Sustainable Energy and Artificial Intelligence

Abstract This paper presents an insulin dosing policy for individuals with Type 1 Diabetes (T1D), utilizing Linear Matrix Inequality (LMI) techniques in combination with the TakagiSugeno (TS) fuzzy approximator. The primary goal is to regulate blood glucose levels by employing robust control strategies for the nonlinear dynamics of the glucose-insulin system, which are described using the Bergman Minimal Model. The proposed approach systematically incorporates insulin pump constraints, such as maximum insulin delivery rates, to ensure practical applicability in real-world scenarios. Simulation results demonstrate that the proposed controller maintains blood glucose levels within a safe range for over 84% of the time, with average glucose levels reduced to as low as 95mg/dL under the least restrictive input constraints. Furthermore, the controller effectively mitigates meal-induced disturbances while minimizing hypoglycemia risks, demonstrating its robustness under varying parameter uncertainties. This research highlights the potential of the proposed method for use in closed-loop insulin delivery systems, offering a promising solution for personalized and adaptive diabetes management.

Abstract A significant portion of electrical energy in power grids is wasted in distribution systems. Distribution systems typically have radially shaped feeders. Today, increased demand resulted in the expansion of distribution systems and their dimensions, which in turn causes greater voltage drop, increased losses, and consequently reduced stability, decreased node voltage, and load imbalance. Nowadays, using modern methods and employing power electronics devices such as flexible alternating current transmission system (FACTS) devices can enhance the quality of electrical power. Additionally, considering the global warming, most power generation companies are inclined towards renewable energies such as photovoltaic panels. One of the suitable FACTS devices used in the PV distribution system is PV-DSTATCOM. These devices are based on reactive power control and use a photovoltaic (PV) system to supply their required energy. Therefore, they should be installed in a way that coordinates with capacitor banks installed in the distribution network and improves power quality parameters, including reduced network losses, improved network performance, deferred investment, increased reliability, and enhanced power quality. In this paper, the problem of locating and sizing of PV-DSTATCOM and shunt supercapacitors is solved based on a simultaneous multi-objective manner, with the objectives focused on power and energy losses, voltage profile, and voltage stability. To solve this multi-objective problem, the Fuzzy-ALPSO algorithm is adopted and implemented on standard IEEE 33- and 69-bus systems. 

Abstract Iris recognition is a widely used biometric technology in various applications. Deep learning methods, particularly convolutional neural networks (CNNs), have been popular in biometric detection due to their ability to generalize well and operate without human intervention. However, CNNs often struggle with image noise and require large datasets for effective training. To address these limitations, Capsule Networks (CapsuleNet) have been introduced, offering improved performance on small and noisy datasets. This paper presents a CapsuleNet architecture with an enhanced routing algorithm tailored for iris recognition. To further refine the learning process, VGG16 and InceptionV3 models are integrated into the CapsuleNet, enabling it to learn effectively from a limited number of samples. The proposed network is structured into a series of subnets, corresponding to its main building blocks. Experiments conducted on the CASIA-V4 Lamp iris dataset demonstrate that the optimized CapsuleNet architecture outperforms traditional networks, offering greater stability and robustness for iris recognitions. Results on the CASIA-V4 dataset show a 6% improvement in accuracy.

Abstract This paper proposes a new active impedance source inverter topology. The proposed inverter provides a continuous source current, appropriate boost-factor, a few numbers of components, low voltage stress on active switch and diodes, low voltage stress across capacitors, and low volume and size. All operating modes of the proposed inverter are analyzed in details, and the design considerations of active and passive components are done. Also, a control method to generate the desired output waveform is proposed. In the following, the comprehensive comparison between the proposed topology and some conventional topologies from different points of view, such as the boost-factor, total voltages across capacitors, total blocked voltages across impedance side diodes, active switch, and volume of passive components are calculated. Ultimately, the simulation results are obtained from a prototype in the PSCAD software. The obtained results from simulation part validate the correctness of operation and performance of the proposed topology as well the given in theoretical part.

Abstract Electric motors are vital in energy systems for converting energy efficiently. Their reliability ensures consistent performance, making them indispensable in renewable energy applications and industrial processes. Bearings are crucial for motor operation, yet they are vulnerable to faults that can impair performance and reduce lifespan. Conventional fault detection methods require high sampling rate vibration data and expensive sensors. This paper presents a cost-effective solution by introducing a hybrid model combining Multiscale Convolutional Neural Networks (MSCNN) with Long Short-Term Memory (LSTM) networks for bearing fault diagnosis using low sampling rate data. We developed dedicated IoT-based hardware and a web server for real-time monitoring. Our approach leverages MSCNN for spatial feature extraction and LSTM for temporal pattern recognition, achieving high diagnostic accuracy with lower resolution data. Experimental results show that our MSCNN-LSTM model provides diagnostic accuracy comparable to or surpassing that of high sampling rate methods, offering a robust and economical solution for bearing fault detection.

Abstract Every year, many deaths occur due to various strains of viral diseases, especially during the Arbaeen procession, numerous crowds in blessed places, and the beginning of the academic year of schools and universities in the country. The spread of compliance with health protocols was considered the best solution to stop this disease, and each individual's commitment to these protocols played a crucial role. Since manually monitoring compliance with health protocols is time-consuming, laborious, and error-prone, using an intelligent monitoring system to check people's mask coverage and identify symptomatic people regardless of quarantine regulations is strongly required in public environments. This article proposes an automatic hardware/software system based on artificial intelligence to identify people's mask cover and measure body temperature, which performs face recognition, mask cover detection and body temperature measurement, respectively, using the Viola Jones algorithm, convolutional neural network and temperature sensor. The Viola Jones converts a gray-level image to face area detected of the image. CNN model that optimized through transfer learning for classifying images is the ResNet type. If a person has not used a mask or the person’s body temperature is higher than 37.5 degrees, the system will issue a warning. The proposed model was obtained accuracies of 99% and 98% for the training and validation sets at Epoch 18. In the field evaluation, this system was able to achieve 96% accuracy in face recognition and 100% accuracy in mask cover detection. This smart system can be used to monitor compliance with health protocols in public centers.