ISLAB/CAISR - User contributions [en]

Smart Alarm

2020-10-12T11:47:57Z

Mahmoud:

{{StudentProjectTemplate
|Summary=Data-driven alarm prediction using sensor data
|Supervisor=Hadi Fanaee-T(www.fanaee.com), Mahmoud Rahat
|Level=Master
|Status=Open
}}
This is a fantastic opportunity to work with Alfa Laval, a world's leader and pioneer in producing separator machines. This project aims to investigate the application of Machine Learning for analysis of onboard sensor data from separator machines. The separators purify oil and water supplies onboard marine vessels.

The main objective of this project is to predict and analyze alarms of separator machines. The automation software of machines produces faults and warming messages during its operation. Currently, these alarms are produced based on fixed, predetermined thresholds. It is interesting to explore timeseries forecasting methods and machine learning models to predict alarms beforehand. The benefits of more in-depth exploration are both in terms of technical and business value, including among the others: property damage control, oil Loss reduction, overall machine health, and fuel quality control.

Predicting the status of machines with vibration data

2020-10-12T11:47:36Z

Mahmoud:

{{StudentProjectTemplate
|Summary=Predicting the status of Alfa Laval's separator machines with vibration data
|Keywords=machine learning, predictive maintenance, time series analysis, vibration analysis
|Supervisor=Hadi Fanaee-T (www.fanaee.com), Mahmoud Rahat
|Level=Master
|Status=Open
}}
This is a fantastic opportunity to work with Alfa Laval, a world's leader and pioneer in producing separator machines. They are collecting large amount of data on vibration of rotating parts. The idea is to use the vibration data to predict the status of the machines. If the status of the machine is known, it can be predicted when next service is needed and if parts needs to be exchanged. The condition of the machine is very important to the customer to make sure there will be no unplanned stop in the production. The machines are used in important processes in various industries as food, cleaning of water, refineries and pharma. The application is used by Alfa Laval to make reports that are sent to the customer on a regular basis. There are hundreds of machines worldwide collecting data for reporting. The machine has five measurement points and each measurement point has a number of frequency ranges to be measured.

Mahmoud Rahat

2020-10-12T11:46:10Z

Mahmoud:

{{Person
|Family Name=Rahat
|Given Name=Mahmoud
|Title=PhD
|Cell Phone=+46721686494
|Position=Postdoctoral Researcher
|Email=mahmoud.rahat@hh.se
|Office=E510
}}

Mahmoud Rahat

2020-10-12T11:42:00Z

Mahmoud:

{{Person
|Family Name=Rahat
|Given Name=Mahmoud
|Title=PhD
|Cell Phone=+46721686494
|Position=Postdoctoral Researcher
|Email=mahmoud.rahat@hh.se
|Image=mahmoud.rahat.jpg
|Office=E510
}}
{{AssignProjects
|project=CAISR+
}}
{{AssignProjects
|project=HEALTH
}}
{{AssignSubjectAreas
|SubjectArea=Machine Learning
}}
{{AssignSubjectAreas
|SubjectArea=Data Mining
}}
{{AssignApplicationAreas
|ApplicationArea=Intelligent Vehicles
}}
[[Category:Staff]]

Mahmoud Rahat

2020-10-12T11:41:23Z

Mahmoud: Created page with "Mahmoud Rahat Artificial Intelligence Ph.D. in Artificial Intelligence from National University of Iran Postdoctoral Researcher at Halmstad University, Sweden He joined the Ce..."

Mahmoud Rahat
Artificial Intelligence
Ph.D. in Artificial Intelligence from National University of Iran
Postdoctoral Researcher at Halmstad University, Sweden
He joined the Center for Applied Intelligent Systems Research (CAISR) in 2019 as a postdoctoral
researcher. Before joining Halmstad University, he has been a researcher at Northern Illinois
University. Mahmoud has worked in a variety of different research areas within the field of
Artificial Intelligence. His research interests include Machine Learning, Deep Learning,
Predictive Maintenance, Natural Language Processing, Computer Vision, and Robotics.

