الفهرس 
chapter 1Only 14 pages are availabe for public view
 |  | from | 102 |  |  |
| | | from | 102 | | |
Abstract
Power quality has become a critical issue due to the increased consumption of electrical load and the increased use of sensitive devices connected to power systems. Nonetheless, the complexity of modern life and the rising use of semiconductors make a serious risk to power quality. The modern power supply, which is based on developing renewable sources such as solar, wind, and nuclear energy, has increased power quality disturbances to huge levels. Power quality disruptions must be detected and identified correctly and precisely to ensure reliability and maintain power quality. Thus, Detection algorithms aid decision-makers in resolving and mitigating disturbances, thereby protecting the power network from significant economic loss.
The economic and social effects of electric disruptions have increased. They can cause current and voltage interferences and result in disasters. Therefore, an online assessment of transit stability is essential for the control of power systems, which can prevent transmission of power disturbances and enable operators to decide on emergency control measures. This thesis proposes a power-disturbance-detection system to analyze online voltage signals and prevent the transmission of disturbance signals. The system consists of three stages. The first stage is data augmentation, which adds noisy signals and more generated signals with multiple events with different magnitudes of disturbance time and locations. We used data augmentation to make it seem like real data. The second stage is stage constructs convolutional neural network (CNN) architecture which consists of six convolutional layers, max-pooling layer, one flatten layer, and three dense layers. Then trained the model it. The third stage develops the proposed CNN model into an electronic circuit to analyze the online voltage signal and prevent transmission of the disturbance signal by cutting the power. To investigate the suggested stages of the detection system, several experiments are conducted. The experimental results demonstrate that the proposed data augmentation stage provides a beneficial influence, and the proposed CNN model is architecturally efficient. It is not affected by the noise, period of modifications, or displacements of the disturbance signals in the real-time detection process. In addition, it obtained a clear improvement rate in the weighted accuracy metrics over the state-of-the-art algorithms.