Irudayaraj, Andrew Xavier RajWahab, Noor Izzri AbdulPremkumar, Manoharan A.Radzi, Mohd Amran Mohd...
17页
查看更多>>摘要：This paper proposes an improved form of chaotic based atom search optimization (IASO) algorithm by adapting one-dimensional (1D) chaotic map (tent, sine and logistic) to improve the search ability by intensifying the exploration and exploitation phase. The IASO avoids premature convergence and trapping into local optima. Initially, the proposed IASO is validated using a classical benchmark function and its performance is compared with ASO algorithm. Test results indicate that the proposed algorithm outperforms in terms of mean, standard deviation, and best values. Further, the proposed technique is used to design the parameters of fractional-order proportional integral derivative controller for automatic load frequency control (ALFC) of multi-area, multi-source hybrid power system (HPS) by minimizing the integral time absolute error. The results obtained show that the proposed control scheme improves the frequency response of the system by 48 %, 70 %, 15 % and 69 % in terms of settling time, peak undershoot, steady state error value and control effort, respectively compared to ASO. Moreover, the sensitivity analysis is carried out by considering +/- 25 % variation in HPS parameters and the real-time applicability is tested with Malaysian meteorological data of solar radiation and wind speed variation. These analysis indicates that the transient oscillations are damped out with minimum settling time and the system regains to stable operating conditions. Further, the evaluation of transient and steady-state performance indices shows that the tent map-based IASO is found to be more efficient for obtaining the optimal solution in solving the ALFC problems. In addition, the stability of the system is analysed by approximating the fractional-order transfer function based on the oustaloup filter in frequency domain. (C) 2022 Elsevier B.V. All rights reserved.
查看更多>>摘要：In recent years, deep learning has been developed very quickly, and related research has shown a blossoming scene. Inception-v4 is a wide and deep network with good classification performance. The network structure is very complex, with convolution operations of different sizes, but there are two limitations: the inability to adaptively select the convolution kernel according to the characteristics of the image and the feature extraction from the high-level layer is not strong. This paper focuses on the investigation on the Inception-v4 model and has made several improvements. The improved Inception-v4 model is named BeIn-v4, which integrates the ideas of the Selective Kernel Network (SKNet) into the Inception-v4 network, and adjusts the network structure to achieve improvements. A number of comparative experiments have been carried out on the network before and after the improvements. The experimental results show that BeIn-v4 can obtain better classification results on the tested image datasets than Inception-v4. (c) 2022 Elsevier B.V. All rights reserved.
Viana, Breno M. F.Pereira, Leonardo T.Toledo, Claudio F. M.dos Santos, Selan R....
16页
查看更多>>摘要：This paper presents a search-based solution for the generation of dungeon levels with barrier mechanics and the placement of challenges and rewards in the levels' rooms. The barrier is a feature that temporarily blocks the player's progression, where one or more keys will unblock the way. The placement of barriers and keys must satisfy some constraints since the player cannot be stuck during the gameplay. Feasible-Infeasible Two-Population Genetic Algorithm (FI2Pop GA) evolves a grid representation that handles the level dependencies of barrier mechanics. We propose the concept of ordered regions to control the availability of keys better in the levels and procedures to create levels with more diversity in their contents. Data to measure the variety of the generated content is collected based on map linearity, mission linearity, leniency, and path redundancy. We analyzed our results through expressive range analysis, and it shows that our approach can generate a wide variety of playable levels. (C)& nbsp;2022 Elsevier B.V. All rights reserved.
查看更多>>摘要：This research work focuses on navigational strategy of humanoid robots in complex environments using a fuzzy embedded neural network based controller. The obstacle distances are measured from robot's current position and referred as front obstacle distance, right obstacle distance and left obstacle distance. These obstacle distances are served as input variables to the neural network model, and target angle is obtained as output parameter. The target angle obtained from neural network is fed to the Mamdani fuzzy system along with the obstacle distances as input variables to obtain the effective target angle for the humanoid robot. A Petri-net controller is embedded with developed neuro-fuzzy controller to perform dynamic path analysis in complex workspaces Single as well as multiple humanoid robots are used to analyze simulation and experimental navigation in different complex environments using developed neuro-fuzzy-petri-net controller. Various simulations are carried out using V-REP simulation software and similar scenario as per simulation is developed under laboratory conditions for various experimental navigation. The results from both the scenarios are related and are found to be in good covenant with each other having permissible range of errors. Simulation and experimental results in relation to navigational parameters shows the robustness of the developed controller. Surface plots and contour plots developed from the designed controller shows the effectiveness and efficacy in achieving global path during motion planning through optimizing target angle. To validate the results and to find out the effectiveness, the developed controller is compared with existing techniques such as IDQ and substantial progress of 16.66% in relation to path length is observed. (c) 2022 Elsevier B.V. All rights reserved.
查看更多>>摘要：Blade tip timing (BTT) is a promising non-contact measurement method for blade vibration monitoring. The limited number of probes leads to the inherent under-sampled problem of BTT signal. The quality of the analysis result of under-sampled signal depends on the probe layout. However, the application of existing probe layout optimization methods requires domain knowledge about blade vibration. Additionally, the lack of comparisons of probe layouts generated by different probe layout optimization methods makes it difficult to evaluate the performance of different methods. In this paper, we proposed a new probe layout optimization method based on a concrete autoencoder to reduce the reliance on the domain knowledge. The probe layouts were compared using an independently trained reconstructor in the time domain. And the spectral distance was defined to evaluate the performance of different probe layouts in the frequency domain. The validation results showed that the proposed method has superior time domain reconstruction performance and decent frequency domain reconstruction performance compared to other methods. The most informative and representative probe layout can be effectively selected by the proposed method. (c) 2022 Elsevier B.V. All rights reserved.