{{Person
|Family Name=Rahat
|Given Name=Mahmoud
|Title=PhD
|Cell Phone=+46721686494
|Position=Postdoctoral Researcher
|Email=mahmoud.rahat@hh.se
|Image=mahmoud.rahat.jpg
|Office=E510
}}
{{AssignProjects
|project=CAISR+
}}
{{AssignProjects
|project=HEALTH
}}
{{AssignSubjectAreas
|SubjectArea=Machine Learning
}}
{{AssignSubjectAreas
|SubjectArea=Data Mining
}}
{{AssignApplicationAreas
|ApplicationArea=Intelligent Vehicles
}}
[[Category:Staff]]

Predicting the status of machines with vibration data

2020-10-12T11:34:40Z

Mahmoud:

{{StudentProjectTemplate
|Summary=Predicting the status of Alfa Laval's separator machines with vibration data
|Keywords=machine learning, predictive maintenance, time series analysis, vibration analysis
|Supervisor=Hadi Fanaee-T (www.fanaee.com) and Mahmoud Rahat
|Level=Master
|Status=Open
}}
This is a fantastic opportunity to work with Alfa Laval, a world's leader and pioneer in producing separator machines. They are collecting large amount of data on vibration of rotating parts. The idea is to use the vibration data to predict the status of the machines. If the status of the machine is known, it can be predicted when next service is needed and if parts needs to be exchanged. The condition of the machine is very important to the customer to make sure there will be no unplanned stop in the production. The machines are used in important processes in various industries as food, cleaning of water, refineries and pharma. The application is used by Alfa Laval to make reports that are sent to the customer on a regular basis. There are hundreds of machines worldwide collecting data for reporting. The machine has five measurement points and each measurement point has a number of frequency ranges to be measured.

Automatic Generation of Descriptive Features for Predicting Vehicle Faults

2020-03-02T12:23:45Z

Mahmoud:

{{StudentProjectTemplate
|Summary=Automatic Generation of Descriptive Features for Predicting Vehicle Faults
|Programme=Embedded and Intelligent Systems
|TimeFrame=Fall 2019
|Prerequisites=Artificial Intelligence, Learning Systems, Data Mining
|Supervisor=Mahmoud Rahat, Reza Khosh
|Level=Master
|Status=Ongoing
|Author=Vandan Revanur, Ayodeji Olanrewaju Ayibiowu
}}
Whenever a Volvo truck visits an authorised workshop, aggregated vehicle statistics (parameters) such as average vehicle speed, total fuel consumption, etc. are stored in a database called Logged Vehicle Data (LVD). This information has great potential for understanding how the vehicle is used, what is its current condition, and therefore what are the most likely faults that can occur in the near future.

However, the data is very noisy, often inaccurate and with a lot of duplication. The aggregated statistics in their raw form, at any given point in time, are not the correct input for diagnostics and predictive maintenance. Instead, they should be combined across the whole lifetime of the vehicle, to capture the main relevant usage patterns.

The goal of this project is to generate descriptive features that can be used for predicting vehicle faults. To this end students will evaluate a number of models, including regression models, random forests, and deep neural networks.

Preliminary steps for this project are as follows:
# create models for each LVD parameter based on a single data readout
# create models for each LVD parameter based on the complete vehicle history
# combine those models to find commonalities between different parameters
# extract descriptive features from the combined models

Automatic Generation of Descriptive Features for Predicting Vehicle Faults

2020-03-02T12:20:23Z

Mahmoud:

{{StudentProjectTemplate
|Summary=Automatic Generation of Descriptive Features for Predicting Vehicle Faults
|Programme=Embedded and Intelligent Systems
|TimeFrame=Fall 2019
|Prerequisites=Artificial Intelligence, Learning Systems, Data Mining
|Supervisor=Sepideh Pashami, Sławomir Nowaczyk, Mahmoud Rahat
|Level=Master
|Status=Ongoing
}}
Whenever a Volvo truck visits an authorised workshop, aggregated vehicle statistics (parameters) such as average vehicle speed, total fuel consumption, etc. are stored in a database called Logged Vehicle Data (LVD). This information has great potential for understanding how the vehicle is used, what is its current condition, and therefore what are the most likely faults that can occur in the near future.

However, the data is very noisy, often inaccurate and with a lot of duplication. The aggregated statistics in their raw form, at any given point in time, are not the correct input for diagnostics and predictive maintenance. Instead, they should be combined across the whole lifetime of the vehicle, to capture the main relevant usage patterns.

The goal of this project is to generate descriptive features that can be used for predicting vehicle faults. To this end students will evaluate a number of models, including regression models, random forests, and deep neural networks.

