SAM

Data Augmentation for Regression Machine Learning Problems in High Dimensions

GUILHAUMON, Clara — Thu, 01 Feb 2024 00:00:00 GMT

Data Augmentation for Regression Machine Learning Problems in High Dimensions GUILHAUMON, Clara; HASCOËT, Nicolas; LAVARDE, Marc; CHINESTA SORIA, Francisco; DAIM, Fatima Machine learning approaches are currently used to understand or model complex physical systems. In general, a substantial number of samples must be collected to create a model with reliable results. However, collecting numerous data is often relatively time-consuming or expensive. Moreover, the problems of industrial interest tend to be more and more complex, and depend on a high number of parameters. High-dimensional problems intrinsically involve the need for large amounts of data through the curse of dimensionality. That is why new approaches based on smart sampling techniques have been investigated to minimize the number of samples to be given to train the model, such as active learning methods. Here, we propose a technique based on a combination of the Fisher information matrix and sparse proper generalized decomposition that enables the definition of a new active learning informativeness criterion in high dimensions. We provide examples proving the performances of this technique on a theoretical 5D polynomial function and on an industrial crash simulation application. The results prove that the proposed strategy outperforms the usual ones.

Topological data analysis for lamb waves based shm method in operational conditions

LEJEUNE, Arthur — Sun, 01 Jan 2023 00:00:00 GMT

Topological data analysis for lamb waves based shm method in operational conditions LEJEUNE, Arthur; HASCOËT, Nicolas; RÉBILLAT, Marc; MECHBAL, Nazih; MONTEIRO, Eric Structural Health Monitoring (SHM) based on Lamb wave propagation is a promising solution to optimize maintenance, safety and enlarge service life of aeronautical structures. However, it remains a signiﬁcant challenge to solve requirements for performance and accuracy. In this paper, an original method based on Topological Data Analysis (TDA) is intro- duced. TDA is a multi-dimensional method which can extract the topological features from time series and point cloud. First, the TDA tool is applied to raw 1D data in order to detect damages. Then, speciﬁc pre-processing of the measured time-series based on slicing is developed to improve the persistence homology perception and to leverage topological descriptors to classify different damages. Using a Lamb wave based SHM approach, it is shown that with speciﬁc pre-processing of the measured time-series data, the topological analysis (persistent homology) for damage detection and classiﬁcation can be greatly improved. The temperature of the material has an impact on wave propagation and attenuation properties. It is important to ensure the capacity to detect and classify the damages on material on operational conditions of aerospace structures. The proposed approach enables to consider a priori physical information and provides another way to categorize damages than the traditional approaches. This work aims to characterize the temperature inﬂuence on the TDA performance to cluster damages. Finally, a strategy robust to temperature evolution is suggested to classify the plate health state. The dataset used to apply both methods comes from experimental campaigns performed on aeronautical composite plates with embedded piezoelectric transducers where different damage types have been investigated such as delamination and different impacts. In summary, this paper demonstrates that manipulating the topological the features of time-series signals using TDA provides an efﬁcient mean to separate and classify the damage natures. It opens the way for further developments on the use of TDA in SHM.

Open-Loop Control System for High Precision Extrusion-Based Bioprinting Through Machine Learning Modeling

ARDUENGO, Javier — Fri, 01 Mar 2024 00:00:00 GMT

Open-Loop Control System for High Precision Extrusion-Based Bioprinting Through Machine Learning Modeling ARDUENGO, Javier; HASCOËT, Nicolas; CHINESTA SORIA, Francisco; HASCOET, Jean-Yves Bioprinting is a process that uses 3D printing techniques to combine cells, growth factors, and biomaterials to create biomedical components, often with the aim of imitating natural tissue characteristics. Typically, 3D bioprinting adopts a layer-by-layer method, using materials known as bio-inks to build structures resembling tissues. This study introduces an open-loop control system designed to improve the accuracy of extrusion-based bioprinting techniques, which is composed of a specific experimental setup and a series of algorithms and models. Firstly, a method employing Logistic Regression is used to select the tests that will serve to train and test the following model. Then, using a Machine Learning Algorithm, a model that allows the optimization of printing parameters and enables process control through an open-loop system was developed. Through rigorous experimentation and validation, it is shown that the model exhibits a high degree of accuracy in independent tests. Thus, the control system offers predictability and adaptability capabilities to ensure the consistent production of high-quality bioprinted structures. Experimental results confirm the efficacy of this machine learning model and the open-loop control system in achieving optimal bioprinting outcomes. © 2024, Editorial Institution of Wrocaw Board of Scientific.

