and scrapping as the sample data for model construction. The XGBoost algorithm classification method is used to establish an intelligent electric energy meter quality analysis model. Taking the dismantled electricity meter data of a certain power company as an example, this paper conducts modeling analysis and prediction of various quality issues of smart electricity meters, and conducts on-site verification. Based on the verification results, the model is continuously optimized. The results indicate that: The model was optimized using cross validation and grid search methods, and the final model achieved an accuracy rate of 0.74 and a recall rate of 0.82 on the validation set. This method can meet the actual needs of power grid business and objectively reflect the quality situation of key links in smart energy meters.

Vehicle dynamic safety warning system based on data fusion is researched and designed in this paper. Dynamic information of vehicle, road and environment can be collected in real time. By analyzing and processing these data, active safety warning information is achieved for individual vehicle on different work condition. Safety parameters of vehicle and road condition are detected. Warning information is transferred by wireless network in real time. Research work in this paper is applied to road traffic control area. This vehicle warning system is of value to removing incipient fault of running vehicle, improving reliability, rational maintaining and proper inspection.