Preliminary steps for this project are as follows:
# create models for each LVD parameter based on a single data readout
# create models for each LVD parameter based on the complete vehicle history
# combine those models to find commonalities between different parameters
# extract descriptive features from the combined models

Deep feature analysis and extraction on Logged Vehicle data for the task of predictive maintenance

2019-10-02T04:58:07Z

Mahmoud:

{{StudentProjectTemplate
|Summary=This project is about applying supervised/unsupervised methods of feature selection on Logged Vehicle data (LVD) from Volvo trucks and investigate the contribution in model construction for different predictive maintenance tasks
|References=• Doquet, Guillaume, and Michele Sebag. "Agnostic feature selection." The European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases 2019
• Prytz, Rune, et al. "Predicting the need for vehicle compressor repairs using maintenance records and logged vehicle data." Engineering applications of artificial intelligence 41 (2015): 139-150.
|Prerequisites=Machine Learning
|Supervisor=Mahmoud Rahat
|Status=Draft
}}
This project is about applying supervised/unsupervised methods of feature selection on Logged Vehicle data (LVD) from Volvo trucks and investigate the contribution in model construction for different predictive maintenance tasks. The LVD dataset is collected by storing aggregated vehicle statistics (parameters) such as average vehicle speed, total fuel consumption, and so on. These parameters provide information about truck usage and its current condition. Though, the data is very noisy, often inaccurate and contains many redundant, and uninformative features, which can be identified and removed without incurring much loss of information.

Embedding DNN models on mobile robots for object detection

2019-10-01T15:23:53Z

Mahmoud:

{{StudentProjectTemplate
|Summary=The idea in this project is to employ transfer learning methods to teach a mobile robot to detect a handful of everyday objects in the real-world environment, and investigate the challenges and difficulties that are faced to this end
|Keywords=Mobile robots, Transfer Learning, Object Detection
|References=• Pan, Sinno Jialin, and Qiang Yang. "A survey on transfer learning." IEEE Transactions on knowledge and data engineering 22.10 (2009): 1345-1359.
• Yosinski, Jason, et al. "How transferable are features in deep neural networks?." Advances in neural information processing systems. 2014.
|Prerequisites=Machine Learning
|Supervisor=Mahmoud Rahat
|Status=Open
}}
The idea in this project is to employ transfer learning methods to teach a mobile robot to detect a handful of everyday objects in the real-world environment and investigate the challenges and difficulties that are faced to this end. One to say is the hardware/software limitation on mobile robots.

Deep feature analysis and extraction on Logged Vehicle data for the task of predictive maintenance

2019-10-01T11:19:52Z

Mahmoud: Created page with "{{StudentProjectTemplate |Summary=This project is about applying supervised/unsupervised methods of feature selection on Logged Vehicle data (LVD) from Volvo trucks and invest..."

{{StudentProjectTemplate
|Summary=This project is about applying supervised/unsupervised methods of feature selection on Logged Vehicle data (LVD) from Volvo trucks and investigate the contribution in model construction for different predictive maintenance tasks
|References=• Doquet, Guillaume, and Michele Sebag. "Agnostic feature selection." The European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases 2019
• Prytz, Rune, et al. "Predicting the need for vehicle compressor repairs using maintenance records and logged vehicle data." Engineering applications of artificial intelligence 41 (2015): 139-150.
|Prerequisites=Machine Learning
|Supervisor=Mahmoud Rahat
}}
This project is about applying supervised/unsupervised methods of feature selection on Logged Vehicle data (LVD) from Volvo trucks and investigate the contribution in model construction for different predictive maintenance tasks. The LVD dataset is collected by storing aggregated vehicle statistics (parameters) such as average vehicle speed, total fuel consumption, and so on. These parameters provide information about truck usage and its current condition. Though, the data is very noisy, often inaccurate and contains many redundant, and uninformative features, which can be identified and removed without incurring much loss of information.

Embedding DNN models on mobile robots for object detection

2019-10-01T10:53:27Z

Mahmoud: Created page with "{{StudentProjectTemplate |Summary=The idea in this project is to employ transfer learning methods to teach a mobile robot to detect a handful of everyday objects in the real-w..."

{{StudentProjectTemplate
|Summary=The idea in this project is to employ transfer learning methods to teach a mobile robot to detect a handful of everyday objects in the real-world environment, and investigate the challenges and difficulties that are faced to this end
|Keywords=Mobile robots, Transfer Learning, Object Detection
|References=• Pan, Sinno Jialin, and Qiang Yang. "A survey on transfer learning." IEEE Transactions on knowledge and data engineering 22.10 (2009): 1345-1359.
• Yosinski, Jason, et al. "How transferable are features in deep neural networks?." Advances in neural information processing systems. 2014.

|Prerequisites=Machine Learning
|Supervisor=Mahmoud Rahat
}}
The idea in this project is to employ transfer learning methods to teach a mobile robot to detect a handful of everyday objects in the real-world environment and investigate the challenges and difficulties that are faced to this end. One to say is the hardware/software limitation on mobile robots.