fuzzy-logic

In analyzing phenomena around us, clustering is among the most commonly used techniques in machine learning for comparing, and categorizing them into different groups based on intrinsic features. One of the main challenges facing clustering algorithms is selecting a suitable representative for each cluster. Existing algorithms often choose a single representative, which can lead to suboptimal performance on many datasets (especially asymmetric datasets). This process is completely dependent on the type of internal distribution of the clusters, and that single point may not be a suitable representative for that cluster. The proposed algorithm for dealing with datasets, inspired by the fuzzy ALM method and avoiding complex formulas, and calculations, initially breaks the system down into simpler (two-dimensional) systems. After spreading ink drops, by finding the vertical Narrow path and the horizontal narrow path, it selects a set of points as the representation of each cluster. The proposed algorithm, unlike many conventional algorithms, provides a representative set for each cluster and also enhances the algorithm's performance in dealing with datasets that have an asymmetric structure by introducing a new distance measure based on the KNN method and utilizing the set of prime numbers. The Accuracy, F1-Score, and AMI achieved when working with many low-dimensional, and high-dimensional datasets has been higher compared to algorithms such as FUALM, HiDUALM, K-Means, DBSCAN, DENCLUE and IRFLLRR and in some cases, the achieved accuracy has been equal to 100 percent.

In islanded microgrids, circulating currents among parallel inverters pose significant challenges to system stability and efficient power distribution. Traditional droop control methods often struggle to manage these currents effectively, leading to inefficiencies and potential system damage. This study introduces an advanced fuzzy-robust droop control strategy that integrates fuzzy logic with robust droop control to address these challenges. By incorporating fuzzy logic, the proposed strategy enhances the adaptability of droop control to varying system conditions, improving the management of circulating currents and ensuring more accurate power sharing among inverters. Comprehensive mathematical modeling and extensive simulation analyses validate the performance of this control strategy. The results show that the fuzzy-robust droop control method significantly outperforms conventional approaches, achieving up to a 70% reduction in circulating currents. This improvement leads to a substantial reduction in power losses and enhances the dynamic response under varying load conditions. Additionally, the strategy improves voltage and frequency regulation, contributing to the overall stability and reliability of the microgrid. The findings provide a robust solution to the longstanding issue of circulating currents, optimizing microgrid operations, and paving the way for more efficient and resilient distributed energy systems. The advanced control strategy presented in this study not only addresses critical challenges but also demonstrates the potential for innovative methodologies to meet the growing demands of future energy infrastructures, where reliability and efficiency are essential.

The limitation of government resources, especially in the field of budget, for the development of infrastructure on the one hand and the possibility of defining multiple plans and projects on the other hand, makes the issue of prioritizing and choosing the best portfolio of feasible projects vital. The basic question is which projects should be done and how the projects will be managed. The model used to select the project portfolio should be realistic, capable, flexible, affordable, and simple. It is natural that the correct choice of the prioritization model and the selection of projects with the help of economic and non-economic criteria can help in the development of infrastructures as quickly as possible and in achieving the goals. In this research, we are trying to prioritize projects using the TOPSIS method and the innovative method of fuzzy group decision making. Among the main indicators identified in this research, we can mention financial indicators, technical indicators, risk indicators, environmental indicators and political-social indicators; Each of these indicators has sub-criteria. As a case study, we rated four dam construction projects using this decision-making method. The sensitivity analysis of the ranking based on changing the weights of the sub-criteria showed that the ranking has high robustness. The results show that this method can be used to select projects of the same type in other project portfolios.

Vehicular ad-hoc networks are a type of ad-hoc network that lacks a fixed infrastructure. The network's nodes are vehicles that self-organize and perform various network operations, including packet routing and network management. These networks enable intelligent autonomous behavior in vehicles, particularly in situations such as accidents. Each vehicle in a vehicular ad-hoc network acts as a network node, and these nodes can collaborate to enhance network efficiency. Nowadays, technologies like vehicular ad-hoc networks are widely employed to enhance traffic flow and transportation in urban areas. Routing in vehicular ad-hoc networks remains a fundamental challenge in these networks. This article presents a routing method specifically designed for vehicular ad-hoc networks operating in urban environments. Given that urban environments consist of numerous roads and intersections, the proposed approach is divided into three phases. The first phase introduces an intersection detection method that does not require a city map. It classifies the network nodes into two categories: those located at intersections and those outside intersections. The second phase presents a routing method for nodes outside intersections, while the third phase outlines a method for determining routes for nodes within intersections. To evaluate the performance of the proposed method, key parameters such as packet delivery ratio, routing overhead, throughput, and end-to-end delay have been analyzed. The results indicate that the proposed method outperforms other existing methods.

