Download as PDF, PPTX
Uploaded byYan Xu
Detecting anomalies on rotating equipment using Deep Stacked Autoencoders - by by Ziad Katrib
The document discusses the use of deep stacked autoencoders for anomaly detection in rotating equipment, highlighting the shortcomings of existing frameworks, specifically a third-party system used by Calpine. It emphasizes a combined approach using deep learning and classical machine learning algorithms to improve anomaly detection and fault classification. Additionally, it outlines a framework for equipment monitoring that includes data collection, model architecture, and continuous performance optimization.
More Related Content
Similar to Detecting anomalies on rotating equipment using Deep Stacked Autoencoders - by by Ziad Katrib
Detecting anomalies on rotating equipment using Deep Stacked Autoencoders - by by Ziad Katrib
- 1. Deep Stacked Autoencoders ForAnomaly Detection Deep learning workshop Ziad Katrib
- 2. How important isequipment prognostics • 80+ Plants, several thousand rotating equipment, translating to several thousand potential failures – Better monitor thy equipment • Since 2013 Calpine has been using a 3rd party Anomaly Detection framework based on NASA’s ORCA algorithm – Blind spots – No Fault Detection or Remaining Useful to speak of • Improving on existing methods through a combination of DL and classical ML algorithms
- 3. Anomaly detection vsFault Classification Unknown unknowns Unknown knowns Anomaly Detection, 2018 PHM Society Annual Conference Tutoria, Neil Eklund
- 4. Anomaly detection incontext Maslow Pyramid of needs Sensor Data Collection Maintenance Labeling Remaining Useful Life Anomaly Detection Fault Detection Prognostics
- 5.
- 6. Framework overview Design Build PublishModel architecture Get New Data QC Data (Prep Training Data) Deploy Model Read Model Retrieve / Receive New Data Score & Alert / Store Data Science Team Analysts
- 7. Framework overview: AnalystPerspective
- 8. http://wasp-sweden.org/custom/uploads/2017/08/ViditSaxena-DenoisingAutoencoder-20170811.pd KNN Regressor Modelvs actual threshold Existing Algorithm New supplemental Algorithm based on Autoencoders Model vs actual threshold Isolation Forest Normal/ Abnormal Improving existing Anomaly Detection Methodology Reconstruction Error
- 9. Hands on MachineLearning with Scikit-Learn and TensorFlow Anomaly Detection and Fault Disambiguation in Large Flight Data: A Multi-modal Deep Auto-encoder Approach Kishore K. Reddy, Soumalya Sarkar, Vivek Venugopalan, Michael Giering Denoised Stacked Autoencoders Reconstruction Error Why Autoencoders: - Semi Supervised - Easy to train - Can use on multi modal and heterogeneous without feature engineering
- 10. Isolation Forest forone class classification(OCC) Isolation Forest Reconstruction Error
- 11. Isolation Forest Normal/ Abnormal Optimizing with OCC Reconstruction Error Hyperparemetertuning is done with entire pipeline in mind, i.e. optimizing for False Positives, True positives and not just Autoencoder re-construction error
- 12.