Explore the vast range of titles available.

The utilization of surrogate models to approximate complex systems has recently gained increased popularity. Because of their capability to deal with black-box problems and lower computational requirements, surrogates were successfully utilized by researchers in various engineering and scientific fields. An efficient use of surrogates can bring considerable savings in computational resources and time. Since literature on surrogate modelling encompasses a large variety of approaches, the appropriate choice of a surrogate remains a challenging task. This review discusses significant publications where surrogate modelling for finite element method-based computations was utilized. We familiarize the reader with the subject, explain the function of surrogate modelling, sampling and model validation procedures, and give a description of the different surrogate types. We then discuss main categories where surrogate models are used: prediction, sensitivity analysis, uncertainty quantification, and surrogate-assisted optimization, and give detailed account of recent advances and applications. We review the most widely used and recently developed software tools that are used to apply the discussed techniques with ease. Based on a literature review of 180 papers related to surrogate modelling, we discuss major research trends, gaps, and practical recommendations. As the utilization of surrogate models grows in popularity, this review can function as a guide that makes surrogate modelling more accessible.

Federated learning ( FL ) is a promising decentralized deep learning technology, which allows users to update models cooperatively without sharing their data. FL is reshaping existing industry paradigms for mathematical modeling and analysis, enabling an increasing number of industries to build privacy-preserving, secure distributed machine learning models. However, the inherent characteristics of FL have led to problems such as privacy protection, communication cost, systems heterogeneity, and unreliability model upload in actual operation. Interestingly, the integration with Blockchain technology provides an opportunity to further improve the FL security and performance, besides increasing its scope of applications. Therefore, we denote this integration of Blockchain and FL as the Blockchain-based federated learning ( BCFL ) framework. This paper introduces an in-depth survey of BCFL and discusses the insights of such a new paradigm. In particular, we first briefly introduce the FL technology and discuss the challenges faced by such technology. Then, we summarize the Blockchain ecosystem. Next, we highlight the structural design and platform of BCFL . Furthermore, we present the attempts ins improving FL performance with Blockchain and several combined applications of incentive mechanisms in FL . Finally, we summarize the industrial application scenarios of BCFL .

Human activity recognition aims to determine actions performed by a human in an image or video. Examples of human activity include standing, running, sitting, sleeping, etc . These activities may involve intricate motion patterns and undesired events such as falling. This paper proposes a novel deep convolutional long short-term memory (ConvLSTM) network for skeletal-based activity recognition and fall detection. The proposed ConvLSTM network is a sequential fusion of convolutional neural networks (CNNs), long short-term memory (LSTM) networks, and fully connected layers. The acquisition system applies human detection and pose estimation to pre-calculate skeleton coordinates from the image/video sequence. The ConvLSTM model uses the raw skeleton coordinates along with their characteristic geometrical and kinematic features to construct the novel guided features. The geometrical and kinematic features are built upon raw skeleton coordinates using relative joint position values, differences between joints, spherical joint angles between selected joints, and their angular velocities. The novel spatiotemporal-guided features are obtained using a trained multi-player CNN-LSTM combination. Classification head including fully connected layers is subsequently applied. The proposed model has been evaluated on the KinectHAR dataset having 130,000 samples with 81 attribute values, collected with the help of a Kinect (v2) sensor. Experimental results are compared against the performance of isolated CNNs and LSTM networks. Proposed ConvLSTM have achieved an accuracy of 98.89% that is better than CNNs and LSTMs having an accuracy of 93.89 and 92.75%, respectively. The proposed system has been tested in realtime and is found to be independent of the pose, facing of the camera, individuals, clothing, etc . The code and dataset will be made publicly available.

The important part of mechanical equipment is rotating machinery, used mostly in industrial machinery. Rolling element bearings are the utmost dominant part in rotating machinery, so even small defects in these components could result in catastrophic system failure and enormous financial losses. Hence, it is crucial to create consistent and affordable condition monitoring and fault diagnosis systems that estimate severity level and failure modes and to create an appropriate maintenance strategy. The studies reveal that the fault diagnostic system focuses on single fault diagnosis of the shaft-bearing system. However, in real scenarios, the occurrence of a single fault is very unlikely. Thus, multifault diagnosis of the shaft-bearing system is of greater significance. This paper aims at steadily and broadly summarizing the development of the intelligent multifault diagnostic and condition monitoring systems. In addition, there is a rapid development of application of Internet of things, cloud computing and artificial intelligence techniques for fault diagnosis. In this paper, we summarize the study of various fault diagnostic system built on the architecture and application of these cutting-edge technologies for predictive maintenance of mechanical equipment.

Crop protection is the prime hindrance to food security. Plant diseases destroy the overall quality and quantity of agricultural products. Grape is an important fruit and a major source of vitamin C nutrients. The automatic decision-making system plays a paramount role in agricultural informatics. This paper aims to detect the diseases in grape leaves using convolutional capsule networks. The capsule network is a promising neural network in deep learning. This network uses a group of neurons as capsules and effectively represents spatial information of features. The novelty of the proposed work relies on the addition of convolutional layers before the primary caps layer, which indirectly decreases the number of capsules and speeds up the dynamic routing process. The proposed method has experimented with augmented and non-augmented datasets. It effectively detects the diseases of grape leaves with an accuracy of 99.12%. The method's performance is compared with state-of-the-art deep learning methods and produces reliable results.

