Mekki Ksouri

Model Predictive Control of Fractional Order Systems

This paper provides the model predictive control (MPC) of fractional order systems. The direct method will be used as internal model to predict the future dynamic behavior of the process, which is used to achieve the control law. This method is based on the Grünwald-Letnikovs definition that consists of replacing the noninteger derivation operator of the adopted system representation by a discrete approximation. The performances and the efficiency of this approach are illustrated with practical results on a thermal system and compared to the MPC based on the integer ARX model.

LMI formulation for static output feedback design of discrete-time switched systems

This paper concerns static output feedback design of discrete-time linear switched system using switched Lyapunov functions (SLFs). A new characterization of stability for the switched system under arbitrary switching is first given together with γ-performance evaluation. The various conditions are given through a family of LMIs (LinearMatrix Inequalities) parameterized by a scalar variable which offers an additional degree of freedom, enabling, at the expense of a relatively small degree of complexity in the numerical treatment (one line search), to provide better results compared to previous one. The control is defined as a switched static output feedback which guarantees stability and γ-performance for the closed-loop system. A numerical example is presented to illustrate the effectiveness of the proposed conditions.

Wavelets: An efficient tool for lung sounds analysis

The objective of this paper is to use adaptive wavelets for lung sounds analysis and show that wavelets with one vanishing moment can successfully detect pathological changes of the lung which produce sounds with measurable regularities. Local regularity measures allow us to detect some significant components of adventitious sounds which are difficult to detect by the physician ears due to their short duration. This paper will concentrate on a development of lung sounds pattern recognition features. The key properties of pattern recognition features, Lipschitz regularity at any point of wavelet transform modulus maxima along the maxima lines converging to this point, regularity of some adventitious lung sounds such as Crackles and Wheezes will be analyzed. Numerical results prove that normal lung sound is more regular than crackles lung sounds.

Lung sound extraction from mixed lung and heart sounds FASTICA algorithm

The reduction of interface by heart sounds in lung sounds, in order to improve the analysis of lung sounds has been discussed since the earliest work in 1986 [1] and numerous numbers of papers were developed further. In fact, a large amount of methods was proposed to quantify pure lung sounds, compared with heart sounds interference [2-5]. This paper deals with the performance of lung and heart sound separation based on an objective method, which consist on a mathematical independent component analysis algorithm applied to a preprocessed waveform and spectra of lung sounds recorded with a digital stethoscope. However, such objective evaluation allows to interpret the significance of the noise reduction and to conclusively compare the performance of this particular method with other methods. The quantitative performance measure of those techniques is almost based on the NR ratio or NIR ratio, which is the noise to input ratio, in some references and also the S/N ratio that is the signal to noise ratio. In this investigation we will introduce the bases of the ICA technique used to lung and heart signal separation, we will expose some results of lung and heart signal separation that we have obtained based on the methodology described above. This method was successfully used to separate heart sound and lung sound from recorded mixed sounds. A number of illustrations are exposed here and a heart to lung sound signal ratio H/L is calculated for different cases.

Computer based analysis for heart and lung signals separation

In this paper, two methodologies are proposed to enhance the automatic noise cancellation and signal separation between heart and lung sounds. In fact, transient signals such as heart and lung signals may undergo abrupt or sharp change in the first and second derivatives. A real separation between such two interfering mixed signals needs an efficient approach to avoid losing important information in both signals. Rhythmic cardiac signal contains important characteristics which can be exploited to develop adaptive based algorithms that allow efficient separation between lung and heart signals when they are mixed in a recorded signal. In the first proposed methodology we have developed an algorithm based on adaptive filtering technique and build using multiple filtering functions with coefficients correlated to the mixed source signal. In the second methodology, fast independent component analysis was developed to cancel heart sound in lung mixed sound. Both methods are well detailed in this work, and a comparative study is achieved to evaluate the efficiency of each method. A high accuracy of the new proposed algorithms is found and many applications are used to quantify the performances of these techniques.

A chattering free control approach based on dynamical multi-sliding surfaces

in this work a new control technique combining integral sliding mode and multi-model concept is developed. It is based on a finite set of dynamic commutation surfaces which allow to have a completely robust behavior with respect to certain classes of uncertainties and to completely eliminate the chattering phenomenon. Simulations results show a notable improvement relatively to the classical sliding mode control.

Behavior Graphs for Hybrid Systems Monitoring

Hybrid Dynamical Systems (HDS) constitute a wide class of common industrial applications, where the behavior is determined by the interaction between continuous and discrete dynamics, i.e. behavioral modes succession. The general principle of model-based Fault Detection and Isolation (FDI) algorithms is to compare the expected behavior of the system, given by a model, with its actual behavior, known through on-line observations. Faults in HDS may corrupt the two dynamics. In that paper, we propose to limit the set of possible mode candidates by using a priori information on the discrete evolution under normal and faulty hypothesis. Two kinds of graphs are derived from the initial hybrid model, namely Normal Behavior Graphs (NBG), Faulty Behavior Graphs (FBG). Using these graphs allows us not only to identify efficiently the actual mode but also to directly interpret (diagnose) the discrete faulty evolution in terms of faults. The whole FDI methodology is described and applied to a two tanks system example.

Extended CNP Framework for the Dynamic Pickup and Delivery Problem Solving

In this paper, we investigate in the applicability of the Contract Net Protocol negotiation (CNP) in the field of the dynamic transportation. We address the optimization of the Dynamic Pickup and Delivery Problem with Time Windows also called DPDPTW. This problem is a variant of the Vehicle Routing Problem (VRP) that may be described as the problem of finding the least possible dispatching cost of requests concerning the picking of some quantity of goods from a pickup to a delivery location while most of the requests continuously occur during the day. The use of contract nets in dynamic and uncertain domains such as ours has been proved to be more fruitful than the use of centralized problem solving [9].We provide a new automated negotiation based on the CNP. Negotiation process is adjusted to deal intelligently with the uncertainty present in the concerned problem.

TSP based Evolutionary optimization approach for the Vehicle Routing Problem

Vehicle Routing and Flexible Job Shop Scheduling Problems (VRP and FJSSP) are two common hard combinatorial optimization problems that show many similarities in their conceptual level [2, 4]. It was proved for both problems that solving techniques like exact methods fail to provide good quality solutions in a reasonable amount of time when dealing with large scale instances [1, 5, 14]. In order to overcome this weakness, we decide in the favour of meta heuristics and we focalize on evolutionary algorithms that have been successfully used in scheduling problems [1, 5, 9]. In this paper we investigate the common properties of the VRP and the FJSSP in order to provide a new controlled evolutionary approach for the CVRP optimization inspired by the FJSSP evolutionary optimization algorithms introduced in [10].

Integral pseudo sliding mode control for nonlinear systems

This paper reports a new technique of control combining the concepts of sliding mode control and the multimodel approach. Baptized Integral Pseudo Sliding Mode Control, this control strategy, reduces the library of models whose number can block the establishment of the multi model control strategy in real time. It also allows, on the one hand, having a robust behavior compared to certain classes of uncertainties, and, on the other hand eliminating completely the chattering phenomenon.