Search

This study investigates the use of a multivariate approach, based on Principal Component Analysis PCA), as software sensor for fault detection and reconstruction of missing measurements in on-line monitoring of sewer water quality. The analysis was carried out on a 16-months dataset of five commonly available on-line measurements (flow, turbidity, ammonia, conductivity and temperature). The results confirmed the great performance of PCA (up to 10 weeks after parameter estimation) when estimating a measurement from the combination of the remaining four variables, a useful feature in data validation. However, the study also showed a dramatic drop in predictive capability of the software sensor when used for reconstructing missing values, with performance quickly deteriorating after 1 week since parameter estimation. The software sensor provided better results when used to estimate pollutants mainly originated from wastewater sources (such as ammonia) than when used for pollutants affected by several processes (such as TSS). Overall, this study provides a first insight in the application of multivariate methods for software sensors, highlighting drawback and potential development areas. A combination of (i) advanced methods for on-line data validation, (ii) frequent parameter estimation, and (iii) automatic method for classification of dry/wet periods may provide the needed background for a successful application of these software sensors.

We present a prototype for a new software tool which enables quantification of impacts of water sensitive urban design (WSUD) plans in a simplifying manner. The tool is designed to fill a gap between the needs of utility companies for assessing WSUD performance and available urban drainage simulation tools. Emphasis is put on reducing complexity in order to help drainage engineers communicate their priorities to other stakeholders. The tool outputs include two key indicators: The first is the amount of runoff held back in stormwater control measures on event basis, presented graphically against three distinct decision domains. The second is the percentage of runoff managed locally on an annual basis, presented graphically as a water budget. The tool concept was tested on several smaller case studies in Denmark, and we plan to have it ready for full-scale testing ultimo 2015.

Flooding produced by high-intensive local rainfall and drainage system capacity exceedance can have severe impacts in cities. In order to prepare cities for these types of flood events – especially in the future climate – it is valuable to be able to simulate these events numericallyboth historically and in real-time. There is a rather untested potential in real-time prediction of urban floods. In this paper radar data observations with different spatial and temporal resolution, radar nowcasts of 0-2 hours leadtime, and numerical weather models with leadtimes up to 24 h are used as inputs to an integrated flood and drainage systems model with the purpose to investigate the potential for predicting future floods. The system is tested on a small town Lystrup in Denmark, which has been recently flooded. Results show that it is possible to generate detailed flood maps in real-time with high resolution radar rainfall data, but rather limited performance in predicting floods with leadtimes more than half an hour. Flooding produced by high-intensive local rainfall and drainage system capacity exceedance can have severe impacts in cities. In order to prepare cities for these types of flood events – especially in the future climate – it is valuable to be able to simulate these events numericallyboth historically and in real-time. There is a rather untested potential in real-time prediction of urban floods. In this paper radar data observations with different spatial and temporal resolution, radar nowcasts of 0-2 hours leadtime, and numerical weather models with leadtimes up to 24 h are used as inputs to an integrated flood and drainage systems model with the purpose to investigate the potential for predicting future floods. The system is tested on a small town Lystrup in Denmark, which has been recently flooded. Results show that it is possible to generate detailed flood maps in real-time with high resolution radar rainfall data, but rather limited performance in predicting floods with leadtimes more than half an hour.