The trajectory planning for Unmanned aerial vehicles (UAVs) in the dynamic environments is a challenging task. Many restrictions should be taken into consideration, including dynamic terrain collision, no-fly zone criteria, power and fuel criteria and so on. However, some methods treat dynamic restrictions as static in order to reduce cost and obtain efficient and acceptable paths. To achieve optimal, efficient and acceptable paths, we first use a high-dimension matrix, extended hierarchical graph, to model the unexplored dynamic grid environment and to convert weather conditions to passable coefficient. The original shortest path planning task is then translated to plan the safest path. Therefore, we propose a new forward exploration and backward navigation algorithm, matrix alignment Dijkstra (MAD), to pilot UAVs. We use matrix alignment operation to simulate the parallel exploration process of all cells from time t to t + 1 . This exploration process can be accelerated on a GPU. Finally, we recall the optimal path according to the navigator matrix. In addition, we can achieve UAV path planning from one source cell to multiple target cells within one single run. We validate our method on a real dynamic weather dataset and get competitive results in both efficiency and accuracy. We analyse the performance of MAD on a grid-based benchmark dataset and artificial maze data. Simulated results show MAD is especially helpful when the grid-based environment is large-scale and dynamic.

To achieve the more effective and robust trajectory tracking performance for the omnidirectional mobile robotic chassis (OMRC), a control system is designed by combining the linear active disturbance rejection control (LADRC) and proportional-integral-derivative (PID) methods, while a novel modified fruit fly optimization algorithm (Le-OFFO) is proposed to achieve the optimal parameters for the designed controller. The PID controller is utilized to the outer loop for trajectory tracking and LADRC controller is applied to the inner loop for the control of torque of each motor. A fitness function is established to evaluate the cost of control system, with several metrics and constraints taking into consideration. The proposed Le-OFFO algorithm combines the levy flight and the opposition-based learning (OBL) operators, in which the levy flight can help to escape from local optimum and the OBL-based mutation operator can enhance the exploration ability. The optimal parameters combination for the designed control system can be achieved by using the Le-OFFO to minimize the objective function. Numerous results show that the Le-OFFO has better performance of convergence speed and exploitation capability compared with other optimization algorithms in solving the controller parameter tuning problem. In addition, the effectiveness of the optimized LADRC controller is validated by results of experiments when compared with basic PID controller and original ADRC controller.

This paper presents a risk-averse stochastic programming model for the optimal design of a home energy system that integrates renewable energy generation from photovoltaic panels and a battery energy storage system. Prosumer’s loads are classified into base and programmable loads and the possibility of exploiting the flexibility of these latter is considered in the optimal design. Uncertainties associated with weather-related variables, load demand, and electricity tariffs are considered, through the definition of scenarios representing possible joint evolutions of these factors. The model incorporates a risk measure to control the cost variability and the objective function, by the conditional value-at-risk, aims at minimizing the expected costs that the prosumer may incur in a given percentage of worst-case scenarios. The approach is applied to a real case study in the residential sector calibrated on data of the Italian electricity market. Through numerical experiments and sensitivity analysis, optimal system sizing and operational strategies are derived under different risk preferences. Results demonstrate that the risk-averse stochastic programming approach leads to robust decisions, providing a balance between cost-effectiveness and reliability in the management of the home energy system.

About 300,000 leukemia cases are diagnosed every year, with the total number of active cases rising to 2.3 million in 2015. Although the number of adults diagnosed with leukemia is pretty high, this is the most common type of cancer found in children in developed countries. Its ability to recur and expensive diagnostic process make patients unable to undergo the diagnosis on a timely basis and consequently can prove fatal for many. The proposed novel model PyraNet aims to tackle the requirement of high-precision machinery and human expertise, as there might not be enough resources for the latter. Proportionate fine-tuning and construction make the model to accurately and precisely detect the presence of leukemic blast cells and classify them into their respective class types. Also, this novel architecture proposed here is a step towards solving the problem of unbalanced classes that often arises when the quantitative distribution of data within different classes is highly biased. As an initiative to tackle it, we have used multi-model-layer training. The analysis of experimental results shows that the proposed model is capable of correctly predicting a higher number of classes with better accuracy.

In various applications, such as access to mobile device, web application access, etc., user’s voice biometric authentication is one of the easy and excellent authentication factors for these applications. It can increase user’s usability, be believed to provide enhanced security unlike PINs, and improves customer experience. However, generally, in the authentication process, false acceptance is one of the fatal weaknesses since it leads to system access for the unauthorized person. Especially, in the case of the mobile environment with only a small training dataset, it is very hard to reduce the fault acceptance rate. To address this limitation of user’s voice biometric authentication, in this paper, we propose a novel approach that dramatically reduces the weakness of mis-acceptance from a given deep learning-based voice authentication with a small training dataset. To prove this improvement, we experimentally show that all test samples that were mis-accepted in a given deep learning-based voice authentication trained with a small dataset are correctly validated after applying our technique.