Solar energy, in the form of photovoltaic (PV) panels, is important for achieving clean energy solutions. The photovoltaic health index must be monitored and improved because of the high demand for green energy. Unfortunately, defective solar cells are a significant source of performance degradation in photovoltaic (PV) systems. Experts often manually analyze electroluminescence (EL) images by visually inspecting them, which is personal, time-consuming, and requires extensive expertise. This work presents a comparative analysis of YOLOv8 and an Improved YOLOv5 for an automatic PV defect detection system in EL images in which Global Attention Module (GAM) is incorporated into the traditional YOLOv5s model for better object representation. Adaptive Feature space fusion (ASFF) was added to YOLOv5’s original structure for feature fusion. The Distance Intersection over Union (Non-Maximum) Suppression (DIoU-NMS) is aggregated to produce a more accurate bounding box. The ELDDS1400C5 dataset was used to train and evaluate the proposed system. Experiments on the ELDDS1400C5 test set revealed that the Improved YOLOv5 algorithm achieved a mean Average Precision of 76.3% (mAP@0.5), which is a 2.5% improvement over the standard YOLOv5 algorithm for detecting faults in PV modules in EL images. Furthermore, the experimental results demonstrated that Test Time Augmentation (TTA) significantly increased the mAP@0.5 to 77.7%, surpassing the YOLOv8 model, which achieved 77.5% under the same conditions.

Solar power is a clean and sustainable energy source that does not emit greenhouse gases or other atmospheric pollutants. The inherent variability in solar energy due to random fluctuations introduces novel attributes to the power generation and load dynamics of the grid. Consequently, there has been growing attention to developing an accurate forecast model using various machine and deep learning techniques. Temporal attention mechanisms enable the model to concentrate on the critical components of the input sequence at each time step, thereby enhancing the accuracy of the prediction. The suggested GRU–temporal fusion transformer (GRU-TFT) model was trained and validated employing the “Daily Power Production of Solar Panels” Kaggle dataset. Furthermore, an innovative loss function termed DILATE is introduced to train the proposed model specifically for multistep and nonstationary time series forecasting. The outcomes have been subjected to a comparative analysis with alternative algorithms, such as neural basis expansion analysis for interpretable time series (N-BEATS), neural hierarchical interpolation for time series (N-HiTS), and extreme gradient boosting (XGBoost), using several evaluation metrics, including the absolute percentage error (MAE), mean square error (MSE), and root mean square error (RMSE). The model presented in this study exhibited significant performance improvements compared with traditional statistical and machine learning techniques. This is evident from the achieved values of MAE, MSE, and RMSE, which were 1.19, 2.08, and 1.44, respectively. In contrast, the machine learning approach utilizing the Holt–Winters method for time series forecasting in additive mode yielded MAE, MSE, and RMSE scores of 4.126, 29.105, and 5.3949, respectively.

Proper surveillance and maintenance of photovoltaic (PV) systems are crucial to ensure continuous power generation and prevent operational downtimes. However, manual analysis of electroluminescence (EL) images is subjective, time-intensive, and requires significant expertise. To address this issue, a comprehensive deep learning architecture has been developed for the semantic segmentation of 29 different features and defects within EL images of PV panels. The SegNet architecture encoder has been replaced with the VGG16 encoder, which incorporates pre-trained weights to leverage transfer learning during the feature extraction stage. A Convolutional Block Attention Module (CBAM) block has also been introduced to enhance the decoder’s ability to generate fine-grained segmentations. Additionally, the suggested architecture has been evaluated through the application of three different loss functions: weighted categorical cross-entropy loss, categorical cross-entropy, and focal loss. The Attention-Based SegNet architecture proposed with a weighted categorical cross-entropy loss exhibits superior performance in terms of accuracy, F1 score, intersection over union (IoU), precision, recall, mean IoU (mIoU), specificity, Jaccard index, and Dice coefficient. It achieves a Dice coefficient of 0.9408 and an mIoU of 0.9101, outperforming the state-of-the-art SEiPV-Net trained on the same dataset by 8.77% and 4.97%, respectively.

This paper discusses the usage of wireless sensor networks in irrigation control by a smart watering system in which the irrigation process is controlled by valves. Water supplies became pollute and scarce. During the last years, the necessity to use water resources such as rivers, ground water and rain water efficiently has increased rapidly. Population growth, rapid urbanization and climate change have stressed water resources on the planet. The application of wireless soil-moisture sensor networks to detect water content in the soil can utilize water resources very efficiently. Water requirements depend on the type of plants and the soil as well as the season. A study has been made on the clay soil to observe its behavior and its different characteristics. By this study the time of excitation of the sensor could be known and the period of irrigation could be detected. This will be more efficient in terms of the time in which the sensor will be excited and the quantity of water that will be used.

Condition monitoring and fault diagnostics in rotorcraft have significant effect on improving safety level and reducing operational and maintenance costs. In this paper, a new method is proposed for fault detection and diagnoses of AH-64D (Apache helicopter) tail rotor drive-shaft problems. The proposed method depends on decomposing signal into different frequency ranges using mother wavelet. The most informative part of the vibration signal is then determined by calculating Shannon entropy of each part. Bispectrum is calculated for this part to investigate quadratic nonlinearities in this segment. Then, search algorithm is used to extract minimum number of indicative features from the bispectrum, which are then fed to classification algorithms. In order to quantitatively evaluate the proposed method, six classification algorithms are compared against each other such as fine K-nearest neighbor (KNN), cubic KNN, quadratic discriminant analysis, linear support vector machine (SVM), Gaussian SVM, and neural network. Comparison criteria include accuracy, precision, sensitivity, F score, true alarm, recall, and error classification accuracy (ECA). The proposed method is verified using real-world vibration data collected from a dedicated AH-64D helicopter tail rotor drive train (TRDT) research test bed. The proposed algorithm proves its ability in finding minimum number of indicative features and classifying the shaft faults with superior performance.

“deaf/dumb” and blind people face problems in communicating with others with difficulties in dealing with the communication technology. The goal of this paper is to design a desktop human computer interface application that is used to facilitate communication between normal, “deaf/dumb” and blind people. SVBiComm system helps blind person to hear voice saying the word gestured by the” deaf/dumb” while the deaf will receive a gesture representing the word said by the blind. SVBiComm works in two directions, the first direction is processing from video to speech. The animated word gestures are mapped with language knowledge base into text. Then, the relevant audio is generated using Text-to-Speech (TTS) API. The second direction is processing from speech to video. The voice from blind is converted into its corresponding text using Speech-to-Text (STT) API. Then, the natural language is mapped from the database to “deaf/dumb” in a relevant sign language form by using a 3D graphical model. The system was evaluated using a set of 113 sentences with 244 signs. In voice recognition; system recognized words with a percentage of 90% from 19 different persons. For Image recognition the system recognized images with a percentage of 84 % for 21 different persons. SVBiComm system provides many facilities with low cost that could be used in many areas.