Search

The introduction of a multi-population structure in differential evolution (DE) algorithm has been proven to be an effective way to achieve algorithm adaptation and multi-strategy integration. However, in existing studies, the mutation strategy selection of each subpopulation during execution is fixed, resulting in poor self-adaptation of subpopulations. To solve this problem, a dynamic multi-strategy integrated differential evolution algorithm based on reinforcement learning (RLDMDE) is proposed in this paper. By employing reinforcement learning, each subpopulation can adaptively select the mutation strategy according to the current environmental state (population diversity). Based on the population state, this paper proposes an individual dynamic migration strategy to “reward” or “punish” the population to avoid wasting individual computing resources. Furthermore, this paper applies two methods of good point set and random opposition-based learning (ROBL) in the population initialization stage to improve the quality of the initial solutions. Finally, to evaluate the performance of the RLDMDE algorithm, this paper selects two benchmark function sets, CEC2013 and CEC2017, and six engineering design problems for testing. The results demonstrate that the RLDMDE algorithm has good performance and strong competitiveness in solving optimization problems.

Day-ahead scenario generationof renewable power plays an important role in short-term power system operations due to considerable output uncertainty included. In this paper, a deep renewable scenario generation model using style-based generative adversarial networks followed by a sequence encoder network, is developed to generate accurate and reliable day-ahead scenarios directly from historical data through different-level scenario style controlling and mixing, thus achieving better characterization of renewable spatial-temporal dynamics. Meanwhile, the integration of meteorological information serving as conditions enables our model to precisely capture the complex diurnal pattern and seasonality difference of renewable power. From wind and photovoltaic power perspectives, the effectiveness of the proposed model is validated on two real-world datasets reflecting region aggregation level and distributed power station level respectively. Numerical results demonstrate the superiority of model performance through both the statistical and power system scheduling analysis, compared to three benchmarks.

Cloud solutions accelerate the large‐scale acceptance of IoT projects. By diminishing the need for maintaining on‐premises infrastructure, the cloud has enabled corporations to surpass the traditional applications of IoT (e.g., in‐home appliances) and opened the doors for large‐scale deployment of IoT applications on the cloud. However, shifting legacy systems to the cloud environment can be considerably difficult. Accordingly, this article proposes a method that may support organizations in deciding to modernize their legacy systems. The main concept of this study is to discuss the modernization strategies in detail and to support organizations in selecting the most accurate and appropriate cloud migration strategy, based on their requirements of the legacy system. This article introduces a novel research process, called the K‐means cosine cloud clustering method (K3CM). K3CM is a statistical knowledge‐based method for identifying and clustering the most relevant and similar cloud migration strategies. The quality of a cluster is evaluated by measuring intra‐cohesiveness. Simulation experiments statistically analyzed, evaluated, and verified the quality of K3CM clusters. Correspondence analysis explored the similarity and relationship among cloud migration frameworks and validated the proposed technique. The statistical and simulation results of this study focus on the analytics and decision support system implementation that provides a reliable, valid, and robust clustering method for modernizing the legacy system. The main concept of this study is to discuss the modernization strategies in detail and to support organizations in selecting the most accurate and appropriate cloud migration strategy, based on their requirements of the legacy system. This article introduces a novel research process, called the K‐means cosine cloud clustering method (K3CM). K3CM is a statistical knowledge‐based method for identifying and clustering the most relevant and similar cloud migration strategies.

Whatever the exact figures, world energy consumption, particularly electricity consumption, can increase significantly during couple of decades. It is not possible solely due to demographic pressure, but also due to expansion of living standards within the less developed countries. Hydropower has reached high levels of technical sophistication in power generation as compared with other renewable energy sources. The paper discusses recent hydropower optimization research and development activities using metaheuristic approach. The article discusses emerging attempts to promote the optimization of hydropower. It provides a comprehensive analysis of recent attempts to extend the operating policies of hydraulic structures to reach extraordinary degrees of versatility, a subject of many recent research projects using metaheuristic algorithms. In addition, groundbreaking technologies for hydroelectric energy production are also discussed with the study based on reservoir operation and scheduling of flow and energy.

