DC10 Damiano Gianotti- Zensor
At Zensor we strive to make civil infrastructure maximally sustainable, by making their life as long, as safe and as financially beneficial as possible. To achieve this we offer continuous, in-use monitoring of civil assets. This monitoring is based on continuously collecting data from various types of sensors distributed over the asset, combined with advanced analysis through various dedicated algorithms. As an outcome we can detect phenomena that may lead to degradataion in an early stage and detect when changing environmental or usage loads have a detrimental effect on the useful life of the asset. In the end we try to make sure that the asset has a maximal operational life.
Keywords: Data science, data analytics, monitoring, cloud, sustainability
Research field: Computer science
Click here to watch Damiano introduce himself as well as his PhD subject.
Civil structures represent the largest asset class when considered in monetary units. Maintaining these assets in optimal
condition for as long as possible can only be achieved through continuous and automated monitoring coupled with appropriate interventions (repairs, maintenance) at an early stage. Continuous monitoring depends on the reliability of monitoring systems: automatically tracking data integrity and quality therefore becomes essential. Automatic predictions, event detection and warnings are necessary to achieve a successful and economic package in the years to come. The objective of this project is to generate a generic framework for monitoring the quality of data flows generated by multiple types of sensors installed on civil assets before being used for (advanced) analysis tasks. Algorithms will be developed and then combined with code based on physics, including background knowledge on the sensor type and the corresponding working principle, together with more black box (neural-network based) model approaches.
The end result will combine various functionalities inside a single framework:
- Deploying on continuous data flows (pipeline)
- Monitoring and assessing the data from various sensors measuring the structural integrity of an asset
- Assessing sensor health based on its data flow and taking into account the data generated by (neighbouring) sensors
- Assessing whether the data are physically correct taking into account the specific conditions of the sensor installation (measuring principle, assembly, failure modes)
- Predicting damage or degradation of hardware based on historical sensor readouts and by comparing data from multiple sensors of different types on the same asset.
- CEA, Sylvain Magne, 2 months, M11-M12: FBG data acquisition and analysis.
- Inductosense, ChengHuan Zhong, 2 months, M20-M21: Ultrasonic methods.
- Airbus, Jaime GarcĂa-Alonso, 2 months, M29-M30: Managing large datasets
Contact
Contact
DC10 Damiani Gianotti
Contact
Industrial local supervisor Yves Van Ingelgem
Contact
Academic co-supevisor Sylvain Magne
Contact
PhD supervisor Herman Terryn