ABSTRACT
Shrink and swell is a phenomenon that causes transient variability in water level once boiler load variation occurs. The leading cause of the swell effect is the steam demand changes and the actual arrangement of steam generating tubes in the boiler. Steam bubbles beneath HRSG drum water make the level control very difficult, particularly with significant disturbances in the input heat to HRSG. Plant shutdown may occur in some situations, and combined cycle plant efficiency is diminished. The recently applied control methods in industry are single-element and three-element control with PID controllers, but these methods are not well suited for substantial load changes. The main aim of this paper is to investigate the shrink and swell phenomenon inside HRSG power plants. In addition to the existing PID loops, two different standalone controllers, namely, the FOPID controller and fuzzy controller, are implemented with the HRSG model. Besides, Artificial Bee Colony (ABC) algorithm is used to tune FOPID efficiently. Based on overshoot, rise time, ISE, IAE, ITAE as performance measures, the comparison has been held between the three controllers. Simulations show that how the ABC optimization algorithm is efficient with PID, FOPID. It turns out that the proposed method is capable of improving system responses compared to the conventional optimal controller.
ABSTRACT
COVID-19 (Coronavirus) went through a rapid escalation until it became a pandemic disease. The normal and manual medical infection discovery may take few days and therefore computer science engineers can share in the development of the automatic diagnosis for fast detection of that disease. The study suggests a hybrid COVID-19 framework (named HMB-HCF) based on deep learning (DL), genetic algorithm (GA), weighted sum (WS), and majority voting principles in nine phases. Its segmentation phase suggests a lung segmentation algorithm using X-Ray images (named HMB-LSAXI) for extracting lungs. Its classification phase is built from a hybrid convolutional neural network (CNN) architecture using an abstractly-designed CNN (named HMB1-COVID19) and transfer learning (TL) pre-trained models (VGG16, VGG19, ResNet50, ResNet101, Xception, DenseNet121, DenseNet169, MobileNet, and MobileNetV2). The hybrid CNN architecture is used for learning, classification, and parameters optimization while GA is used to optimize the hyperparameters. This hybrid working mechanism is combined in an overall algorithm named HMB-DLGA. The study experiments implemented the WS approach to evaluate the models' performance using the loss, accuracy, F1-score, precision, recall, and area under curve (AUC) metrics with different pre-defined ratios. A collected, combined, and unified X-Ray dataset from 8 different public datasets was used alongside the regularization, dropout, and data augmentation techniques to limit the overall overfitting. The applied experiments reported state-of-the-art metrics. VGG16 reported 100% WS metric (i.e., 0.0097, 99.78%, 0.9984, 99.89%, 99.78%, and 0.9996 for the loss, accuracy, F1, precision, recall, and AUC respectively) concerning the highest WS. It also reported a 99.92% WS metric (i.e., 0.0099, 99.84%, 0.9984, 99.84%, 99.84%, and 0.9996 for the loss, accuracy, F1, precision, recall, and AUC respectively) concerning the last reported WS result. HMB-HCF was validated on 13 different public datasets to verify its generalization. The best-achieved metrics were compared with 13 related studies. These extensive experiments' target was the applicability verification and generalization.
