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.     

Coupled steering control of a low torsional torque capsule robot in the intestine

In order to avoid the distortion dangers to the colon, caused by spiral-type capsule robot, a Low Torsional Torque Capsule Robot (LTTCR) is developed, capable of non-contact steering swimming motion in a tortuous pipe, filled with silicone oil. Due to the non-contact and suspension state of motion, the robot is isolated from the colon and inner pipe wall, achieving low torsional effect on the intestines. Based on the coupling principle of magnetic field, a spatial magnetic torque model is derived, while a method for detecting the critical coupling magnetic torque, in tortuous environment, is proposed. The fluid resistance torque in straight and tortuous pipes is studied, deriving the scope of stable steering. Simulation and experimental results have shown that higher commutation and slip angles are beneficial to robot steering, and the swimming speed of LTTCR is higher and the twist impact on the colon, during the steering process, is lower. The LTTCR has a promising prospect in non-invasive examination applications in the sectors of diagnosis and treatment, inside the human body.     

State of the art advancement in rational design of g-C3N4 photocatalyst for efficient solar fuel transformation,environmental decontamination and future perspectives

Recently, the graphite based heterogeneous photocatalysts has attained tremendous research attention in various environmental applications. Among them, the graphitic carbon nitride (g-C₃N₄) is categorized as a unique solar active particle with its outstanding intrinsic properties i.e., adequate band configuration, excellent light absorptivity and thermo-physical durability, which make it highly useful and reliable for revenue transformation and ecological concerns. Considering the intrinsic potential of g-C₃N₄ in photocatalysis, so far, no report has been done in literature for its extraordinary configuration, morphological characteristics and perspective tuning for said applications. To overcome this research gap, our primary emphasis of this review regarding photocatalysis is to provide layout as well as the advancement of visible-light-fueled materials as highly stabilized and extremely effective ones for pragmatic implementation. Thus, this existing comprehensive assessment conducts a systematic survey over visible light driven non-metal novel g-C₃N₄. The major advancement of this evaluation is the fabrication of well-designed nanosized g-C₃N₄ photocatalysts with unique configurable frameworks and compositions. Furthermore, alternative techniques in order to customize the analogue band configuration and noticeable cultivation such as metal (cation), nonmetal (anion) doping, worthy metal activating, and alloy initiation with certain semiconductors are discussed in detail. In addition to this, g-C₃N₄ photocatalytic functionalities towards photocatalytic hydrogen evolution, CO₂ photoreduction, biological metal ions deterioration as well as bacterial sanitization are also presented and discussed in detail. Therefore, we believe that such a pivotal compact assessment can provide a roadmap in several perspectives on the currently underway obstacles in the innovation of effective g-C₃N₄ catalytic design processes. Moreover, this critical assessment will ultimately serve as a useful supplement in the research area of g-C₃N₄ nanosized photocatalysts and for the researchers working on its key aspects in diverse range of natural, chemistry, engineering and environmental applications.     

A novel solution of deep learning for endoscopic ultrasound image segmentation: enhanced computer aided diagnosis of gastrointestinal stromal tumor

Multimedia Tools and Applications - Gastrointestinal stromal tumor is one of the critical tumors that doctors do not suggest to get frequent endoscopy, so there is a need for a diagnosis system...     

Looking ahead: Joint small group detection and tracking in crowd scenes

Small group detection and tracking in crowd scenes are basis for high level crowd analysis tasks. However, it suffers from the ambiguities in generating proper groups and in handling dynamic changes of group configurations. In this paper, we propose a novel delay decision-making based method for addressing the above problems, motivated by the idea that these ambiguities can be solved using rich temporal context. Specifically, given individual detections, small group hypotheses are generated. Then candidate group hypotheses across consecutive frames and their potential associations are built in a tree. By seeking for the best non-conflicting subset from the hypothesis tree, small groups are determined and simultaneously their trajectories are got. So this framework is called joint detection and tracking. This joint framework reduces the ambiguities in small group decision and tracking by looking ahead for several frames. However, it results in the unmanageable solution space because the number of track hypotheses grows exponentially over time. To solve this problem, effective pruning strategies are developed, which can keep the solution space manageable and also improve the credibility of small groups. Experiments on public datasets demonstrate the effectiveness of our method. The method achieves the state-of-the-art performance even in noisy crowd scenes.     

Exactly satisfying initial conditions neural network models for numerical treatment of first Painlevé equation

In this paper, novel computing approach using three different models of feed-forward artificial neural networks (ANNs) are presented for the solution of initial value problem (IVP) based on first Painlevé equation. These mathematical models of ANNs are developed in an unsupervised manner with capability to satisfy the initial conditions exactly using log-sigmoid, radial basis and tan-sigmoid transfer functions in hidden layers to approximate the solution of the problem. The training of design parameters in each model is performed with sequential quadratic programming technique. The accuracy, convergence and effectiveness of the proposed schemes are evaluated on the basis of the results of statistical analyses through sufficient large number of independent runs with different number of neurons in each model as well. The comparisons of these results of proposed schemes with standard numerical and analytical solutions validate the correctness of the design models.     

Monomer casting nylon-6-b-polyether amine copolymers: Synthesis and properties

MC nylon-6-b-polyether amine copolymers were prepared with macro-initiator based on amino-terminated polyether amine functionalized with isocyanate via in-situ polymerization. It was found that the introduction of polyether amine delayed the polymerization process of caprolactam by increasing apparent activation energy and pre-exponential factor, resulting in the decrease of molecular weight of nylon-6. The motion of molecular chain of the copolymers was easy because of the decreased hydrogen bonds and weakened inter-molecular forces. The physical entanglement of molecular chains of the copolymers was significant and strong which increased the entanglement density. Only the nylon-6 phase crystallized in the copolymers and the crystal grain size, spherulite size and crystallinity of the copolymers decreased. A small amount of γ crystal formed at high polyether amine content. The copolymers presented obvious strain hardening behavior in stress-strain curves and the loss factor dramatically increased while the glass transition temperature and storage module decreased. The fracture surface of the copolymers became rough and presented hairy structure, indicating that the toughening mechanism of the copolymers corresponded to the multi-layer crack extension mechanism.     

Remarkably enhanced energy storage properties of lead-free Ba0.53Sr0.47TiO3 thin films capacitors by optimizing bottom electrode thickness

The lead-free Ba_0.53Sr_0.47TiO₃ (BST) thin films buffered with La_0.67Sr_0.33MnO₃ (LSMO) bottom electrode of different thicknesses were fabricated by pulsed laser deposition method on a (001) SrTiO₃ substrate. It was found that the roughness of electrode decreases and substrate stress relaxes gradually with the increase of LSMO thickness, which is beneficial for weakening local high electric field and achieving higher E_b. Therefore, the recoverable energy density (W_rec) of BST films can be greatly improved up to 67.3 %, that is, from 30.6 J/cm³ for the LSMO thickness of 30 nm up to 51.2 J/cm³ for the LSMO thickness of 140 nm after optimizing the LSMO thickness. Furthermore, the thin film capacitor with a 140 nm LSMO bottom electrode shows an outstanding thermal stability from 20 °C to 160 °C and superior fatigue resistance after 10⁸ electrical cycles with only a slightly decrease of W_rec below 1.6 % and 3.7 %, respectively. Our work demonstrates that optimizing bottom electrodes thickness is a promising way for enhancing energy storage properties of thin-film capacitors.