CHAIMA Chabira

SEFIANE Nour elislam , SAIDANE Sarra, SEGHIOUR Chaima

Mini Drone Pédagogique (Contrôle de l'Altitude).

Mekki Ahmed Tayeb , Laraba Ala eddine, Benlatrache islam

Smart Trash Car

GHECHAM MERYEM EL BATOUL , GANA SAMAH, FOUDILI KAOUTHER

Un système d’irrigation intelligent innovant utilisant une carte à base de microcontrôleur précis basée sur des capteurs pour améliorer l’arrosage des plantes

Water distribution networks (WDNs) are critical infrastructure for supplying water to communities but are susceptible to leaks, leading to significant water losses, infrastructure damage, and increased operational costs. Effective monitoring and maintenance solutions are essential to address these challenges. This study presents an advanced leak detection system incorporating an artificial intelligence (AI)-based algorithm to enhance accuracy in identifying water leaks. Real-time pressure variations are sensed using pressure transducers, designed to resist environmental noise, thereby improving system reliability. To facilitate deep data learning (DL), a prototype PEHD hydraulic pipeline was built, measuring 100 m in length and 40 mm in diameter with a thickness of 2.4 mm, on which two pressure transmitters were installed at previously known positions linked to a dSPACE acquisition system. Pressure data, including leakage and no-leakage scenarios, is collected and transformed into 2D images using continuous wavelet transform (CWT). These 2D images are used to train convolutional neural networks (CNNs), optimized for extracting spatial and temporal features from image-based data. Transfer learning techniques, leveraging pretrained models like DenseNet, MobileNet, ResNet, and Xception, further enhance feature extraction. Additionally, machine learning (ML) classifiers such as SVM, KNN, and XGBoost are integrated with the CNN framework to improve classification accuracy under dataset splits (80%-20%). A hybrid approach, combining deep feature extraction with traditional classification methods, is proposed as an efficient and accurate solution for leak detection in WDNs.

Citation

CHAIMA CHABIRA , ,(2025-01-15), Enhancing Leak Detection in WDNs with CWT and CNN Applied to Pressure Signals,The First National Conference on Emerging Trends in Engineering and Technology (NC2ET’25),Mascara

Water Distribution Networks (WDNs) are essential infrastructures responsible for delivering water to communities. However, they are highly susceptible to leaks, which result in considerable water losses, increased operational costs, and potential damage to the infrastructure. Accurate and efficient leak detection is therefore critical for ensuring the sustainability of these systems. This study introduces the Empirical Mode Decomposition (EMD) method as an advanced approach to enhance pipeline leak detection. Leaks in pipelines generate acoustic and vibration signals that propagate at low frequencies, exhibiting nonlinear and nonstationary characteristics. To address this, we constructed a prototype hydraulic circuit made of Polyethylene High-Density (PEHD) material, measuring 100 m in length and 40 mm in diameter. The circuit was equipped with vibration transducers and an advanced dSPACE based acquisition system to capture high-resolution data. Traditional signal analysis techniques, such as Fast Fourier Transform (FFT), correlation, and cepstrum analysis, often miss crucial information due to their dependence on narrow frequency bands and the assumption of signal stationarity. EMD, a time-domain decomposition method designed for nonstationary signals, resolves these issues by extracting Intrinsic Mode Functions (IMFs), enabling the separate analysis of signal components. Our experiments involved testing the EMD method on various leak scenarios and comparing it with conventional techniques. The results demonstrate that EMD provides superior leak detection capabilities by effectively analyzing the complex and nonstationary nature of pipeline signals. This approach offers a promising solution for real-time monitoring and maintenance of WDNs, ensuring their operational efficiency and sustainability.

Citation

CHAIMA CHABIRA , ,(2024-12-23), Leak Detection in WDNs with PEHD Pipes Using Vibration Signals and EMD,5th International Conference on Modern and Advanced Research ICMAR 2025,Turkey