Quantum computing-inspired metaheuristic algorithms have emerged as a powerful computational tool to solve nonlinear optimization problems. In this paper, a quantum-behaved bat algorithm (QBA) is implemented to solve a nonlinear economic load dispatch (ELD) problem. The objective of ELD is to find an optimal combination of power generating units in order to minimize total fuel cost of the system, while satisfying all other constraints. To make the system more applicable to the real-world problem, a valve-point effect is considered here with the ELD problem. QBA is applied in 3-unit, 10-unit, and 40-unit power generation systems for different load demands. The obtained result is then presented and compared with some well-known methods from the literature such as different versions of evolutionary programming (EP) and particle swarm optimization (PSO), genetic algorithm (GA), differential evolution (DE), simulated annealing (SA) and hybrid ABC_PSO. The comparison of results shows that QBA performs better than the above-mentioned methods in terms of solution quality, convergence characteristics and computational efficiency. Thus, QBA proves to be an effective and a robust technique to solve such nonlinear optimization problem.

Wafer defect detection using an intelligent system is an approach of quality improvement in semiconductor manufacturing that aims to enhance its process stability, increase production capacity, and improve yields. Occasionally, only few records that indicate defective units are available and they are classified as a minority group in a large database. Such a situation leads to an imbalanced data set problem, wherein it engenders a great challenge to deal with by applying machine-learning techniques for obtaining effective solution. In addition, the database may comprise overlapping samples of different classes. This paper introduces two models of evolutionary fuzzy ARTMAP (FAM) neural networks to deal with the imbalanced data set problems in a semiconductor manufacturing operations. In particular, both the FAM models and hybrid genetic algorithms are integrated in the proposed evolutionary artificial neural networks (EANNs) to classify an imbalanced data set. In addition, one of the proposed EANNs incorporates a facility to learn overlapping samples of different classes from the imbalanced data environment. The classification results of the proposed evolutionary FAM neural networks are presented, compared, and analyzed using several classification metrics. The outcomes positively indicate the effectiveness of the proposed networks in handling classification problems with imbalanced data sets. [ABSTRACT FROM PUBLISHER]

In the field of services marketing, consumers can be segmented into separate subgroups with homogeneous service requirements in the process of consumer market segmentation. In this process, a computable number of consumers and the services that they can select are not difficult to handle. However, when there is a large consumer base and the services that the consumers can select have multiple shared and distinct characteristics, measurement and segmentation in polynomial time are extremely difficult. Therefore, in this paper, we propose a biologically systematic computational method that would be appropriate for the segmentation of consumer and service characteristics. We also demonstrate the application of this biologically systematic computational method to a services marketing problem model.

Real-life applications face simultaneously hybrid uncertainty namely fuzziness and randomness, or ambiguous and vague information that makes the existing decision-making model incapable of handling such uncertainties. This paper presents the possibilistic programming for decision-making using a fractile approach. Some real world problems are formulated as a necessity measure model to deal with the uncertainties, which come from vague aspiration and ambiguous coefficients. Thus, the proposed methodology is important in building the model and finding the solution. The vagueness and ambiguity are properly treated in the paper and the fractile approach is used to solve fuzzy linear programming problem. An illustrative example explains the proposed model. The analytical results of the proposed method reveal the improvement of conventional decision-making approaches to appropriately handle inherent uncertainties contained in the real world situation.

Exploring an intranet or internet database enables us to discover useful knowledge. In this process, a search engine plays a pivotal role. To this end, various search engines have been proposed to heighten information accuracy by exploiting key content relations in semantic web resources. But a general-purpose search engine always includes useless or irrelevant web pages in the search results. The next generation of web architecture, known as Semantic Web, can build a layered architecture to possibly mitigate this deficiency by decreasing the noisy data in a searched result. The objective of this paper is to propose a Probabilistic Latent Semantic Indexing (PLSI) method used in semantic web search engines. The method can better return appropriate information for user queries; in particular, a novel ranking strategy is provided to measure the relevance score of an annotated set of web results by considering user queries, data annotation, and the underlying ontology.

