Andreea Udrea

A Dual Approach for Modelling Urban Traffic

Abstract The paper presents the first results of a national budgeted “Intelligent Techniques for the Modeling, Analysis, and Optimization of Urban Traffic” research project concerning the modeling, evaluation and optimization of urban traffic. For the modeling phase a dual approach was used – using both Hybrid Petri Nets and Matlab/Simulink models. Models were realized starting from the same assumptions and ontology. The paper will present the modeling methodology, as well as a comparison between the two kinds of models, using the same set of input data. Also, for the considered example, a solution is given, that eliminates traffic blockage.

Implementation of a telemedicine system for optimal on site medical response in case of disasters and for emergency situations management

This paper presents a telemedicine support system developed by implementing an integrated disaster management protocol, and providing specialized expertise in local/remote affected regions (affected civilians, soldiers, rescuers, explorers, people in remote areas, etc.). The system has a modular structure consisting in: a containerized management system (regional / local), a mobile monitoring-rescue system including an individual acquisition and monitoring system and a communication handheld/portable monitoring stretcher and an integrated communication module and other software facilities.

Enhanced visualization of cerebral blood vessels for X-ray angiograms

Neurologists frequently use X-ray angiographic techniques in order to establish the optimal treatment or intervention procedure. Neurosurgeries performed at blood vessels level are difficult procedures because of the low accessibility in the human brain. Therefore, the establishment of a correct diagnosis based on angiographies is of paramount importance. Enhanced visualization of blood vessels by separating veins and arteries from background can help, in many cases, in giving a more accurate diagnosis. In this article, three methods for arteries and veins classification and enhanced visualization are presented. From our study, it results that, using coefficients of wavelet transform as parameters for the classification, a better visual representation of small vessels is achieved.

Precise ratio control structure for nonlinear blend processes

This paper proposes a precise structure solution for the ratio control of nonlinear processes blend stations. The main feature of the proposed structure is to give a good ratio performance for both set-point changes and load disturbances that might occur in any one of the loops. For each loops (two or more) a nonlinear structure based on a multiple model and nonlinearity compensator is proposed. The main advantages consist in a classic tuning procedure for each of the controllers and a very simple procedure for nonlinearity determination, suitable for industrial applications. All components of control structure ware implemented on real time control application.

Integrated modeling and control platform for urban traffic networks

The paper presents respectively a modeling and a control approach of urban traffic networks, integrated into a software application. These approaches are based on an agent-oriented methodology and have as a goal the possibility of simulating, evaluating and optimizing the traffic in crowded areas, such as large cities, using both off line scenario policies and fine tuning scenario control.

Multi model control solution for some classes of hysteretic processes

The paper proposes a multiple model structure as control solution for some classes of hysteretic processes. This is developed using the nonlinear geometric characteristic of the process and a classical multiple model control algorithm. The components of the proposed structure are designed based on experimental tests, classic identification and closed loop pole placement methods. The applicability of the different implementations of the structure, suitable for each class of studied hysteretic processes, is demonstrated with a real-time control application implementing RST numerical algorithms. The software and the obtained results are presented and commented in terms of efficiency, advantages and disadvantages.

Hysteresis Control of a (Ba/Sr)Tio3 Based Actuator: A Comparison of Prandtl-Ishlinskii and Nonlinear Compensator Numerical Methods

Abstract The paper presents an experimental study on (Ba,Sr)TiO3 based actuators hysteretic comportment. The purpose of the study is to model the nonlinearity introduced by the actuators material in positioning systems for better tracking performances and efficient numerical implementation. The classic and modified Prandtl – Ishlinskii techniques are used such that numerical control can be achieved, by using the inverse model of the nonlinearity. This method is compared to a proposed nonlinear compensator based control structure that uses the geometric characteristic of the process, a lookup table and a robust controller. The hysteretic comportment of the material is evaluated for the normal (symmetric hysteretic curve) and extended functioning domain (asymmetric hysteretic curve). The control methods are compared in terms of tracking error, number of operations and used memory from numerical implementation point of view. The results show that the proposed method needs lesser resources only in the case of extended functioning domain while the tracking errors for the two methods are in general of the same magnitude.

On the relevance of nonlinear analysis of time series associated to CT kidney images

The goal of this paper is to present a set of algorithms — based on nonlinear time series analysis methods, use them for discrimination between normal and modified kidney tissue and conclude if this analysis is trustworthy. A set of 100 CT images of normal and benign/malign affected renal tissue was used. The classification procedure implies the association of a time series to every CT image and computation of the correlation dimensions of the reconstructed attractors over a range of embedding spaces. The paper concludes that the properties of the associated time series can be used to discriminate between normal and modified tissue. This classification tool can be utilized as diagnostic confirmation support and the results can be improved by a further added database.

Intelligent Precision Control for Air Heater Systems

Abstract Air heater systems and, in general, thermo conditioning systems are among the most used systems in industrial and domestic areas. Superior performances in exploitation are real sources of energy savings. This paper presents some aspects on the intelligent precision control of air heater processes. The proposed solution consists in a general decentralized control structure for pressure/flow and temperature processes, the main phenomena in thermal systems. The important influences between pressure/flow and temperature loops are solved using an intelligent combined multiple model - feedforward control structure. Its applicability is analyzed by means of a real-time structure implementation using a RST control algorithm. Based on the software implementation, an analysis of the advantages and disadvantages of the proposed solution is made.

Modeling and simulation of the dynamics of hydro-energetic systems

Abstract This paper presents a strategy for modeling and simulation of the hydro-energetic dam dynamics. The useful information for the evaluation of the mathematical models is represented by a data set collected for twenty years from one of the hydro-energetic systems in Romania. The results of our work allow the evaluation of the present state of this system and offer some important information for the computer-aided design of similar systems. These results have been obtained using software tools dedicated to data acquisition, modeling and identification.