Data-driven Framework Research Articles

CNN-based automated approach to crack-feature detection in steam cycle components

Periodic manual inspection by trained specialists is an important element of asset management in the nuclear industry. Detection of cracks caused by stress corrosion is an important element of remote visual inspection (RVI) in power plant steam generator components such as boilers, superheaters and reheaters. Challenges exist in the interpretation of RVI footage, such as high degree of concentration for reviewing lengthy and disorienting footage due to narrow field of view offered by endoscope. Deep learning is considered useful to automate crack detection process for improved efficiency and accuracy, and has enjoyed success in related applications. This article utilises a new application of automated crack feature detection in steam cycle components to demonstrate a transferrable data-driven framework for a variety of anomaly inspections in such structures. Specifically, a case study of superheater (a type of reactor pressure vessel head) anomaly inspection is presented to automatically detect regions of crack-like features in inspection footage with a good accuracy of 92.97 % using convolutional neural network (CNN), even in challenging cases. Due to the black-box nature of the CNN classification, the explicability of the classification results is discussed to enhance the trustworthiness of the detection system.

A hierarchical enhanced data-driven battery pack capacity estimation framework for real-world operating conditions with fewer labeled data

Battery pack capacity estimation under real-world operating conditions is important for battery performance optimization and health management, contributing to the reliability and longevity of battery-powered systems. However, complex operating conditions, coupling cell-to-cell inconsistency, and limited labeled data pose great challenges to accurate and robust battery pack capacity estimation. To address these issues, this paper proposes a hierarchical data-driven framework aimed at enhancing the training of machine learning models with fewer labeled data. Unlike traditional data-driven methods that lack interpretability, the hierarchical data-driven framework unveils the “mechanism” of the black box inside the data-driven framework by splitting the final estimation target into cell-level and pack-level intermediate targets. A generalized feature matrix is devised without requiring all cell voltages, significantly reducing the computational cost and memory resources. The generated intermediate target labels and the corresponding features are hierarchically employed to enhance the training of two machine learning models, effectively alleviating the difficulty of learning the relationship from all features due to fewer labeled data and addressing the dilemma of requiring extensive labeled data for accurate estimation. Using only 10% of degradation data, the proposed framework outperforms the state-of-the-art battery pack capacity estimation methods, achieving mean absolute percentage errors of 0.608%, 0.601%, and 1.128% for three battery packs whose degradation load profiles represent real-world operating conditions. Its high accuracy, adaptability, and robustness indicate the potential in different application scenarios, which is promising for reducing laborious and expensive aging experiments at the pack level and facilitating the development of battery technology.

A data-driven framework for thermal comfort assessment method based on user interaction

The rapid development of smart home systems requires a transition of thermal comfort studies. A feasible, user-friendly, consumer-oriented thermal comfort assessment method for large-scale residential applications is extremely urgent. However, most studies of thermal comfort models have focused on large commercial buildings. The application and deployment of such methods often rely on multiple prerequisites like relatively steady thermal conditions, regular energy consumption schedule, and additional sensor information. While situations are different in residential scenarios. In this paper, over 100 thousand air conditioner users were analyzed and a fundamental but overlooked question in thermal comfort research was established. In the household environment, the thermal condition is more complicated. Difference in age, gender, and usage preference leads to diversity in individual thermal comfort. Furthermore, introducing more sensors for better comfort prediction performance is not acceptable both from the cost perspective and the user perspective. To bridge the above missing gap, a novel thermal comfort assessment framework combining user interaction and metric learning was constructed. The proposed framework can be used to construct thermal comfort assessment systems by exploiting user interaction actions, which is a low-cost alternative and complementary to the traditional methods based on thermal equilibrium. The ease of construction makes the framework easy to integrate into smart home systems, addressing the difficulty of applying thermal comfort studies in residential scenarios.

Forecasting the dynamics of correlations in complex systems

Investigating the correlations between time series is a fundamental approach to reveal the hidden mechanisms in complex systems. However, the estimated correlations often show time-dependent behaviors, which may create uncertainty for decision-making in various scenarios. Thus, forecasting the evolution of these varying correlations may be helpful, but it is still unsolved entirely. We bridge this gap by proposing a data-driven framework: (a) we first embed all the pairwise correlations within a complex system into multivariate correlation-based series by sliding windows; (b) we then identify two different low-dimensional representations of multivariate correlation-based series through delay embedding and dimensionality reduction; (c) finally, multistep ahead predictions of varying correlations can be achieved by training a mapping between two low-dimensional representations. Both model and real-world systems are used to illustrate our framework, including finance, neuroscience, and climate. Our framework is robust and has the potential to be used for other complex systems. Hopefully, forecasting the evolution of correlations in complex systems can be a useful complementary, since existing works mainly focus on the predictions of components within the systems.

