Permanent magnetic synchronous motor (PMSM) drive is one of the very famous AC motor drives in the industry. This motor is being used in various platform because of its most advantageous compared to other AC drives like small in size, light weight, wide speed range, high power factor, high efficiency and large overload capacity. An approach for PMSM drive to control the drive using space vector pulse width modulation (SVPWM) technique is used based on the field orientation control (vector control). As the PMSM-drive is having characteristic of non-linearity, strong coupling and time-varying parameters which is not self-tuning for PI controller with the variable system states as PI controller comes with a large overshot, steady state error, torque ripples and slow response with variable states which has been overcome in this paper using FUZZY-PI controller. This paper proposes the Fuzzy PI controller with changing PI online accordingly with the system states, this is then compared with the standard PI controller. The simulation result has been showed for the speed and torque dynamics of the PMSM drive for comparison.
IoT -based applications have helped to popularize the Fog computing paradigm, which enables the seamless integration of mobile edge and cloud computing resources. The effective scheduling of application operations is difficult under these circumstances because of the restricted resource capabilities, IoT versatility qualities, asset heterogeneity, network ordered progression, and stochastic conduct. Thus, current heuristics and Reinforcement Learning-based approaches are inadequate for this issue. This kind of system only works well in centralized settings since it is incapable of dealing with variations in workload over time. When it comes to adapting quickly to dynamic situations, Offbeat Advantage-Actor-Critic (A3C) learning stands apart for its capacity to utilize the Residual Recurrent Neural Network (R2N2) to refresh model boundaries. This outcomes in an A3C-based scheduler for stochastic Edge Cloud frameworks, which permits a few specialists to advance simultaneously in a decentralized manner. We utilize the R2N2 engineering to store a wide number of host and assignment boundaries, as well as fleeting patterns, to settle on the most ideal booking choices. In order to meet the demands of the application, the proposed model may be adapted by changing different hyper-parameters. We may justify our decision to use hyper-parameters by doing a sensitivity analysis. Studies on true information show a lessening in energy utilization, reaction time, Service- Level-Agreement, and functional expenses of 14.4 %, 7.74 %, 31.9%, and 4.64 percent when contrasted with present status of-the-craftsmanship calculations. An Actor-Critical Advantage Actor-Critical Edge Computing Cloud Computing Deep Reinforcement Learning Task Scheduling Recurrent Neural Network.
Fish categorization and identification in the wild continue to be challenging because of a number of characteristics connected to manual labor, which leads to inferior or false reporting. Without direct interaction with humans, the camera is used by the proposed deep neural fish classification system to detect fish automatically. This research work has produced a You Only Look Once (YOLO) architecture-based method for classifying fish species. YOLO is a neural network-based object detection technique that will only be used once. Unlike the generic convolutional networks, which require thousands of networks and sufficiently long computing times to generate an image. In the first phase of development the proposed classification system focuses on the variation in fish scale and position and seeks out distinguishing traits to differentiate fish. The fishes are detected, positioned correctly, and fed to classifiers. Assuming that the sorts of fish captured are related to the surrounding environmental conditions, such as a boat, fishing gear, daylight, and weather, to infer the type of fish the second phase uses context information from the scenario is used. The final prediction is then estimated with the help of averaging and reweighting the two phases of the initial categorization predictions. It aids in determining the accuracy and producing suggestions tailored to particular species. Images of gilt head bream, trout, red sea bream, red mullet, sea bass, striped, red mullet, shrimp, horse mackerel, and black sea sprat from nine distinct classes are featured in the dataset utilized in this research. Underwater classification presents difficulties including background noises, image distortion, the existence of other water bodies in photos, image quality, and occlusion, the majority of algorithms now in use concentrate on classifying fishes outside of water. Compared to the previously suggested ways, the proposed strategy guarantees a significant improvement in discrimination accuracy. The proposed method outperforms the existing models by 6% and achieves an accuracy of 99.89%. This strongly implies that the approach would still work with a larger dataset.
Existing cellular communications and future communication networks requires very low latency, high reliability standards, increased capacity, enhanced security, and efficient user communication. The ability to accommodate several independent devices is a feature that mobile operators are seeking for a programmable solution, comparable functional networks technical foundation. Through the use of the Network Slicing concept, 5G networks enable end-to-end deployment of network resources (NS). Due to the surge in traffic and the acceleration of 5G network performance, emerging communication networks will demand data-driven strategic planning. This paper has to implement machine learning based network slicing algorithm to divide 5G network IoT devices into effective network slices such as eMBB, mMTC, URLLC for the traffic. The GNB and B-GNB algorithms are used to classify the usecase devices under the three network slices. This work developed bagging integrated with GNB algorithm and its performance metrics have been analysed. The B-GNB algorithm works well for prediction of best slice and strategic recommendations even there is network interruption, be able to predict the best network slice and implement strategic recommendations. The performance metrics such as sensitivity, F-score, precision and accuracy have also been analyzed. The comparative analysis shows B-GNB classify the slices with 86% of accuracy.
In the field of medicine, fuzzy set analysis is essential. Fuzzy matrices are used in a variety of models to address various challenging elements of medical diagnosis. Fuzzy matrices were used in the diagnostic modelsusing to show the relationship between symptoms and diseases in medicine. The most significant and rapidly expanding field of uses of uncertain set theory. The type of medical records and ambiguous data acquired through the usage of a fuzzy hexagonalmatrix. In this research, fuzzy matrices are defined, and an algorithm for medical diagnostic and treatment phases is devised. A fuzzy membership matrix is used to construct patient and drug information. The suggested algorithms are used to locate patients who have a disease.
