Overview and details of the sessions of this conference. Please select a date or location to show only sessions at that day or location. Please select a single session for detailed view (with abstracts and downloads if available).
Plenary 10: Monitoring and Associated Technologies 2 - Data and Uncertainties
1:30pm - 1:50pm
Machine learning to improve understanding of pipe failures
University of Sheffield, United Kingdom
Historical data collected from a sewer network covering a town with a population of about 50k, with 180 km of predominantly combined sewers is analysed to understand potential causes of incidents, such as blockage and flooding, on the network. For this purpose, Machine Learning (ML) models are developed to identify the major relationships between different elements of the system, and then estimate the risk of incidents through quantifying the identified relationships. This risk relationship can be used to plan pro-active inspection of pipes to reduce the likelihood of failures occurring.
1:50pm - 2:10pm
Using centrality measures, network cross k-function and geographically weighted regression as decision support for operational issues and redesigning sewers
Lulea University of Technology, Sweden
The topology of Sanitary Sewer Networks (SSNs) can play an influential role in the occurrence and magnitude of operational failures such as blockages and basements flooding. It could be argued that the spatial behaviour of operational failures may be related to the topology of SSNs. This article explored this argument by investigating the spatial association between the location of recurrent blockages and influential nodes within the network topology using centrality measures and the network cross-K-function. Results from a preliminary application to the SSN of one municipality (total network length 500 km, »40 people/km) using its historical blockage data are presented.
2:10pm - 2:30pm
Image Based Quantification of Solids Transport and Transformation Processes in Sewer Pipes
Technion – Israel Institute of Technology, Israel
2:30pm - 2:50pm
How reusable are your data? - Towards truly FAIR open data for urban drainage
1Swiss Federal Institute of Aquatic Science and Technology (Eawag), Switzerland; 2Swiss Federal Institute of Technology Lausanne (EPFL); Switzerland; 3SINEF SA, Givisiez, Switzerland; 4The University of Sheffield, United Kingdom
Generating new insight from existing data is a major cornerstone of the scientific process. Re-using existing observations with a fresh idea, possibly including complementary datasets, may answer questions that the initial investigators did not even consider. Unfortunately, in the urban drainage area only very few Open Research Datasets (ORD) exist. Reusing observations in our field is challenging, among other things, because the required meta-data, e.g. on sensor calibration and maintenance, are lacking. In this contribution, we critically review some examples of urban drainage ORD. We find that bottlenecks are mostly concerned with the interoperability and the reusability of the data.