A data-driven framework for designing microstructure of multifunctional composites with deep-learned diffusion-based generative models

This paper introduces a novel integrated microstructure design methodology that replaces the common existing design approaches for multifunctional composites: 1) reconstruction of microstructures, 2) analyzing and quantifying material properties, and 3) inverse design of materials. The problem of microstructure reconstruction is addressed using the diffusion-based generative model (DGM), which is a state-of-the-art generative model formulated with a Markovian diffusion process. Then, the conditional formulation of DGM is introduced for guidance to the embedded desired material properties with a transformer-based attention mechanism, which enables the inverse design of multifunctional composites. Furthermore, a convolutional neural network (CNN)-based surrogate model is utilized to facilitate the prediction of linear/nonlinear material properties for building microstructure-property linkages. Combined, the proposed artificial intelligence-based design framework enables large data processing and database construction that is often not affordable with resource-intensive finite element method (FEM)-based direct numerical simulation (DNS) or iterative reconstruction methods. It is worth noting that the proposed DGM-based methodology is not susceptible to unstable training or mode collapse, which are common issues in generative models that are often difficult to address even with extensive hyperparameter tuning. An example case is presented to demonstrate the effectiveness of the proposed approach, which is designing mechanoluminescence (ML) particulate composites. The results show that the designed ML microstructure samples with the proposed generative and surrogate models meet the multiple design requirements (e.g., volume fraction, elastic constant, and light sensitivity). This assessment demonstrates that the proposed integrated methodology provides an end-to-end solution for practical material design applications.

Optimized variable resource allocation framework for scheduling of fast-track industrial construction projects

Efficient scheduling and resource allocation for large-scale industrial projects is challenging due to their size and complexity, especially with fast-track contracts, which often lack detailed information during the early planning phase. This paper introduces a data-driven workface planning framework to enhance scheduling and resource allocation while accommodating uncertainties and constraints (e.g., minimum and maximum resource allocation curves, dynamic predecessor relationships, congestion limits). This framework employs an integrated approach, combining time-stepped simulation with graph-based optimization. By leveraging historical data and expert knowledge, the data-driven framework mitigates certain subjective assumptions, including durations and resource allocations. In practical application, the framework generates near-optimal schedules, even with limited information. Applying the framework to a fast-track industrial construction project case study demonstrated enhanced resource allocation. These findings offer practical benefits to industrial projects regarding time and cost savings and serve as a foundation for future research in data-driven project planning approaches.

Research on composition-process-property prediction of die casting Al alloys via combining feature creation and attention mechanisms

In industrial applications, the mechanical properties of die casting Al alloy parts are influenced by a combination of material composition and forming process. This study establishes a performance data-driven framework that takes into account balanced weights of alloy compositions and process parameters. First, a compositional coefficient normalization method is proposed, in which the electronegativity, thermal, and physical descriptors are dot-multiplied into the coefficient matrix to build alloy factors for coupled compositional information. Higher prediction accuracy and lower risk of overfitting validate the feasibility of the method. Subsequently, to improve the performance of machine learning (ML) model for prediction of mechanical properties, a feature engineering method embedded in attention mechanism is used. The experimental results with high accuracy confirm the ability of the ML model to reproduce the relationships between alloy properties and balance their sensitivities. Finally, contour plots of the characterized parameters are constructed for predicting mechanical property intervals and guiding parameter design. This study innovatively applies the normalization method to the field of die casting aluminum alloy feature optimization. And combined with the attention mechanism to improve the prediction accuracy between material features and target properties, which may also be extended to other alloys.