Water Distribution Networks (WDNs) are critical infrastructure for delivering water to populations. However, leaks in these networks can result in significant water losses, increased operational costs, and damage to infrastructure. Traditional leak detection methods often lack the precision for timely and accurate identification. This study proposes an innovative approach combining Discrete Wavelet Transform (DWT) and Convolutional Neural Networks (CNN) for detecting leaks using pressure signals recorded from WDNs. DWT is employed to decompose pressure signals into multi-scale components, enabling the isolation of features relevant to detecting leaks. By converting raw pressure data into a structured representation. DWT enhances the clarity of signal patterns associated with leakage events. These decomposed signals are subsequently analyzed using a CNN, which extracts spatial and temporal features to classify signals into leaky and non-leaky categories. The integration of DWT and CNN ensures that both fine-grained signal variations and broader contextual patterns are effectively captured. The proposed method was validated on a dataset of pressure signals collected from real-world WDNs, demonstrating high accuracy, precision, and recall. Compared to existing techniques, this approach offers superior performance in distinguishing between leakage and non-leakage scenarios, even in complex environments. The results highlight the potential of combining advanced signal processing and deep learning (DL) methods to improve leak detection capabilities significantly reduce water losses and enhance system reliability. Future work will focus on expanding the dataset, integrating additional sensors, and exploring further improvements to detection accuracy and robustness.

Citation

CHAIMA CHABIRA , ,(2024-12-13), Advanced Leak Detection in WDNs Using DWT and CNN,4th International Conference on Frontiers in Academic Research ICFAR,Turkey

Water Distribution Networks (WDNs) are essential for ensuring the reliable delivery of water to communities. However, they are susceptible to leaks, which can result in substantial water losses, increased operational costs, and damage to the system. Early detection and mitigation of leaks are therefore crucial for maintaining the efficiency and sustainability of WDNs. To detect leaks using pressure signals recorded from WDNs, this study proposes a new method that combines Convolutional Neural Networks (CNN) and Continuous Wavelet Transform (CWT). The CNN can extract significant spatial and temporal information by preprocessing the pressure signals and converting them into time-frequency representations using CWT. The proposed model was trained and evaluated on a dataset of pressure signals from real WDNs, achieving high accuracy, precision, and recall. The results demonstrate the efficacy of integrating CWT for signal processing and CNN for feature extraction to enhance leak detection. This approach provides a promising framework for improving urban infrastructure leak management and maintenance strategies. Future work will focus on validating the model on larger datasets and exploring the integration of additional sensors to enhance detection robustness. This approach not only aids in timely leak identification but also supports proactive maintenance strategies, ultimately reducing water losses and enhancing system reliability. Future research will focus on validating the model on larger and more diverse datasets.

Citation

CHAIMA CHABIRA , ,(2024-12-13), Enhancing Leak Detection in WDNs with CWT and CNN Using Pressure Signals,4th International Conference on Frontiers in Academic Research ICFAR,Turkey

Detecting leaks in pipeline networks, particularly in water distribution systems, is a critical yet challenging task. Numerous techniques have been proposed for leak detection; however, none have proven entirely suitable for industrial applications due to limitations such as low efficiency or high costs. This study focuses on employing cepstrum analysis as a powerful signal processing technique to detect and locate leaks in pipelines based on pressure signal measurements. The proposed method involves preprocessing pressure signals to remove noise using orthogonal wavelets, followed by the application of cepstrum analysis to extract key features indicative of leaks. Earlier studies have demonstrated that this approach can effectively pinpoint the location of leaks and assess their severity in pipeline systems. Experiments were conducted on a controlled fluid-filled pipeline network, where pressure transducers recorded transient pressure drops triggered by simulated leaks. The analysis demonstrated the ability of cepstrum-based technique to detect accurately the location of leaks across varying pipeline distances. Pressure transducers were used to record pressure signals, with leakage initiated via a push-button mechanism acting on a solenoid valve. The transient pressure drop was captured during the recordings, which were conducted at varying pipeline distances from the leakage point.

Citation

CHAIMA CHABIRA , ,(2024-12-09), Leak Detection in WDNs Using Cepstrum Analysis of Pressure Signals,The 1st National Conference of Advanced Systems in Electrical Engineering (NCASEE'24),Boumardas

In recent years, the convergence of the Internet of Things (IoT) has driven innovations in pressure sensor development, enhancing accuracy, responsiveness, and efficiency in data handling. IoT connects sensors in real-time, enabling them to transmit vast amounts of pressure data to centralized systems for analysis and decision-making. The development of IoT, the everincreasing number of internet users, and the advancement of data rates and services have produced enormous amounts of data that can be utilized to enhance service delivery. IoT provides opportunities for both gathering and applying this data. This paper explores relationship between IoT and sensors, emphasizing their roles in optimizing various sectors such as Autonomous Vehicles, Industrial Automation, Smart Homes, Healthcare and Medical, Smart City and Agriculture. By integrating advanced technologies, including Machine Learning (ML) algorithms, we also examine the relationship between pressure sensors and IoT in leak detection within Water Distribution Networks (WDNs).

