International Journal of Fuzzy Systems最新文献

Fuzzy C-means (FCM) clustering has the virtue of simple structure and easy implementation; however, it relies on the initial cluster centers and is sensitive to noise. To overcome these problems, this paper presents a novel FCM clustering method with slime mold and a Fisher score. First, logistics chaotic mapping is introduced to initialize the slime mold population and increase the population diversity. Modifying the convergence factor of the slime mold enhances the convergence speed and accuracy of the slime mold algorithm (SMA). Second, an adaptive weight is introduced into the SMA to promote the transition between exploration and development. Then, this optimal solution for SMA initializes the cluster center of FCM to avoid initialization sensitivity. Third, when considering the influence of feature differentiation degrees on the samples, the feature evaluation criteria of the Fisher score is constructed and then the importance of the feature is ranked to identify noise. The square root error criterion selects the most effective features to improve the clustering effect. Finally, by constructing uncertainty relations and introducing information entropy, the objective function of FCM is constructed to effectively solve the issue of FCM being sensitive to noise. The experimental results on 13 benchmark test functions for optimization, and 25 datasets for clustering show that the proposed algorithm outperforms other compared algorithms in terms of several evaluation metrics.

Regularization techniques have been widely applied in the context of fuzzy regression models, primarily tailored to triangular fuzzy outcomes. While this approach effectively handles fuzzy data in explicit interval data formats, its adaptability to various data types commonly encountered in practical applications is limited. To address this gap, we introduce the new fuzzy (alpha )-cut Lasso, extending the classical Lasso to encompass two essential data formats for fuzzy outcomes: explicit interval data formats and implicit formats with multiple measurements. Leveraging (alpha )-cuts, this model can extract richer insights from the data regarding the shape of fuzzy numbers. The model shows flexibility in handling fuzzy outputs and fuzzy regression coefficients of the LR-type, encompassing specific examples such as triangular and Gaussian types.

Intuitionistic fuzzy sets (IFSs) are more useful than FSs for modelling uncertain data of realistic problems by allowing the hesitancy degree. Due to which different extensions of IFSs are available with different domains and objectives. One of the important generalization of IFS is quaternion IFS (QIFS) based on the concept of quaternion numbers. Despite the significance of QIFS, there’s been a notable gap in exploring the aggregation of QIFS information which is the motivation of the presented study. Consequently, this manuscript effort to established some aggregation operators under QIFS environment. Their important properties are also analysed. Firstly, a new novel order relation of QIF numbers (QIFNs) is introduced to address limitations in existing model. Secondly, a sequence of aggregation operators termed as “the QIFA, the QIFG, the QIFOA and the QIFOG” are proposed under QIFS environment. Then, these approaches are applied to the classification of photo-voltic (PV) solar power plants. The obtained results are compared with some other existing models in details to show its supremacy. Finally, the conclusions of the presented study are listed.

This paper investigates the consensus of T–S fuzzy positive multi-agent systems by combining compensator and disturbance observer. First, a fuzzy compensator and a disturbance observer are proposed for the systems. Then, a fuzzy event-triggered control protocol is designed by using an event-triggered compensator information and adding an additional constant term. A novel practical consensus framework is constructed by integrating the constant term into error dynamic variables. Moreover, fuzzy copositive Lyapunov function and linear programming are employed to analyze the consensus and design the control protocol, respectively. Under the designed control protocol, all states of agents converge to a bounded scope rather than fixed finite values. The main contributions lie in that (i) A positive compensator and a positive disturbance observer are presented, (ii) A practical consensus protocol is established and the corresponding results are more practicable than existing asymptotic consensus, and (iii) Linear programming is utilized for the protocol design and it has less computational burden than other approaches. Finally, an example is provided to verify the effectiveness of the proposed design.

The issue of command filter-based fault-tolerant fast finite-time control is explored for manipulator systems with actuator faults. By using finite-time control and a finite-time command filter, all the signals in the closed-loop system are bounded and converge to the bounded regions in finite time. In addition, the problems of complicated calculation and influence of filtering errors are solved by the introduction of the finite-time command filter and a compensation mechanism. It is especially emphasized that the main contribution of this paper is as follows: (1) Several advanced control methods are integrated, which takes into account the speed, reliability, and adaptability of the controlled system. (2) In the last step of the design based on backstepping, an intermediate variable is designed which can simplify the proposed control algorithm. In the end, with the help of a numerical simulation example, it is shown that better transient/steady state performance and the fast finite-time stability of the closed-loop system can be obtained. As a result, it is concluded that our scheme is effective.

