查看更多>>摘要：Power transformers are an indispensable equipment in power transmission and distribution systems, and failures or hidden defects in power transformers can cause operational and downtime issues in power supply, resulting in economic and resource losses. Therefore, it is highly desirable to put in place intelligent preventive maintenance measures to diagnose and evaluate the condition of power transformers. Although conventional methods have achieved success in detecting problems associated with power transformers, their adoption rate in practical environments is still far from universal. The advent of Computational Intelligence (CI) models offers useful potential to complement the existing diagnostic practices of power transformers. In this paper, we provide a review on various computational intelligence techniques for fault detection and diagnosis pertaining to preventive maintenance of power transformers. An overview of each representative CI approach is presented to facilitate researchers in selecting an appropriate method for a specific problem at hand. We carry out a broad discussion on numerous concerns and challenges that are missing from the current literature, which, nevertheless, need to be addressed seriously. We identify the research gaps in the literature, and suggest the way forward in research that will in the long run enhance power system reliability by embracing CI approaches into business operations in an effort to realize the Sustainable Development Goal (SDGs) advocated by the United Nation, primarily SDG7: Clean and Affordable Energy and SDG9: Industry, Innovation and Infrastructure. (C) 2021 Elsevier B.V. All rights reserved.
查看更多>>摘要：Internal audit is an independent and objective assurance and consulting activity that aims to improve the operations of an organization and add value to them. Planning internal audit by prioritizing the units to be audit is critical in terms of effective use of available audit and financial resources. In this paper, a new ELimination and Choice Translating Reality (ELECTRE) based decision support model is developed for addressing an internal audit prioritization problem. Spherical fuzzy sets are used for modeling the uncertainty in the nature of the problem and three different approaches are proposed within the study. The first approach is constructed with gradual concordance and discordance sets by comparing spherical fuzzy membership, non-membership, and hesitancy degrees of alternatives; the second approach is developed based on a single type of outranking relation obtained by utilizing score and accuracy functions of spherical fuzzy sets, and the third approach provides an increased fuzziness modeling capacity by using interval-valued spherical fuzzy sets. In the application part of the study, the units of an organization are prioritized for internal audit activity based on five components of the internationally recognized Committee of Sponsoring Organizations (COSO) framework. Sensitivity analyses for decision-maker and criterion weights and a comparative analysis with six other state-ofthe-art multi-criteria decision making (MCDM) models are also presented to analyze the consistency and validity of the proposed spherical fuzzy ELECTRE model. (C) 2021 Elsevier B.V. All rights reserved.
查看更多>>摘要：Y Analyzing the relations and patterns that exist in complex data sets is an integral part of the research in complex fuzzy set theory. The main object of study in this paper is the interval-valued complex fuzzy set (IV-CFS) model. This adaptation of complex fuzzy sets can handle datasets with a time-periodic feature, and the partial ignorance that exists in the data as well as the process of assigning values for the membership functions, in addition to modeling multi-dimensional data. This paper focuses on finding the patterns and relations between complex data sets using the properties of interval valued complex fuzzy sets (IV-CFSs). To achieve this objective, this paper establishes the concept of relations and the composition operation for IV-CFSs using the extensive properties of the Cartesian product. Some of the algebraic properties of the relations and compositions are also introduced to define the equivalence relation between IV-CFSs. The proposed method is then applied to an MCDM problem related to customers & rsquo; online shopping preferences and behavior. A detailed case study of this MCDM problem is then presented through the interpretation of the results that were obtained. A brief comparison is then presented between our proposed method and other methods in literature used to analyze patterns between complex data sets. (C) 2021 Elsevier B.V. All rights reserved.
查看更多>>摘要：YY It is hard to obtain sufficient fault samples in most real-world industrial scenarios. This has raised the need of addressing the critical issue of imbalanced fault diagnosis that remains a major challenge for popular fault diagnosis methods such as the autoencoder(AE). In this research, we propose non-revisiting genetic cost-sensitive sparse autoencoder(NrGCS-SAE) solution, which not only incor-porates cost-sensitive learning with sparse autoencoder but also solves the problem of class weights assignment. Specifically, sparse autoencoder is adopted as it has better generalization performance than autoencoder, and genetic algorithm(GA) is employed to optimize class weights that are initially unknown. In addition, a non-revisiting strategy is devised to prevent repeated evaluation of the same individual in different generations, which can help increase exploration ability and decrease computing costs. Computational experiments are used to evaluate the proposed NrGCS-SAE solution on the Tennessee Eastman(TE) dataset and the real plasma etching process dataset, which involves both binary imbalanced fault diagnosis and multi-class imbalanced faults diagnosis. As evidenced in the tests, NrGCS-SAE achieves improved performance and more importantly this improvement is consistent in different settings of experiments. (C) 2021 Elsevier B.V. All rights reserved.
