AHMED FARIS Amiri
أحمد فارس عميري
ahmedfaris.amiri@univ-msila.dz
0658258117
- Departement of ELECTRONICS
- Faculty of Technology
- Grade PHd
About Me
Location
Msila, Msila
Msila, ALGERIA
Code RFIDE- 1996-06-16 00:00:00
-
AHMED FARIS Amiri birthday
- 2024-12-23
-
2024-12-23
Comparative Analysis of Machine Learning Models for Fault Detection in Photovoltaic Systems
Fault detection (FD) in photovoltaic (PV) systems is crucial for ensuring efficient energy production, minimizing maintenance costs, and maintaining system reliability. In this study, we conducted a comprehensive evaluation of several machine learning techniques for FD in PV systems, including Support Vector Regression (SVR), Multi-Layer Perceptron (MLP), Artificial Neural Networks (ANN), k-Nearest Neighbors (KNN), Decision Trees (DT), and Random Forest (RF). The performance of these models was analyzed based on their ability to handle the dynamic and nonlinear behavior of PV systems. Results from our experiments affirm that RF outperformed the other models in terms of robustness to noisy data and overall accuracy. MLP and ANN exhibited strong capabilities in capturing complex patterns, while SVR and KNN showed promise in handling specific data structures. This study offers valuable insights into the application of machine learning techniques for fault detection in PV systems, with RF emerging as the most reliable solution for enhancing system performance and reducing downtime.
Citation
Houcine OUDIRA , Ahmed Faris amiri , Aissa CHOUDER , ,(2024-12-23), Comparative Analysis of Machine Learning Models for Fault Detection in Photovoltaic Systems,5th International Conference on Scientific and Academic Research on 23-24 December in 2024 a,Konya/Turkey.
- 2024-10-31
-
2024-10-31
Enhanced Parameters Extraction Procedure of PV Module Using Electrical Fish Optimization
Accurate and reliable fault detection procedures are crucial to ensure normal operation of photovoltaic (PV) systems. To this end, the use of trusted model is the major step and an essential tool for monitoring and supervision the system under consideration. In his paper, the unknown parameters of the one diode model (ODM) in outdoor conditions are accurately identified using an enhanced methodology. The proposed methodology combines a novel translation method to correct the I-V curves to reference conditions and analytical formulations to derive the considered parameters in any operating condition of irradiance and temperature. For determining the five unknown parameters at standard test conditions, an optimization algorithm namely the electrical fish optimization (EFO) is used. Based on the extracted parameters, the evolution of maximum power point model was modeled and simulated versus measurements of a grid connected real MPP system . The obtain results show the effectiveness of the proposed strategy.
Citation
Houcine OUDIRA , Ahmed Faris amiri , Aissa CHOUDER , ,(2024-10-31), Enhanced Parameters Extraction Procedure of PV Module Using Electrical Fish Optimization,2024 IEEE International Multi-Conference on Smart Systems & Green Process,Tunisia
- 2024-01-05
-
2024-01-05
Faults detection and diagnosis of PV systems based on machine learning approach using random forest classifier
Accurate and reliable fault detection procedures are crucial for optimizing photovoltaic (PV) system performance. Establishing a trustworthy PV array model is the primary step and a vital tool for monitoring and diagnosing PV systems. This paper outlines a two-step approach for creating a reliable PV array model and implementing a fault detection procedure using Random Forest Classifiers (RFCs). Firstly, we extracted the five unknown parameters of the one-diode model (ODM) by combining the current– voltage translation method to predict the reference curve and employing the modified grey wolf optimization (MGWO) algorithm. In the second step, we simulated the PV array to obtain maximum power point (MPP) coordinates and construct operational databases through co-simulations in PSIM/MATLAB. We developed two RFCs: one for fault detection (a binary classifier) and another for fault diagnosis (a multiclass classifier). Our results confirmed the accuracy of the PV array modeling approach. We achieved a root mean square error (RMSE) value of 0.0122 for the ODM parameter extraction and RMSEs lower than 0.3 in dynamic PV array output current simulations under cloudy conditions. Regarding the fault detection procedure, our results demonstrate exceptional classification accuracy rates of 99.4% for both fault detection and diagnosis, surpassing other tested models like Support Vector Machines (SVM), K-Nearest Neighbors (KNN), Neural Networks (MLP Classifier), Decision Trees (DT), and Stochastic Gradient Descent (SGDC).
Citation
Ahmed Faris amiri , Houcine OUDIRA , Aissa CHOUDER , , (2024-01-05), Faults detection and diagnosis of PV systems based on machine learning approach using random forest classifier, Energy Conversion and Management, Vol:301, Issue:1, pages:1-15, Elsevier
- 2023-10-28
-
2023-10-28
Prediction Model of PV Module Based on Artificial Neural Networks for the Energy Production
The accurate characterization and prediction of current-voltage characteristics of photovoltaic (PV) modules under different operating conditions is the main step for energy prediction and an important tool for monitoring and supervision the system. However, one of the problems of this technology is that as yet there are no models in the literature to directly calculate the daily dynamic maximum power of these kinds of PV systems. The development of models is an important task to support the application of this technology because it allows the prediction of the energy yield. In this paper a model based on artificial neural networks has been developed to address this important issue. The model takes into account the main important parameters that influence the electrical output of these kinds of systems which are direct irradiance, and module temperature. Comparative study with The simulated dynamic MPP model using the single diode model is presented to demonstrate the effectiveness of the considered approach. The obtained results show that the proposed model can be used for estimating the maximum power of a grid connected system located in the Centre de Developpement des Energies Renouvelables (CDER) in Algiers with an adequate margin of error.
Citation
Ahmed Faris amiri , Houcine OUDIRA , Aissa CHOUDER , ,(2023-10-28), Prediction Model of PV Module Based on Artificial Neural Networks for the Energy Production,5th Novel Intelligent and Leading Emerging Sciences Conference (NILES),Egypt
- 2022-11-26
-
2022-11-26
Faults Detection of PV Systems Based on Extracted Parameters using Modified Grey Wolf Algorithm
Accurate and reliable fault detection procedures are crucial to ensure normal operation of photovoltaic (PV) systems. To this end, the use of trusted model is the major step and an essential tool for monitoring and supervision the system under consideration. In his paper a suggested procedure based on three main steps is presented. Firstly, the unknown parameters of the one diode model (ODM) are accurately identified using modified grey wolf (MGW) algorithm. Subsequently, based on the extracted parameters, the evolution of maximum power point model was modeled and simulated versus measurements of a grid connected real MPP system . Finally, the PV array is simulated to take out the MPP coordinates by using a PSIMTM/MatlabTM co-simulation, as well as an efficient fault detection process based on simple approach is implemented. The obtain results show the effectiveness of this method in detecting and diagnosing faults for real time application.
Citation
Ahmed Faris amiri , Houcine OUDIRA , Aissa CHOUDER , ,(2022-11-26), Faults Detection of PV Systems Based on Extracted Parameters using Modified Grey Wolf Algorithm,the 2022 International Conference of advanced Technology in Electronic and Electrical Engineering (ICATEEE),M'sila University, Algeria,