Citation

CHAIMA CHABIRA , ,(2024-12-09), Integration of IoT for Effective Leak Detection in WDNs,The 1st National Conference of Advanced Systems in Electrical Engineering (NCASEE'24),Boumardas

The convergence of the Internet of Things (IoT) with pressure sensor technology has revolutionized data acquisition and management, enabling unprecedented accuracy and efficiency in various applications. IoT facilitates real-time connectivity among sensors, allowing them to transmit vast amounts of pressure data to centralized systems for analysis and decision-making. This seamless integration has been fueled by the rapid growth in internet users, advancements in data rates, and enhanced service delivery capabilities. IoT not only enables the collection of large datasets but also opens avenues for their application across diverse industries. This paper highlights the relationship between IoT and pressure sensors, examining their transformative impact on sectors such as autonomous vehicles, industrial automation, smart homes, healthcare, smart cities, and agriculture. By leveraging advanced technologies like Machine Learning (ML), IoT enhances system optimization and operational efficiency. A focal point of this study is the application of IoT and pressure sensors in leak detection with in Water Distribution Networks (WDNs). Integrating ML algorithms with IoT-connected pressure sensors allows for the identification and localization of leaks, ensuring effective system monitoring and maintenance. This approach significantly reduces water loss, operational costs, and downtime, contributing to the sustainability and reliability of WDNs.

Citation

CHAIMA CHABIRA , ,(2024-12-03), The Role of IoT and Pressure Sensors for Leak Detection in WDNs,3rd International Conference on Recent Academic Studies ICRAS 2024,Turkey

Detecting leaks in pipeline networks, particularly in water distribution systems, is a critical yet challenging endeavor. Despite the availability of various detection techniques, many are unsuitable for industrial applications due to limitations such as inefficiency and high costs. This study explores the use of cepstrum analysis, a robust signal processing technique, to detect and locate pipeline leaks by analyzing pressure signal measurements. The method begins with preprocessing pressure signals to eliminate noise using orthogonal wavelets, ensuring the accuracy of subsequent analysis. Cepstrum analysis is then applied to extract key features indicative of leaks. Previous research has shown that this approach can effectively identify leak locations and assess their severity in pipeline systems. Controlled experiments were conducted on a fluid-filled pipeline network, where pressure transducers recorded transient pressure drops caused by simulated leaks. The leaks were initiated using a push-button mechanism that operated a solenoid valve, with pressure signals captured at various pipeline distances from the leakage point. The results demonstrated that the cepstrum-based technique reliably detects and locates leaks with high precision, regardless of the distance from the leak. This study underscores the potential of cepstrum analysis as a valuable tool for improving leak detection methodologies. By providing a cost-effective and efficient solution, this technique can enhance the integrity and operational efficiency of pipeline systems, addressing a critical need in water distribution networks (WDNs).

Citation

CHAIMA CHABIRA , ,(2024-12-03), Cepstrum Analysis of Pressure Signals for Effective Leak Detection in WDNs,3rd International Conference on Recent Academic Studies ICRAS 2024,Turkey

This paper examines the analysis and modeling of vector-controlled Permanent Magnet Synchronous Motors (PMSMs) for electric vehicles (EVs) applications. PMSMs are widely used in EVs due to their high efficiency and torque density. However, acquiring best performance is requires suitable control systems. The research focuses on developing a dynamic model for PMSMs with Field-Oriented Control (FOC), which allows for exact torque and speed regulation. PMSM performance is simulated under different operating conditions. The impact of control strategies on system response is considered and torque stability. The findings provide valuable insights for optimizing PMSM drive systems in electric vehicles, aiming to enhance performance and energy efficiency. Results is studied using MATLAB/SIMULINK.

