Seminars and Events at automatic control
All seminars are held at the Department of Automatic Control, in the seminar room M:2112B on the second floor in the M-building, unless stated otherwise.
MSc, Basim Elessawi and Van Duy Dang: Improving Temperature Estimation Models using Machine Learning Techniques
Seminarium
From:
2024-06-03 14:00
to
15:00
Place: Seminar Room M 3170-73 at Dept. of Automatic Control, LTH
Contact: richard [dot] pates [at] control [dot] lth [dot] se
Date & Time: June 3rd, 14:00 - 15:00
Location: Seminar Room M 3170-73 at Dept. ofAutomatic Control, LTH
Author: Basim Elessawi and Van Duy Dang
Title: Improving Temperature Estimation Models using Machine Learning Techniques
Supervisor: Richard Pates
Examiner: Bo Bernhardsson
Abstract:
Temperature estimation models are crucial for various products manufactured by BorgWarner. These models often require manual calibration, where experts adjust parameters to ensure accuracy. However, this process can be slow and prone to errors. This thesis investigates how Machine Learning techniques can be used to improve accuracy and efficiency of temperature estimation models. Both black-box and grey-box approaches are used to evaluate the effectiveness of machine learning-based calibration. The black-box model employs techniques such as Decision Trees, Random Forests, and Neural Networks to predict temperature directly from raw input data, bypassing traditional temperature estimation processes. The grey-box model, on the other hand, uses Deep Q-learning to adjust the calibration automatically.
Results show that the black box model achieves better performance compared to conventional temperature estimation methods. Meanwhile, the grey-box model significantly improves accuracy and reduces the need for manual calibration in temperature estimation models.