A Data-Driven Framework for Environmental Monitoring: Integrating Machine Learning and Analytics

Effective environmental monitoring and conservation activities are very important, as evidenced by the quickening rate of environmental degradation and the pressing need for sustainable management of natural resources. Scale, accuracy, and real-time data gathering are three areas where traditional approaches frequently fall short. Artificial Intelligence (AI) has become a disruptive force in this setting, providing novel solutions that greatly improve the capacity to observe, evaluate, and deal with environmental concerns. Artificial intelligence (AI)-driven technologies, such as machine learning, computer vision, and data analytics, have the potential to completely transform environmental monitoring by making it possible to gather and analyse enormous amounts of data from a variety of sources, such as sensor networks, satellite imagery, and citizen science projects. Examining the most recent developments in AI applications for several environmental domains, including pollution management, biodiversity conservation, and climate change mitigation, this study investigates the role of AI in advancing environmental monitoring and conservation efforts. The study also addresses the difficulties and moral issues surrounding the application of AI in these settings, highlighting the necessity of interdisciplinary cooperation and strong regulatory frameworks to guarantee the ethical and efficient application of AI technology. The accuracy, mean absolute error (MAE), and root mean square error (RMSE) of the suggested approach are 96.2%, 0.401, and 0.204, respectively. AI can help create more resilient and sustainable ecosystems, which will ultimately benefit the world and its people, by being included into environmental initiatives.

An intelligent framework for forecasting and investigating corrosion in marine conditions using time sensor data

Corrosion of marine steel structures can be regarded as a time-dependent process that might result in critical strength loss and, eventually, failures. The availability of reliable forecasting models for corrosion would be useful, enabling intelligent maintenance program management, and increasing marine structure safety, while lowering in-service expenses. In this study, an intelligent framework based on a data-driven model is developed that employs a group method of data handling (GMDH) type neural network to forecast free atmospheric corrosion as time-series problem. Therefore, data from sensor data with a 30-min interval over a 110 day period that includes free atmospheric corrosion as well as environmental factors are used. In addition, the Shapley additive explanations (SHAP) technique is used to investigate the impact of the surrounding environmental factors on free atmospheric corrosion. For the performance evaluation of the proposed intelligent framework, selected comparative metrics are used. Findings demonstrate the high accuracy and efficiency of the time series data-driven framework for tackling free atmospheric corrosion progression in marine environments.

Data-driven framework for evaluating digitization and artificial intelligence risk: a comprehensive analysis

Data-driven framework for evaluating digitization and artificial intelligence risk: a comprehensive analysis

A Physics-Assisted Data-Driven Framework for Predicting Crystal Chord Length Distributions under Arbitrary Experimental Conditions

A Physics-Assisted Data-Driven Framework for Predicting Crystal Chord Length Distributions under Arbitrary Experimental Conditions

A new computational workflow to guide personalized drug therapy

Objective:Computational models are at the forefront of the pursuit of personalized medicine thanks to their descriptive and predictive abilities. In the presence of complex and heterogeneous data, patient stratification is a prerequisite for effective precision medicine, since disease development is often driven by individual variability and unpredictable environmental events. Herein, we present GreatNectorworkflow as a valuable tool for (i) the analysis and clustering of patient-derived longitudinal data, and (ii) the simulation of the resulting model of patient-specific disease dynamics. Methods:GreatNectoris designed by combining an analytic strategy composed of CONNECTOR, a data-driven framework for the inspection of longitudinal data, and an unsupervised methodology to stratify the subjects with GreatMod, a quantitative modeling framework based on the Petri Net formalism and its generalizations. Results:To illustrate GreatNectorcapabilities, we exploited longitudinal data of four immune cell populations collected from Multiple Sclerosis patients. Our main results report that the T-cell dynamics after alemtuzumab treatment separate non-responders versus responders patients, and the patients in the non-responders group are characterized by an increase of the Th17 concentration around 36 months. Conclusion:GreatNectoranalysis was able to stratify individual patients into three model meta-patients whose dynamics suggested insight into patient-tailored interventions.

Silicone/ broadleaf wood fiber /MWCNTS composite stretchable strain sensor for smart object identification

Stretchable strain sensors have been rapidly developing in recent years. Fabricating strain sensors with both large workable strain range and high sensitivity via a low-cost and straightforward process is still challenging. Here, this work created a multi-walled carbon nanotubes (MWCNTs) / broadleaf wood fiber (BWF) film, the highly micro-porous network of which allows a slow infiltration of liquid silicone in a vacuum environment to form an elastic composite, serving as a stretchable strain sensor. The sensor showed an ultra-wide strain of up to 440% and high sensitivity of GF (gage factor) of 3.3 at 0∼200% strain and of 7.8 at 200∼440% strain. In addition, the sensor is able to endure 5500 cyclic loading/unload at 100% strain with no damage in its mechanical structure and small variation in electrical resistance. To demonstrate its practicality, the sensor was incorporated into a soft pneumatic griper and utilized as a bending sensor to detect the griper deformation. The system demonstrated a high accuracy of 99.33% in the classification of lab beakers using a data-driven framework built with a trained Support Vector Machine (SVM) algorithm.

