Browsing by Author "Grouni, Said"

Now showing 1 - 17 of 17

Advanced control algorithm : applications to industrial processes
(2015) Ramdani, Ammar; Grouni, Said; Traïche, M.
As in the most industrial systems, a control of the input of the systems including a classic regulator is a key point. The Proportional-Integral-Derivative controllers are commonly used in many industrial control systems and appeared suitable to stable the control of the majority of real processes. But in some cases like a non-minimum-phase plant or a plant with a dead-time proceed to a thin regulating of coefficients until to get a system respecting the conditions specified. It is possible also to present a problem of overtaking with the increase of the gain or seems impotent for systems having a big delay and the use of sophisticated process controllers is required. Model predictive control is an important branch of automatic control theory, it refers to a class of control algorithms in which a process model is used to predict and optimize the process performance. MPC has been widely applied in industry. Dynamic Matrix Control Algorithm belongs to the family of Model predictive control Algorithms where these algorithms only differ between themselves in the model that represents the process, disruptions and the function of cost. In this paper the study of the Dynamic Matrix Control Algorithm are interested while applying him on processes of water heating and mechanical rotations of steering mirrors in a Light Detection and Ranging system as a second application. The objective of this work consists of solving the problem of prediction of the output and input of the process by fixing a horizon finished N, and while considering the present state like initial state, to optimize a cost function on this interval, while respecting constraints. Therefore, the future reference is known and the system behavior must be predictable by an appropriate model. It results an optimal sequence of N control of it among which alone the first value will be applied effectively. As the time advances, the horizon of prediction slips and a new problem of optimization is to solve while considering the state of the system updating. In summary, every moment, it is necessary to elaborate an optimal control sequence in open loop, refined systematically by the present measure arrival
Advanced trajectory tracking control applied to dynamic system with disturbance
(2014) Ramdani, Ammar; Grouni, Said; Bouallegue, K.
This paper deals to present an advanced control in predictive control application. This control method is mainly based on the prediction model and the objective function to drive the nearest output possible of the trajectory in the sense of least square within. We are interested in evaluating the performance of these control technique and their applications on dynamic systems
Application of predictive controller tuning and a comparison study in terms of PID controllers
(Elseier, 2016) Ramdani, Ammar; Grouni, Said; Souﬁ, Youcef
Dynamic angle stability of an industrial turbo generator connected in power system
(AIP Publishing, 2014) Grouni, Said; Hallak, M.; Aibeche, Abderrazak; Ramdani, Ammar; Bouallegue, K.
Dynamic matrix control and generalized predictive control, comparison study with IMC-PID
(Elsevier, 2017) Ramdani, Ammar; Grouni, Said
Improved Pi-Sigma Neural Network for nonlinear system identification
(2017) Ladjouzi, Samir; Grouni, Said; Kacimi, Nora; Soufi, Youcef
In this paper, we propose a modified architecture of a Pi-Sigma Neural Network (PSNN) based on two modifications: extension of the activation function and adding delays to neurons in the hidden layer. These new networks are called respectively Activation Function Extended Pi-Sigma (AFEPS) and Delayed Pi-Sigma (DPS) are obtained first by adding an activation function to all hidden neurons and secondly by modifying the PSNN so its hidden layer outputs are fed to temporal adjustable units that permit to this new network to be capable to identify nonlinear systems. Architecture and dynamic equations of these networks are given in details with their training algorithm. To ensure the effectiveness of our proposed networks, examples of nonlinear system identification are provided. The obtained results show the capacity of HONNs for the nonlinear systems identification. In particular, the proposed neural architectures (AFEPS and DPS) provide better results due to the modifications made on them
Improved sensorless backstepping controller using extended Kalman filter of a permanent magnet synchronous machine
(2022) Kirad, Abderrahmen; Grouni, Said; Soufi, Youcef
This paper deals with an improved backstepping control strategy for sensorless control of a permanent magnet synchronous motor (PMSM) based on field-oriented control (FOC), using a backstepping controller to improve its performances. However, this control requires the precise knowledge of some machine’s variables which could not be available. In electric drives control, sensors are generally used as the main devices for feedback information. Some practical constraints could affect the system performances, due to the lack of measurement material or maintenance difficulties caused by the dysfunction or faults of the used sensors such as: encoder or resolver sensors of speed-position. In this paper, a sensorless control is proposed based on a dynamic backstepping method and an extended Kalman filter (EKF) which uses the state space formulation with a set of mathematical equations to recursively estimate future observations and minimizes the mean square error of the estimated variables (rotor speed position and torque) to design controllers for nonlinear systems. The proposed control scheme achieves the asymptotically uniformed stability. The effectiveness of this method is illustrated by the stabilization and tracking numerical examples and the obtained simulation results show the effectiveness and the feasibility of the proposed controller using Lyapunov approach.
