Determining reliability of modelling of drinking water warming

Reliable application of computational models

With a view to the microbial water quality, the drinking water utilities in the Netherlands are legally obligated to deliver drinking water at a temperature not exceeding 25 °C. As a result of climate change, urbanisation and intensification of subsurface infrastructures, the likelihood that the temperature will exceed this threshold is increasing.

Earlier research indicated the degree to which soil temperature is affected by above-ground factors, the subsurface and the drinking water pipe. The application of appropriate control measures to the entire distribution network (grass, shade, sufficiently distanced heating network, deep installation) is however not always feasible because of practical constraints and high costs. For this reason, a deeper insight is required.

Water utilities want to know which specific locations in the mains network, and during which periods, the risks arise, and which targeted measures they can take in order to limit these risks. To gain a  deeper insight is challenging, because of the numerous interacting variables. These include for instance  weather conditions, above-ground surroundings, soil, mains network and hydraulic conditions. The risks and control measures are not easily determined on the basis of expert judgment alone, which is why the use of a validated computational model is essential. KWR has earlier developed – in the TKI project Engine – among others – a two-staged model to calculate the influence of the surroundings on drinking water temperature in mains networks. Although there is confidence in the functioning of the Engine model, there are also points of concern that hinder application. For this reason, it is essential that this be evaluated and a strategic plan be developed for application.

Evaluation and application

The project consists of the following activities:

• Sensitivity study of Engine model
  A sensitivity study reveals which model parameters have the strongest influence on the outcomes and their reliability. Recommendations for the water utilities will be formulated on the basis of the outcomes, so that the utilities will know which input data need to be collected and how much scope there is for the development of rules of thumb for specific input parameters.
• Validation of coupling of sub-models in Engine model (BTM+ and WTM+) using a 3D model
  A 3D variant of the BTM+/WTM+ model system will be created using finite element methods. This system is no longer divided into separate sub-models, so that it is also no longer subject to simplification, which is needed to couple these sub-models. This makes the 3D model usable for assessing the impact and validity of the simplified coupling.
• Develop project plan for application
  A project plan for follow-up research will be developed in consultation with the drinking water utilities. The plan will focus on detecting risk locations for warming in the distribution network, and identifying control measures. Depending on the outcomes of the sensitivity study and the analysis of the model coupling, it may also be desirable to include a refining and/or thorough experimental validation of the model.

Well-founded recommendations

With the critical evaluation in this project, a precise quantitative foundation will be possible with regard to what may be expected from the staged Engine model. The evaluation will produce recommendations for modifications to the model, when necessary. We will also develop a deeper insight into the relationship between the warming of drinking water during distribution and relevant parameters, such as flow volume and pipe diameter. The development of a project plan will provide room for the formulation of concrete, broadly supported objectives and of an approach, and to anticipate practical and organisational aspects, such as the collection of data, the organisation of sensor measurements, and the use of software.