International Journal of Fuzzy Systems最新文献

The livestock sector has exacerbated the problems of ensuring global food safety and greenhouse gas emissions. The rapid increase in livestock production has called to shed light on decision-support tools that develop sustainable production strategies. In this context, this study aims to expand the application of multiple-criteria decision analysis (MCDM) methods to assign weights to criteria and classify decision support tools for livestock with a high degree of certainty. In order to begin serious steps to address the global sustainability problem, this study extended the PIPRECIA method with a high-certainty fuzzy environment called Z-cloud rough numbers (ZCRNs) to record the weight of 19 criteria for decision support tools in livestock farming. An innovative and advanced method called CoCoSo has been utilized to rank decision-support tools for livestock farming. The methodology included two stages. The first phase involved developing the decision matrix. The second phase encompassed developing MCDM methods by clarifying the steps of the PIvot Pairwise RElative Criteria Importance Assessment (PIPRECIA) method for assigning weight to criteria, in addition to highlighting the steps of the CoCoSo method for classifying decision support tools in the livestock industry. The results of the PIPRECIA method extended to the fuzzy environment of ZCRNs confirmed that visualization and herd characteristics received the highest weight compared to the rest of the criteria of decision support tools. The CoCoSo results provided insight into ranking alternatives for livestock decision support tools. AgRECalc has the highest ranking, and FCFC has the lowest ranking. This study conducted an evaluation test to increase the chances of generalizing the results of ranking decision-support tools of the livestock industry.

This paper addresses the problem of guaranteed cost output feedback control for a class of networked interval type-2 Takagi–Sugeno (IT2 T-S) fuzzy systems with adaptive event-triggered stochastic communication protocol (AETSCP) scheduling and hybrid attacks. A novel AETSCP scheduling is designed to judge whether or not data are triggered as well as to determine which node transmits data to the controller. Meanwhile, the security problem of hybrid attacks with respect to denial-of-service (DoS) attacks and deception attacks on the system is considered. The quadratic boundedness (QB) technique is employed to depict the closed-loop stability of the concerned networked control systems (NCSs). Two adequate theorems are given based on Lyapunov stability theory for designing the observer-based and dynamic output feedback-based controllers, which guarantee the stability and robust performance of the required system. In the end, a simulation example of the mass-spring-damping system is provided to confirm the effectiveness of the presented control strategy.

The O-index presented here allows a statement about the percentage deviation between two fuzzy numbers. For this purpose, one fuzzy number is defined s the reference. This fuzzy number is described with the help of its core value and its support. The deviation of the other fuzzy number, which is defined as the comparison number, is then quantified via the area differences of the left and right limits of the two numbers. In addition, the case when decomposed fuzzy numbers are involved, which are given as (alpha )-cuts is taken into account. Thus, the O-index-(alpha ) can be used to calculate a separate percentage deviation for each (alpha )-cut and thus generate additional knowledge. The O-index then allows a very detailed description of the deviation between two fuzzy numbers. One application of the O-index is the estimation of the accuracy of a surrogate model in relation to a reference model in the context of uncertainty quantification. This is illustrated by a mechanical example, a bending beam.

The class imbalance problem is one of the critical research areas of machine learning and deep learning and has received widespread attention from researchers. To solve the class imbalance problem, current typical methods only use positive samples to generate synthetic samples that are similar to the minority class while ignoring the characteristic information of negative samples. Therefore, when the number of positive samples is too small and has highly similar features, it will cause the classifier to have fitting problems. In response to the above problems, we propose a new positive sample enhancement algorithm (PENH) to solve the class imbalance by simulating the process of chromosome cross-fusion. We select the fuzzy negative sample set around the positive sample by the K-nearest neighbor algorithm and adopt the beyond empirical risk minimization (Mixup) to randomly hybridize the positive sample with the negative sample of the set. To overcome the problem of sample imbalance, we adopt the One-class SVM with overfitting of positive samples to select the newly generated unlabeled samples to obtain the balanced dataset. We construct multiple experiments in 20 open datasets. The results show that our PENH outperforms the other six baseline methods in multiple evaluation indicator.

Fuzzy logic-based systems are nowadays commonly used in nonlinear function approximation when incoming data are available. Their main advantage is that the resulting rules can be interpreted understandably. Nevertheless, when the data are noisy an overfitting may occur which leads to poor accuracy and generalization ability. Prior information about the nonlinear function may improve fuzzy system performance. In this paper the case when the function is monotonic with respect to some or all variables is considered. Sufficient conditions for the monotonicity of first-order Takagi–Sugeno fuzzy systems with raised cosine membership functions are derived. Performance of the proposed fuzzy system is tested on two benchmark datasets

When a person feels some emotion such as happiness, sadness, or love, the person consciously knows that such feelings appear, or it is done unconsciously, and sometimes it appears as a compound result of the two. In this paper, these two are defined as cognition and noncognition. By applying these two to the fuzzy love model using complex fuzzy numbers, we observe the chaotic behavior that appears in this model. We verify chaotic behaviors in the love model with fuzzy triangular and trapezoidal external forces using phase portrait and bifurcation diagram. The love model is known as the differential equation that can represent how a person feels love with respect to the time when the positive or negative external force is changed. Because love is a person’s feeling that includes vagueness and ambiguity of human emotion. Even more the external force also can express some external influence that is also can be human’s response. Because a person’s feeling is vague and ambiguous, fuzzy valued sinusoidal functions and Gaussian have been used to express those feelings and external forces.