Fire accidents are a disaster that can cause loss of life, property damage and permanent disability to the affected victim. Firefighting is a very important and dangerous job. Firefighters must extinguish the fire quickly and safely to prevent further damage and destruction. Detecting and extinguishing fires is a dangerous task that always puts the lives of firefighters at risk. One of the most effective tools for early fire extinguishing is the firefighting robot. Fire sensing in most industries is absolutely essential to prevent catastrophic losses. Robots with this type of embedded system can save the lives of engineers in industrial sites with hazardous conditions. This project aims to design and implement a solar-powered  with artificial intelligent of mobile fire detection robot to detect fires in disaster-prone areas and thus reduce human work effort and level of destruction. Design a robot capable of moving using a rotary motor, finding a flame using a flame sensor, and extinguishing a fire using a water spray using a pump, all of which is controlled by an Arduino Uno microcontroller and programmed using an artificial intelligence (fuzzy) logic technology) using MATLAB, the inputs It has two variations:: flame and gas with three organic functions, each of which has a gas variable (low, medium, high), flame sensor (small, normal, large), and the output is a pump, (pump off , pump on ) with 9 rules. In addition to the experimental setup of the proposed system which demonstrates the performance of sensors (gas, flame) using fuzzy and implemented logic tools. The performance of the solar panels was first tested using MATLAB software as well as experimentally under different weather conditions. The pump's performance is being tested experimentally, and the robot is also being tested to detect and extinguish fires. The process of designing and implementing robotics involves creating mechanical and electrical systems. The results showed the effect of temperature change on the solar panel, as when it increases, the panel’s production capacity decreases, as well as the effect of decreased solar radiation resulting from clouds and other things, and the extent of its effect. Impact on the performance efficiency of solar panels, and observing the pump performance in terms of flow rate and height. Hence, it can be noted that the robot designed in the project is capable of discovering fire sources and extinguishing them using fire-fighting systems equipped with a water tank and a controllable pump to spray the water necessary for the process. From this study, can be concluded that the designed model is able to work according to its initial design  with artificial intelligence  with the least amount of errors, and therefore it can be applied in industrial applications, avoiding fire damage and extinguishing it when it occurs for the first time.

The Power Quality (PQ) issue refers to the occurrence of irregular voltage, current, or frequency that leads to failure or incorrect functioning of equipment used by end users. The PQ meter is utilized to monitor a diverse range of power supply characteristics, all of which possess the capacity to impact the effectiveness of both operational procedures and machinery. The dynamic voltage restorer (DVR) performs the role of a specialized power device employed to mitigate the voltage drop experienced at the terminal of a sensitive load. DVR can be controlled by various control designs. This work conducts a comparative analysis on a normally managed voltage system and a medium-power DVR controlled by a neural network (NN), fuzzy logic (FL), or adaptive neuro-fuzzy inference system (ANFIS) by utilizing an output voltage regulator. The identification and rapid compensation of voltage perturbations, such as voltage sag, are essential elements in monitoring and controlling DVRs. The conventional PI controller is commonly employed in regulating DVRs. While the traditional controller possesses certain merits, it is not free of limitations. One such downside pertains to its utilization of constant gains, which can impede its capability to provide optimal control performance in instances where system parameters undergo fluctuations. Possible solutions have been proposed to effectively tackle this issue, such as the use of NNs, FL, or ANFIS controllers. Furthermore, to attain both rapid dynamic response and robustness, a modified d-q converted three-phase voltage regulator was adopted. Instead of employing a conventional three-phase regulator, this particular regulator is operated by means of an NN, FL, ANFIS, or PI controller. The suggested voltage regulator offers a prompt solution for rectifying voltage irregularities, such as voltage sag, by promptly restoring the voltage to the nominal magnitude. The primary source of power adopted in this study is a wind turbine unit.

In this research paper, a multivariable prediction control method based on direct vector control is applied to command the active power and reactive power of a doubly-fed induction generator used into a wind turbine system. To obtain high energy performance, the space vector modulation inverter based on fuzzy logic technique (fuzzy space vector modulation) is used to reduce stator currents harmonics and active power and reactive power ripples. Also the direct vector control model of the doubly-fed induction generator is required to ensure a decoupled control. Then its classic proportional integral regulators are replaced by the multivariable prediction controller in order to adjust the active and reactive power. So, in this work, we implement a new method of control for the doubly-fed induction generator energy. This method is carried out for the first time by combining the MPC strategy with artificial intelligence represented by Fuzzy SVM-based converter in order to overcome the drawbacks of other controllers used in renewable energies. The given simulation results using Matlab software show a good performance of the used strategy, particularly with regard to the quality of the energy supplied.