The objective of this paper is to provide an improved fuzzy clustering approach to data mining. The method consists of fuzzy clustering and rough set model that together deal with the uncertainty of data. To describe the proposed method, the rough set model is used to optimize the mined knowledge, then after embedding the sample data for data mining, fuzzy clustering is applied to cluster the target data by sample data to extract desired data. The process behind algorithm and its applicability are illustrated through an application of the proposed method to the knowledge mining of accident cases, and this shows that improved fuzzy clustering should have wider-ranging applications.

Abstract: This paper studies a facility location model with fuzzy random parameters and its swarm intelligence approach. A Value-at-Risk (VaR) based fuzzy random facility location model (VaR-FRFLM) is built in which both the costs and demands are assumed to be fuzzy random variables, and the capacity of each facility is unfixed but a decision variable assuming continuous values. Under this setting, the VaR-FRFLM is inherently a two-stage mixed 0–1 integer fuzzy random programming problem, to which analytical nonlinear programming methods are not applicable. A hybrid modified particle swarm optimization (MPSO) approach is proposed to solve the VaR-FRFLM. In this hybrid mechanism, an approximation algorithm is utilized to compute the fuzzy random VaR objective, a continuous Nbest–Gbest-based PSO and a genotype–phenotype-based binary PSO vehicles are designed to deal with the continuous capacity decisions and the binary location decisions, respectively, and two mutation operators are incorporated into the PSO to further decrease the possibility of becoming trapped in the local optima. A numerical experiment illustrates the application of the proposed hybrid MPSO algorithm and lays out its robustness to the system parameter settings. The comparison shows that the hybrid MPSO exhibits better performance than that when hybrid regular continuous-binary PSO and genetic algorithm (GA) are used to solve the VaR-FRFLM. [Copyright &y& Elsevier]

This paper demonstrates a satisfaction-based method for solving multi-objective programming problems in a multi-level decision structure through the renewed use of a weighted additive fuzzy goal programming (weighted additive-FGP) approach. The weighted additive-FGP enables us to indicate decision maker's preferences toward the unit's goals and decision vectors that are controlled by the decision maker in the pth-level. This paper presents the algorithm reflecting the influence of a higher-level decision on the decision of a lower level, with the respective decision maker having the authority to place emphasis on certain decision variables. A numerical example clearly illustrates the proposed solution procedure.

In this paper, a hybrid neural network approach to solve mixed integer quadratic bilevel programming problems is proposed. Bilevel programming problems arise when one optimization problem, the upper problem, is constrained by another optimization, the lower problem. The mixed integer quadratic bilevel programming problem is transformed into a double-layered neural network. The combination of a genetic algorithm (GA) and a meta-controlled Boltzmann machine (BM) enables us to formulate a hybrid neural network approach to solving bilevel programming problems. The GA is used to generate the feasible partial solutions of the upper level and to provide the parameters for the lower level. The meta-controlled BM is employed to cope with the lower level problem. The lower level solution is transmitted to the upper level. This procedure enables us to obtain the whole upper level solution. The iterative processes can converge on the complete solution of this problem to generate an optimal one. The proposed method leads the mixed integer quadratic bilevel programming problem to a global optimal solution. Finally, a numerical example is used to illustrate the application of the method in a power system environment, which shows that the algorithm is feasible and advantageous.

DNA-based computation is one of the latest computation paradigms. Compared to conventional methods that obtain their end results via electronic processes, aDNA-based approach obtains its result from bio-chemical reactions. It is essential in this approach for all experimental processes to be performed without fault. However, some errors may occur while carrying out these bio-chemical experiments. Consequently, it is necessary to overcome their weaknesses. The aim of this study is to examine the robustness of DNA-based techniques in solving a clustering problem. In the broadest sense, robustness can be defined as being able to withstand stresses, pressures, or changes in procedure or circumstance. To examine the robustness of the approach, this research examined the impact of error or added noise on DNA-based procedure results. Comparative studies of different error sets are also provided here. Additionally, two well-known conventional clustering algorithms (Fuzzy C-means and k-means) were applied to the same error sets, to study the reliability and validity of results when comparing DNA-based clustering.