查看更多>>摘要：As the Internet of Things (IoT) becomes more and more intelligent, a new computing paradigm, predictive intelligence is incorporated into many IoT applications. The devices of predictive intelligence in IoT applications must consider the power and delay consumption. As the Power-Driven X-Routing (PDXR) problem model under the advanced semiconductor design, the length-restricted condition in the multi-dynamic voltage model is introduced to save power consumption and the X-architecture is introduced to better reduce the wirelength to optimize the chip delay. To this end, an effective particle swarm optimization-based power-driven length-restricted X-routing algorithm is proposed for predictive intelligence in IoT applications. Firstly, a pre-calculated lookup table is designed to provide fast information query for the subsequent algorithm flow. Secondly, an improved particle swarm optimization algorithm is presented for the discrete PDXR problem. Thirdly, in the adjustment phase, the choice of intermediate nodes is expanded, and is no longer limited to the corner points of obstacles. Fourthly, a removal strategy of redundant points is proposed to optimize the routing path. Finally, the wirelength is further reduced by a local topology optimization strategy. The experimental results show that the proposed algorithm can achieve the best wirelength cost at a very fast speed under the constraint of restricted wirelength, so as to better satisfy the demand of the power and delay performance of semiconductor design for predictive intelligence in IoT applications. (C) 2021 Elsevier B.V. All rights reserved.
查看更多>>摘要：Interval many-objective optimization problems (IMaOPS) are ubiquitous in practical applications. Therefore, it is of great significance to study the solving method for IMaOPS. However, there are fewer solving methods due to the uncertain interval of the objective function. In this paper, an improved NSGA-III algorithm (named LFOA-NSGA-III) is proposed to effectively solve these problems. Due to the uncertain interval in the IMaOPs, the original NSGA-III algorithm can ineffectively evaluate the relationship between the interval solution set and the reference point. So the matter-element extension model is introduced, which can make the optimized solution set close to the Pareto optimal solution. Furthermore, in order to improve the optimization performance and population diversity of the improved algorithm, the K-mean algorithm is used to solve the initial solution set, as well as a local fruit fly optimization algorithm (FOA) is combined with the genetic algorithm (GA). Finally, the LFOA-NSGA-III algorithm is empirically evaluated on eleven interval benchmark test problems and an unmanned aerial vehicles (UAVs) path planning problem. Through simulation comparisons with other different algorithms, it is concluded that the hyper-volume value, the imprecision value and the IGD value indicators are significantly better than other comparison algorithms. In addition, from a simulation experiment in application of the multi-UAVs path planning problem, it can be seen that the LFOA-NSGA-III algorithm is more effective and applicative in the IMaOPs. (C) 2021 Elsevier B.V. All rights reserved.
查看更多>>摘要：In this work, we backtest and compare, under the VaR risk measure, the fitting performances of three classes of density distributions (Gaussian, Stable and Pareto) with respect to three different types of emerging markets: Egypt, Qatar and Mexico. We also propose a new technique for the estimation of the Pareto tail index by means of the Threshold Accepting (TAVaR) and the Hybrid Particle Swarm Optimization algorithm (H-PSOVaR). Furthermore, we test the accuracy and robustness of our estimates demonstrating the effectiveness of the proposed approach. (C) 2021 Elsevier B.V. All rights reserved.
查看更多>>摘要：Recently, a new category of problems known as multi-factorial optimization (MFO) is gaining momentum in the field of evolutionary computation (EC). This paper aims at improving the aggregate performance of an underlying EC model in a multi-tasking environment by implementing simple strategies with minimal parameter tuning effort. Firstly, an enhanced Simulated Binary Crossover (SBX)-based unary variation operator to improve EC performance is devised. Secondly, to overcome the challenge in parameter tuning and operators selection, an adaptive control strategy underpinned by information entropy is proposed. We study the measure of entropy to quantify the uncertainty of evolutionary search and use the information to adapt the algorithmic parameters. Thirdly, the MFO problems are solved using the proposed methods. Three experiments are carried out to attest the methodology efficacy. The first experiment benchmarks the proposed method against twelve state-of-the-art single objective optimization algorithms in the CEC2014 competition. The second experiment compares the performance of the proposed method using a recent benchmark MFO problem. We further extend the investigation into the analysis of parameter sensitivity and solicit insights pertaining to the algorithm characteristics. Overall, the empirical results on various benchmark problems are promising. The third experiment offers a solution to a multi-mode resource-constrained project scheduling problem in the real-world construction industry. The results indicate improvements of 2.29% in project quality and 8.23% in project duration subject to an increase of 4.78% in project cost, which are well within the acceptance limits of the decision makers. (C) 2021 Elsevier B.V. All rights reserved.
Christoforou, AndreasAndreou, Andreas S.Garriga, MartinBaresi, Luciano...
17页
查看更多>>摘要：Microservice architectures foster the development of applications as suites of small, autonomous and conversational services, which are then easy to understand, deploy and scale. However, one of the problems nowadays is that microservices introduce new complexities to the system and, despite the hype, many factors should be considered when deciding whether to use them or not. This paper introduces a novel decision and analysis model with enhanced interpretative and explanatory capabilities. The model is conceived by identifying the key concepts and factors in deciding whether to adopt microservice architectures, or not, through literature review and experts' feedback from the industry and academia. These concepts are organized as a Multi-Layer Fuzzy Cognitive Map (MLFCM), a graph-based computational intelligent model. A new formulation is proposed, along with a novel genetically evolved algorithm, both aiming at improving the model in terms of performance, bias resilience and explainability. The model is evaluated and calibrated through a series of executions over real and synthetic scenarios. The application of static and dynamic analyses, in conjunction with the incorporation of the evolutionary approach, guide the identification of the prevailing factors that regulate the adoption of a microservice architecture and allow the interpretation of the importance of each concept. Finally, an industrial scenario leverages the assessment of the model's applicability and efficacy, highlighting some interesting results. (C) 2021 Published by Elsevier B.V.
NSTL
Elsevier