A Modified elman network with memory units for system identification
(2018) Ladjouzi, Samir; Grouni, Said; Soufi, Youcef
In this paper we propose a modified Elman network structure called memory Elman neural network. The idea of this new architecture is based on adding memory units to the neurons of the classic Elman network. These memory units are trainable temporal elements that make the output history-sensitive. By virtue of this capacity, this new architecture can take into account the past information of the neurons and use them in order to accomplish the task of the network. In order to show the performance of this new network, some dynamical systems are used for identification and results are compared with the conventional Elman network
The National Conference on Computational Engineering, Artificial Intelligence and Smart Systems- NCCEAISS’2024 December 10th to 12th, 2024, Tamanrasset, Algeria : Book of abstracts
(2024) Recioui, Abdelmadjid; Dekhandji, Fatma Zohra; Bentarzi, Hamid; Benazzouz, Djamel; Grouni, Said; Yakhelef, Yassine
A neural MPPT approach for a wind turbine
(IEEE, 2017) Ladjouzi, Samir; Grouni, Said; Djebiri, Mustapha; Soufi, Youcef
New improved hybrid MPPT based on neural network-model predictive control-kalman filter for photovoltaic system
(2020) Kacimi, Nora; Grouni, Said; Idir, Abdelhakim; Boucherit, Mohamed Seghir
In this paper, new hybrid maximum power point tracking strategy for photovoltaic systems has been proposed. The proposed technique for control based on a novel combination of an artificial neural network with an improved model predictive control using Kalman Filter. In this paper the Kalman Filter is used to estimate the converter state vector for minimized the cost function then predict the future value to track the maximum power point with fast changing weather parameters. The proposed control technique can track the in fast changing irradiance conditions and a small overshoot. Finally, the system is simulated in the MATLAB/Simulink environment. Several tests under stable and variable environmental conditions are made for the four algorithms, and results show a better performance of the proposed compared to conventional perturb and observation neural network based proprtional integral control and neural network based model predictive control in terms of response time, efficiency and steady-state oscillations
A new training method for solving the XOR problem
(2017) Ladjouzi, Samir; Grouni, Said; Kirat, Abderrahmen; Soufi, Youcef
Training of Artificial Neural Networks (ANN) is an important step to make the network able to accomplish the desired task. This capacity of learning in such networks makes them applied in many applications as modeling and control. However, many of training algorithms have some drawbacks like: too many parameters to be estimated, important calculus time. In this paper, we propose a very simple method to train a Single Hidden Layer Perceptron (SHLP) based on replacing the traditional ANN’s phase training by another approach called Neural Least Mean Square (NLMS) problem resolution. The key of this method is to compute some ANN’s weights by the Least Mean Square (LMS) formula, and to leave others weights to their initial values. This new training method is applied to the classical XOR problem and the results are compared with the conventional Backpropagation algorithm. The obtained results were satisfactory and the comparison made with the classical algorithm revealed that our method allowed to reduce several parameters in the learning, namely: the computation time, the overall value of the error squared, the number of iterations and the number of weights to be adjusted
PID Control of DC Servo Motor using a Single Memory Neuron
(IEEE, 2018) Ladjouzi, Samir; Grouni, Said; Soufi, Youcef
In this paper, a novel approach to determine the optimal values of a PID controller is presented. The proposed method is based on using a single memory neuron which its weights represent the PID parameters. These weights are updated by the well-known bio-inspired algorithm: the particle swarm optimization. To show the efficiency of our method, we have applied it to control a DC servo motor which is used as an actuator for an arm robot manipulator. The obtained results are compared with those a fuzzy logic controller.
PID controller parameters adjustment using a single memory neuron
(Elsevier, 2020) Ladjouzi, Samir; Grouni, Said
The work presented in this paper presents the use of a single memory neuron to find optimal gains for a PID controller. The adopted strategy with the principal equations is discussed. The efficiency of the proposed method is shown for two usual problems which frequently occur in the industry: a single tank and a boiler and heat exchanger applications. A comparison with the Ziegler–Nichols method is presented in order to prove the effectiveness of our approach
Predictive functional control approach, study and simulation
(2015) Ramdani, Ammar; Grouni, Said; Bouallegue, K.
Quadcopter nonsingular finite-time adaptive robust saturated command-filtered control system under the presence of uncertainties and input saturation
(Springer Science and Business Media B.V., 2021) Eliker, Karam; Grouni, Said; Tadjine, Mohamed; Zhang, Weidong
The nonsingular finite-time adaptive robust saturated command-filtered control problem for quadcopter unmanned aerial vehicles is investigated in this paper. Firstly, an adaptive robust command-filtered control, based on backstepping command-filtered and nonsingular fast terminal sliding mode control, is developed. Secondly, the parametric and nonparametric uncertainties are estimated by using a small number of adaptive laws. Also, a projector function is used to ensure the estimation of quadcopter parameters within an admissible set. Thirdly, error compensating signals are employed to tackle the undesirable effect of command filters. Finally, saturation compensator signals are developed to deal with the adverse effect of saturation in the system. The proposed control strategy can cope with the “explosion of complexity” and “singularity” problems. In addition, it can alleviate the chattering phenomenon and satisfy practical finite-time stability. The numerical simulation results and comparison display the effectiveness of the proposed technique over the other control methods
A Single-Neuron-Based Temperature Control of a Continuous Stirred Tank Reactor
(Springer Nature, 2024) Ladjouzi, Samir; Grouni, Said
In this paper, a new technique to determine the best values of a PID controller is presented. The proposed scheme is based on using a single-neuron controller which its weights represent the PID parameters. Weight’s adjustment is accomplished with a recent meta-heuristic algorithm called the DragonFly Algorithm. To show the effectiveness of our method, we have applied it to control a Continuous Stirred Tank Reactor. The obtained results are compared with several algorithms: the Ziegler–Nichols, Genetic Algorithm, and Particle Swarm Optimization.