Got (Optimal) Milk? Pooling Donations in Human Milk Banks with Machine Learning and Optimization

Problem definition: Human donor milk provides critical nutrition for millions of infants who are born preterm each year. Donor milk is collected, processed, and distributed by milk banks. The macronutrient content of donor milk is directly linked to infant brain development and can vary substantially across donations, which is why multiple donations are typically pooled together to create a final product. Approximately half of all milk banks in North America do not have the resources to measure the macronutrient content of donor milk, which means pooling is done heuristically. For these milk banks, an approach is needed to optimize pooling decisions. Methodology/results: We propose a data-driven framework combining machine learning and optimization to predict macronutrient content of donations and then optimally combine them in pools, respectively. In collaboration with our partner milk bank, we collect a data set of milk to train our predictive models. We rigorously simulate milk bank practices to fine-tune our optimization models and evaluate operational scenarios such as changes in donation habits during the COVID-19 pandemic. Finally, we conduct a year-long trial implementation, where we observe the current nurse-led pooling practices followed by our intervention. Pools created by our approach meet clinical macronutrient targets approximately 31% more often than the baseline, although taking 60% less recipe creation time. Managerial implications: This is the first paper in the broader blending literature that combines machine learning and optimization. We demonstrate that such pipelines are feasible to implement in a healthcare setting and can yield significant improvements over current practices. Our insights can guide practitioners in any application area seeking to implement machine learning and optimization-based decision support. History: This paper has been accepted as part of the 2023 Manufacturing & Service Operations Management Practice-Based Research Competition. Supplemental Material: The e-companion is available at https://doi.org/10.1287/msom.2022.0455 .

Fault diagnosis of marine machinery via an intelligent data-driven framework

Large amounts of ship accidents caused by marine machinery indicate the significance of fault diagnosis for marine machinery. Since numerous onboard sensor signals contain feature information that reflects the status of marine machinery and considering the time-consuming and experience-dependent of previous data-driven fault diagnosis methods, an intelligent data-driven framework based on a multi-head attention neural network (MANet) is proposed for the fault diagnosis of marine machinery in this work. MANet integrates the multi-head attention (MHA) mechanism, convolutional layers, and residual (Res) structure. Therein, the global correlation features in long-term operating signal sequences extracted by MHA can effectively characterize the status variation of marine machinery; the local fault features in operating signals extracted by convolutional layers can effectively characterize status information at certain moments for marine machinery; the application of Res structure in MANet is conducive to alleviating the gradient propagation anomaly. With a case study for the fault diagnosis of marine machinery, the superiority of MANet in terms of the accuracy and reliability of fault diagnosis is validated by comparing it with two recently published methods. Moreover, the results under different cases of engine loads and application scenarios validate that MANet possesses promising generalization performance for fault diagnosis. Hence, the proposed method provides a new approach for the real-time fault diagnosis of marine machinery in engineering applications.

A data-driven framework to assess population dynamics during novel coronavirus outbreaks: A case study on Xiamen Island, China.

The outbreak of the Coronavirus Disease 2019 (COVID-19) has profoundly influenced daily life, necessitating the understanding of the relationship between the epidemic's progression and population dynamics. In this study, we present a data-driven framework that integrates GIS-based data mining technology and a Susceptible, Exposed, Infected and Recovered (SEIR) model. This approach helps delineate population dynamics at the grid and community scales and analyze the impacts of government policies, urban functional areas, and intercity flows on population dynamics during the pandemic. Xiamen Island was selected as a case study to validate the effectiveness of the data-driven framework. The results of the high/low cluster analysis provide 99% certainty (P < 0.01) that the population distribution between January 23 and March 16, 2020, was not random, a phenomenon referred to as high-value clustering. The SEIR model predicts that a ten-day delay in implementing a lockdown policy during an epidemic can lead to a significant increase in the number of individuals infected by the virus. Throughout the epidemic prevention and control period (January 23 to February 21, 2020), residential and transportation areas housed more residents. After the resumption of regular activities, the population was mainly concentrated in residential, industrial, and transportation, as well as road facility areas. Notably, the migration patterns into and out of Xiamen were primarily centered on neighboring cities both before and after the outbreak. However, migration indices from cities outside the affected province drastically decreased and approached zero following the COVID-19 outbreak. Our findings offer new insights into the interplay between the epidemic's development and population dynamics, which enhances the prevention and control of the coronavirus epidemic.