Citation

CHAIMA CHABIRA , ,(2024-11-22), Analysis and Modeling of Vector-Controlled Permanent Magnet Synchronous Motors for Electric Vehicles,2nd International Conference on Trends in Advanced Research ICTAR 2024,Turkey

Water distribution networks (WDNs) are important infrastructures that provide water to communities. However, they do absorb, causing large amounts of water loss and damage to infrastructure. The methods of detecting and reporting leaks in WDNs are usually laborious and time-consuming. Recently, machine learning has demonstrated its ability to improve the efficiency and accuracy of leak detection methods. In this paper, we have proposed the use of pressure transducers due to their high accuracy and resistance to ambient noise compared to those available in the international market. We fabricated a new 100 m long PEHD sample pipe with a diameter of 40 mm, equipped with two transducers. In addition, we used the purchase dispositif of the DSPACE commercial research purchase order, specifically the MicroLabBox model. In this case the collected Data were processed, to ensure a complete picture of the leak signals. Various ML algorithms, including Support Vector Machines (SVM), Naïve Bayes (NB) and XGBoost, have been analyzed to classify the collected dataset into leak our no_leak classes accurately Overall, the obtained results show that XGBoost has an optimal performance of 90.91%. had the highest accuracy, outperforming other algorithms such as SVM (77.27%) and NB (68.18%) thus demonstrating the potential of ML methods to improve flows in WDNs.

Citation

CHAIMA CHABIRA , ,(2024-06-12), Effective Leak Detection in Water Distribution Networks using Pressure Transducers and Machine Learning,3rd International Conference on Frontiers in Academic Research ICFAR 2024,Turkey

Water is a precious, sometimes scarce resource that is crucial for all kinds of life. In the last two decades, the water demand has far exceeded the supply in many countries. It is also worth mentioning that distribution networks are constantly increasing. According to an international study 20-50% of the produced quantities are lost due to leaks. These leaks can cause significant economic losses and multiple water contaminations that are carried as a major health risk for the citizen. Therefore, network managers are always looking for fast and inexpensive harmless leak detection systems. The rapid detection of a leak in underground pipes is widely taken into account in the performance evaluation of water supply systems. In addition, detection methods should not interfere with the normal operation of water transport. To minimize the effect of false alarms (FA) that have a costly effect on infrastructure. These false alarms are produced by the use of the most widely used acoustic detectors in the world, which are usually based on the signal correlation technique to know the exact location of the leak in relation to one of the sensors. In this work, we present a new leak detector applied to a prototype pipe using highly sensitive pressure transmitters. For this purpose, we have applied the wavelet technique (DWT) for denoising our non-stationary signals from these transmitters. The STFT (Short time Fourier transform) will be used for the analysis of these non-stationary and non-linear signals coming from the leaks to know the exact position of the leak. Validation tests have proven the efficiency of our detector.

Citation

CHAIMA CHABIRA , ,(2023-07-15), DWT and STFT applied to Detect and locate leaks in WDNs,(hybrid ) International Conference on Nonlinear Science and Complexity (ICNSC23,) July 10-15, 2023, Istanbul-Turkey,Istanbul-Turkey

Water distribution networks around the world suffer from leaks. These are due to the movements of various agents such as ground movements, and vibrations caused by road traffic without forgetting the nature of the ground. For this, permanent control is needed. In this work, we have developed a LabVIEW interface for the acquisition and processing of signals received from pressure transmitters. A Wifi system for transmitting and receiving signals from the pressure transmitters is used. To analyze signals to remotely detect anomalies that may occur on the network. The work is divided into two parts: a practical or hard part and a soft part. The hard one is used to obtain the signals containing the leakage information. The soft part is based on the application of signal processing techniques for the detection and location of the exact position of the leakage and therefore minimize the amount of water that is lost.

Citation

CHAIMA CHABIRA , ,(2023-07-15), Design of leaks detection and localization in WDNs based on LabVIEW and pressure signals,(hybrid ) International Conference on Nonlinear Science and Complexity (ICNSC23,) July 10-15, 2023, Istanbul-Turkey,Istanbul-Turkey

The pipeline system is the most important part of the transportation of potable water. In other words, pipes are the most important component of the water distribution system. Because of several factors, they suffer from holes in which water is lost, and therefore an economic loss to states and humanity. Responsible for the water distribution networks always try to minimize the damage. By working to find the most effective way to detect leaks in these networks. This article centers on a mathematical model of a hydraulic system that aims to locate leakage in water transmission pipes. Several optimization techniques can be used. We opted for the BBO (biogeography-based optimization) method because of its performance, in particular the execution time of its algorithm. Using this method, two parameters of the model (friction and effective flow area factor) are determined to obtain the exact position of the leak.

