WatEner
The 5 keys to water and energy efficiency in supply networks

The 5 keys to water and energy efficiency in supply networks

 

Through Water Idea S.A., a subsidiary specialising in Smart Water Management, Grupo Inclam presents a new technological solution called WatEner. WatEner is an ICT solution that enhances the operation, maintenance and management of drinking water networks, within the framework of the water-energy nexus.

Grupo Inclam’s technical experience in the development of decision-making support systems and models for large water utilities has enabled the company to identify a necessary change of approach. An approach that integrates, know-how, tools and expertise for the purpose of achieving optimum global water supply management, whilst focusing on the importance of the interrelationship between water and energy consumption.

The traditional approach to solving efficiency issues in drinking water network has focused on leak detection and anomalies in the operation of the main system parameters (pressure, flows, consumption, etc.). Running in parallel to this approach, efforts have also focused on optimising energy consumption through the implementation of mathematical models of networks and their pumping systems. Ultimately, water supply utilities have sought to avail to the utmost of the data they generate, their IT tools and the models already in place. There is also a need to take advantage of the know-how of expert staff members with a view to enhancing day-to-day operations.

All these aspects have been taken into account in the development of the WatEner platform, with the aim of helping water supply utilities to achieve global, intelligent  enhancements in operations and maintenance management.

WatEner affords a number of key enhancements to water and energy management in supply networks:

Global vision of water-energy nexus to reduce costs and improve efficiency 

Improving the efficiency of complex supply systems whilst maintaining the quality of service to users can only be achieved with an integrated solution that takes account of the

interrelationships between water and energy, and manages these interrelationships holistically.

Smart approach that incorporates and enhances expert knowledge through machine learning and a training module

Expert staff knowledge is entered into a Knowledge Database by means of Artificial Intelligence tools, such as Machine Learning. The improved knowledge is achieved by means of the training module, which enables analysis of alternatives in diverse operating scenarios. System parameters are simulated and the results obtained are displayed.

Synergies in the integration of available data, information and IT tools

The platform enables the integration of data, information and existing models, allowing storage, display and graphical analysis. The use of existing resources optimises the power of the platform and facilitates synergies.

User experience with an intuitive, flexible and portable system

A user-friendly environment and easy access to information is achieved by means of an intuitive website, which can be accessed from any search engine, laptop computer or tablet. The system is flexible in order to  incorporate control panels, KPIs, and visual displays with permissions in accordance with user roles. Functionalities such as unSIG, a document manager, weather forecasting and consumption demand forecasting, real-time display of network information and simulations, etc. all serve to facilitate and enhance user experience.

Environmental sustainability aimed at reducing carbon footprint

The improved energy efficiency of production processes enables a reduction in CO2 emissions, thereby contributing to a reduction in greenhouse gases and helping in the fight against climate change.

Similarly, water efficiency contributes to responsible resource use and the implementation of good environmental practices. The functioning of WatEner is based on a Decision-making Aid nucleus that analyses requests for  information and provides operating recommendations and alerts. It learns from expert users by means of knowledge inference techniques such as Pattern Recognition Technique (PRT) or Business Rules Technique (BRT).

The aim of PRT is to analyse a situation in the network, obtain the main characteristics of this situation and search the knowledge base (pattern base) in order to detect the most similar patterns based on these characteristics. In other words, it searches for similar patterns in a knowledge database and proposes an operating scenario. This method is much more efficient than the mathematical optimisation algorithms commonly used (which do not always function correctly in highly complex networks). It enables maximum advantage to be taken of the expert knowledge contained in the decisions of the technical team that operates the network and “stores” its know-how.

The BRT is the second engine and it operates by imposing operational rules, policies, priorities and restrictions on the network. These must be carefully chosen and parameterised so that they can be simulated in the hydraulic model and enable enhancement of the recommendations and alerts provided by the platform. The BRT has a consistent mathematical basis, which is based on propositional or predicate logic. Deductive inference is the only accepted, valid way in mathematics and computation to carry out checks and draw conclusions and it has been successfully used in most critical decision-making aid tools (clinical diagnosis and diagnosis in the areas of communications, urban transport, aerospace, finance, etc.). The platform enables a hydraulic model to be run in just a few seconds to simulate network behaviour. It can simulate real-time network behaviour or the behaviour of the scenario proposed by the Decision-making Aid nucleus. In this way, operating and maintenance conclusions can be obtained for the user. The platform also has a drinking water consumption forecasting model and a weather forecasting model with different time horizons. Unlike other solutions, the WatEner design incorporates the experience and knowledge of the expert operator and takes all variables and patterns into account when making decisions. Ultimately, we are no longer facing a challenge associated with lack of data or information. The challenge now is to endow our systems with the intelligence necessary to help people make decisions.


About the Author