AI specialization for pathways of economic diversification

The growth in AI is rapidly transforming the structure of economic production. However, very little is known about how within-AI specialization may relate to broad-based economic diversification. This paper provides a data-driven framework to integrate the interconnection between AI-based specialization with goods and services export specialization to help design future comparative advantage based on the inherent capabilities of nations. Using detailed data on private investment in AI and export specialization for more than 80 countries, we propose a systematic framework to help identify the connection from AI to goods and service sector specialization. The results are instructive for nations that aim to harness AI specialization to help guide sources of future competitive advantage. The operational framework could help inform the public and private sectors to uncover connections with nearby areas of specialization.

Data-driven turbulence modelling of inherently unsteady flow in stratified water storage tanks

The flow and thermal characteristics of thermal stratification appears in a wide range of industrial applications related to energy storage. In a previous study (Xu et al. submitted to Flow, Turbulence and Combustion), it has been shown that linear closure models in unsteady Reynolds averaged Navier–Stokes (RANS) calculations are unable to accurately capture the transient flow and thermal characteristics occurring in stratified water storage tank configurations. In the current study, we extend a data-driven framework to develop turbulence and heat flux closures for a three-dimensional (3D) geometry of a stratified water storage tank, using data from a highly resolved large eddy simulation. The transient high-fidelity data set is partitioned into several time-intervals used for closure development and testing. In detail, we first present the model formulations for 3D buoyant flow. We then use the reference data to extract the most suitable input features, i.e. variables used to regress the data driven model. In order to reduce the complexity of the resulting models and accelerate the training process, a recursive basis function elimination is performed, using alignment of the different basis functions with the target turbulence stress and heat flux as metric. Using the identified features and the subsets of basis functions, novel closures are then developed using a symbolic regression framework, based on Gene Expression Programming (GEP). Models are developed either for each time interval or for all intervals at once. The different machine-learnt turbulence and heat-flux closure models are then tested for two different tank configurations where the thermal stratification is affected by positively/negatively buoyant jets within simple or complex three-dimensional geometries. It is demonstrated that the extended GEP-based data-driven turbulence modelling approach is able to create models that can accurately predict flow with inherent unsteadiness in three dimensional complex-geometry cases.

A Data-driven Framework to Reduce Diesel Spillages in Underground Mines

Several methodologies have been developed to manage diesel in open-cast mining due to its high demand and increasing diesel prices. Although the use of diesel-powered equipment in underground mines has increased over the years, effective management thereof has not received the same attention. With the advent of Industry 4.0, data can be utilised more effectively by modern businesses to identify and solve problems in a structured manner. In this study, an underground mine was used as a case study to determine whether a Data, Information, Knowledge, Wisdom (DIKW) method for diesel management could be coupled with the Six Sigma Define, Measure, Analyse, Improve, Control (DMAIC) tool to make more informed decisions and gain new insights to help reduce diesel wastage underground. The new integrated methodology identified diesel spillages and highlighted the biggest contributors to these underground spillages. The Six Sigma DMAIC domain utilised root cause analysis to determine the reason for recent systems failures, followed by the identification of practical solutions to eliminate up to 200 ML (megalitres) of diesel spillage. With this information, the case study mine stands to save over USD 175,000 per annum.

Improvement of the algebraic stress model for separated flows based on field inversion and machine learning

Due to the difficulties of precisely calculating the turbulence intensity within the separated shear layer using traditional turbulent models, computing strongly separated flows is a key task of considerable interest. In this paper, the Reynolds stress functional expression is improved toward an explicit algebraic stress model for separated flows that are similar to hump flows using a data-driven framework of field inversion and machine learning that can take model consistency into account during the model correction process. The iterative Kalman filter algorithm is utilized to address the inversion problem, and the inversion results are used as training data for correction models that are trained using random forest regression. For model verification and validation, we employ both the curved backward-facing step and bump cases. The findings indicate that the inversion produces favorable outcomes, and the enhanced model developed utilizing the inversion data exhibits good generalizability.