Machine learning algorithm selection for windage alteration fault diagnosis of mine ventilation system

Machine learning algorithms have been widely used in mine fault diagnosis. The correct selection of the suitable algorithms is the key factor that affects the fault diagnosis. However, the impact of machine learning algorithms on the prediction performance of mine fault diagnosis models has not been fully evaluated. In this study, the windage alteration faults (WAFs) diagnosis models, which are based on K-nearest neighbor algorithm (KNN), multi-layer perceptron (MLP), support vector machine (SVM), and decision tree (DT), are constructed. Furthermore, the applicability of these four algorithms in the WAFs diagnosis is explored by a T-type ventilation network simulation experiment and the field empirical application research of Jinchuan No. 2 mine. The accuracy of the fault location diagnosis for the four models in both networks was 100%. In the simulation experiment, the mean absolute percentage error (MAPE) between the predicted values and the real values of the fault volume of the four models was 0.59%, 97.26%, 123.61%, and 8.78%, respectively. The MAPE for the field empirical application was 3.94%, 52.40%, 25.25%, and 7.15%, respectively. The results of the comprehensive evaluation of the fault location and fault volume diagnosis tests showed that the KNN model is the most suitable algorithm for the WAFs diagnosis, whereas the prediction performance of the DT model was the second-best. This study realizes the intelligent diagnosis of WAFs, and provides technical support for the realization of intelligent ventilation.     

Digital twin-based analysis of volumetric error mapping procedures

The evaluation of the volumetric accuracy of a machine tool is an open challenge in the industry, and a wide variety of technical solutions are available in the market and at research level. All solutions have advantages and disadvantages concerning which errors can be measured, the achievable uncertainty, the ease of implementation, possibility of machine integration and automation, the equipment cost and the machine occupation time, and it is not always straightforward which option to choose for each application. The need to ensure accuracy during the whole lifetime of the machine and the availability of monitoring systems developed following the Industry 4.0 trend are pushing the development of measurement systems that can be integrated in the machine to perform semi-automatic verification procedures that can be performed frequently by the machine user to monitor the condition of the machine. Calibrated artefact based calibration and verification solutions have an advantage in this field over laser based solutions in terms of cost and feasibility of machine integration, but they need to be optimized for each machine and customer requirements to achieve the required calibration uncertainty and minimize machine occupation time.This paper introduces a digital twin-based methodology to simulate all relevant effects in an artefact-based machine tool calibration procedure, from the machine itself with its expected error ranges, to the artefact geometry and uncertainty, artefact positions in the workspace, probe uncertainty, compensation model, etc. By parameterizing all relevant variables in the design of the calibration procedure, this simulation methodology can be used to analyse the effect of each design variable on the error mapping uncertainty, which is of great help in adapting the procedure to each specific machine and user requirements. The simulation methodology and the analysis possibilities are illustrated by applying it on a 3-axis milling machine tool.     

Designing optical glasses by machine learning coupled with a genetic algorithm

Engineering new glass compositions have experienced a sturdy tendency to move forward from (educated) trial-and-error to data- and simulation-driven strategies. In this work, we developed a computer program that combines data-driven predictive models (in this case, neural networks) with a genetic algorithm to design glass compositions with desired combinations of properties. First, we induced predictive models for the glass transition temperature (T_g) using a dataset of 45,302 compositions with 39 different chemical elements, and for the refractive index (n_d) using a dataset of 41,225 compositions with 38 different chemical elements. Then, we searched for relevant glass compositions using a genetic algorithm informed by a design trend of glasses having high n_d (1.7 or more) and low T_g (500 °C or less). Two candidate compositions suggested by the combined algorithms were selected and produced in the laboratory. These compositions are significantly different from those in the datasets used to induce the predictive models, showing that the used method is indeed capable of exploration. Both glasses met the constraints of the work, which supports the proposed framework. Therefore, this new tool can be immediately used for accelerating the design of new glasses. These results are a stepping stone in the pathway of machine learning-guided design of novel glasses.     

Learning shape and texture progression for young child face aging

Face aging (FA) for young faces refers to rendering the aging faces at target age for an individual, generally under 20s, which is an important topic of facial age analysis. Unlike traditional FA for adults, it is challenging to age children with one deep learning-based FA network, since there are deformations of facial shapes and variations of textural details. To alleviate the deficiency, a unified FA framework for young faces is proposed, which consists of two decoupled networks to apply aging image translation. It explicitly models transformations of geometry and appearance using two components: GD-GAN, which simulates the Geometric Deformation using Generative Adversarial Network; TV-GAN, which simulates the Textural Variations guided by the age-related saliency map. Extensive experiments demonstrate that our method has advantages over the state-of-the-art methods in terms of synthesizing visually plausible images for young faces, as well as preserving the personalized features.     

