StockMarket Forecasting Using Hidden Markov Model: A New Approach
ISDA '05 Proceedings of the 5th International Conference on Intelligent Systems Design and Applications
A fusion model of HMM, ANN and GA for stock market forecasting
Expert Systems with Applications: An International Journal
A hidden Markov model-based algorithm for fault diagnosis withpartial and imperfect tests
IEEE Transactions on Systems, Man, and Cybernetics, Part C: Applications and Reviews
Hi-index | 0.00 |
The progressive development in high-voltage direct current (HVDC) transmission enhances the need to implement an efficient protection scheme to distinguish the minimum faulty part of the system and to relieve the stressed equipment. This paper proposes a new classification method based on Hidden Markov Models and wavelet transform to discriminate HVDC converter faults. In the proposed technique, probabilistic characteristics of signals discriminate fault signals without any deterministic index, so more flexible classification in different system conditions is achieved. Based on this method, high-speed protection decisions with small computational time could be performed in approximately 5 ms for severe system faults, and in 12.5 ms for faults that are restricted by protective control decision. PSCAD/EMTDC software simulations demonstrate suitable performance of this scheme for different fault types and system conditions in the International Council for Large Electric Systems (CIGRE) HVDC benchmark. All simulation results validate the stability and robustness of proposed scheme in different conditions, such as noisy systems.