Groumpos Petros

Bio: Groumpos Petros is an academic researcher from University of Patras. The author has contributed to research in topics: Fuzzy cognitive map. The author has an hindex of 1, co-authored 1 publications receiving 11 citations.

New concerns on fuzzy cognitive maps equation and sigmoid function

Abstract: Effective decision making is the fundamental factor for the appropriate operation of any system. Several methods have been developed in recent years, and Fuzzy Cognitive Maps (FCMs) is one of them. Basics of FCMs are briefly presented. They have been used in a variety of applications with a very good degree of success. All these years some of their drawbacks have been detected and discussed by several researchers, who are searching for ways to improve FCMs performance and broaden their implementation area. An effort towards the improvement of FCMs response is made, implementing for first time a new equation for the calculation of concept values and replacing the sigmoid function with Anti Windup control method. Results are presented and discussed, and further thoughts leading towards new research directions are presented.

Pathways for the transition of the Polish power sector and associated risks

Abstract: In Poland, coal has been a key driver of energy security and, despite its declining role, still accounts for most of the country’s power generation mix. Transition potential for the fossil fuel-based power sector can be reflected in a heated debate between a coal-oriented, evolutionary process of slowly reducing emissions; and a more ambitious renewable energy-driven pathway. Both pathways feature barriers; some are evident in standard economic analyses, while others are not. Here, we use a bottom-up energy model to identify the optimal energy mix for the two pathways, and a dynamic stochastic general equilibrium macroeconomic model to assess their socioeconomic implications. We then employ fuzzy cognitive maps to explore which agenda appears to be favoured by stakeholders, and identify omitted risk channels associated with both pathways. Finally, based on these insights, we reiterate the modelling analyses and assess the socioeconomic implications of some of the identified risk channels.

Intelligence and Fuzzy Cognitive Maps: Scientific Issues, Challenges and Opportunities

Abstract: Today humankind is facing an unprecedented set of many difficult problems and challenges that have never been encountered before. They urgently need realistic and long lasting solutions. Nowadays, practically all systems have become more and more complex and dynamic in nature. The modelling and controlling challenge of today’s complex dynamic systems (CDS) is carefully reviewed. The fuzzy cognitive maps (FCM) approach is used to model CDS. Drawbacks and deficiencies of classical FCMs are identified and solutions to overcome them are mathematically developed. Two basic scientific fields are studied and analyzed on a joint synergistic way for the very first time. Intelligence and intelligent control (IC), cognition and cognitive control (CC) are two challenging scientific fields. A new modelling and control theory is proposed for addressing the complex dynamic systems (CDS).

Why Model Complex Dynamic Systems Using Fuzzy Cognitive Maps

Abstract: Throughout the natural and artificial world one observes phenomena of great complexity. Yet research in physics and to some extent biology and other fields has shown that the basic components of many systems are quite simple. It is now a crucial problem for many areas of science to elucidate the mathematical mechanisms by which large numbers of such simple components, acting together, can produce behavior of the great complexity observed. Therefore today’s systems have become more and more complex and dynamic. The concept of complex dynamic systems (CDS) arises in many scientific fields and technological areas. Modelling and controlling complex dynamic systems is a very difficult and challenging task. As a result “complex systems theory” cuts across the boundaries between conventional scientific disciplines. It makes use of ideas, methods and examples from many different fields.

Fuzzy Cognitive Maps and Fuzzy Logic applied in industrial processes control

Abstract: In this work, we present intelligent controllers of two industrial processes. First, a first order heat exchanger (Heatex) is designed using a Fuzzy Logic based Controller (FLC) and a Fuzzy Cognitive Map based controller. The second process is a fourth-order alcoholic fermenter with two tuning mechanisms (FCM-PID and Fuzzy-PID) acting on proportional and derivative gains of a PID controller, adjusted heuristically through experts’ knowledge. We analyzed the IAE and ISE errors, the settling time, the overshoot and control signals for both processes. The performance of the proposed intelligent controllers was compared to the original conventional controllers. The results revealed that the FLC’s performance is comparable to the one of the conventional PI Heatex. However, in the case of the alcoholic fermenter process, the FCM-PID mechanism obtained the most satisfactory responses according to the analyzed parameters.

Semi-Unknown Environments Exploration Inspired by Swarm Robotics using Fuzzy Cognitive Maps

Abstract: This paper aims to present a fuzzy cognitive map (FCM) controller on multiple autonomous mobile robots (multirobot system) in order to complete a foraging task in semiunknown environments, and to compare the FCM results with a fuzzy logic controller (FLC). In this work, the limits of the searching area are known. However, everything within it (goals, obstacles and position of other robots during navigation) is unknown. The foraging task simulates a real-life application of robots on rescue missions, where the main objective is rescuing the victims of a tragedy, or lost in a forest, or even victims of a major accident in an industry. Although it is a simulated environment, we tested the autonomy and cooperation among all robots for this kind of operation in different simulated scenarios. We used a reactive (real-time) architecture to enhance the robots’ global robustness on dealing with unpredictable situations. The results imply that the FCM approach presented, in general, less simulation processing time and distance traveled, without significant prejudice in the explored area. In this way, we expect to collaborate and serve as inspiration for future works on this research area.