An enhanced topological analysis for Lamb waves based SHM methods

LEJEUNE, Arthur — Sun, 01 Oct 2023 00:00:00 GMT

An enhanced topological analysis for Lamb waves based SHM methods LEJEUNE, Arthur; HASCOËT, Nicolas; RÉBILLAT, Marc; MONTEIRO, Eric; MECHBAL, Nazih Topological data analysis (TDA) is a powerful and promising tool for data analysis, but yet not exploited enough. It is a multidimensional method which can extract the topological features contained in a given dataset. An original TDA-based method allowing to monitor the health of structures when equipped with piezoelectric transducers (PZTs) is introduced here. Using a Lamb wave based Structural Health Monitoring (SHM) approach, it is shown that with specific pre-processing of the measured time-series data, the TDA (persistent homology) for damage detection and classification can be greatly improved. The TDA tool is first applied directly in a traditional manner in order to use homology classes to assess damage. After that, another method based on slicing the temporal data is developed to improve the persistence homology perception and to leverage topological descriptors to discriminate different damages. The dataset used to apply both methods comes from experimental campaigns performed on aeronautical composite plates with embedded PZTs where different damage types have been investigated such as delamination, impacts and stiffness reduction. The proposed approach enables to consider a priori physical information and provides a better way to classify damages than the traditional TDA approach. In summary, this article demonstrates that manipulating the topological the features of PZTs time-series signals using TDA provides an efficient mean to separate and classify the damage natures and opens the way for further developments on the use of TDA in SHM.

Data-Driven Modelling of Polyethylene Recycling under High-Temperature Extrusion

CASTÉRAN, Fanny — Fri, 18 Feb 2022 00:00:00 GMT

Data-Driven Modelling of Polyethylene Recycling under High-Temperature Extrusion CASTÉRAN, Fanny; DELAGE, Karim; HASCOËT, Nicolas; AMMAR, Amine; CHINESTA SORIA, Francisco; CASSAGNAU, Philippe Two main problems are studied in this article. The first one is the use of the extrusion process for controlled thermo-mechanical degradation of polyethylene for recycling applications. The second is the data-based modelling of such reactive extrusion processes. Polyethylenes (high density polyethylene (HDPE) and ultra-high molecular weight polyethylene (UHMWPE)) were extruded in a corotating twin-screw extruder under high temperatures (350 °C < T < 420 °C) for various process conditions (flow rate and screw rotation speed). These process conditions involved a decrease in the molecular weight due to degradation reactions. A numerical method based on the Carreau-Yasuda model was developed to predict the rheological behaviour (variation of the viscosity versus shear rate) from the in-line measurement of the die pressure. The results were successfully compared to the viscosity measured from offline measurement assuming the Cox-Merz law. Weight average molecular weights were estimated from the resulting zero-shear rate viscosity. Furthermore, the linear viscoelastic behaviours (Frequency dependence of the complex shear modulus) were also used to predict the molecular weight distributions of final products by an inverse rheological method. Size exclusion chromatography (SEC) was performed on five samples, and the resulting molecular weight distributions were compared to the values obtained with the two aforementioned techniques. The values of weight average molecular weights were similar for the three techniques. The complete molecular weight distributions obtained by inverse rheology were similar to the SEC ones for extruded HDPE samples, but some inaccuracies were observed for extruded UHMWPE samples. The Ludovic® (SC-Consultants, Saint-Etienne, France) corotating twin-screw extrusion simulation software was used as a classical process simulation. However, as the rheo-kinetic laws of this process were unknown, the software could not predict all the flow characteristics successfully. Finally, machine learning techniques, able to operate in the low-data limit, were tested to build predicting models of the process outputs and material characteristics. Support Vector Machine Regression (SVR) and sparsed Proper Generalized Decomposition (sPGD) techniques were chosen to predict the process outputs successfully. These methods were also applied to material characteristics data, and both were found to be effective in predicting molecular weights. More precisely, the sPGD gave better results than the SVR for the zero-shear viscosity prediction. Stochastic methods were also tested on some of the data and showed promising results.

Learning the Parametric Transfer Function of Unitary Operations for Real-Time Evaluation of Manufacturing Processes Involving Operations Sequencing

LOREAU, Tanguy — Fri, 01 Jan 2021 00:00:00 GMT

Learning the Parametric Transfer Function of Unitary Operations for Real-Time Evaluation of Manufacturing Processes Involving Operations Sequencing LOREAU, Tanguy; CHAMPANEY, Victor; HASCOËT, Nicolas; MOURGUE, Philippe; DUVAL, Jean-Louis; CHINESTA SORIA, Francisco For better designing manufacturing processes, surrogate models were widely considered in the past, where the effect of different material and process parameters was considered from the use of a parametric solution. The last contains the solution of the model describing the system under study, for any choice of the selected parameters. These surrogate models, also known as meta-models, virtual charts or computational vademecum, in the context of model order reduction, were successfully employed in a variety of industrial applications. However, they remain confronted to a major difficulty when the number of parameters grows exponentially. Thus, processes involving trajectories or sequencing entail a combinatorial exposition (curse of dimensionality) not only due to the number of possible combinations, but due to the number of parameters needed to describe the process. The present paper proposes a promising route for circumventing, or at least alleviating that difficulty. The proposed technique consists of a parametric transfer function that, as soon as it is learned, allows for, from a given state, inferring the new state after the application of a unitary operation, defined as a step in the sequenced process. Thus, any sequencing can be evaluated almost in real time by chaining that unitary transfer function, whose output becomes the input of the next operation. The benefits and potential of such a technique are illustrated on a problem of industrial relevance, the one concerning the induced deformation on a structural part when printing on it a series of stiffeners.