Citation

CHAIMA CHABIRA , ,(2023-04-17), Parameterization and Validation of the Physical Coefficients of a WDNs by BBO,The 2022 International Conference of Advanced Technology in Electronic and Electrical Engineering (ICATEEE,M'sila (Algeria)

The design of leak detection systems that occur on water pipes is a priority area of applied research that has an economic and health impact on the future of any nation. The various control systems and tools that currently exist throughout the world are designed to ensure permanent and effective monitoring of natural resources that have become rare and precious. The determining factor for the choice of a good detector is the cost in the first place, flexibility, and speed of processing. In this work, the basic idea is to simultaneously acquire from a new inexpensive electronic device two signals from two pressure transmitters installed on a prototype pipe carried out at the laboratory. These signals are usually immersed in noise. For this, denoising by an appropriate digital filter is indispensable. In our case, the Savitzky-Golay filter (S.G) presents its efficiency. The denoising performances are obtained from the calculation of SNR. The denoised signals are analysed to confirm the presence of the leak in the case of its existence. Mathematical equations are applied to determine the exact position of the leak with regards to one of the sensors. Validation tests are required to determine the position of the leak when the difference time between the signals is known.

Citation

CHAIMA CHABIRA , ,(2023-04-17), S.G Filter And Speed of Pressure Wave Applied to locate leak in water pipe networks,The 2022 International Conference of Advanced Technology in Electronic and Electrical Engineering (ICATEEE,M'sila (Algeria)

Water distribution systems (WDNs) suffer from leakage problems. The latter can cause damage to infrastructure and also act on public health by the penetration of microbes of waterborne diseases transmissible through the orifices of leaks as soon as a drop in pressure occurs. For this purpose, the network managers always look for the best devices to detect and locate in time the anomaly of the networks as soon as it appears. To realize a system of detection and location of leakage a prototype circuit was realized equipped with pressure sensors and a system of acquisition based on a Digital Signal Processor (DSP). Pressure measurements for specific distances were carried out. A digital filter was applied for denoising. Pairs of pressure values before and after leakage are applied to a mathematical model for localization. We opted to use nonlinear regression (RNL) for the determination of non-measurable physical parameters based on pre-localized leak positions. Validation tests are performed to demonstrate the effectiveness of the model. In our study, a divergence of the model used was found.

Citation

CHAIMA CHABIRA , ,(2023-04-17), RNL used for the verification of the efficiency of a localization model in a real WDNs,The 2022 International Conference of Advanced Technology in Electronic and Electrical Engineering (ICATEEE,M'sila (Algeria)

Water distribution systems (WDNs) suffer from leakage problems. The latter can cause damage to infrastructure and also act on public health by the penetration of microbes of waterborne diseases transmissible through the orifices of leaks as soon as a drop in pressure occurs. For this purpose, the network managers always look for the best devices to detect and locate in time the anomaly of the networks as soon as it appears. To realize a system of detection and location of leakage a prototype circuit was realized equipped with pressure sensors and a system of acquisition based on a Digital Signal Processor (DSP). Pressure measurements for specific distances were carried out. A digital filter was applied for denoising. Pairs of pressure values before and after leakage are applied to a mathematical model for localization. We opted to use nonlinear regression (RNL) for the determination of non-measurable physical parameters based on pre-localized leak positions. Validation tests are performed to demonstrate the effectiveness of the model. In our study, a divergence of the model used was found.

Citation

CHAIMA CHABIRA , ,(2023-04-17), RNL used for the verification of the efficiency of a localization model in a real WDNs,Artificial intelligence in the service of society,M'sila (Algeria)

The function of the water supply is to supply good quality water with a sufficient quantity at an adequate pressure. distribution systems are degraded over time, which can lead to leaks. Efforts must then be made to minimise these leaks, which cause economic losses and public health risks as well as environmental problems. Water balances provide an overall picture of the quantities lost and identify areas of the system where significant leaks occur. Leak detection and its precise location are among the determining factors to reduce this economic loss and achieve good network performance, various techniques are implemented, the most commonly used is the acoustic correlation of two signals from two sensors usually installed on fire hydrants. The latter still need professional users for its localization, because the surrounding noise due to daily activities or road traffic can during their given correlation a peak simulating the existence of a leak.

Citation

CHAIMA CHABIRA , ,(2022), Characterization of signals provided from leak in water distribution network ,Journée doctorale en électronique 2022 (JDE'2022),M'sila (Algeria)