Search

The increased addition of DERs, their intermittent nature, and the high cost of ESSs are the key hurdles for interconnected multi-microgrids to operate economically. It is important to have a system that is simple to adopt, computationally inexpensive, does not share private data. In this paper, an energy management scheme is proposed for multiple interconnected renewable energy resources within the multi-microgrid paradigm. The core objectives aimed at minimizing the operational cost of each microgrid and utilize RER most economically without any load interruptions. The proposed scheme is performed in two steps. In the first step, the load is shifted in accordance with the daily price curve via resource scheduling i.e., charging and discharging of battery storage, day-ahead forecast of RER generation, and respective loads. In the second stage, the cost of energy is further reduced using opportunistic trading with neighboring microgrids in real-time. The mismatch between forecasted generation and load is also adjusted via either energy trading or with efficient scheduling of available power that is stored during off-peak hours. The effectiveness of the proposed scheme is applied to a test system consisting of four interconnected microgrids. Results indicate that daily average power imported from the utility grid and peak demand is changed from 103 kW to 162 kW, and from 661 kW to 527 kW, respectively. Likewise, as per comparison, significantly favorable results have been achieved in terms of efficient available resources utilization, market trading, and reduction of the respective bills.

Open-domain question answering aims to get answers for given questions from a set of documents. Recently, dual encoder architecture is widely adopted to dense passage retrieval for question answering. In-batch negative sampling is typically used to gather extra negative samples during training. In this paper, we propose adaptive batch scheduling to enhance the performance of in-batch negative sampling. The proposed algorithm schedules training batches to increase the difficulty of the sampled negatives by in-batch negative sampling during training. We evaluated the proposed approach on the two well-known document retrieval benchmark datasets MSMARCO and Natural Questions. The evaluation result shows that the proposed adaptive batch scheduling could significantly improve the document retrieval performances of dual encoder architecture document retrieval systems.

Shape from Focus (SFF) has been studied extensively in computer vision for 3D shape and depth recovery. The first stage in SFF methods is to compute the focus value of every pixel by converting the colored images into gray scale and then apply the focus measure operator. Converting colored values in the images into gray scale values may lead to imprecise mapping of pixels with different colored values onto the same gray scale value, this affects the overall accuracy of the system. In a colored image, the focused pixels maintain a considerable color difference from their neighboring pixels as compared to the defocused ones, which are blended into their neighborhood. This article presents an alternative method to measure the degree of focus by directly processing colored images. The color differences of the neighbor pixels with respect to the central pixel are obtained and summed together, this is followed by calculating their spread. The sum and the spread are combined to measure the degree of focus of the pixel in consideration. The proposed focus measure is then used for shape recovery of various simulated and real objects and is compared with previous techniques. The comparison results show the proposed method has the highest correlation and smallest RMSE values confirming the effectiveness of using color images for shape recovery.

Convolutional Neural Networks (CNN) have been very successful in classification and object recognition. A lot of related work has been done to modify the performance of the networks, but they could either perform better with accuracy at high cost or decrease the time taken by a model on training datasets. We propose a deep neural network model ’Multi-Blocked Model’ which tends to decrease this gap and is efficient and accurate with less convolutional layers, easy to deploy online or embedded systems. We choose three datasets that are publicly available and popular for their own uniqueness among the datasets in deep neural networks. These three diverse datasets are: Modified National Institute of Standards and Technology (MNIST), Street View House Number (SVHN) and the Canadian Institute for Advanced Research with 10 Cases (CIFAR-10). Our stateof-the-art Multi-Blocked model is presented well on all three data sets. Dropout is added to overcome the overfitting problem. The multi-blocked model is designed in a way that it uses a minimum number of parameters so that it is able to run on a Graphical Processing Unit (GPU), which requires less power. The experimental results show that our proposed Multi-Blocked model tends to achieve the accuracy of these datasets by 99.40%, 90.8%, 88.07% consuming under 2 GB of graphical memory.

Smart community setups nowadays are subjected to complicated issues such as instability, intermittent integration of the load at the demand side, and lack of intelligent two-way communication process. These issues need to be addressed in terms of a balanced power demand dispatch (DD) in the real-time or day-ahead duplex signal regime under multi-microgrids. This paper offers an intelligent multi-agent-based approach that works between different levels of communication and their respective layers for a community-based system to optimize the power in community-based multi-microgrids model. This will further enhance user personal comfort. Constraints relative to cost minimization also have a relation with this model. A three-level structure with various layers of autonomous agents take intelligent decisions based on prioritized particle swarm optimization (P-PSO), prioritized plug and play (PPnP), and knapsack; considering DD as the main driver of the system to address objectives like price and power consumption uncertainties. Distinct smart home models, depending upon their living habits, are keenly observed providing their power infrastructure and personal comfort. Load appliances considered as load agents are individually contemplated for maximum proficiency. Furthermore, two-way communication between utility and consumers lowers down the risk of the inefficiency of the system.

In the pharmaceutical field, early detection of lung nodules is indispensable for increasing patient survival. We can enhance the quality of the medical images by intensifying the radiation dose. High radiation dose provokes cancer, which forces experts to use limited radiation. Using abrupt radiation generates noise in CT scans. We propose an optimal Convolutional Neural Network model in which Gaussian noise is removed for better classification and increased training accuracy. Experimental demonstration on the LUNA16 dataset of size 160 GB shows that our proposed method exhibit superior results. Classification accuracy, specificity, sensitivity, Precision, Recall, F1 measurement, and area under the ROC curve (AUC) of the model performance are taken as evaluation metrics. We conducted a performance comparison of our proposed model on numerous platforms, like Apache Spark, GPU, and CPU, to depreciate the training time without compromising the accuracy percentage. Our results show that Apache Spark, integrated with a deep learning framework, is suitable for parallel training computation with high accuracy.

The integration of commercial onshore large-scale wind farms into a national grid comes with several technical issues that predominately ensure power quality in accordance with respective grid codes. The resulting impacts are complemented with the absorption of larger amounts of reactive power by wind generators. In addition, seasonal variations and inter-farm wake effects further deteriorate the overall system performance and restrict the optimal use of available wind resources. This paper presented an assessment framework to address the power quality issues that have arisen after integrating large-scale wind farms into weak transmission grids, especially considering inter-farm wake effect, seasonal variations, reactive power depletion, and compensation with a variety of voltage-ampere reactive (Var) devices. Herein, we also proposed a recovery of significant active power deficits caused by the wake effect via increasing hub height of wind turbines. For large-scale wind energy penetration, a real case study was considered for three wind farms with a cumulative capacity of 154.4 MW integrated at a Nooriabad Grid in Pakistan to analyze their overall impacts. An actual test system was modeled in MATLAB Simulink for a composite analysis. Simulations were performed for various scenarios to consider wind intermittency, seasonal variations across four seasons, and wake effect. The capacitor banks and various flexible alternating current transmission systems (FACTS) devices were employed for a comparative analysis with and without considering the inter-farm wake effect. The power system parameters along with active and reactive power deficits were considered for comprehensive analysis. Unified power flow controller (UPFC) was found to be the best compensation device through comparative analysis, as it maintained voltage at nearly 1.002 pu, suppressed frequency transient in a range of 49.88–50.17 Hz, and avoided any resonance while maintaining power factors in an allowable range. Moreover, it also enhanced the power handling capability of the power system. The 20 m increase in hub height assisted the recovery of the active power deficit to 48%, which thus minimized the influence of the wake effect.