In order to make the best decisions in real-world applications, we must deal with optimization and decision-making problems that call for the input of experts and masters. Using an optimization problem with hesitant fuzzy parameters is important in these circumstances. Few studies have been done on the problem of hesitant fuzzy linear programming (HFLP). Therefore, in this article we study HFLP problems with hesitant decision variables and right-hand-side values. In order to solve the mentioned optimization problems, we suggest the hesitant fuzzy simplex algorithm. For this, first state the optimization theorems then express the hesitant fuzzy simplex algorithm using the introduced linear ranking functions. We will finally test the implementation of the suggested strategy by solving two descriptive examples using hesitant fuzzy information.

An adaptive predefined-time fuzzy tracking control problem for non-strict feedback systems is investigated in this paper. Compared with some previous findings on adaptive control for the output constrained systems with input saturation, the most significant feature of this research is that the system convergence time only depends on one design parameter. To approximate the unknown nonlinearity, fuzzy logic systems as a powerful method is introduced. Additionally, by integrating the auxiliary control signal and the barrier Lyapunov function method into the backstepping deduce procedure, a predefined-time adaptive fuzzy control scheme is proposed for such a system. Theoretical analysis proves that all of the system variables are bounded, and the system output converges to a small region near the given signal within a predefined time. Finally, a practical example of the single-link rigid robot system and a numerical example are conducted to verify the effectiveness of the proposed control approach.

In this paper, the problem of adaptive fuzzy control for stochastic biological systems with stage structure is studied. Firstly, considering the species itself in nature will be affected by a variety of uncertain factors, a more realistic stochastic biological model is established. Then, aiming at the unknown nonlinear functions in the stochastic biological system, the fuzzy logic system (FLS) is used to approximate the unknown nonlinear terms. Secondly, the backsteppting method and adaptive fuzzy means are applied to the prey–predator model with stage structure, and the corresponding adaptive fuzzy controller is designed. It is guaranteed that all states in the biological system are semi-globally uniformly ultimately bounded (SGUUB), the juvenile prey density can track the given desired density, and the tracking error converges to a small neighborhood near zero. Finally, a simulation experiment is carried out with reference to the real case of the significant reduction of the number of lampreys. The results show that compared with the general adaptive control method, the control strategy proposed shows higher superiority in the stochastic biological system.

Data envelopment analysis (DEA) is a widely used approach for evaluating the relative efficiency of decision-making units (DMUs) in multiple-input and multiple-output situations. Although traditional DEA models use precise input–output data, real-world problems often involve mixed uncertainties, including fuzziness and stochasticity. This paper focuses on dealing with situations where inputs and outputs have both fuzzy and stochastic characteristics, using DEA models for efficiency evaluation. Through the integration of the α-level approach and chance-constrained programming, novel DEA models with fuzzy stochastic variables (FSVs) are proposed, and deterministic equivalent interval DEA models with linear constraints are provided to address this problem. The main contributions and advantages of the proposed model over existing DEA models with FSVs are fourfold: (1) linear and always-feasible models are proposed; (2) a fixed and uniform production boundary (i.e., the same set of constraints) is used to measure the efficiency of DMUs with fuzzy stochastic input and output; (3) the obtained results can distinguish between efficient and inefficient DMUs; (4) Equivalent interval DEA models were obtained to provide a more comprehensive assessment of the efficiency of the DMUs. Finally, a numerical example is presented to demonstrate the applicability of the proposed models and the feasibility of the obtained solutions.

The adulteration of camellia seed oil with different processes will seriously violate the rights and interests of consumers. The accurate identification of camellia seed oil processes is of great significance to reduce such illegal activities. However, the fatty acid composition of camellia seed oil is complex and the content varies greatly in the same process, while the difference is small in different processes. This multivariate data are easy to lead to the fuzzy characteristics of camellia seed oil, which increases the difficulty of identifying camellia seed oil quality. To solve these problems, we propose a multi-scale interactive attention network (MIANet) for the accurate identification of camellia seed oil. Firstly, a one-dimensional multi-scale convolutional feature extraction method (OMCM) was proposed, which was used to reduce the difference from multivariate fuzzy features and better solve the problem of fuzzy features of camellia seed oil fatty acids with the same process. Secondly, the interactive attention mechanism (IA) was proposed to enhance the deep characteristics of multivariate fatty acids from the fusion of two dimensions, so that the model paid more attention to the subtle differences between different processes, and effectively solved the problem of fuzzy fatty acid characteristics of camellia seed oil in different processes. Finally, in order to verify the effectiveness of MIANet, MIANet is compared with classical machine learning methods such as SVM, KNN, LR, LDA, QDA, classical deep learning method AlexNet, and the most advanced deep learning methods such as DMCNN and HCA-MFFNet. The accuracy of MIANet reached 94.10%, which was better than the eight methods. The experimental results show that MIANet is an effective method for the accurate identification of camellia seed oil data with fuzzy characteristics.