Abstract: This paper studies a renewal reward process with fuzzy random interarrival times and rewards under the -independence associated with any continuous Archimedean t-norm . The interarrival times and rewards of the renewal reward process are assumed to be positive fuzzy random variables whose fuzzy realizations are -independent fuzzy variables. Under these conditions, some limit theorems in mean chance measure are derived for fuzzy random renewal rewards. In the sequel, a fuzzy random renewal reward theorem is proved for the long-run expected reward per unit time of the renewal reward process. The renewal reward theorem obtained in this paper can degenerate to that of stochastic renewal theory. Finally, some application examples are provided to illustrate the utility of the result. [Copyright &y& Elsevier]

AbstractIt is difficult to define the space that all people can feel comfortable. The reason is partly because comfortness relates to many attributes that specify an environment. People have different preferences, and sometimes even the same person changes his/her preference according the state of health, body conditions, working state and so on. Various parameters and attributes of the space should be controlled in order to realize such a comfortable space depending on the data-base of past usages. Information obtained from human bodies such as temperature, blood pressure, a brainwaves and etc. can be employed to adjust the space to the best condition. In order to realize a comfortable space, first we should recognize human states and behaviors, and secondly we provide an appropriate environment for a focal person in his/her state.The objective of this paper is to build a space befitting human behaviors or actions where people can feel comfortable using technologies such as KANSEI engineering. In other words, it is required to create a comfortable environment suitable to our feelings. Kansei Engineering is such a field that offers better products to the customers by employing human sensitivity.

In the study, non-deterministic information systems-Apriori-based (NIS-Apriori-based) rule generation from table data sets with incomplete information, SQL implementation, and the unique characteristics of the new framework are presented. Additionally, a few unsolved new research topics are proposed based on the framework. We follow the framework of NISs and propose certain rules and possible rules based on possible world semantics. Although each rule τ depends on a large number of possible tables, we prove that each rule τ is determined by examining only two τ -dependent possible tables. The NIS-Apriori algorithm is an adjusted Apriori algorithm that can handle such tables. Furthermore, it is logically sound and complete with regard to the rules. Subsequently, the implementation of the NIS-Apriori algorithm in SQL is described and a few new topics induced by effects of NIS-Apriori-based rule generation are confirmed. One of the topics that are considered is the possibility of estimating missing values via the obtained certain rules. The proposed methodology and the environment yielded by NIS-Apriori-based rule generation in SQL are useful for table data analysis with three-way decisions. [ABSTRACT FROM AUTHOR]

This paper develops a hybrid approach integrating an adaptive artificial neural network (ANN) and a fuzzy logic system for tackling class-imbalance problems. In particular, a supervised learning ANN based on Adaptive Resonance Theory (ART) is combined with a Tagaki–Sugeno–Kang-based fuzzy inference mechanism to learn and detect defects of a real large highly imbalanced dataset collected from a semiconductor company. A benchmark study is also conducted to compare the classification performance of the proposed method with other published methods in the literature. The real dataset collected from the semiconductor company intrinsically demonstrates class overlap and data shift in a highly imbalanced data environment. The generalization ability of the proposed method in detecting semiconductor defects is evaluated and compared with other existing methods, and the results are analyzed using statistical methods. The outcomes from the empirical studies positively indicate high potentials of the proposed approach in classifying the highly imbalanced dataset posing overlap class and data shift. [ABSTRACT FROM AUTHOR]

