Number Of Measured Data Points Research Articles

Application of unsupervised deep learning to image segmentation and in-situ contact angle measurements in a CO2-water-rock system

Rock surface wettability is a critical property that regulates multiphase flows in porous media, which can be quantified using the surface contact angle (CA). X-ray micro-computed tomography (μCT) provides an effective approach to in-situ measurements of surface CAs. However, the CA measurement accuracy depends significantly on the quality of CT image segmentation, which is the clustering of CT pixels into separate phases. Inspired by this, we developed a deep learning (DL)-based CA measurement workflow. Motivated by the recent tremendous progress in unsupervised learning techniques and aiming to avoid expensive manual data annotations, an unsupervised DL pipeline for CT image segmentation was proposed and implemented, which includes unsupervised model training and post-processing. The unsupervised model training was driven by a novel loss function constrained with feature similarity and spatial continuity and implemented by iterative forward and backward paths; the former clustered the pixel-wise feature vectors extracted by convolution neural networks, whereas the latter updated the parameters using gradient descent. An over-segmentation strategy was adopted for model training. The post-processing steps based on agglomerative hierarchical clustering (AHC) were implemented to further merge the over-segmented model output to the desired cluster number, which is intended to improve the efficiency of image segmentation. The developed unsupervised DL pipeline was compared with other commonly-used image segmentation methods using pixel-wise and physics-based evaluation metrics on a synthetic raw-image dataset, which had a known ground truth. The unsupervised DL pipeline showed the best performance. Next, the segmented images were input to an automatic CA measurement tool, and the results were validated by comparisons with manual measurements. The CA values from the manual and automatic measurements showed similar distributions and statistical properties. The automatic measurement demonstrated a wider spectrum because of the much larger number of measurement data points. The primary novelty of the unsupervised DL pipeline developed in this study lies in the novel loss function and the over-segmentation strategy associated with AHC post-processing. The workflow has been proven an efficient tool for pore-scale wettability characterization, which has a wide range of applications in fundamental studies of multiphase flows in natural porous media, which have critical implications to geological carbon sequestration, hydrocarbon energy recovery, and contaminant transport in groundwater.

Planar Array Diagnostic Tool for Millimeter-Wave Wireless Communication Systems

In this paper, a diagnostic tool or procedure based on Bayesian compressive sensing (BCS) is proposed for identification of failed element(s) which manifest in millimeter-wave planar antenna arrays. With adequate a priori knowledge of the reference antenna array radiation pattern, a diagnostic problem of faulty elements was formulated. Sparse recovery algorithms, including total variation (TV), mixed ℓ 1 / ℓ 2 norm, and minimization of the ℓ 1 , are readily available in the literature, and were used to diagnose the array under test (AUT) from measurement points, consequently providing faster and better diagnostic schemes than the traditional mechanisms, such as the back propagation algorithm, matrix method algorithm, etc. However, these approaches exhibit some drawbacks in terms of effectiveness and reliability in noisy data, and a large number of measurement data points. To overcome these problems, a methodology based on BCS was adapted in this paper. From far-field radiation pattern samples, planar array diagnosis was formulated as a sparse signal recovery problem where BCS was applied to recover the locations of the faults using relevance vector machine (RVM). The resulted BCS approach was validated through simulations and experiments to provide suitable guidelines for users, as well as insight into the features and potential of the proposed procedure. A Ka-band ( 28.9 GHz ) 10 × 10 rectangular microstrip patch antenna array that emulates failure with zero excitation was designed for far-field measurements in an anechoic chamber. Both simulated and measured far-field samples were used to test the proposed approach. The proposed technique is demonstrated to detect diagnostic problems with fewer measurements provided the prior knowledge of the array radiation pattern is known, and the number of faults is relatively smaller than the array size. The effectiveness and reliability of the technique is verified experimentally and via simulation. In addition to a faster diagnosis and better reconstruction accuracy, the BCS-based technique shows more robustness to additive noisy data compared to other compressive sensing methods. The proposed procedure can be applied to next-generation transceivers, aerospace systems, radar systems, and other communication systems.

Interpretation of soil property profile from limited measurement data: a compressive sampling perspective

Variation of soil properties with depth, i.e., the soil property profile, is a key input in geotechnical design and analysis, and it is determined during geotechnical site characterization. Determination of such a soil property profile requires extensive measurement data points from site characterization. However, the number of measurement data points from geotechnical site characterization is usually sparse and limited. As such, determining the soil property profile from a limited number of measurement points remains a challenge to geotechnical engineers. In engineering practice, the soil property profile is frequently determined with the assistance of engineering experience and judgment or statistical methods when only limited measurement data are available. Because both methods inevitably involve either subjectivity or assumptions that might contradict reality, the derived profile might not reflect the real variation of soil properties with depth. This paper aims to address this problem and develop an objective and rational approach to interpret the soil property profile from limited measurement data. The proposed approach is based on a novel sampling theory, called compressive sampling (or compressive sensing, CS), in mathematics and signal processing. Using compressive sampling, a high-resolution signal (e.g., a soil property profile in this study) can be reconstructed from a limited number of measurement data points. The reconstructed soil property profile is nearly continuous and has a resolution as high as cone penetration test (CPT) data. As it contains a large number of data points, conventional statistical methods can be applied easily. In this paper, the proposed approach is illustrated and validated using a set of real CPT data (i.e., tip resistance, qc). The results show that the proposed approach reasonably reconstructs the complete qcprofiles from a limited number of qcdata points.

Data-driven emission model structures for diesel engine management system development

This article discusses some specific data-driven model structures suitable for prediction of NOx and soot emissions from a diesel engine. The model structures can be described as local linear regression models where the regression parameters are defined by two-dimensional lookup tables. It is highlighted that this structure can be interpreted as a B-spline function. Using the model structure, models are derived from measured engine data. The smoothness of the derived models is controlled by using an additional regularization term, and the globally optimal model parameters can be found by solving a linear least squares problem. Experimental data from a five-cylinder Volvo passenger car diesel engine is used to derive NOx and soot models, using a leave-one-out cross-validation strategy to determine the optimal degree of regularization. The model for NOx emissions predicts the NOx mass flow with an average relative error of 5.1% and the model for soot emissions predicts the soot mass flow with an average relative error of 29% for the measurement data used in this study. The behavior of the models for different engine management system settings regarding boost pressure, amount of exhaust gas recirculation, and injection timing has been studied. The models react to the different engine management system settings in an expected way, making them suitable for optimization of engine management system settings. Finally, the model performance dependence on the selected model complexity and on the number of measurement data points used to derive the models has been studied.

A contrast source inversion method in the wavelet domain

We present a contrast source inversion (CSI) algorithm using truncated wavelet representations. Specifically, we represent the unknown contrast sources and the contrast function in terms of the wavelet basis functions. In order to reduce the number of wavelet coefficients for these unknowns, we apply a progressive multiscale truncation scheme based on the reconstructed contrast function. This approach increases the robustness of the CSI algorithm for noisy or limited data, and decreases the computation time as well as the memory usage. We tested the wavelet-domain CSI method using both synthetic and experimental data. The numerical experiments show that similar results with the regular spatial-domain CSI method are obtained when the number of (independent) measurement data points is comparable to the number of the unknowns in the contrast function. The advantages of the wavelet-domain CSI method become apparent as we deal with cases where the number of measurement data points is smaller than the number of unknowns in the contrast function.