The Lean, Agile, Resilience, and Green (LARG) supply chains are more competitive than conventional ones. Evaluating its performance under current conditions and developing suitable strategies is crucial to enhance LARG. This study aims to create an assessment model for LARG in Iran's hospital medical equipment supply chain, especially in Hamadan. The Fuzzy Inference System (FIS) evaluates LARG across four dimensions: lean, agile, resilient, and green. Key indicators obtained from a comprehensive review of the literature and other published reports in the field of LARG were also confirmed by a focused group of experts in the medical equipment supply chain field. The findings indicate that the value LARG of the medical equipment supply chain is 0.787. Key indicators for the evaluation of LARG in the hospital medical equipment supply chain include reducing overall supply chain costs, optimizing inventory management, shortening supply chain development cycle time, increasing the introduction of new products, promoting information sharing among supply chain members, establishing flexible supply bases and sourcing, reducing fossil fuel consumption, and implementing waste management practices such as reuse and recycling of recyclable materials. This research provides managers with valuable insights into the current state of LARG and serves as a reference for formulating LARG strategies and practices. The study's results enable supply chain actors, particularly in Iran's Hamadan Province, to comprehend the key indicators for improving LARG performance in the hospital medical equipment supply chain. The proposed model can be adapted to other industries and service sectors by adjusting the indicators and assessing data availability.

In a delayed master-slave teleoperation system, if the slave robot interacts with a delicate and sensitive environment, it is essential to control the slave-environment interactions. Variable impedance control has been proposed as a useful method for this aim in the literature. However, changing the impedance parameters based on the system requirements imposes a complex process in the controller design. To address this issue, we propose a variable impedance control strategy for the slave side, where the impedance variables are changed using fuzzy logic. This is carried out based on the environment destruction threshold—defined based on the contact force and the velocity of the slave robot—and system stability range. The proposed method is simulated in MATLAB’s Simulink considering telesurgery conditions and soft tissue environment under an unknown and varying time delay. Simulation results show that the proposed method maintains the velocity of the slave robot and the environment force in the desired interval and performs better in keeping the environment safe compared to the constant-coefficient impedance control.

New technologies have emerged over the last few years, such as IoT and fog computing. IoT devices and the enormous amounts of data generated every minute have led to the vast growth of the Internet of Things (IoT). In order to meet the term "Data Never Sleeps", some IoT applications require real-time services and low bit latency. In order to provide quick processing, storage, and services, Cisco proposed fog computing as an extension of cloud computing. The traditional methods are not capable of addressing the complex scheduling scenarios of fog computing. In this paper, we introduce a novel Fuzzy Reinforcement Learning Scheduling algorithm (FRLS) that enhances schedule accuracy in dynamic computing environments. In order to optimize task scheduling, the FRLS algorithm integrates fuzzy logic with reinforcement learning. In order to prioritize critical tasks, fuzzy logic handles uncertainty and prioritizes tasks according to deadlines, sizes, and file sizes. Then, reinforcement learning schedules the prioritized tasks, continually adjusting to dynamic conditions to ensure the best resource allocation. In addition to improving overall system performance, this combination provides a robust framework that can address the complexity and variability of fog computing environments. FRLS is designed to minimize response time while adhering to resource and deadline constraints in fog-based applications. A comparison of FRLS with existing algorithms shows that it significantly improves load balancing, deadline satisfaction, response time, and waiting time. Combining reinforcement learning and fuzzy logic leads to an efficient scheduling solution. In addition, FRLS outperforms non-prioritized algorithms.

Outsourcing is a key strategy in many industries and service sectors and businesses. Choosing the right outsourcing subject can boost productivity while choosing the wrong one can cause problems. Thus, a scientific model to support outsourcing decisions is needed. Especially, a specific model for the service sectors and businesses. We aim to make a comprehensive model to rank outsourcing options in the service sectors and businesses. This model covers different fields and can be used universally. We used a comprehensive library study, a questionnaire, and the CVR analysis to find and validate outsourcing decision indicators, these comprehensive indicators are one of the advantages of this article. Then, we used Kano’s model to divide them into functional and basic categories for better analysis. Next, we used the fuzzy best-worst model to weigh the indicators. Finally, we used the fuzzy WASPAS model to rank the outsourcing options. We applied the model to a case study at a Hospital as a service business unit. We considered and evaluated and ranked four sectors for outsourcing: "Restaurant", "Pharmacy", "Maintenance and Repair" and "Finance and Accounting".