In this study, we discuss multi-period portfolio selection problems when security returns are described as fuzzy random variables. The main concern of this work is to apply dynamic risk tolerance and expected return levels in mathematical modeling; i.e ., these two indices of each period are influenced by the investment result of the previous period as well as human risk attitudes instead of static values over the entire investment horizon. Essentially, this assumption is based on the reality that investors tend to update targets when their wealth changes. In addition, fuzzy random variables are employed here to incorporate historical data with expert knowledge when estimating security future returns. Based on the above considerations, two multi-period portfolio selection models are built in light of the different risk attitudes. We then provide property analysis on complicated nonlinear optimization problems and derive several equivalents of the models, which can be solved by the existing dynamic programming. In general situations, a fuzzy random simulation-based particle swarm optimization algorithm is developed to search for approximate optima. The performance of this research is exemplified by a real market data-based case study in which the superiority of the dynamic strategy is demonstrated by a comparison with conventional approaches. [ABSTRACT FROM AUTHOR]

The electricity output is worth monitoring because of the rising electricity demand. Ambient temperature, air pressure, relative humidity, and exhaust pressure all impact the production output of a combined cycle power plant. This study proposes the BOAPPE algorithm, which combines the butterfly optimization algorithm (BOA) with the phasmatodea population evolution algorithm (PPE) to estimate power output better and reduce excessive cost waste. When used in conjunction with a support vector regression (SVR) model, such as BOAPPE-SVR, for estimating the power output of a power plant under basic load, it not only enhances the model's prediction accuracy but also successfully avoids the problem of the model entering local optimization too soon. The parallel strategy of the BOAPPE algorithm in this research improves convergence speed, while the random walk strategy prevents the model from sliding into local optimization. The results reveal that the model paired with the BOAPPE algorithm is more accurate and better than the other models in this research when comparing the performance of the parameters such as mean square error, relative error, and correlation coefficient. As a result, the BOAPPE-SVR model is a viable model for power load forecasting. [ABSTRACT FROM AUTHOR]

Scenario generation is a pivotal method for providing system operators with a reasonable quantity of power scenarios that are capable of reflecting uncertainties and spatiotemporal processes to make exact and effective decisions for power systems. Aiming at improving the forecasting performance of renewable generation and capturing uncertainty as well as dependency over renewable site groups in different regions, this paper proposes a data-driven approach for parallel scenario generation. To capture the complex spatiotemporal dynamics of renewable energy sources (RESs), the proposed approach utilizes Gramian angular field (GAF) to process time sequences and constructs style-based super-resolution models that correspond with the idea of multi-model ensembles. Thereafter, a two-stage stochastic optimization strategy is adopted to accomplish scenario forecasting using point forecasts and historical error information as input. Based on two real-world datasets from the National Renewable Energy Laboratory (NREL) and the Belgian transmission operator ELIA, the effectiveness of the proposed approach is verified by methods including statistical analysis, spatiotemporal correlations, power system scheduling, and out-of-sample evaluations. Compared with three advanced benchmarks, the proposed approach has superior forecasting performance and spatiotemporal dynamic capture capability. At a 24-h lead time, the proposed model achieves continuous ranked probability scores (CRPSs) of 4–14% over the other models with consistent performance during the economic dispatching of actual power system operations. [ABSTRACT FROM AUTHOR]

Wind power forecast is playing a significant role in the operation and dispatch of modern power systems. Compared with traditional point forecast methods, interval forecast is able to quantify uncertainties effectively. Unfortunately, the stochastic and intermittent nature of wind power has brought significant challenges to get high quality prediction intervals (PIs). In this paper, a novel interval forecast model based on long short-term memory neural networks (LSTM) is proposed to construct PIs with the lower and upper bound estimation method. Besides, an improved PI evaluation criterion is proposed by considering the estimation error of PIs. Moreover, a multi-objective optimization framework is proposed to investigate the relationship between PI estimation error and average width. To tune the parameters of LSTM, the widely used non-dominated fast sort genetic algorithm is further improved by introducing the competitive learning mechanism. The effectiveness of the proposed model and algorithm is demonstrated by a series of experiments based on a real world wind power dataset. [ABSTRACT FROM AUTHOR]