لطفا منتظر بمانید ...
0% Complete
صفحه اصلی
/
یازدهمین کنفرانس بین المللی کنترل ، ابزار دقیق و اتوماسیون
Attention-Augmented BiLSTM with Ensemble Learning for Fault Diagnosis in Industrial Processes
نویسندگان :
Hamed Soltani
1
Behzad Moshiri
2
1- دانشگاه تهران
2- دانشگاه تهران
کلمات کلیدی :
Fault diagnosis،Recurrent neural networks
چکیده :
Fault detection and diagnosis (FDD) is a critical challenge due to the increasing complexity of systems and the prevalence of high-dimensional, complex industrial processes. This paper proposes a supervised deep learning framework that integrates an attention mechanism with a Bidirectional Long Short-Term Memory (BiLSTM) network To extract discriminative temporal features from multivariate process data. Finally, three classifiers—Support Vector Machine (SVM), Random Forest (RF), and XGBoost—were trained using the extracted features. To enhance classification performance, these classifiers were then combined using the stacking method, and Gradient Boosting was used as the meta-learner in this method. The Tennessee Eastman Process (TEP) dataset was used to evaluate the suggested approach. The result shows that the attention-augmented BiLSTM model outperforms standard BiLSTM in feature extraction and classification framework, achieves an average accuracy of 97% in detecting faults, and demonstrates strong robustness against sensor noise, maintaining high diagnostic performance even under 40% Gaussian noise levels.
لیست مقالات
لیست مقالات بایگانی شده
Swing-up Control of a Double Inverted Pendulum by Combination of Q-Learning and PID Algorithms
Ahmad Zeynivand - Hoda Moodi
The Effect of Observer Position on Networked Control Systems with Random Transmission Delays and Packet Dropouts
Elaheh Mohammadi-Asl - Farzad Hashemzadeh - Mahdi Baradarannia - Peyman Bagheri
A Typical Skill 4.0 Platform for Process Safety Management
Reyhane Mokhtarname - ALi Akbar Safavi - Leonhard Urbas - Fabienne Salimi - Hosein Taghipoor
A Hybrid Inverse Reinforcement Learning and Deep Deterministic Policy Gradient Framework for Model-Free Control of Robotic Manipulators
Samim Kamyab - Ghoncheh Zand - Masoud Shirzadeh
A Control-Oriented Index for Monitoring Sleep Apnea Severity Progression Using Longitudinal Oxygen Desaturation Changes
Elmira Aftabi - Shiva Naghsh - Ali akbar Safavi - Amir Sharafkhaneh
Frequency Regulation Improvement in AC Microgrids: A Fuzzy-Based Extended Virtual Synchronous Generator Control
Arman Jafari - Sharara Rehimi - Hassan Bevrani
Detection and Isolation of False Data Injection Attack in Load Frequency Controller of Power Systems
Seyed Mahdi Mirhadi - Aliakbar Ahmadi - Abolfazl Nateghi
Preassigned-Time Dynamic Surface Control for Nonlinear Systems with Input Disturbances
Mehran Derakhshannia - Yasaman Salmanpour - Seyyed Sajjad Moosapour
Environment Recognition for Controlling Lower-Limb Exoskeletons, by Computer Vision and Deep Learning Algorithm
Marzie Khalili - Sadjaad Ozgoli
A Model Reference Adaptive Control System for Phase- Locked Loop (PLL) in Grid-Connected Power Systems
Mostafa Khavaninzadeh - Ramin Nahavandi - Mjtaba Moeti
بیشتر
ثمین همایش، سامانه مدیریت کنفرانس ها و جشنواره ها - نگارش 44.9.5