Identifying catalyst layer compositions of proton exchange membrane fuel cells through machine-learning-based approach

Membrane electrode assembly (MEA) is considered a key component of a proton exchange membrane fuel cell (PEMFC). However, developing a new MEA to meet desired properties, such as operation under low-humidity conditions without a humidifier, is a time- and cost-consuming process. This study employs a machine-learning-based approach using K-nearest neighbor (KNN) and neural networks (NN) in the MEA development process by identifying a suitable catalyst layer (CL) recipe in MEA. Minimum redundancy maximum relevance and principal component analysis were implemented to specify the most important predictor and reduce the data dimension. The number of predictors was found to play an essential role in the accuracy of the KNN and NN models although the predictors have self-correlations. The KNN model with a K of 7 was found to minimize the model loss with a loss of 11.9%. The NN model constructed by three corresponding hidden layers with nine, eight, and nine nodes can achieve the lowest error of 0.1293 for the Pt catalyst and 0.031 for PVA as a good additive blending in the CL of the MEA. However, even if the error is low, the prediction of PVA seems to be inaccurate, regardless of the model structure. Therefore, the KNN model is more appropriate for CL recipe prediction.     

A new reverse reduce-error ensemble pruning algorithm

Although greedy algorithms possess high efficiency, they often receive suboptimal solutions of the ensemble pruning problem, since their exploration areas are limited in large extent. And another marked defect of almost all the currently existing ensemble pruning algorithms, including greedy ones, consists in: they simply abandon all of the classifiers which fail in the competition of ensemble selection, causing a considerable waste of useful resources and information. Inspired by these observations, an interesting greedy Reverse Reduce-Error (RRE) pruning algorithm incorporated with the operation of subtraction is proposed in this work. The RRE algorithm makes the best of the defeated candidate networks in a way that, the Worst Single Model (WSM) is chosen, and then, its votes are subtracted from the votes made by those selected components within the pruned ensemble. The reason is because, for most cases, the WSM might make mistakes in its estimation for the test samples. And, different from the classical RE, the near-optimal solution is produced based on the pruned error of all the available sequential subensembles. Besides, the backfitting step of RE algorithm is replaced with the selection step of a WSM in RRE. Moreover, the problem of ties might be solved more naturally with RRE. Finally, soft voting approach is employed in the testing to RRE algorithm. The performances of RE and RRE algorithms, and two baseline methods, i.e., the method which selects the Best Single Model (BSM) in the initial ensemble, and the method which retains all member networks of the initial ensemble (ALL), are evaluated on seven benchmark classification tasks under different initial ensemble setups. The results of the empirical investigation show the superiority of RRE over the other three ensemble pruning algorithms.     

Friction compensation controller for load varying machine tool feed drive

The load applied to a machine tool feed drive changes during the machining process as material is removed. This load change alters the Coulomb friction of the feed drive. Because Coulomb friction accounts for a large part of the total friction the friction compensation control accuracy of the feed drives is limited if this nonlinear change in the applied load is not considered. This paper presents a new friction compensation method that estimates the machine tool load in real time and considers its effect on friction characteristics. A friction observer based on a Kalman filter with load estimation is proposed for friction compensation control considering the applied load change. A specially designed feed drive testbed that enables the applied load to be modified easily was constructed for experimental verification. Control performance and friction estimation accuracy are demonstrated experimentally using the testbed.     

Multiple Empirical Kernel Learning with Majority Projection for imbalanced problems

Traditional Multiple Empirical Kernel Learning (MEKL) expands the expressions of the sample and brings better classification ability by using different empirical kernels to map the original data space into multiple kernel spaces. To make MEKL suit for the imbalanced problems, this paper introduces a weight matrix and a regularization term into MEKL. The weight matrix assigns high misclassification cost to the minority samples to balanced misclassification cost between minority and majority class. The regularization term named Majority Projection (MP) is used to make the classification hyperplane fit the distribution shape of majority samples and enlarge the between-class distance of minority and majority class. The contributions of this work are: (i) assigning high cost to minority samples to deal with imbalanced problems, (ii) introducing a new regularization term to concern the property of data distribution, (iii) and modifying the original PAC-Bayes bound to test the error upper bound of MEKL-MP. Through analyzing the experimental results, the proposed MEKL-MP is well suited to the imbalanced problems and has lower generalization risk in accordance with the value of PAC-Bayes bound.     

Structural Cues for Understanding eEF1A2 Moonlighting

Spontaneous mutations in the EEF1A2 gene cause epilepsy and severe neurological disabilities in children. The crystal structure of eEF1A2 protein purified from rabbit skeletal muscle reveals a post-translationally modified dimer that provides information about the sites of interaction with numerous binding partners, including itself, and maps these mutations onto the dimer and tetramer interfaces. The spatial locations of the side chain carboxylates of Glu301 and Glu374, to which phosphatidylethanolamine is uniquely attached via an amide bond, define the anchoring points of eEF1A2 to cellular membranes and interorganellar membrane contact sites. Additional bioinformatic and molecular modeling results provide novel structural insight into the demonstrated binding of eEF1A2 to SH3 domains, the common MAPK docking groove, filamentous actin, and phosphatidylinositol-4 kinase IIIβ. In this new light, the role of eEF1A2 as an ancient, multifaceted, and articulated G protein at the crossroads of autophagy, oncogenesis and viral replication appears very distant from the “canonical” one of delivering aminoacyl-tRNAs to the ribosome that has dominated the scene and much of the thinking for many decades.