Islanding detection using DWT and ANN.pptx
- 1. Under the guidance of Mrs. MAMUN MISHRA Assistant Professor Presented by:- FLEVIE PATTANAIK(2002050086) ROXY RANJAN MALLCK(2002050101) SUPRIYA SAHU(2002050115) BAKI TIRUPATHI(2004050001) SUSMITA DASH(2002050134) SUBHAM S. PANDA(2002050124) ANSHUMAN BEHERA(2004050015) 1 Presentation on ISLANDING DETECTION APPROACH BASED ON DISCRETE WAVELET TRANSFORM AND K-NEAREST NEIGHBORS
- 2. PRESENTATION OUTLINE Section-I Introduction Section-II Motivation & Objective Section-III Problem formulation Section-IV Simulation & Results analysis Section- V Conclusion Section-VI Future Scope Section-VII References 2
- 3. INTRODUCTION Islanding detection is vital for grid safety. 3 Islanding refers to the situation where a portion of the power system becomes electrically isolated from the main grid while still generating and consuming power locally. GOALS EMPHASIS VALIDATION CONTRIBUTION PROJECT FOCUS
- 4. 4 TYPES OF ISLANDING Passive: Monitors grid passively for fault. Hybrid: Combination of methods. Remote: Utilizes external communication links . Active: Actively injects signals to detect faults.
- 5. 1. Develop an effective method for detecting islanding events in power systems. 2. Enhance detection accuracy and efficiency. 3. Minimize false alarms and reduce the non-detection zone. 4. Improve grid stability and safety through timely detection of islanding events OBJECTIVE 5 Why we have chosen hybrid islanding? MOTIVATION
- 6. 6 WHY DISCRETE WAVELET TRANSFORM ? PROBLEM FORMULATION ROCOF ROCOV Transient Representation Noise Robustness Real Time Implementation Adaptability Sparse Representation
- 7. 7 Localized transient Efficiency AND Simplicity Piecewise Constant Signal Benefits of Artificial intellegence: Accuracy Handle noisy data Offer Versatility and Adaptability What are the Specific strategies to minimize non-detection zones - Advantages of Haar Function:
- 8. 8 Figure 1. Schematic Diagram of Test system TEST SYSTEM
- 9. 9 TABLE 1: VALUES OF FREQUENCY TABLE 2: VALUES OF VOLTAGES
- 10. 10 Figure 2: Confusion matrix of frequency Figure 3: KNN Model for frequency Simulation & Results analysis
- 11. 11 Figure 4:Confusion matrix of voltage Figure 5: KNN Model of Voltage
- 12. CONCLUSION 12 1. Investigated islanding detection in power grids with distributed generation. 2. Combined Discrete Wavelet Transform (DWT) for feature extraction and K- Nearest Neighbors (KNN) for classification. 3. DWT effectively captured transient characteristics of islanding events. 4. KNN classifier distinguished between grid-connected and islanding scenarios. 5. Offers promising solution but needs optimization in wavelet functions, decomposition levels, and KNN parameters. 6. Fine-tuning classification threshold crucial for minimizing false alarms and ensuring reliable detection.
- 13. FUTURE SCOPE 13 1. Real-time Implementation: Deploy DWT-KNN on embedded systems for real-time data processing. 2. Advanced Techniques: Explore CNNs and other ML models for better feature extraction and classification. 3. Multi-layer Detection: Extend islanding detection to multi-layer grids with communication between devices. 4. Model Robustness: Address challenges like noise and data imbalances, using strategies like data augmentation and robust ML models.
- 14. REFERENCE 14 1. Hashemi, F., and Mohammadi, M. "Islanding detection approach with negligible non-detection zone based on feature extraction discrete wavelet transform and artificial neural network" In proceedings 2016 of Int. Trans. Electr. Energ. Syst., 26: 2172–2192. doi: 10.1002/etep.2197. 2. M. S. ElNozahy, E. F. El-Saadany and M. M. A. Salama, "A robust wavelet-ANN based technique for islanding detection," In proceedings 2011 of IEEE Power and Energy Society General Meeting, Detroit MI, USA, 2011, pp. 1-8, doi: 10.1109/PES.2011.6039158. 3. Kazemi Karegar H, Sobhani B. "Wavelet transform method for islanding detection of wind turbines", In proceedings 2012 of Renewable Energy; 38(1) , doi:10.1016/j.renene.2011.07.002 4. Y. Lin and J. Wang, "Research on text classification based on SVM-KNN," In proceedings (2014) of IEEE 5th International Conference on Software Engineering and Service Science, Beijing, China, 2014, pp. 842-844, doi: 10.1109/ICSESS.2014.6933697. 5. S. Kumari, B. Bhalja and R. P. Maheshwari, "A new SVM based islanding detection scheme in the presence of distributed generation," In proceeding 2017 of 7th International Conference on Power Systems (ICPS), Pune, India, 2017, pp. 352-357, doi: 10.1109/ICPES.2017.8387319. 6. T. -t. Dai and Y. -s. Dong, "Introduction of SVM Related Theory and Its Application Research," In proceedings of 3rd International Conference on Advanced Electronic Materials, Computers and Software Engineering (AEMCSE), Shenzhen, China, 2020, pp. 230-233, doi: 10.1109/AEM- CSE50948.2020.00056. 7. K. Taunk, S. De, S. Verma and A. Swetapadma, "A Brief Review of Nearest Neighbor Algorithm for Learning and Classification," In proceedings 2019 of International Conference on Intelligent Computing and Control Systems (ICCS), Madurai, India, 2019, pp. 1255-1260, doi: 10.1109/ICCS45141.2019.9065747. 8. Zhang Hao, Berg Alexander, Maire Michael, Malik Jitendra, "SVM-KNN: Discriminative Nearest Neighbor Classification for Visual Category Recognition." In proceedings 2006 of IEEE Computer Society Conference on Computer Vision and Pattern Recognition. 2. 2126-2136. 10.1109/CVPR.2006.301.
- 15. THANK YOU 15