Visible to the public Biblio

Filters: Keyword is event-based  [Clear All Filters]
2023-07-11
Sari, Indah Permata, Nahor, Kevin Marojahan Banjar, Hariyanto, Nanang.  2022.  Dynamic Security Level Assessment of Special Protection System (SPS) Using Fuzzy Techniques. 2022 International Seminar on Intelligent Technology and Its Applications (ISITIA). :377—382.
This study will be focused on efforts to increase the reliability of the Bangka Electricity System by designing the interconnection of the Bangka system with another system that is stronger and has a better energy mix, the Sumatra System. The novelty element in this research is the design of system protection using Special Protection System (SPS) as well as a different assessment method using the Fuzzy Technique This research will analyze the implementation of the SPS event-based and parameter-based as a new defense scheme by taking corrective actions to keep the system stable and reliable. These actions include tripping generators, loads, and reconfiguring the system automatically and quickly. The performance of this SPS will be tested on 10 contingency events with four different load profiles and the system response will be observed in terms of frequency stability, voltage, and rotor angle. From the research results, it can be concluded that the SPS performance on the Bangka-Sumatra Interconnection System has a better and more effective performance than the existing defense scheme, as evidenced by the results of dynamic security assessment (DSA) testing using Fuzzy Techniques.
2022-06-06
Jobst, Matthias, Liu, Chen, Partzsch, Johannes, Yan, Yexin, Kappel, David, Gonzalez, Hector A., Ji, Yue, Vogginger, Bernhard, Mayr, Christian.  2020.  Event-based Neural Network for ECG Classification with Delta Encoding and Early Stopping. 2020 6th International Conference on Event-Based Control, Communication, and Signal Processing (EBCCSP). :1–4.
We present a scalable architecture based on a trained filter bank for input pre-processing and a recurrent neural network (RNN) for the detection of atrial fibrillation in electrocardiogram (ECG) signals, with the focus on enabling a very efficient hardware implementation as application-specific integrated circuit (ASIC). Our already very efficient base architecture is further improved by replacing the RNN with a delta-encoded gated recurrent unit (GRU) and adding a confidence measure (CM) for terminating the computation as early as possible. With these optimizations, we demonstrate a reduction of the processing load of 58 % on an internal dataset while still achieving near state-of-the-art classification results on the Physionet ECG dataset with only 1202 parameters.