MABY5350 Sensor Networks and Model Based Decisions Support Emneplan

The course will give an understanding of the status and potential for the use of sensors and models for decision support and system diagnostics in urban water systems. Since each sensor only measures in a single location, we need models to interpret the sensor data and give us an idea about what is happening throughout the system. The combination of models and sensor data will also be used for forecasting purposes and to increase data security by estimating if sensor data is realistic or not. The student will learn about the types of sensor data available now and in the foreseeable future and how to utilize the data for decision support and diagnostic purposes. The student will learn how to do model-based inference to learn about the state of a system by combining models and sensor data. The students will be exposed to numerous real-life cases. The course will cover the following key aspects: Simple conceptual modelling, Bayesian inference, Ensemble-based system diagnostics, forecasting, Real time control, data in water distribution systems, opportunities and challenges for smart control of urban water systems.

The students will make use of software such as Matlab and Python.

No formal requirements over and above the admission requirements.

After completing the course, the student is expected to have achieved the following learning outcomes defined in terms of knowledge, skills and general competence:

Knowledge

The student will have

•  broad knowledge about the various sensor types used within the different fields of urban water;
• deep knowledge about the most common properties, deficiencies and uncertainties for the various sensor types;
• insight into the current state-of-the-art of sensor and model usage in current urban water management;
• good understanding of the different requirements for models for design, diagnostics and forecasting.

Skills

The student

• can design and implement simple hydrological/hydraulic models;
• can interpret data using a model;
• can use a model to critically assess sensor data and balance conflicting data against each other for systemwide diagnostics.

General competences

The student will know how to

• integrate knowledge into practice in the water industry to apply this to a range of scenarios, to critically analyse, evaluate, interpret and report on information from a range of sources and to solve complex problems systematically and creatively;
• critically assess sensor data.

The teaching will consist of a combination of:

Lectures & discussions

Independent studies including video recordings and online exercises

Coursework assignments

Guest lectures

Practical use of tools and software.

Two individual multiple-choice of 30 minutes each must be passed.

2 project written reports in groups of 2-3 students, maximum 2000 words each.

The student groups can be different on each project.

Each project counts 50% of the grade.

In the event of failed or valid absence of exam, the postponed exam will be given as either an oral or written examination.

All aids permitted.

A grade scale with grades from A to E for pass (with A being the highest grade and E being the lowest pass grade) and F for fail is used in connection with the final assessment. 

One internal examiner. External examiners are used regularly.