This paper studies the input-constrained containment control problem for a class of fractional-order unknown nonlinear heterogeneous multi-agent systems with distributed time-varying delays under a directed communication network topology. To deal with the unknown time-delay function, we extend the signal permutation method to multiple leaders case and represent it as a bounded function with a generalized consensus tracking error function. To handle input saturation and system uncertainties, we design a distributed adaptive controller using interval type-2 fuzzy logic system theory and projection algorithm, which effectively avoids the complexity caused by general model reduction and ensures the boundedness of estimated parameters. To analyze the convergence of the error system, we construct a new Lyapunov-Krasovskii functional that fully considers the effects of system uncertainties and time delay without requiring the Lyapunov matrix to satisfy a special diagonal form. Then, combining with fractional calculus theory and linear matrix inequality (LMI) method, the sufficient conditions for implementing containment control have been proposed. A new controller design method has also been proposed, ensuring that all followers converge within the convex hull spanned by the leaders. The designed controller is fully distributed and easy to implement in practical applications, as each controller only uses its own and neighboring nodes’ information. Finally, simulation example is presented to demonstrate the effectiveness of the proposed methods.

The Adaptative neuro-fuzzy inference system (ANFIS) has shown great potential in processing practical data from control, prediction, and inference applications, reflecting advantages in both high performance and system interpretability as a result of the hybridization of neural networks and fuzzy systems. Matlab has been a prevalent platform that allows to utilize and deploy ANFIS conveniently. On the other hand, due to the recent popularity of machine learning and deep learning, which are predominantly Python-based, implementations of ANFIS in Python have attracted recent attention. Although there are a few Python-based ANFIS implementations, none of them are directly compatible with scikit-learn, one of the most frequently used libraries in machine learning. As such, this paper proposes Scikit-ANFIS, a novel scikit-learn compatible Python implementation for ANFIS by adopting a uniform format such as fit() and predict() functions to provide the same interface as scikit-learn. Our Scikit-ANFIS is designed in a user-friendly way to not only manually generate a general fuzzy system and train it with the ANFIS method but also to automatically create an ANFIS fuzzy system. We also provide four kinds of representative cases to show that Scikit-ANFIS represents a valuable addition to the scikit-learn compatible Python software that supports ANFIS fuzzy reasoning. Experimental results on four datasets show that our Scikit-ANFIS outperforms recent Python-based implementations while achieving parallel performance to ANFIS in Matlab, a standard implementation officially realized by Matlab, which indicates the performance advantages and application convenience of our software.

This paper contributes to the fusion of metaheuristic fuzzy neural network (FNN) and self-tuning autonomous control for omnidirectional mobile platforms in robotic cyber-physical systems (RCPSs). A cyber grey wolf optimization (GWO)-based FNN computing is incorporated with the backstepping control scheme and dynamic modeling to achieve autonomous control for the omnidirectional Mecanum platforms with uncertainties for RCPSs, called GWOFNN. The proposed cyber GWOFNN computing method is employed to address the self-tuning autonomous control problem of RCPS omnidirectional platforms by considering modeling uncertainties and unknown frictions. Numerical simulations and real-time experiments via field-programmable gate array (FPGA) realization are provided to illustrate the efficacy, applicability and merits of the presented RCPS GWOFNN real-time self-tuning cyber control strategy. Through comparison works, the advantages of the proposed GWOFNN computing are validated to accomplish autonomous control for Mecanum mobile RCPSs in polar space.

Pisciculture encounters an array of intricate challenges that span disease management, preservation of water quality, prevention of genetic hybridization, ensuring the integrity of net systems, sourcing sustainable aquatic feed, and comprehending fish growth and reproductive dynamics. Addressing these multifaceted challenges necessitates a comprehensive research approach. This study employs an innovative synergy of fuzzy logic and deep learning techniques, resulting in a robust strategy to tackle these obstacles effectively. Fuzzy logic excels in assessing stressed fish conditions by handling inherent uncertainties. Simultaneously, YOLOv7 with fuzzy color enhancement (YOLOv7FCE) is used to detect damaged fish nets, thereby mitigating losses and upholding the integrity of the net infrastructure. This approach also leverages YOLOv7FCE for identifying Cobia fish within shoals, streamlining the identification process. Subsequently, DeepLabv3 is implemented to meticulously segment the recognized Cobia fish, facilitating precise measurements of their physical attributes. This comprehensive methodology yields profound insights into growth patterns and feeding tendencies within the confined aquatic environment. By embracing this approach, the research presents a versatile and adaptive framework that not only enhances our comprehension of piscine dynamics but also holds the potential to revolutionize the aquaculture industry.