查看更多>>摘要：Seru Production is usually used as a creative mode of production in electronics. It involves two NP hard subproblems, namely seru formation and seru scheduling. To obtain the optimal global solution of multi-objective Seru Production, we develop a multi-objective cooperative coevolution algorithm with a Master-Slave mechanism. In the proposed algorithm, a cooperative mechanism is used to simultaneously optimize seru formation and seru scheduling. In addition, three usually non-dominated solutions are defined and used in the cooperative mechanism to improve the quality of non-dominated solutions, which takes more computational time. To reduce the computational time, three seru scheduling/seru formation populations evolve in parallel with the assistance of the three corresponding non-dominated seru formations/seru schedulings. To further improve the quality of non-dominated solutions, we propose a Master-Slave mechanism. The Master population is designed to store the current non-dominated solutions and communicates with three Slave populations. Extensively tested experiments show that the proposed algorithm outperforms existing algorithms for multi-objective Seru Production. (C) 2022 Elsevier B.V. All rights reserved.
查看更多>>摘要：Micro-expression is a kind of facial feature that reflects the most real emotional state hidden in the human heart. Most of the existing micro-expression recognition methods are based on manual feature extraction of subtle movements of facial muscles. Due to its short duration and weak intensity, the accurate identification of micro-expression remains a challenging task. This paper investigates micro-expression recognition based on deep learning methods and proposes a three-dimensional SE-DenseNet architecture, which fused Squeeze-and-Excitation Networks with a 3D DenseNet and can automatically integrate the spatiotemporal features extracted from each video to increase the weight of valid feature maps. The proposed architecture first obtains apex frames from each video for the most obvious facial muscle movements and then amplifies facial muscle movements using Euler video magnification to significantly alleviate the issue of small sample size and weak intensity of micro-expression recognition. Finally, the pre-processed videos are fed into the 3D SE-DenseNet for further feature extraction as well as to perform micro-expression classification. Experiments are performed on three public datasets. Our best model obtains an overall accuracy of 95.12%, 92.96%, and 82.74% on SMIC, CAS(ME)(2) and CASME-II dataset, respectively. The experimental results show that the proposed methods can well describe the considerable details of micro-expression and outperform most of the state-of-the-art methods on three public datasets. (C) 2022 Elsevier B.V. All rights reserved.
查看更多>>摘要：This paper addresses the problem of classification of object and background in unevenly illuminated images using Decision-Theoretic Rough Set (DTRS) framework. The proposed scheme employs adaptive windowing technique to partition the image into different windows. Thereafter, the proposed DTRS based method is applied on each window to find out the optimal threshold that is used for classification of the window. Determination of optimal threshold of a given window is dependent on the optimal granule size used for the window. The problem of determination of optimal granule size and optimal threshold is cast in optimization framework. The optimal threshold obtained for each window is used to classify the window and the classification of the entire image is the union of classifications over all the windows. Manual tuning of parameters is not required to determine the optimal threshold. The proposed scheme is tested on different images considered from Berkeley image database. The performance of the proposed scheme is compared with other granular and non-granular computing based schemes. Evaluation of different quantitative measures demonstrates the improved performance of the proposed schemes over others. (C)& nbsp;2022 Elsevier B.V. All rights reserved.
查看更多>>摘要：Ozone prediction, a key role for ozone pollution control, is facing the following challenges, i.e., the complex evolution trend of ozone, the cross-interference phenomena between ozone and other pollutants, and the low-quality monitoring data. To overcome the above challenges, we propose a multi-source and multivariate ozone prediction model based on fuzzy cognitive maps (FCMs) and evidential reasoning theory from the perspective of spatio-temporal fusion, termed as ERC-FCM. In this framework, an FCM-based prediction model is introduced to solve the ozone forecasting problem. Inspired by the multivariate time series forecasting, a multivariate ozone prediction problem is modeled as an FCM learned by the real-coded genetic algorithm, in which each node denotes a variable (pollutant). Thus, both the complex evolution trend of ozone and the cross-interference phenomena can be reflected by the FCM. Further, we propose an ensemble theoretical framework based on evidence reasoning theory and the matrix 2 norm. This theoretical framework relieves the negative factors from the low-quality monitoring data and improves the prediction accuracy when facing multi-source and multivariate time series. The performance of ERC-FCM is validated on two real-world datasets. The experimental results demonstrate that our method yields the best prediction performance by comparison with the other classical FCM-based methods on mean absolute error (MAE), mean square error (MSE), and root mean square error (RMSE). In addition, the Friedman test and Nemenyi test show that ERC-FCM gets relatively better prediction accuracy than other models.(c) 2022 Elsevier B.V. All rights reserved.
查看更多>>摘要：The joint extraction of entities and relations is an important task in natural language processing, which aims to obtain all relational triples in plain text. However, few existing methods excel in solving the overlapping triple problem. Moreover, most methods ignore the position and order of the words in the entity in the entity extraction process, which affects the performance of triples extraction. To solve these problems, a joint extraction model with position-aware attention and relation embedding is proposed, named PARE-Joint. The proposed model first recognizes the subjects, and then uses the subject and relation guided attention network to learn the enhanced sentence representation and determine the corresponding objects. In this way, the interaction between entities and relations is captured, and the overlapping triple problem can be better resolved. In addition, taking into account the important role of word order in the entity for triple extraction, the position-aware attention mechanism is used to extract the subjects and the objects in the sentences, respectively. The experimental results demonstrate that our model can solve the overlapping triple problem more effectively and outperform other baselines on four public datasets.(c) 2022 Elsevier B.V. All rights reserved.
NSTL
Elsevier