Gheorghe P

Bio: Gheorghe P is an academic researcher from University of Seville. The author has contributed to research in topics: Natural computing & Membrane computing. The author has an hindex of 1, co-authored 1 publications receiving 26 citations.

Tracing Some Open Problems in Membrane Computing

Abstract: Membrane computing is a branch of natural computing aiming to abstract computing models from the structure and functioning of the living cell, and from the way cells cooperate in tissues, organs, or other populations of cells. This research area developed very fast, both at the theoretical level and in what concerns the applications. During the almost ten years since membrane computing was initiated, several open problems were circulated, sometimes in comprehensive lists prepared for meetings in this area. The present notes intend to survey the research related to some of these problems, mentioning both progresses made in solving them and questions which still wait for research efiorts.

A comprehensive review on parameter estimation techniques for Proton Exchange Membrane fuel cell modelling

Abstract: The widespread use of Proton Exchange Membrane fuel cell for its unique advantages compelled researchers for precise modelling of its characteristics. Since, modelling becomes extremely important for better understanding, simulation, design, analysis and development of high efficiency fuel cell system. However, due to its non-linearity, multivariate and strongly coupled characteristics; mathematical modelling based on empirical equations was widely adopted. But, the shortage of data, complexity in modelling, and number of unknown parameters favored the use of optimization methods. Many optimization methods have been endeavored to model Proton Exchange Membrane fuel cell characteristics. However, no prior attempt has been made to consolidate the contributions. Hence, this paper comprehensively describes and discusses the various Artificial Intelligence/bio inspired methods applied for fuel cell parameter estimation problem. The methods background theory and its application to the problem is elaborated. It is envisioned that, this review will be a one stop solution to the researchers and engineers working in the area of fuel cell systems.

A Modified Membrane-Inspired Algorithm Based on Particle Swarm Optimization for Mobile Robot Path Planning

Abstract: To solve the multi-objective mobile robot path planning in a dangerous environment with dynamic obstacles, this paper proposes a modified membraneinspired algorithm based on particle swarm optimization (mMPSO), which combines membrane systems with particle swarm optimization. In mMPSO, a dynamic double one-level membrane structure is introduced to arrange the particles with various dimensions and perform the communications between particles in different membranes; a point repair algorithm is presented to change an infeasible path into a feasible path; a smoothness algorithm is proposed to remove the redundant information of a feasible path; inspired by the idea of tightening the fishing line, a moving direction adjustment for each node of a path is introduced to enhance the algorithm performance. Extensive experiments conducted in different environments with three kinds of grid models and five kinds of obstacles show the effectiveness and practicality of mMPSO.

A membrane algorithm with quantum-inspired subalgorithms and its application to image processing

Abstract: This paper presents a membrane algorithm, called MAQIS, by appropriately combining concepts and principles of membrane computing and quantum-inspired evolutionary approach. MAQIS has four distinct features from the membrane algorithms reported in the literature: initial solutions are only inside the skin membrane; different regions separated by membranes have different components of the algorithm; all the components inside membranes cooperate to produce offspring in a single evolutionary generation; communication rules are performed in a single evolutionary step. Extensive experiments conducted on knapsack problems show that MAQIS outperforms five counterpart approaches and our previous work. The effectiveness of MAQIS is also verified in the application of image processing.